Bacteriophage: Genetics and Molecular Biology [1 ed.] 190445514X, 9781904455141

Table of contents :
Contents
List of Contributors
1 The New Phage Biology: from Genomics to Applications • Olivia McAuliffe, R. Paul Ross, and Gerald F. Fitzgerald
2 Bacteriophage Bioinformatics and Genomics • Carlos A. Canchaya, Marco Ventura, and Douwe van Sinderen
3 Bacteriophage in the Environment • Markus G. Weinbauer, Martin Agis, Osana Bonilla-Findji, Andrea Malits, and Christian Winter
4 Bacteriophages and Food Fermentations • Eric Emond and Sylvain Moineau
5 Bacteriophages in Medicine • Andrzej Górski, Jan Borysowski, Ryszard Międzybrodzki, and Beata-Weber-Dąbrowska
6 Phage Therapy: the Western Perspective • Harald Brüssow
7 Bacteriophage–Host Interaction in Lactic Acid Bacteria • Christina Skovgaard Vegge, John Gerard Kenny, Lone Brøndsted, Stephen Mc Grath, and Douwe van Sinderen
8 Transfer of DNA from Phage to Host • Lucienne Letellier, Laure Plançon, and Pascale Boulanger
9 Prophages and their Contribution to Host Cell Phenotype • W. Michael McShan and Joseph J. Ferretti
10 Prophage Induction of Phage λ • John W. Little
11 Phage φ29: Membrane-associated DNA Replication and Mechanism of Alternative Infection Strategy • Wilfried J.J. Meijer, Daniel Muñoz-Espín, Virginia Castilla-Llorente, and Margarita Salas
12 Release of Progeny Phages from Infected Cells • Carlos São-José, João G. Nascimento, Ricardo Parreira, and Mário A. Santos
Index
Colout Plate

Citation preview

Bacteriophage

Essential reading for all phage researchers and of interest to molecular biologists and microbiologists working on bacteria in academia, biotechnology and pharmaceutical companies, and in the food and other industries.

Genetics and Molecular Biology

Written by eminent international researchers actively involved in the disparate areas of bacteriophage research, this book focuses on the current rapid developments in this exciting field. The book opens with an excellent chapter that provides a broad overview of the topics and also highlights the multifaceted nature of bacteriophage research. This is followed by a series of reviews that focus on the current most cutting-edge topics, including bioinformatics and genomics, phage in the environment, bacteriophage in medicine, transfer of phage DNA to the host, contribution to host phenotype and much more.

Bacteriophage Genetics and Molecular Biology Edited by Stephen Mc Grath and Douwe van Sinderen

Mc Grath van Sinderen

I S B N 978-1-904455-14-1

www.caister.com 9

781904 455141

Caister Academic Press

Bacteriophage Genetics and Molecular Biology Edited by Stephen Mc Grath and Douwe van Sinderen

Caister Academic Press

Copyright © 2007 Caister Academic Press Norfolk, UK www.caister.com British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library ISBN: 978-1-904455-14-1 Description or mention of instrumentation, software, or other products in this book does not imply endorsement by the author or publisher. The author and publisher do not assume responsibility for the validity of any products or procedures mentioned or described in this book or for the consequences of their use. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without the prior permission of the publisher. No claim to original U.S. Government works. Printed and bound in Great Britain

Contents

1

List of Contributors

v

The New Phage Biology: from Genomics to Applications

1

Olivia McAuliffe, R. Paul Ross, and Gerald F. Fitzgerald

2

Bacteriophage Bioinformatics and Genomics

43

Carlos A. Canchaya, Marco Ventura, and Douwe van Sinderen

3

Bacteriophage in the Environment

61

Markus G. Weinbauer, Martin Agis, Osana Bonilla-Findji, Andrea Malits, and Christian Winter

4

Bacteriophages and Food Fermentations

93

Eric Emond and Sylvain Moineau

5

Bacteriophages in Medicine

125

Andrzej Górski, Jan Borysowski, Ryszard Mię dzybrodzki, and Beata-Weber-Dą browska

6

Phage Therapy: the Western Perspective

159

Harald Brüssow

7

Bacteriophage–Host Interaction in Lactic Acid Bacteria

193

Christina Skovgaard Vegge, John Gerard Kenny, Lone Brøndsted, Stephen Mc Grath, and Douwe van Sinderen

8

Transfer of DNA from Phage to Host

209

Lucienne Letellier, Laure Plançon, and Pascale Boulanger

9

Prophages and their Contribution to Host Cell Phenotype

229

W. Michael McShan and Joseph J. Ferretti

10

Prophage Induction of Phage L John W. Little

251

iv

|

Contents

11

Phage F29: Membrane-associated DNA Replication and Mechanism of Alternative Infection Strategy

273

Wilfried J.J. Meijer, Daniel Muñoz-Espín, Virginia Castilla-Llorente, and Margarita Salas

12

Release of Progeny Phages from Infected Cells

307

Carlos São-José, João G. Nascimento, Ricardo Parreira, and Mário A. Santos

Index Colour Plate

335

Contributors

Martin Agis Laboratoire d’Océanographie de Villefranche Microbial Ecology and Biogeochemistry Group CNRS-UPMC Villefranche-sur-Mer France Osana Bonilla-Findji Laboratoire d’Océanographie de Villefranche Microbial Ecology and Biogeochemistry Group CNRS-UPMC Villefranche-sur-Mer France

Lone Brøndsted Faculty of Life Sciences Department of Veterinary Pathobiology University of Copenhagen Frederiksberg Denmark [email protected] Harald Brüssow Nestlé Research Center Food and Health Microbiology Nestec Ltd. Lausanne Switzerland [email protected]

Jan Borysowski Department of Clinical Immunology Transplantation Institute The Medical University of Warsaw Warsaw Poland [email protected]

Carlos A. Canchaya Department of Microbiology Biosciences Institute National University of Ireland Cork Ireland [email protected]

Pascale Boulanger Institut de Biochimie et Biophysique Moléculaire et Cellulaire CNRS Université Paris Sud Orsay France [email protected]

Virginia Castilla Llorente Instituto de Biología Molecular “Eladio Viñuela” (CSIC) Centro de Biología Molecular “Severo Ochoa” (CSIC-UAM) Universidad Autónoma Canto Blanco Madrid Spain [email protected]

vi

|

Contributors

Éric Émond Assistant Director, Biotechnology Choisy Laboratoires Louiseville Québec CANADA [email protected]

Lucienne Letellier Institut de Biochimie et Biophysique Moléculaire et Cellulaire CNRS Université Paris Sud Orsay France [email protected]

Joseph J. Ferretti Department of Microbiology and Immunology University of Oklahoma Health Sciences Center Oklahoma City, OK USA Gerald F. Fitzgerald Department of Microbiology University College Cork Cork Ireland [email protected] Andrzej Górski Department of Clinical Immunology Transplantation Institute The Medical University of Warsaw Warsaw; Bacteriophage Laboratory Institute of Immunology and Experimental Therapy, Polish Academy of Sciences Wroclaw Poland [email protected] John Gerard Kenny School of Biological Sciences University of Liverpool Liverpool UK [email protected]

John W. Little Department of Biochemistry and Molecular Biophysics and Department of Molecular and Cell Biology University of Arizona Tucson, AZ USA [email protected] Olivia McAuliffe Teagasc Moorepark Food Research Centre Fermoy Co. Cork Ireland olivia.mcauliff[email protected] Stephen Mc Grath Microchem Laboratories Ltd Clogherane, Dungarvan Co. Waterford Ireland [email protected] W. Michael McShan College of Pharmacy Stonewall Oklahoma City, OK USA [email protected] Andrea Malits Laboratoire d’Océanographie de Villefranche Microbial Ecology and Biogeochmistry Group Villefranche-sur-Mer France

Contributors

Wilfried J.J. Meijer Instituto de Biología Molecular “Eladio Viñuela” (CSIC) Centro de Biología Molecular “Severo Ochoa” (CSIC-UAM) Universidad Autónoma, Canto Blanco Madrid Spain

Ricardo Parreira Unidade de Virologia Instituto de Higiene e Medicina Tropical Universidade Nova de Lisboa Lisboa Portugal

[email protected]

Laure Plançon Institut de Biochimie et Biophysique Moléculaire et Cellulaire CNRS Université Paris Sud Orsay France

Ryszard Międzybrodzki Phage Therapy Unit Institute of Immunology and Experimental Therapy, Polish Academy of Sciences Wroclaw Poland [email protected] Sylvain Moineau Département de biochimie et de microbiologie Faculté des sciences et de génie Université Laval Québec city Province of Québec Canada [email protected] Daniel Muñoz Espín Instituto de Biología Molecular “Eladio Viñuela” (CSIC) Centro de Biología Molecular “Severo Ochoa” (CSIC-UAM) Universidad Autónoma Canto Blanco Madrid Spain [email protected] João Nascimento Instituto de Ciência Aplicada e Tecnologia Faculdade de Ciências da Universidade de Lisboa Campo Grande Lisboa Portugal [email protected]

[email protected]

[email protected] R. Paul Ross Teagasc Moorepark Food Research Centre Fermoy Co. Cork Ireland Margarita Salas Instituto de Biología Molecular “Eladio Viñuela” (CSIC) Centro de Biología Molecular “Severo Ochoa” (CSIC-UAM) Universidad Autónoma Canto Blanco Madrid Spain [email protected] Mário A. Santos Instituto de Ciência Aplicada e Tecnologia Faculdade de Ciências da Universidade de Lisboa Campo Grande Lisboa Portugal [email protected]

|

vii

viii

|

Contributors

Carlos São José Instituto de Ciência Aplicada e Tecnologia Faculdade de Ciências da Universidade de Lisboa Campo Grande Lisboa Portugal

Beata Weber-Dąbrowska Phage Laboratory Institute of Immunology and Experimental Therapy Polish Academy of Sciences Wroclaw Poland

[email protected]

[email protected]

Douwe van Sinderen Department of Microbiology and Alimentary Pharmabiotic Centre Biosciences Institute National University of Ireland Cork Ireland

Marco Ventura Department of Genetics, Biology of Microorganisms, Anthropology, Evolution University of Parma Parma Italy [email protected]

[email protected] Christina Skovgaard Vegge Department of Veterinary Pathobiology The Royal Veterinary and Agricultural University Frederiksberg Denmark

Markus Weinbauer Laboratoire d’Océanographie de Villefranche Microbial Ecology and Biogeochemistry Group CNRS-UPMC Villefranche-sur-Mer France

[email protected]

[email protected] Christian Winter Laboratoire d’Océanographie de Villefranche Microbial Ecology and Biogeochemistry Group CNRS-UPMC Villefranche-sur-Mer France

The New Phage Biology: from Genomics to Applications

1

Olivia McAuliffe, R. Paul Ross, and Gerald F. Fitzgerald

Abstract Bacterial viruses, or bacteriophages, are estimated to be the most widely distributed and diverse entities in the biosphere. From initial research defining the nature of viruses, to deciphering the fundamental principles of life, to the development of the science of molecular biology, phages have been “model organisms” for probing the basic chemistry of life. With more recent advances in technology, most notably the ability to elucidate the genome sequences of phages and their bacterial hosts, there has been a resurgence of interest in phages as more information is generated regarding their biology, ecology and diverse nature. Phage research in more recent years has revealed not only their abundance and diversity of form, but also their dramatic impact on the ecology of our planet, their influence on the evolution of microbial populations, and their potential applications. This review focuses on this new post-genomic era of phage biology, from information emerging from genomics and metagenomics approaches through to applications in agriculture, human therapy and biotechnology. Introduction Bacterial viruses (or bacteriophages) are found in all habitats in the world where bacteria proliferate. They are estimated to be the most widely distributed biological entity in the biosphere, with an estimated viral population of greater than 1031 or approximately 10 million per cubic centimeter of any environmental niche where bacteria or archaea reside (Hendrix, 2003). Most of these viruses are bacteriophages. The dsDNA tailed phages, or Caudovirales, account for 95% of all the phages reported in the scientific literature, and possibly make up the majority of phages on the planet. However, there are other phages that occur abundantly in the biosphere, phages with different virions, genomes and lifestyles (Table 1.1). Over the past three decades, phage research has revealed the abundance of phages in nature, the diversity of their genomes, their impact on evolution of microbial diversity, their control of infectious diseases and their influence in regulating the microbial balance in every ecosystem where this has been explored (Kutter, 2005). These findings have led to a resurgence of interest in phage research. Since their discovery in 1915 and 1917 by Fredrick Twort and Felix d’Herelle respectively, bacteriophages have been studied in many laboratories and used in a variety of practical applications. Indeed, phage research

2

|

McAuliffe et al.

Table 1.1 ICTV classification of phages Order

Family

Morphology

Nucleic acid

Caudovirales

Myoviridaea

Non-enveloped, contractile tail

Linear dsDNA

Siphoviridaea

Non-enveloped, long non-contractile tail

Linear dsDNA

Podoviridaea

Non-enveloped, short noncontractile tail

Linear dsDNA

Tectiviridaeb

Non-enveloped, isometric

Linear dsDNA

Corticoviridaeb

Non-enveloped, isometric

Circular dsDNA

Lipothrixviridae

Enveloped, rod-shaped

Linear dsDNA

Plasmaviridae

Enveloped, pleomorphic

Circular dsDNA

Rudiviridae

Non-enveloped, rod-shaped

Linear dsDNA

Fuselloviridae

Non-enveloped, lemon-shaped

Circular dsDNA

Inoviridae

Non-enveloped, filamentous

Circular ssDNA

Microviridae

Non-enveloped, isometric

Circular ssDNA

Leviviridae

Non-enveloped, isometric

Linear ssRNA

Cystoviridaeb

Enveloped, spherical

Segmented dsRNA

a

Tailed phages.

b

Lipid containing.

has played a central role in some of the most significant discoveries in biological sciences, from the identification of DNA as the genetic material, to the deciphering of the genetic code, to the development of the science of molecular biology over 50 years ago. Research on phages has continually broken new ground in our understanding of the basic molecular mechanisms of gene action and biological structure. In more recent times, phage genomics is revealing novel biochemical mechanisms for replication, maintenance and expression of the genetic material and is providing new insights into origins of infectious disease and the potential use of phage gene products and even whole phage as therapeutic agents. As mentioned above, the ubiquity and prevalence of bacteriophages in nature and the diversity of their genomes are just two of the reasons for the renewed interest and excitement in phage research. Studies to date have revealed that phages are incredibly varied in their properties, from host range, genetic content, regulatory mechanisms, and physiological effects (Brussow and Kutter, 2005). Indeed, most of the genes identified on genome sequencing of cultured phages and metagenomic analysis of uncultured phage communities are unidentifiable, i.e. they show no similarity to anything in the currently available databases. At one level, there is diversity in the types of phages that infect individual or interrelated bacterial species. At another level, there is diversity among genomically related phages that do not share the same bacterial hosts. A classic example is the well-studied group of dsDNA T4 phages. This group of phages contains relatives that infect species such as Aeromonas,

Phage Genomics and Applications

Vibrio, Acinetobacter, marine and other bacteria. The genomes of a few T4-like phages have been sequenced and found to share homologies with T4 itself, but to also differ from one another in size, organization of the T4-like genes and content of other putative genes (Miller et al., 2003b). The evidence suggests that phage families like the T4-related phages have learned to cross bacterial species barriers through possession of dynamic genomes that acquire and lose genetic cassettes. Since phages share a long evolutionary history with their hosts (Hambly and Suttle, 2005), it is tempting to speculate that the genomes of the dsDNA phages are repositories of the genetic diversity of all microorganisms in nature. In addition to their role as carriers of genetic information, bacteriophages have a strong influence on microbial populations: these populations fluctuate with nutritional inputs and there is a dynamic relationship between phage population sizes and host numbers and physiology (Day, 2004). In addition, phages have an impact on the performance of microbial food webs and biogeochemical cycles (Weinbauer, 2004), affecting nutrient cycling, system respiration, biodiversity and species distribution, and genetic transfer (Fuhrman, 1999). In the oceans, phages exert significant control on marine bacterial and phytoplankton communities, with respect to both biological production and species composition, influencing the pathways of matter and energy transfer in these systems (Fuhrman, 1999). Studies on the evolution of phages and their role in natural ecosystems are on the increase. With the data emerging from these studies, researchers can ask more practical questions, such as how to use phages to combat infectious diseases that are caused by bacteria, how to eradicate phage pests in the food and agricultural sectors and what role they have in the causation of human diseases (Campbell, 2003). Interest in phage and phage gene products as potential therapeutic agents is increasing rapidly and is likely to have profound impact on the pharmaceutical industry and biotechnology in general over the coming years. There is a general sense that the best is yet to come out of phage research; in this review, we will focus on the current rapid developments in this field. A historical perspective Since the discovery of phages by Frederick Twort in 1915 (Twort, 1915) and independently in 1917 by Felix d’Herelle (d’Herelle, 1917), phages have been key players in scientific research. D’Herelle pioneered two important areas of phage research. He observed at an early stage that phages had the potential to kill bacteria that cause diseases in humans, as well as in agriculturally important plants and animals, and advocated phages as therapeutic agents in the pre-antibiotic era (for review, see Sulakvelidze et al., 2001). In 1933, he co-founded an institute for phage research in the Soviet Republic of Georgia, together with George Eliava. This establishment continued to supply phage for therapeutic uses to the entire Soviet Union until its recent breakup. In the West, research on such “phage therapy” was abandoned when penicillin and other chemical antibiotics were discovered in the 1940s, though there has been some renewed interest in phage therapy in recent years as antibiotic resistance of pathogenic bacteria has become a more prominent threat to public health. D’Herelle’s second research program concentrated on the biological nature of the bacteriophage itself, proving the concept that phages are obligate intracellular parasites. Concurrent with d’Herelle’s work, research on the nature of phages continued. The viral nature of the bacteriophage was clearly established, the chemical composition of the

|

3

4

|

McAuliffe et al.

virions was confirmed as protein and DNA, and progress was made in understanding the phage life cycle. One of the first scientific applications of the electron microscope was the visualization of bacteriophages and their interaction with bacterial cells (Anderson, 1966). This research demonstrated that specific viruses have characteristic morphologies and led to further studies on the morphogenesis and supramolecular assembly of phages. However, the “modern” era of bacteriophages in biology is said to have begun with the work of Max Delbrück in the late 1930s when he established the lytic mechanism by which some bacteriophages replicate and studied the genetic changes that occur when phages infect bacteria (Cairns et al., 1966). Until 1952, scientists did not know which part of the virus, the protein or the DNA, carried the information regarding viral replication. It was then that Al Hershey, working with Martha Chase at the Carnegie Laboratory of Genetics, performed a series of experiments using bacteriophages, proving that DNA is the molecule that transmits genetic information (Hershey and Chase, 1952). For these discoveries concerning the structure and replication of viruses, Delbrück, Hershey and Salvadore Luria shared the Nobel Prize for Physiology or Medicine in 1969. From this point forward, it was possible to ask and answer complex biological questions using bacteriophage as a model, such as what is the nature of a gene, how do mutations affect genes, how do mutations arise, how do genes replicate, and how are genes expressed. Around the early 1970s, the world of biological research began to be transformed by the “recombinant DNA revolution.” The suite of laboratory techniques that made this revolution possible was developed largely through research on phages. The science of molecular biology has produced some profound changes in bacteriophage research, as in all other areas of biological research. For one thing, the number of researchers working primarily on phages decreased dramatically as it became possible to study the genes of more complex, particularly eukaryotic, organisms with nearly the same ease as had been possible previously with simpler organisms, such as phages and bacteria. At the same time, the number of biological researchers using some form of phage in their research has increased substantially, since many of the tools of modern molecular biological research are phages or phage-derived. Thus, for those scientific problems where phages provide advantageous experimental systems, bacteriophage research is still vigorous and in many cases leading the field. Phage genomics and evolution The genomic era began in 1977 with the elucidation of the first phage genome sequence, that of FX174 (Sanger et al., 1977). Since then, over 200 completed phage genomes have become available in public databases and that number is being added to daily. With the increasing amount of information, the contribution of phages to evolution, ecology, virulence and medicine becomes more and more apparent. The majority of cultivated phages sequenced to date have been temperate phages, such as lambdoid ( Juhala et al., 2000), mycobacterial (Pedulla et al., 2003), dairy siphoviridae (Brussow and Desiere, 2001) and prophages found in bacterial genomes (Casjens, 2003; Desiere et al., 2002). Although the presently available genome sequence data reflects a minute proportion of the phage biosphere, analysis of this data gives a perspective of the extensive genetic diversity in this population. This amazing diversity is even more evident on analysis of data obtained from metagenomic analysis of

Phage Genomics and Applications

viral populations (Edwards and Rohwer, 2005) in marine, soil and gut communities. Such analysis has shown that most of the genes identified are novel, up to 68% in some studies (Cann et al., 2005). These observations suggest that uncultured phage communities are the largest untapped source of genomic information in the biosphere. Phage genomics has the potential to be as important for fundamental genomics since phage genomics represents all the problems that are also currently discussed in bacterial genomics: unity or diversity of origin, vertical versus horizontal transmission of genetic information, non-orthologous gene displacement, tree versus web-like phylogeny, synteny versus instability of gene order, gene-splitting versus domain accretion (Desiere et al., 2002; Koonin et al., 2000) Revelations from comparative genomic studies The dsDNA-tailed bacteriophages are probably the largest evolving group in the biosphere and are possibly a very ancient group. Comparative genomic studies of phage groups with many members sequenced have revealed insights into the mechanisms of phage evolution and have provided compelling evidence for the role of horizontal genetic exchange among genomes, mediated by both rampant non-homologous recombination between different genomes and by reassortment of the variant sequences so created through homologous recombination (Hendrix, 2003). It is apparent that all the dsDNA tailed phages, or Caudovirales, sequenced to date are highly modular in nature, with access to an enormous common genetic pool (Hendrix et al., 1999). High frequency horizontal gene transfer is primarily responsible for this genetic mosaicism (Canchaya et al., 2003). The modules are usually individual genes, but can be large blocks of genes, up to half the genome. This is especially true of the temperate phages where possibilities exist for exchange between phages and prophages, infecting phages and the bacterial genomes in which they reside (Brussow and Kutter, 2005). There is much debate about the relative contribution of vertical and horizontal transmission in phage evolution. Despite rampant horizontal exchange between temperate phage genomes, vertical lines of phage evolution are thought to play key roles in many phage families, particularly in relation to virulent phages (Brussow and Kutter, 2005). Virulent phages may recombine with other phages during simultaneous infection or with homologous sequences in the host genome or on resident plasmids. There is some evidence that suggests that virulent phages form biologically coherent groups with limited exchange of sequences outside their group. The T4-like phages, for example, form quite a distinct group with little obvious genetic exchange; however, there is evidence of lateral exchange between non-essential regions of these genomes (Chibani-Chennoufi et al., 2004a; Miller et al., 2003a; Miller et al., 2003b). In contrast, comparisons of the temperate lambdoid coliphages have shown that they are examples of the modular theory of phage evolution, i.e. they are mosaic with respect to each other. In some lambdoid phages, short conserved sequences have been observed at the boundaries of functional units (Clark et al., 2001), suggesting homologous or site-specific recombination as the driving force for horizontal gene transfer between these phages. However among others, it is evident that non-homologous recombination is rampant followed by selection of functional phages, as in the case of HK97 and HK022 ( Juhala et al., 2000). Comparable mechanisms of genetic exchange have been demonstrated in the phages of the lactic acid bacteria (Brussow and Desiere, 2001) and the mycobacteriophages (Ford et al., 1998).

|

5

6

|

McAuliffe et al.

Bacteriophages from the dairy industry tend to be more closely interrelated than any other group of phages because of their very specific ecological niche. Undoubtedly, strong elements of vertical evolution exist particularly in the structural gene clusters; the many sequenced genomes of these phages are remarkably similar in size, organization, and sequence (Brussow and Desiere, 2001; Brussow and Hendrix, 2002). This may be a consequence of the narrow range of habitats from which most of these phages were derived (commercial dairy fermentors), or alternatively, it may indicate a fundamental difference in their evolutionary mechanisms. The dairy phages show quantitatively less mosaicism than the other sequenced phage groups (Brussow and Desiere, 2001), but whether this reflects a real difference in the biology of these phages or simply a narrower sampling of the population than in the other groups remains to be determined. There is, however, clear evidence of horizontal exchange. The study by Brussow and Desiere, mentioned previously, compared Streptococcus thermophilus phages and demonstrated several horizontal gene transfer events (Brussow and Desiere, 2001). An example was the Sfi21-type DNA replication module which is very highly conserved, probably reflecting its recent acquisition and rapid lateral spread in the S. thermophilus phage population. Similar data was generated for replication genes in lactococcal phages. Cases of horizontal gene transfer were also observed between phage and plasmid DNA in S. thermophilus: two distinct non-coding regions of phage Sfi19 (cos-site, origin of replication) shared > 80% DNA identity with a cryptic S. thermophilus plasmid (Desiere et al., 1997; Lucchini et al., 1999). However, there appears to be no evidence for lateral gene transfer between S. thermophilus phages and the host genome. In contrast, lactococcal phages were shown to acquire chromosomal genes but interestingly, to date, all laterally acquired genes were of prophage origin. Comparative analysis of sequence data of Siphoviridae infecting five distinct Lactobacillus species (Lb. casei, delbrueckii, gasseri, johnsoni and plantarum) revealed only an extremely limited sharing of DNA sequences (Desiere et al., 2000). When all Siphoviridae were taken into consideration, close DNA sequence similarity was limited to phages that belong to an interbreeding phage population. The occurrence of lateral gene transfer among mycobacteriophage populations is also very frequent, even more so than that apparently occurring between mycobacteriophage and their hosts or other bacteriophage. This group looks very much like the lambdoid phages in terms of their mosaicism, evidence of large-scale non-homologous recombination and strong conservation of genetic organization in regions like the head genes (Papke and Doolittle, 2003). Comparative analysis of the sequenced mycobacteriophages suggests that horizontal gene transfer through illegitimate recombination is the major driving force for evolution of this group (Pedulla et al., 2003). Hendrix proposed that illegitimate recombination (by whatever means) occurs arbitrarily throughout the phage genome, and that the non-random appearance of recombination events is the result of natural selection that eliminates progeny that have DNA inserted in the middle of useful relevant genes (Hendrix et al., 2000). At least three recombination events could be identified on analysis of 14 mycobacteriophage genomes that appear to have arisen by illegitimate recombination. They appeared to have occurred sufficiently recently that the recombinant boundaries were not obscured by further evolutionary events (Pedulla et al., 2003). Two of these recombination events were found within a highly mosaic segment of Che8 and the third event—involving

Phage Genomics and Applications

mycobacteriophages Che8 and Corndog—involves quite different parts of the genomes but the common segments are 100% identical (Pedulla et al., 2003). These observations of genetic exchange through illegitimate recombination, giving rise to new module junctions and generate coding sequences, indicate that this is a very creative process and a dominant force in genome evolution. In addition to the mosaic genomic structure, Pedulla et al. found that up to 13% of the total open reading frames defined are homologous to open reading frames from non-mycobacteriophage species (Pedulla et al., 2003). Problems in phage taxonomy The recent revelations of phage genome sequencing regarding the mosaic nature of phage genomes and the obvious extent of horizontal gene transfer has made taxonomy extremely complex and the existing methods outdated. Spirited debate continues on what comprises a phage “species” and what overall classification scheme is best. Several formal classification schemes are currently in place, such as the ICTV system (van Regenmortel, 2000) which emphasizes the nature of the nucleic acids, nucleotide sequence and particle structure, and which is based on vertical transmission of genetic characteristics. The ICTV phage classifications are shown in Table 1.1. Nonetheless, it is generally agreed that there is a need for the development of a new system of phage taxonomy, one that reflects genomic data as a primary component. The difficulty involved in this, however, is the fact there is no suitable candidate gene that could be useful as a phylogenetic indicator for all tailed phages (Casjens, 2005), analogous to ribosomal DNA in bacteria. Despite this, a number of sequence-based systems of viral classification have been proposed. An approach put forward by the Pittsburgh Bacteriophage Institute involves divisions according to genome type (dsSNA, ssDNA, dsRNA, ssRNA) and a further division separating phages on the basis of morphological characteristics (Lawrence et al., 2002). The classification is then guided by three basic principles: members of a group should exhibit similarity that has resulted from one or more clearly defined cohesion mechanisms, including independent origins, frequency of genetic exchange, ecological isolation and periodic selection; for any given phage, a significant amount of genome sequence data should be available for meaningful taxonomic assignment; and groups may be reticulate, i.e. all phage within a particular group will bear the prominent features of that group, but any one phage could belong to more than one group at the same taxonomic level (Lawrence et al., 2002). Another classification system proposed is based on comparative genomics of a single, but prominent, structural gene module (Proux et al., 2002). The authors of this study suggest that there are some arguments in favor of basing a phage taxonomy system on the comparative genomics analysis of the DNA packaging-head gene cluster, since the evolutionary analysis of the morphogenesis module is the most advanced of the different phage modules, and the structural gene order is apparently very conserved (Brussow and Desiere, 2001) Alternatively, a phylogeny-based system could be adopted using phage tail genes (excluding tail fiber genes). This would ensure the greatest overlap with the current phage taxonomy, which differentiates the majority of the isolated phages according to phage tail morphology. Rohwer and Edwards (2002) have proposed a third system, the Phage Proteomic Tree, as the basis of a genome-based taxonomical system for phage. This system is based on an

|

7

8

|

McAuliffe et al.

algorithm that uses every phage genome to determine an average distance between pairs of phage, grouping phage both relative to their near neighbors and in the context of all other phages. Unlike other systems, this classification does not require direct visualization of the free virion, information about host range, or lifestyle information about the phage. It was shown that the Phage Proteomic Tree resolved a number of classification anomalies associated with the ICTV system, in particular the ICTV classification of Salmonella FP22 as a podovirus due to its morphology (i.e. it has a short tail), despite a complete lack of DNA sequence relatedness to other members of this family. With a system based on protein distance such as this, P22 groups with the other lambdoid phage which more accurately reflects its biology. However, phage groups such as the Siphoviridae present major problems for the proposed system, probably because of rampant lateral gene transfer. In the Phage Proteomic Tree, the ICTV Siphoviridae are spread into at least five separate groups. An example is the recent comparison by Trotter et al. of the lactococcal phage 4268 with other dairy phages (Trotter et al., 2006). Comparative analysis at the DNA sequence level suggested a close relationship between 4268 and BK5-T phage and the bIL286 prophage from L. lactis IL1403. Phages BK5-T and bIL286 had been previously grouped within the proposed Sfi21-like phages, distinguished by their structural gene cluster (Proux et al., 2002). Inclusion of phage 4268 in Rohwer and Edwards proteomic tree resulted in the grouping of 4268 with the BK5-T like phages in agreement with the analysis of Proux et al.; however, phage bIL286 did not group with 4268 and BK5-T on this occasion but appeared to be more closely related to the P335/Sfi11 group of phage than to the Sfi21-like phages (Trotter et al., 2006). The analysis also separated the lactococcal phages into several distinct groups, with the BK5-T and P335 phage clusters forming distinctly separate groups. Also, some phages of the c2 type were grouped with phages of the 936 type. Analysis by other authors suggest the former, BK5-T and P335, should be grouped together (Labrie and Moineau, 2002), while morphologically, c2 and 936 phages are very distinct from each other. The study by Trotter et al. highlights the difficulties in deriving a single classification system that can accommodate all the information now available. In the majority of cases, the different systems outlined here categorize phages within the same groups in which they are currently classified under the ICTV rules. The real taxonomic value of one method over another will very much depend on the diversity of the bacteriophage genomes encountered in the future. Phage metagenomics A technology recently employed to studying uncultured viral communities at the genome level is metagenomics. Metagenomic analysis involves the study of environmental population genomes by obtaining gene sequences directly from environmental samples of DNA with no cultivation or characterization of individual species (Figure 1.1) (Streit and Schmitz, 2004). Such an approach provides a greater understanding of a community’s diversity and function. Recently, metagenomic analysis of uncultured phage communities has provided insights into the composition and structure of the viruses within them through shotgun cloning of the community. To date, there are five published viral metagenomic libraries containing dsDNA viruses; two from near-shore marine water samples (Breitbart et al., 2002), a marine sediment sample (Breitbart et al., 2004), a human fecal sample (Breitbart et al.,

Phage Genomics and Applications

Environmental Samples Construction of Metagenomic Library DNA Extraction

Cloning of DNA fragments

Transformation into Host Bacterium

Metagenomic Library

Sequencing of Random Clones

Screening for Particular Phenotypes

Screening for Individual Sequences by PCR or Hybridization

Figure 1.1 Schematic representation of metagenomic analysis of uncultured virus communities. Fig. 1 Adapted from Riesenfeld et al. (2004).

2003) and an equine fecal sample (Cann et al., 2005). Examination of the data from these studies by Monte-Carlo analysis, which reconstructs the structure of uncultured communities, revealed that marine sediment viral communities are the most diverse biological systems characterized to date containing between 10 000 and 1 million viral genotypes, the most dominant virus making up less than 0.1% of the total community (Breitbart et al., 2004). The fecal sample contained approximately 1000 genotypes while the seawater samples most probably contained 5000 viral genotypes, each containing a different dominant genotype that represented approximately 1% of the total community (Breitbart et al., 2003; Breitbart et al., 2002). What was also immediately evident from the data emerging from these studies was the high proportion of unidentifiable genes; approximately 65–75% of all the sequences obtained showed no similarity to anything in the databases. Analysis carried out two years later on the same data revealed that most of the sequences were still unique, despite the fact that the GenBank database had more than doubled in size (Edwards and Rohwer, 2005). Since only about 10% of the sequences in microbial metagenomes and cultured microbial genomes are novel when analyzed in a similar fashion, it has been suggested that the unique genes in microbial genomes have been acquired from the phage genomic pool (Daubin and Ochman, 2004). This, of course, poses a problem for existing bioinformatic techniques; it is unclear whether this is a result of limitations of the search algorithms or limitations of the diversity represented in GenBank or both. It is envisaged that this problem will be alleviated somewhat with continued sampling and sequencing and the development of new computational methods. However, it is also clearly obvious from these studies that the global phage metagenome remains largely uncharacterized (Edwards and Rohwer, 2005) and that further studies are required to increase the coverage of this sequence space.

|

9

10

|

McAuliffe et al.

Evolution of bacterial hosts: the role of phages The evolution of bacteria is associated with the continuous generation of novel genetic variants. Extensive bacterial whole genome sequencing has led us away from the idea of bacterial genomes as static structures towards viewing them as relatively variable and plastic structures that are constantly evolving (Casjens, 1998). The major forces driving this process are point mutations, genetic rearrangements, and horizontal gene transfer. Horizontal gene transfer produces extremely dynamic genomes in which substantial amounts of DNA are introduced to and deleted from the chromosome. The sequencing of multiple strains from the same bacterial species has clearly demonstrated that horizontal gene transfer accounts for the majority of intraspecies genome differences (Hambly and Suttle, 2005). Comparative analysis of these bacterial sequences has revealed the movement of phage DNA by lateral transfer as having significant impact on the evolution of the host genomes. The genomes of bacteria are littered with both functional and defective phage genomes (Casjens, 2003); in fact, 3–10% of the genome of many bacterial strains sequenced to date are prophages (Desiere et al., 2002). Therefore, lysogeny (i.e. the integration of the phage DNA into the bacterial chromosome as a prophage) is an extremely common phenomenon (Canchaya et al., 2003). Two-thirds of the sequenced low GC Gram-positive and Gram-negative bacteria contain identifiable prophages (Canchaya et al., 2003) and many of these organisms are polylysogenic (contain multiple prophages). Prophages are the major contributors to genome diversity in some species, whereas in others, integrative plasmids, transposons and pathogenicity islands play a role. The presence of prophage DNA in a genome increases the metabolic burden on the host cell and there is the possibility that the lysogen will be destroyed by the phage should induction occur. Thus, it seems fitting that these temperate phages confer some advantage to the cell (Brussow and Hendrix, 2002). Some of the genes expressed from the prophage in a lysogen can alter the host phenotype; these are referred to as “lysogenic conversion genes” (Brussow et al., 2004). These may include genes that can increase the fitness of the cell and range from genes which encode products that protect the host from further phage infection to increasing the virulence of the host. Bacteria containing six or more prophages in their genomes are primarily pathogens. An extreme example is the case of E. coli 0157; between 15 and 18 prophages are present in these serotypes which are not found in the non-pathogenic K12 strain (Perna et al., 2001). Therefore, lysogenic conversion can, in effect, change the ecological and/or pathogenic character of a bacterial species. Contribution of phage to virulence Bacterial pathogens differ from related non-pathogenic species in a variety of ways, most obviously in their gene content (Groisman and Casadesus, 2005). Chromosomal gene clusters encoding toxin and other virulence genes—or pathogenicity islands—are found in pathogenic bacteria but are absent from related non-pathogenic species. Research over the past decades has revealed that phages play a major role in the introduction of gene clusters that contribute to pathogen virulence in a process known as phage conversion. Bacteriophage-encoded virulence genes can convert their bacterial host from a non-pathogenic strain to a virulent strain, or to a strain with increased virulence, by providing mechanisms for the invasion of host tissues and the avoidance of immune defenses (Boyd, 2005). It was

Phage Genomics and Applications

discovered in the early 1950s that in some important bacterial diseases, such as diphtheria (Corynebacterium diphtheriae), botulism (Clostridium botulinum), and scarlet fever (Streptococcus pyogenes), pathogenicity depends on the presence of certain prophages. In more recent times, whole genome sequencing of pathogenic bacteria has revealed the prevalence of prophages encoding proven or suspected virulence factors that are major contributors to the genetic individuality of these strains. These factors include ADP-ribosyl transferase toxins, superantigens, lipopolysaccharide-modifying enzymes, type III effector proteins, detoxifying enzymes, hydrolytic enzymes and proteins conferring serum resistance (Brussow et al., 2004). Prophage genes encoding factors such as these are often referred to as “morons” (Hendrix, 2003). Moron-encoded functions confer a selective benefit on the host, such as an increase in virulence, but do not appear to confer a benefit to the phage. The most widely recognized bacterial pathogen characteristic linked to bacteriophage infection is the production of exotoxins, and toxin genes are an example of one class of morons. Exotoxin production is the major pathogenic mechanism of several bacterial pathogens, including Vibrio cholerae (Waldor and Mekalanos, 1996), Clostridium botulinum (Barksdale and Arden, 1974), Corynebacterium diphtheriae (Freeman, 1951), E. coli (O’Brien et al., 1984), S. aureus (Betley and Mekalanos, 1985) and S. pyogenes (Weeks and Ferretti, 1984). In the case of V. cholerae, multiple phages have contributed to its pathogenicity. The pathogenic serogroups of V. cholerae (O1 and O139) produce cholera toxin (or CT), encoded by the filamentous bacteriophage CTXF. This phage is maintained as a prophage within the V. cholerae genome and does not excise to form extracellular particles, unlike other coliphages (Waldor and Mekalanos, 1996). Instead, CTXF is generated by a replicative process that requires tandem elements within the chromosome, either CTXCTX prophages or CTX-RS1 prophages (Davis et al., 2000). CT is the principal virulence factor of V. cholerae and is a classical AB toxin that induces rapid chloride and water efflux into the intestine resulting in secretory diarrhea, the primary symptom of epidemic cholera. Another key virulence factor of V. cholerae is the toxin-coregulated pilus (TCP), which is critical for intestinal colonization. The genetic element encoding TCP has been described as the genome of another filamentous phage, VPIF or TCPF, but the phage nature has been disputed recently. Interestingly, while both of these virulence genes are phage-encoded, their expression is controlled by three master regulators that are host-encoded, i.e. ToxR (Bina et al., 2003), AphA and AphB (Kovacikova and Skorupski, 1999). In contrast, the Shiga toxin-producing phages of Escherichia coli (EHEC; enterohemorrhagic E. coli) are an example of a case where the bacteriophage life cycle can exert control over virulence factor production. EHEC causes diarrheal disease as well as more severe clinical manifestations such as hemorrhagic colitis and hemolytic-uremic syndrome (Karch et al., 2005). The Shiga toxin Stx, released by bacteria in the intestine, is thought to be responsible for these symptoms. Two bacteriophages encode the Shiga toxins, Stx1 and Stx2 respectively. The toxin genes achieve their pathogenically effective expression only through transcription from a phage promoter during lytic growth following prophage induction (Wagner et al., 2002; Wagner et al., 2001). Indeed, phage-inducing agents such as mitomycin C have been shown to increase Stx production by EHEC. In addition to toxin production, phage-encoded products can also influence bacterial adhesion, colonization and invasion. Phage-mediated adhesion is observed in Streptococcus

|

11

12

|

McAuliffe et al.

mitis, a causative agent of infective endocarditis (Bensing et al., 2001). Genes for the surface proteins responsible for the ability of this organism to adhere to platelets and fibrin are encoded on an inducible prophage, SM1. In Salmonella enterica serovar Typhimurium, a leading cause of human gastroenteritis, genes encoded on the temperate phage, SopEF, facilitate critical steps in the infection process, namely uptake by the intestinal epithelial cells (sopE) and survival in the Peyer’s patches (Mirold et al., 1999). Other phage-encoded products result in enhanced bacterial resistance to serum and phagocytes. Staphylococcus aureus produce a number of proteins involved in phagocyte evasion, including the recently discovered chemotaxis inhibitory protein (CHIPS). The chp gene and other genes encoding immune evasion molecules (sak and sea) are located on B-hemolysin-converting bacteriophages (van Wamel et al., 2006). Phages can also alter bacterial susceptibility to antibiotics (Ubukata et al., 1975) and enhance pathogen transmission (Vibrio cholerae). Prophage in lactic acid bacteria The contribution of prophage sequences to bacterial virulence is very evident; however, the role of prophages in other organisms, e.g. non-pathogenic species, is less obvious. It is suggested that these sequences may contribute to the fitness of the lysogenic clone in its specific ecological niche (Ventura et al., 2003), not just through the presence or absence of certain genes, but also through altered expression of host genes (Desiere et al., 2002). Interestingly, comparative genomics has demonstrated that lysogenic conversion genes may also be encoded by temperate phages infecting dairy cultures. Open reading frames of no known phage-related function are often encoded on temperate phages and prophages from dairy organisms at positions in the genome where pathogenic streptococci and staphylococci encode lysogenic conversion genes (Desiere et al., 2002), suggesting a physiological function for these genes in the bacterial host. However, it is difficult at this time to speculate as to what the phenotypic consequences of phage genes in dairy bacteria might be. As mentioned previously, the majority of bacterial species containing high numbers of prophage are pathogenic organisms. An exception is Lactococcus lactis, an organism under extreme pressure from phages in an industrial fermentation setting. The 2.3 Mb genome of the sequenced L. lactis IL1403 contains six prophage elements, all but one of which can be excised from the genome (Chopin et al., 2001). Three of the six prophages, bIL286 and bIL309, showed a genome organization typical of the temperate dairy phages (Chopin et al., 2001). The gene content and organization of chromosomes of Lactococcus lactis and other dairy bacteria probably results from a strong evolutionary pressure toward optimal growth of these microorganisms in milk. Prophage sequences are also abundant in other lactic acid bacteria genomes, including some pathogenic species (Ferretti et al., 2001) and commensals of the alimentary tract (Ventura et al., 2003). More data from other lactic acid bacterial genomes is required to assess the contribution of prophages to phenotype in this industrially and medically important group of bacteria. Phage ecology In all but the most extreme environments, large numbers of different bacteria and phages are found. The abundance of phages varies strongly in different environments and is related to bacterial abundance (Weinbauer, 2004); there is a dynamic relationship between phage

Phage Genomics and Applications

population sizes and host numbers and physiology (Day, 2004). Phage are universally observed in open and coastal waters, marine sediments, terrestrial ecosystems such as soil, and the bodies of humans and animals, their presence having been detected in feces (Breitbart et al., 2003), saliva (Bachrach et al., 2003; Hitch et al., 2004), and human and bovine serum. While information on phage diversity is still limited, early indications from genomic sequencing and metagenomic analysis indicate that natural phage communities are reservoirs of the greatest uncharacterized genetic diversity on Earth, with an enormous variety of environmental niches and survival strategies (Weinbauer, 2004), from the temperate phage lambda at one end of the spectrum to the highly virulent phage T4 at the other (Day, 2004). Phages can affect host diversity by structuring bacterial communities through the control of dominant species, can cause lysis of a significant portion of the ocean biomass on a daily basis, can mediate genetic exchange among host organisms through transduction and lysogenic conversion, and are a component in nutrient cycling in ecology, moving nutrients between particulate and dissolved phases (Hambly and Suttle, 2005). It is becoming increasingly obvious that the behavior of phages is very intimately involved, and moderated by, the physiological stresses in the life cycle of bacteria (Day, 2004). Clearly, viruses possess an enormous potential for having a global impact on microbial ecology. Aquatic environments Much of our understanding of phage ecology to date comes from studies performed on marine ecosystems. The importance of microbial-driven processes in the oceans has become increasingly apparent in recent times; more than half of the total flux of matter and energy in marine food webs passes through heterotrophic microbes by means of dissolved organic matter (DOM; Fuhrman, 1999). The importance of phages in marine ecological and biogeochemical processes and the ecological significance of a viral shunt for organic carbon and nutrient fluxes has also begun to be appreciated (Figure 1.2) (Bergh et al., 1989; Fuhrman, 1999; Proctor, 1997), following the discovery that these dynamic members of aquatic communities are by far the most abundant. There are an estimated 1027 phage particles in the ocean (or 10 million per ml), comprising approximately 270 Mt of C (Bergh et al., 1989). Viral abundance tends to increase with the productivity of the system, with the result that it is lowest in the deep sea (104–105/ml) and highest in coastal environments (106–107/ml; Early electron microscopy studies demonstrated the morphological diversity of marine phages (Torrella and Morita, 1979), most resembling podoviruses, although myoviruses are most commonly isolated (Proctor, 1997). The majority of aquatic phages are DNA viruses, with only one RNA phage isolate known from marine systems (Proctor, 1997). Members of two cyanobacterial genera, Synechococcus and Prochlorococcus, are dominant within the world’s oceans and comprise an important part of the prokaryotic component of the picophytoplankton, contributing significantly to global photosynthetic productivity (Mann, 2003). In many oligotrophic environments, such as in the open ocean, cyanobacteria are responsible for 20–80% of the carbon fixation. Bacterial production of dissolved organic carbon (DOC) averages approximately 50% of the primary production (Weinbauer, 2004). Phages that infect this ecologically important group of bacteria are known as cyanophages and cyanophages that infect Synechococcus were among the first viruses isolated that infect marine primary producers (Suttle, 2005). It is believed that infection by this phage group

|

13

14

|

McAuliffe et al.

The ‘Viral Shunt’ Grazing Food Chain Primary Producers

Grazers

100%

Carnivores

1% 2-10%

Inorganic Nutrients Dissolved Organic Matter

Phage and other Viruses

6-26% Viral Shunt

3-15% Microbial Loop

Heterotrophic Prokaryotes

Figure 1.2 The “Viral Shunt.” A short-circuit in the marine food web, viral lysis influences biogeochemical cycles. Adapted from Wilhem and Suttle (1999).

in the oceans exerts a significant selection pressure on the evolution of both phageFig. and host 2 and community structure (Bailey et al., 2004). Virulent marine phages play a role in the formation of DOM through cell lysis, thus fuelling heterotrophic activity and potentially increasing atmospheric carbon dioxide levels (Fuhrman, 1999). Temperate marine phages can also potentially increase DOM levels during prophage induction events, but are also capable of altering host phenotypes through genetic exchange processes. These processes have major effects on the genetic structure and evolution of the global population of marine bacteria (Fuhrman, 1999). Recently, core photosynthetic genes, such as psbA that encodes the core reaction protein D1 in photosystem II, have been identified in all three of the fully sequenced cyanophage genomes that infect Prochlorococcus and in one Synechococcus phage genome (Mann et al., 2003). D1 is a rate-limiting photosynthesis protein, and host-derived D1 concentrations dramatically drop during phage infection (Lindell et al., 2004). The presence of these phage in the marine environments may ensure that photo-inhibition (i.e. the decline in photosynthetic viability of oxygen-evolving photosynthetic organisms due to excessive illumination) is prevented in infected cells, allowing photosynthesis to continue and therefore provide the energy needed by the virus for its replication (Mann et al., 2003). Similarly, many marine phages encode enzymes involved in phosphate metabolism, e.g. phoH (Chen and Lu, 2002; Sullivan et al., 2005). Cyanophages have acquired these ecologically important genes to help them to adapt to their marine habitat. Exciting findings are also being made from genome studies of uncultured aquatic phage communities. Recent metagenomic analysis of a marine viral water column and an uncultured near-shore marine-sediment viral community revealed that almost 75% of the sequences uncovered were not related to anything previously reported (Breitbart et al., 2004; Breitbart et al., 2002). In the sediment study, temperate phages were more common than lytic phages (Breitbart et al., 2004). This would suggest that lysogeny may be an important lifestyle for sediment viruses. Comparisons between the two communities showed

Phage Genomics and Applications

that certain phage phylogenetic groups were abundant in all marine viral communities, while other phage groups were under-represented or absent, suggesting that marine phage are derived from a common set of ancestors (Breitbart et al., 2004). Other studies such as these have resulted in viruses being isolated in various aquatic environments including solar salterns (Guixa-Boixereu et al., 1999), acidic hot springs (Rice et al., 2001), alkaline lakes ( Jiang et al., 2004), and in polar lakes (Kepner et al., 1998). This type of analysis offers more than just sequence information; it can offer insights into biogeographical distribution, community structure and ecological dynamics. Terrestrial environments In contrast to the information available on marine environments, the abundance and distribution of viruses in soils and other terrestrial systems is almost completely unknown. The most extensive field studies were carried out by Ashelford et al. and examined the population dynamics of six naturally occurring bacteriophages competing for an introduced Serratia liquifaciens host in the rhizosphere of sugar beets (Ashelford et al., 1999a; Ashelford et al., 2003; Ashelford et al., 2002; Ashelford et al., 1999b; Ashelford et al., 2000). These studies revealed that the temporal dynamics of phage and host populations were almost identical over three consecutive years. More recently, Keel et al. examined phage activity in soil and found that the efficacy of the widely used biocontrol bacterium Pseudomonas fluorescens CHA0 was completely eradicated by an epidemic of autochthonous phage in the rhizosphere of cucumbers (Keel et al., 2002). A recent study described the abundance and diversity of autochthonous viruses in six Delaware soils: two agricultural soils, two coastal plain forest soils, and two piedmont forest soils (Williamson et al., 2005). Virus communities extracted from the samples were dominated by phages, demonstrating a wide range of morphologies, including filamentous forms and phages with elongated capsids. The authors reported that land use was a significant factor influencing viral abundance and diversity in soils (Williamson et al., 2005). Phages have also been isolated from extreme terrestrial environments including the surface sands of the Sahara Desert (Prigent et al., 2005). Physiological environments—the human intestine The human gastrointestinal tract harbors more bacterial cells than we have human cells, an estimated 1014 microorganisms (Zhang et al., 2005). Most of these microbes are symbiotic to the human host and beneficial to food digestion (Nicholson et al., 2005). The GI microbiota also contains enteric viruses, including a number of known human viruses, uncharacterized viruses and a variety of bacteriophages. These bacteriophages exert a strong influence on the diversity and population structure of this bacterial community. A recent study by Breitbart et al. described the population structure and genome size distribution of a human fecal phage community using partial shotgun sequencing. The majority of the newly generated sequences (59%) were not significantly similar (E value of < 0.001) to anything previously reported; among the phage matches were siphophages (81%) and prophages within bacterial genomes. The most common phage matches were those that infect gram-positive bacteria, with phage A118 of Listeria monocytogenes, phage E125 of Burkholderia thailandensis and phage bIL285 of Lactococcus lactis being the most prevalent.

|

15

16

|

McAuliffe et al.

An average phage genome size of 30 kb was assumed from population modeling and it was calculated that the community contained approximately 1200 genotypes (Breitbart et al., 2003). The fecal viral community contained very few matches with the T7-like podophages and L-like siphophages. Phages in the intestine can also influence food digestion by regulating microbial communities in the human GI tract through lytic and lysogenic replication (Zhang et al., 2005). As a particular strain becomes dominant, phages can attack them (Chibani-Chennoufi et al., 2004b; Gorski and Weber-Dąbrowska, 2005). Also, through lysogenic conversion of members of the resident microflora, phages may introduce new fitness factors, such as antibiotic resistance determinants or the ability to produce certain enterotoxins. Phages that infect E. coli, Salmonella spp. and Bacteroides fragilis have been isolated from human fecal samples at high concentrations, up to 105 phages per g of feces in some cases (Calci et al., 1998; Cornax et al., 1994; Furuse et al., 1983; Gantzer et al., 2002; Kai et al., 1985). Interestingly, Havelaar and colleagues reported that patients with functional gastrointestinal disturbances had lower frequencies of E. coli K12 phage than patients with infectious diarrhea, suggesting that phage can act as biocontrol agents in the population dynamics of the intestinal bacteria in health and disease (Havelaar et al., 1986), a type of “natural phage therapy.” However, other groups have suggested that E. coli found in the gastrointestinal tract may be non-replicating or have an altered metabolism, and therefore are poor targets for phage infection (Chibani-Chennoufi et al., 2004b). Recently, the immunosuppressive properties of phage when administered in vivo were demonstrated (Gorski et al., 2006). Therefore, in addition to elimination of pathogens in the GIT and controlling the commensal bacterial load, phage may inhibit local immune reactions (Gorski and Weber-Dąbrowska, 2005). Since immune responses in the intestine remain in a state of controlled inflammation, this active suppression could play an important role in normal homeostasis (Makita et al., 2004). Thus, phages could protect against the development of gut inflammation in healthy individuals and lack of specific phages could contribute to conditions such as inflammatory bowel disease as well as disturbances caused by overgrowth of intestinal bacteria e.g. bacterial overgrowth syndrome (Gorski and Weber-Dąbrowska, 2005). Phage in medicine and therapeutics Phage therapy With the recent development of antibiotic resistance within the microbial population, the need for new antibacterials and alternative strategies to control microbial infections is of increasing urgency. One possible option is the use of bacteriophage as antimicrobial agents. Lytic phage kill bacteria via mechanisms that differ from those of antibiotics, and therefore, can be considered as antibacterials with a “novel mode of action,” a concept desired for all new antibacterial agents. The use of phages to treat bacterial infections in animals and humans is an old idea (Sulakvelidze et al., 2001). In Eastern Europe and the former Soviet Union, phage therapy has been used successfully to treat bacterial dysentery (Babalova et al., 1968; Tolkacheva et al., 1989), staphylococcal lung infections (Ioseliani et al., 1980), surgical wound infections (Zhukov-Verezhnikov et al., 1978), among others. Phage therapy was exploited for both diarrheal disease and the treatment of traumatic infections during

Phage Genomics and Applications

and after World War II (Sulakvelidze, 2005). During the 1920s and 1930s, therapeutic phage applications spread rapidly in response to a desperate need for treatment of bacterial infections in Western Europe and the USA. Orally administered phage preparations were reported to effectively treat patients infected with dysentery. Patients suffering from staphylococcal septicemia were also successfully treated by intravenous administration of anti-staphylococcal phages. Phages were reported to reduce the severity of staphylococcal meningitis and eliminate S. aureus from the cerebrospinal fluid. However, with the development of antibiotics for the treatment of infections in the early 1940s and their concomitant widespread use, early clinical trials were abandoned in the West. There is a current renewed interest in bacteriophage therapy in Western Europe and the USA in light of the emergence of drug-resistant pathogenic bacteria and there are strong indications that phages may yet have an important role to play in the treatment of bacterial infection in western countries. Phage therapy: human applications The majority of studies demonstrating the efficacy of phage therapy in clinical settings came from research groups in Eastern Europe and the former Soviet Union and were published in non-English journals (Sulakvelidze et al., 2001). In Poland, Ślopek et al. published a series of papers on the effectiveness of phages against infections caused by several bacterial pathogens, including drug-resistant strains (Ślopek et al., 1984; Ślopek et al., 1983a; Ślopek et al., 1983b; Ślopek et al., 1985a; Ślopek et al., 1985b; Ślopek et al., 1985c; Ślopek et al., 1987). A total of 550 patients with bacterial septicemia were treated with phages when antibiotics were shown to be ineffective in 95% of cases. Isolation of the etiologic agents led to selection of specific phages for treatment against staphylococci, Pseudomonas, Escherichia, Klebsiella and Salmonella. Phages were administered either orally or locally and treatment was continued for 1–16 weeks, depending on the severity of infection. The overall rate of success was 92% and even higher for the group where antibiotics were ineffective (94%). A summary of other studies conducted in Poland and in the former Soviet Union are found in a review by Sulakvelidze et al. (2001). However, control groups not treated with phage were not included in the majority of these studies and the effectiveness of phage therapy was not questioned. Phage therapy is still practised routinely in the former Soviet Union as an alternative, combinatory, and complimentary form of treatment in conjunction with, or in lieu of, antibiotics. Several studies have shown that carefully selected phage preparations may be as efficacious as antibiotics in the treatment of certain bacterial diseases (Kochetkova et al., 1989; Meladze et al., 1982; Sakandelidze, 1991). In one study of 340 patients with purulent disease of the lung, complete recovery from the disease was higher in patients treated with phage (82% recovered) compared to those treated with antibiotics (64% recovered; Meladze et al., 1982). Combining phage and antibiotic therapy was also shown to be beneficial for patients suffering from lung infections (Sakandelidze, 1991). This approach may also help to reduce the problem associated with the emergence of antibiotic- or phage-resistant bacterial pathogens since resistance to antibiotics and phage arise through different mechanisms (Sulakvelidze, 2005). In the West, the prospect of using phage therapy for the treatment and prevention of many bacterial diseases especially those caused by drug-resistant pathogens, is again a topic

|

17

18

|

McAuliffe et al.

for debate and has stimulated much discussion and controversy in recent times. There are a number of key considerations in applying bacteriophages as therapeutic agents in a clinical setting. Notable characteristics that make phage potentially attractive therapeutic agents include their high specificity towards their target bacterium; their self-replicating, self-limiting and also self-propagating nature; the ability to rapidly modify phage to combat the emergence of newly arising bacterial threats; and their history of safe use, as emphasized by their extensive use in Eastern Europe (Sulakvelidze, 2005). However, despite phage therapy’s long history in Eastern Europe and the former Soviet Union, therapeutic preparations would have to undergo a rigorous approval process by various regulatory agencies before they could be commercialized and used to treat humans in the western world. The efficacy of lytic phages in a clinical setting is still a controversial issue. Specific problems associated with using phages as therapeutic phages in humans include the failure to establish rigorous proof of efficacy and the specific mode of action responsible for the therapeutic effect (potentially an enzymatic effect rather than a lytic one). Also questions regarding the distribution of phages in the body, the clearance of phages by the immune system through the development of phage-neutralizing antibodies, development of bacterial resistance to bacteriophage and the potential for phage conversion of bacterial targets have yet to be adequately addressed. Phage technologies and products are certain to gain a foothold in clinical applications, if only to compensate for the ever-increasing decline in the efficacy of antibiotics. In addition, there is still an untapped, immeasurable market for phage applications in veterinary and agricultural settings. However, to date, there have been very few rigorous and controlled studies on any phage bacterial system in humans, or indeed animals and some skepticism exists within the scientific and medical communities. There will be many complex issues associated with using phage or phage products to treat disease, principally the stringent and lengthy regulatory processes required to go from promising molecule to drug on the market. While this technology is still some way from commercialization, there is a growing awareness in Western Europe and the United States of the market potential of phage therapy and its applications in the treatment of human infections. To this end, a number of biotech start-up companies have been formed to develop, manufacture, and achieve regulatory approval of phage pharmaceutical products for the treatment of antibiotic-resistant and other bacterial infections. Phage therapy: veterinary applications Although phage therapy has been historically associated with the use of phages in human medicine, applications in veterinary medicine and other agricultural settings have also been examined. The first demonstrated therapeutic use of phages in animals was the treatment of lethal fowl typhoid in chickens using phages isolated from birds infected with the diseasecausing agent, Salmonella gallinarum. More recently, animal infections caused by Salmonella (Berchieri et al., 1991), E. coli (Barrow et al., 1998; Huff et al., 2002a; Huff et al., 2002b; Smith and Huggins, 1982; Smith and Huggins, 1983; Smith et al., 1987), Campylobacter (Atterbury et al., 2005), vancomycin-resistant Enterococcus faecium (Biswas et al., 2002), V. cholerae (Sarkar et al., 1996), Clostridium difficile, and numerous others have been successfully treated with phage preparations.

Phage Genomics and Applications

A number of studies have demonstrated the possible therapeutic value of phage in treating E. coli in animals. E. coli can cause non-invasive enteritis and septicemia in various animal species. The best known of these studies came from the laboratories of Smith and Huggins in the early 1980s, which effectively reopened the field of phage therapy research in the West. Using a K1 E. coli meningitis mouse model, they demonstrated that low doses of intra-muscularly administered phage protected the mice against massive doses of the pathogen (Smith and Huggins, 1982). This treatment was shown to be at least as effective as multiple intramuscular injections of streptomycin, tetracycline, ampicillin, or chloramphenicol. Other observations included the multiplication of the phage in vivo, the persistence of phages in the bloodstream for 24 hours and in the spleen for several days and the identification of few phage-resistant mutants (Smith and Huggins, 1982). Subsequent studies by this group examined the effect of phage in the treatment of neonatal enteritis in calves, piglets and lambs (Smith and Huggins, 1983). Calves were orally infected with ca. 109 cfu of the O9:K30:99 enterotoxigenic strain of E. coli. Animals subsequently treated with ca. 1011 pfu of a phage mixture did not become ill, whereas the mortality rate in the untreated animals was 93–100% (Smith and Huggins, 1983). Similar studies in broiler chickens demonstrated the therapeutic effect of phage in the prevention of fatal E. coli respiratory infections (Huff et al., 2002b). There are also potential phage therapy application in aquaculture; bacteriophages are candidates as therapeutic agents in cultured fish and shellfish (Nakai and Park, 2002). As in the case of land animals, chemotherapy has resulted in the emergence of drug-resistant strains of bacteria responsible for causing infectious diseases in fish. Phages to some fish pathogens, including Aeromonas salmonicida (Ishiguro et al., 1983), Aeromonas hydrophila (Chow and Rouf, 1983), Edwardsiella tarda (Nakai and Park, 2002) and Yersinia ruckeri (Stevenson and Airdrie, 1984), have been isolated to date. Recent studies have shown phage effects against experimentally induced bacterial infections of cultured fish, particularly Lactococcus garvieae infections of yellowtail Seliora quinqueradiata (Nakai et al., 1999) and Pseudomonas plecoglossicida infection of ayu, Plecoglossus altivelis (Park et al., 2000). The lethality and specificity of phages for particular bacteria, the ability of phages to replicate within infected animal hosts, and the safety of phages make them efficacious antibacterial agents. Although there are still several hurdles to be overcome, it appears likely that phage therapy will regain a role in the veterinary treatment of infectious diseases, especially in the scenario of emerging antibacterial resistance. Phage lysins as antimicrobials A number of recent studies have shown the enormous potential of the use of phage endolysins, rather than the intact phage, as potential therapeutics. Phage endolysins, or lysins, are enzymes that damage the cell walls” integrity by hydrolyzing the four major bonds in its peptidoglycan component. The majority of phage lysins studied to date are modular in structure, composed of at least two distinctly separate functional domains: a C-terminal cell-wall binding domain, which directs the enzyme to its target, and an N-terminal catalytic domain. The catalytic domain can comprise one or more of the following types of peptidoglycan hydrolases: endopeptidases, muramidases (lysozyme), N-acetylmuramoylL-alanine amidases and glucosamidases. Most of the lysins studied to date are amidases.

|

19

20

|

McAuliffe et al.

What makes lysins attractive as potential therapeutics is their ability to cause “lysis from without,” i.e. they have the ability to degrade peptidoglycan and lyse their specific host even when applied externally (Fischetti, 2005). Recently, Loeffler et al. have studied Cpl-1, the lytic enzyme of a pneumococcal phage (Loeffler et al., 2003). This enzyme is specific for groups A, C, and E streptococci, with little or no activity toward oral streptococci or other commensal organisms tested. Using purified lysin in vitro, the authors have shown that 1000 units (10 ng) of enzyme is sufficient to sterilize a culture of approximately 107 group A streptococci within 5 seconds. Furthermore, when Cpl-1 was given orally to heavily colonized mice, no detectable streptococci were observed 2 hours after lysin treatment. Further studies with Cpl-1 have demonstrated its ability to also treat deep-rooted infections caused by S. pneumoniae, such as endocarditis in rats (Entenza et al., 2005). The in vivo efficacy of treatment with various phage lysins has also been examined using mice challenged with the following pathogens: Streptococcus pyogenes (Nelson et al., 2001), Bacillus anthracis (Schuch et al., 2002) and group B streptococci (Cheng et al., 2005). Recently, the lysin from the anti-staphylococcal phage K, LysK, was shown to inhibit drug-resistant strains of S. aureus, including MRSA, VRSA (vancomycin-resistant S. aureus) and VISA (vancomycin-intermediate S. aureus) strains as well as teicoplanin-resistant variants (O’Flaherty et al., 2005). Thus, the modular nature of endolysins lends itself to the creation of “designer lysins” through domain swapping studies. Such studies have been performed with pneumococcal phage lysins where it was reported that when N- and C-terminal domains of the pneumococcal amidase and lysozyme were interchanged, active enzymes were obtained (Lopez et al., 1995; Lopez et al., 1997). Indeed, further studies involved the construction of a trifunctional pneumococcal peptidoglycan hydrolase (Sanz et al., 1996). It can be envisaged that future research into phage endolysins will aid in the development of other fully active chimeric lysins displaying novel catalytic activities, broader host ranges and which can be designed to target any pathogen. Phage display Phage display technology is a particularly powerful molecular tool that has had a major impact on drug discovery, pharmacology, immunology and plant science. It is a technique by which foreign peptides, proteins or antibody fragments are expressed at the surface of phage particles (Smith, 1985). The heterologous peptide or protein is cloned into a phage or phagemid genome as a transcriptional fusion with one of the coat protein genes. These phages then become vehicles for expression that not only carry within them the nucleotide sequence encoding the expressed proteins, allowing the gene sequence to be retrieved, but also have the capacity to replicate. When the recombinant phages replicate, the new fusion proteins are expressed on the phage surface (Willats, 2002). The most widely used phage display methods are based on the use of the M13 filamentous phage of E. coli, but others including phage L (Sternberg and Hoess, 1995; Willats, 2002) and T7 (Yamamoto et al., 1999) have also been used. Using phage display, vast numbers of variant nucleotide sequences may be converted into populations of variant peptides and proteins which may be screened for desired properties. This is generally accomplished by affinity selection (or bio-panning) whereby phage populations are exposed to targets in order to selectively cap-

Phage Genomics and Applications

ture binding phage, and subsequently isolate monoclonal phage populations with desired specificities (Willats, 2002). Phage display technologies have proven to be a powerful enabling technology in genomics and drug development. The directed evolution of proteins can be engineered for specific properties and selectivity. This provides an approach for the engineering of human antibodies, as well as protein ligands, and for such diverse applications as arrays, separations, and drug development. Furthermore, for certain classes of proteins, libraries of antibody fragments provide a rich source of fully human therapeutic precursors (Winter et al., 1994). Antibody phage display has been the leading tool in antibody engineering during the past decade (Hoogenboom et al., 1998). Although initially applied to the isolation of antibodies against various target molecules, the technology has evolved into an efficient tool for the in vitro manipulation of antibody affinity, specificity and stability. The most successful application of antibody phage display has been the isolation of monoclonal antibodies using large phage antibody libraries (Winter et al., 1994). Other biotechnological applications of phage display technology include enzyme display (Fernandez-Gacio et al., 2003), vaccine development (Clark and March, 2004a) and the use of phage display fusions to target phage to specific cell types in order to deliver nucleic acids for gene therapy purposes. Enzyme technology has benefited greatly from phage display. In this case, it has proven more difficult to set up selection processes than for antibodies, processes in which efficient catalytic activity could be coupled to affinity selection. Several strategies have now been devised, each adapted to different classes of catalytic mechanisms. This should now allow the handling of very large libraries of evolved enzymes. The role of phage display in the development of vaccines will be discussed below. Vaccines A novel and exciting use of phages is the use of whole phage particles to deliver vaccines in the form of immunogenic peptides attached to modified phage coat proteins, or as delivery vehicles for DNA vaccines. Phage display is useful for the identification of immunogenic epitopes or mimotopes on displayed peptides which could, in turn, become the basis of peptide vaccines. A study carried out comparing the humoral immune response of animals immunized with a recombinant hepatitis B vaccine or with mimotopes generated by phage display demonstrated that the mimotopes could induce a response similar to that induced by the original antigen; in fact, the mimotopes induced the most reproducible and potent response (Meola et al., 1995). Bastien et al. investigated whether a recombinant phage displaying a known protective epitope to the human syncytial virus could protect against infectious challenge in mice (Bastien et al., 1999). The authors reported that complete protection against the corresponding pathogen could be elicited through mucosal delivery of a filamentous phage displaying the vaccine peptide. This study supports the usefulness of phage display of defined epitopes in prophylactic vaccination. Vaccination with phagedisplaying immunogenic peptides has a number of advantages over the use of recombinant peptides, such as the stimulation of both the cellular and humoral arms of the immune system (Gaubin et al., 2003; Manoutcharian et al., 1999). The studies described here and others have demonstrated that phage display shows great promise for vaccine delivery;

|

21

22

|

McAuliffe et al.

however, not all conformationally active epitopes will be preserved when delivered in this manner (i.e. a phage coat-vaccine peptide fusion; (Clark and March, 2004a)). Consequently, this may limit the scope of phage display in vaccine delivery to certain pathogen and/or diseases. An alternative approach to phage display vaccines and standard naked DNA vaccination is based on the delivery of DNA vaccines via phage particles to allow intracellular antigen expression (Clark and March, 2004a). In particular, temperate phages are particularly appealing, since their ability to form stable lysogens involves site specific insertion of the viral DNA into the bacterial chromosome. The gene encoding the antigen of choice is cloned into the phage genome under the control of a suitable eukaryotic promoter and the whole phage particle is used to inoculate the host (Figure 1.3) (Clark and March, 2004a). Whole phage particles have proved to be highly efficient DNA vaccine delivery vehicles and have been shown to induce significant antibody responses in both lab mice and larger animals such as rabbits and sheep (Clark and March, 2004b). The first study describing phagemediated genetic immunization involved intramuscular injection of mice with L-gt11 containing the gene for the hepatitis B surface antigen (HBsAg) under the control of the cytomegalovirus promoter (PCMV). All mice immunized in this fashion gave significantly higher anti-HBsAg responses than the appropriate controls, in excess of 150 IU/ml (Clark and March, 2004b). The results suggested that direct targeting of antigen-presenting cells (APCs) by phage vaccines occurs, leading to enhanced immune responses compared to naked DNA delivery. Indeed, vaccination of rabbits (March et al., 2004) and sheep has also demonstrated that antibody and cellular immune responses produced as a result of phage

+

Vaccine gene cloned into phage DNA

Package DNA into phage particles

Cell expresses vaccine protein encoded on DNA on surface

Purify phage

Phage broken down and vaccine-containing DNA released

Inject phage into mouse

Immune cells engulf phage particles

Figure 1.3 Schematic representation of the basic principles of phage DNA vaccination. The exact cells involved in protein expression and the mechanisms of expression are still unknown. Fig. 3 Adapted from Clark and March (2004).

Phage Genomics and Applications

delivery of a DNA vaccine are long lasting and significantly higher than those found after vaccination with naked DNA and in some cases, comparable with those produced after vaccination with recombinant proteins. Recently, immunogenic phage constructs against Mycoplasma mycoides subsp. mycoides, the etiologic agent of contagious bovine pleuropneumonia (CBPP), were isolated from a whole bacteriophage L semi-random genomic library of the organism (March et al., 2006). A number of clones were tested for their immunogenicity and protective effect in mouse models. Excellent cellular priming by phage particles was observed; challenge of these mice with M. mycoides subsp. mycoides resulted in a reduced level of mycoplasmemia indicating a trend towards protection (March et al., 2006). There are several advantages to this approach over standard DNA vaccines: large expression constructs can be used (up to 20 kb in standard L vector) suggesting that phage DNA vaccines could be multivalent; the vaccine DNA is protected from degradation within the phage matrix; phage delivery of vaccine DNA appears to target APCs; phage do not carry antibiotic resistance genes, removing potential barriers to regulatory approval; phage vaccines can be easily and inexpensively prepared; and oral delivery is a possibility (Clark and March, 2004b). However, while much promise exists for this technology, only proofof-concept research has been conducted thus far; further work is required to determine the scientific and commercial viability of this method of vaccination. Detection of pathogens The specific interaction of a bacteriophage and its host lends itself to using phages for the detection of bacteria, in particular, pathogenic bacteria. Unlike other detection systems such as ELISA and PCR, detection with phage is a natural system whereby the phages specifically recognize and bind to their host cells. One of the first applications of this phage specificity was in phage typing where phages with different lytic spectra are used to discriminate between different isolates of a bacterial species or genus, according to their ability to infect the isolate. Subsequently, detection techniques were developed that exploited this specific binding ability of phage, including the production of fluorescently labeled antibodies against phage particles (Watson and Eveland, 1965) or fluorescently labeled phage in a fluorescent bacteriophage assay (FBA) (Goodridge et al., 1999). With increased knowledge of phage genetics and structure, recombinant phages were used. An example is the Luciferase Reporter Phage (LRP; Figure 1.4A). Initially, the bacterial bioluminescence (lux) gene was introduced into L phage so that phage infection could be easily detected using a bioluminometer (Ulitzur and Kuhn, 2000). LPR have been used successfully to detect the presence of Salmonella in eggs (Chen and Griffiths, 1996), enteric bacteria in meat (Kodikara et al., 1991) and Mycobacterium tuberculosis in sputum samples (Trollip et al., 2001). A Listeria lux phage developed by Loessner et al. (1996) to detect the presence of Listeria monocytogenes in food samples demonstrated that the specificity of phage–host interaction allowed the detection of low numbers of pathogens in the presence of high numbers of natural competitors without successive rounds of pre-enrichment and selective plating (Loessner et al., 1996). The use of other reporter genes has also been described including the bacterial ice nucleation protein, ina, to detect Salmonella (Wolber and Green, 1990), and green fluorescent protein, GFP, to detect the presence of E. coli (Funatsu et al., 2002).

|

23

24

|

McAuliffe et al.

A.

B.

Step 1. Phage Infection

Step 2. Addition of Virucide

Step 1. Insertion of lux Gene in Phage Genome Transpostion or Homologous Recombination

Phage genome inserted into bacterial cell

Internalised phage genome protected

luxAB gene

Step 2. Infection of Target Cell in Sample

Step 3. Phage Amplification Step 4. Phage Released in Target Cells

luxAB gene

Step 3. Expression of Marker Gene Detected Step 5. Phage Amplification in Sensor Cells Plaque Assay luxAB gene Luciferase

Light detected with bioluminometer Plaques – Positive Result

No Plaques – Negative Result

Figure 1.4 Phage-based detection assays. (A) Amplification technology. (B) Luciferase reporter phage. Adapted from Kalantri et al. (2005).

Other phage-based detection systems are based on amplification technology (Figure 1.4B): the amplification of phages after infection of the host and detection of the progeny phages by plaque formation (Kalantri et al., 2005). McNerney et al. investigated the detection of tuberculosis (TB) with this method. Using a high concentration of phage D29 (high multiplicity of infection), the authors optimized the phage inoculate and incubation times, allowing highly sensitive detection of M. bovis BCG (fewer than 10 cfu (100 cfu/ml) detected (McNerney et al., 2004)). Detection of M. tuberculosis in sputum samples using the phage-based assay was then compared to smear microscopy and culture on Lowenstein–Jensen medium for the diagnosis of pulmonary TB. The phage replication assay demonstrated sensitivity slightly greater than that of direct smear microscopy but was less specific (McNerney et al., 2004). Recently, a number of phage-based detection assays for M. tuberculosis were compared to the conventional diagnostic techniques (Kalantri et al., 2005). This review suggested that while phage-based assays have higher specificity, their sensitivity was modest and quite variable. Because of their overall low sensitivity, the similarity of these assays to sputum microscopy with respect to accuracy and the need for a laboratory infrastructure that may not be necessarily available in primary care settings, the authors concluded that at the current time, phage-based assays cannot replace conventional diagnostic tests for TB (Kalantri et al., 2005). However, since phage-based assays are also

Phage Genomics and Applications

more specific than the competing PCR-based methods (93% compared to 85%), further research to enhance the sensitivity of these assays may result in the development of more effective phage-based detection methods for TB. Phages can also be used as indicators of fecal contamination of the environment, since coliphages are ubiquitous in the feces of humans and other warm-blooded animals (Dhillon et al., 1976) and the numbers of these phages in fresh and marine waters appears to be related to the density of coliforms. Coliphages also exhibit great resistance to environmental stresses and, thus, are useful for detecting remote sources of fecal pollution. In addition to coliforms, coliphages can also be used as indicators of enteric viruses in water (Lasobras et al., 1999). Other biotechnological applications of phage Biocontrol in foods Many foods support complex microbial ecosystems and can harbor relatively large numbers of bacteria that reflect the unique history of the foods, e.g. its origin, composition, human and environmental contact, processing, packaging, storage etc. These bacteria may include those related to food production, psychrotrophic spoilage organisms, and coliforms (Greer, 2005). Since phages are associated with most bacterial species, it is therefore not surprising that phages can be readily recovered from foods (Kennedy and Bitton, 1988). Many reports have been published detailing the isolation of phages from various foods and food processes, suggesting phages are normal inhabitants of food ecosystems (Brussow et al., 1994; Davis et al., 1985; Lu et al., 2003; Yoon et al., 2002). Phages have been isolated from many retail foods including cheese (Suarez et al., 2002), yoghurt (Binetti and Reinheimer, 2000; Suarez et al., 2002) and other dairy products, vegetables such as cucumber (Lu et al., 2003), lettuce (Kennedy et al., 1986), mushrooms (Kennedy et al., 1986) and a variety of meats including pork, chicken and beef. A number of phage–host systems in foods have been studied in detail; phage–host interactions during industrial fermentations of dairy products provided the first evidence that phages exist in the food production environment (Greer, 2005). As in other ecosystems, indigenous phages are likely to influence the diversity of the food microflora by controlling the indigenous bacterial species. Some beneficial applications of phage–host interactions in foods include the potential inhibition of spoilage and/or pathogenic bacteria in refrigerated or perishable foods through phage biocontrol (Hudson et al., 2005) and the potential for development of phage detection methods as hygiene indicators for assessing food quality (Kennedy and Bitton, 1988). In contrast, inhibition of bacteria used for the fermentation of foods is an example of a detrimental implication of the presence of phage in food systems. Phages have a wide range of potential applications in non-clinical settings where regulations for their use as biocontrol agents in foods may not be as stringent. Phages have been utilized to control bacteria during the production of foods of plant and animal origin (pre-harvest) and in perishable foods during storage at refrigeration temperatures (post-harvest; Table 1.2) (Greer, 2005). Phages utilized for the control of pathogens in foods and food processes tend to originate from environmental samples such as soil, feces, sewage and other effluents (Connerton et al., 2004; Dykes and Moorhead, 2002; O’Flynn

|

25

26

|

McAuliffe et al.

Table 1.2 Use of bacteriophages to control bacterial pathogens pre- and post-harvest Food production system/food

Disease/symptom

Phage host strain

Reference

Cultivated tomatoes

Bacterial blotch

Pseudomonas tolaasii

Greer et al., 2005

Tomatoes

Bacterial spot

Xanthomonas campestris pv. vesicatoria

Flaherty et al., 2000

Apples

Fire blight

Erminia amylovara

Erskine, 1973

Stone fruits

Prunus bacterial spot

X. campestris pv. pruni

Greer et al., 2005

Sprouts

Seed contamination

Salmonella Enteriditis

Greer et al., 2005

Fish

Red fin disease

Aeromonas hydrophila Chow and Rouf, 1983

Haemorrhagic ascites

Pseudomonas plecoglossicida

Park et al., 2000

Cecal Salmonella

Salmonella Enteriditis

Fiorentin et al., 2005

Lethal infection

Salmonella Typhimurium

Berchieri et al., 1991

Respiratory infection

Escherichia coli

Huff et al., 2002a

Growth depression

Streptococcus faecium

Greer et al., 2005

Beef cattle

Bacterial shedding

E. coli O157:H7

Greer et al., 2005

Calves, pigs, lambs

Diarrhoea, lethal infection

Enteropathogenic E. coli

Barrow et al., 1998

Sheep

Bacteria in rumen, colon

E. coli O157:H7

Smith and Huggins, 1983

Dairy cows

Mastitis

Staphylococcus aureus

Greer et al., 2005

Pigs

Salmonella

Salmonella Typhimurium

Berchieri et al., 1991

Listeria monocytogenes

Leverentz et al., 2004

Salmonella Enteriditis

Leverentz et al., 2001

Staphylococcus aureus

Greer et al., 2005

Pseudomonas fragi

Greer et al., 2005

Salmonella Enteriditis

Greer et al., 2005

Pre-harvest

Chickens

Post-harvest Melon and apple slices

Milk

Cheese

Phage Genomics and Applications

Table 1.2 Continued Food production system/food

Phage host strain

Reference

Campylobacter jejuni

Atterbury et al., 2005

Salmonella Enteriditis

Goode et al., 1999

Retail chicken

Salmonella Typhimurium DT104

Higgins et al., 2005

Chicken frankfurters

Salmonella Typhimurium DT104

Higgins et al., 2005

Beef steaks

E. coli O157:H7

O’Flynn et al., 2004

Pseudomonas sp.

Greer et al., 2005

Vacuum-packed beef

L. monocytogenes

Dykes and Moorehead, 2002

Pork fat

Brocothrix thermosphacta

Greer and Dilts, 2002

Chicken skin

Disease/symptom

et al., 2004), whereas those used to control spoilage organisms usually come from the food itself (Greer, 1983; Greer and Dilts, 2002; Kennedy and Bitton, 1988). Administration of phages directly to food animals has been suggested as an intervention strategy to specifically eliminate pathogens. However, until recently, preslaughter intervention was not considered to improve food safety and many studies focus on postslaughter strategies. Fresh fruit, vegetables and other foods of plant origin represent prime targets for bacterial pathogen contamination since these foods may not undergo any more processing (i.e. cooking). Research to date suggests that phage have the potential to control pathogens associated with food producing crops or the food itself. In this respect, Leverentz et al. studied phage–host interactions on fresh-cut apples and honeydew melons (Leverentz et al., 2001). A cocktail of four distinct Salmonella enteritidis phages produced a significant reduction in the numbers of Salmonella recovered from deliberately inoculated melon slices under refrigeration and abusive temperatures. This group also demonstrated the ability of phage to control pathogens on honeydew melons in combination with a bacteriocin and found that pathogen numbers decreased by up to 5.7 logs when applied in combination with nisin (Leverentz et al., 2003). Phage application via aerosol has also been examined (Leverentz et al., 2004). Most of the phage biocontrol studies to date concern the control of bacteria in foods such as dairy products, poultry and meat. A number of publications have described pathogen control by phage in poultry (Atterbury et al., 2005; Fiorentin et al., 2005; Goode et al., 2003; Higgins et al., 2005; Loc Carrillo et al., 2005; Wagenaar et al., 2005). The organisms particularly targeted in these studies were Salmonella and Campylobacter. Colonization of broiler chickens by the enteric pathogen Campylobacter jejuni is widespread and difficult to prevent (Loc Carrillo et al., 2005) and Salmonella enteritidis serovar Enteritidis associated with chickens and other poultry products is an important source of infection for human salmonellosis. Atterbury et al. demonstrated reduced numbers of C. jejuni post-harvest on

|

27

28

|

McAuliffe et al.

inoculated chicken skin when treated with phage (Atterbury et al., 2005). Since phages are ubiquitous in commercial poultry, the authors argued that their use as biocontrol agents would not constitute the introduction of a foreign entity into the product. Recently, Loc Carillo et al. performed such a study in broiler chickens; the authors reported that pre-harvest phage treatment of C. jejuni-colonized birds resulted in Campylobacter counts falling between 0.5 and 5 log10 cfu/g of cecal contents compared to untreated controls over a 5-day period post-administration (Loc Carrillo et al., 2005). These reductions were dependent on the phage-Campylobacter combination, the dose of phage applied, and the time elapsed after administration. Phages isolated from free-range chickens were tested as a therapeutic agent for reducing the concentration of Salmonella enterica serovar Enteritidis phage type 4 in the ceca of broilers (Fiorentin et al., 2005). Five days after treatment, the bacteriophagetreated group showed a reduction of 3.5 orders of magnitude on colony-forming units of S. Enteritidis PT4 per gram of cecal content. O’Flynn et al. demonstrated that a cocktail of E. coli O157:H7 phages isolated from bovine farmyard slurry samples could be exploited to control the pathogen on the surface of deliberately inoculated beefsteak (O’Flynn et al., 2004) suggesting that phage cocktails could reduce the numbers of O157:H7 found on beef hides and carcasses, or could potentially be used to reduce carriage and shedding in live animals. Other groups reported that phages could control E. coli O157:H7 in ruminal fluid and that oral administration of phage mixtures can reduce the duration of O157:H7 shedding by calves. However, studies in an artificial rumen system and in an inoculated sheep were not so encouraging; although phages could eliminate O157:H7 in vitro in rumen fluid, they had no effect on fecal shedding by deliberately inoculated lambs. There are several reasons why this may be the case including time of application, inactivation of the phage in vivo or the use of a single phage. It has been demonstrated the phage efficacy is enhanced when mixtures of phages are administered. A successful attempt to control E. coli shedding in calves required the pooling of six or seven phages. The advantages of using a phage biocontrol strategy for foods include the fact that phage are natural, self-perpetuating, have a history of safe use, can be highly host-specific and replicate only in the presence of their host (Greer, 2005). However, there are also several disadvantages including the efficacy of phages as antimicrobial agents ( Joerger, 2003), genetic transfer of undesirable characteristics, and the emergence of phage-resistant bacteria. Phage biocontrol strategies in food appear to be a more natural alternative to traditional food safety and preservation technologies. However, most of the data generated comes from studies with artificially inoculated crops, animals or foods and real world conditions have not been examined as a rule. The technology must be transferred to the field if it is to be adopted in the preharvest and postharvest stages of food production. Phages in industrial fermentations Any industry or process that relies on bacterial fermentation is at risk of infection by phages. Industries affected may include those involved in enzyme production such as amylases and proteases, acid production and antibiotic production (Moineau, 2005). One industry in particular that has openly acknowledged its phage problem is the dairy industry. For almost 70 years, dairy scientists have tried to control, if not eliminate, phages that interfere

Phage Genomics and Applications

with the manufacture of fermented milk products as bacteriophage infection of the starter cultures used in dairy fermentations is the main cause of disturbance in the manufacture of products such as cheese, yogurt, etc. (Coffey and Ross, 2002; Moineau, 1999). Siphophages of the species c2, 936, and P335, that infect Lactococccus lactis, are most commonly found in dairy plants (Bissonnette et al., 2000; Madera et al., 2004). Only virulent representatives of the c2 and 936 groups are known, while P335 includes both temperate and lytic viruses. Of the three, phages belonging to the 936 quasi-species are most frequently isolated from dairy environments, followed by phages belonging to the c2 group (Madera et al., 2004). These phages may originate from the starter strains themselves as these frequently harbor prophages (Chopin et al., 2001).Another possibility is that these phages are introduced with the milk to be used in the fermentation process; in fact, Madera et al. demonstrated that milk is probably the most important phage contamination source in dairy factories (Madera et al., 2004) Also, mixed starters may harbor phages that coexist with the bacterial strains, resulting in continuous recontamination of the dairy environment. Recognition of the phage problem in the dairy industry led to the design and application of a variety of practical measures for its alleviation, such as direct inoculation of the starters in closed fermentation vats, use of antiphage media for starter propagation, and rotation of starter cultures (Daly, 1983). However, this has invariably resulted in a very limited number of strains capable of withstanding intensive use in industry. More recently, application of genetic techniques to improve the phage-resistance of starter cultures has been successful in controlling phage populations in the environment of the dairy plant. One such approach is a more traditional “fire-fighting” tactic, whereby bacteriophage-insensitive mutants (BIMs) of the “phaged out” culture are generated in the laboratory. BIMs are spontaneous resistant mutants that survive extended exposure to high levels of lytic phage in factory whey and they can be used as replacement cultures when phage problems arise. Another approach is that of transferring naturally occurring host-encoded phage resistance systems to starter strains by food-grade means. Individual members of four such mechanisms have been characterized in detail and these have been extensively transferred to a variety of strains used in industrial fermentations to create phage-insensitive variants (Coffey and Ross, 2002). These systems include adsorption inhibition injection blocking mechanisms, restriction/modification and abortive infection (Forde and Fitzgerald, 1999). However, the application of phage resistance mechanism in starter cultures places an evolutionary pressure on phage to circumvent the mechanisms and this has led to the emergence of new phages which can overcome these defense mechanisms. Methods to counteract this include stacking of multiple phage resistance systems in a single starter strain ( Josephsen and Klaenhammer, 1990; Klaenhammer, 1991; Sing and Klaenhammer, 1993). As a consequence of all these measures, total loss of the final product is infrequent nowadays, although phages are still responsible for quality defects that affect the flavor, texture, and even safety of dairy foods. New approaches will need to be taken in the future and given the recent advancements in the genomics of starter bacteria and starter phage, investigations into phage–host interactions using genomics approaches such as microarrays may provide the key to the development of new host defense systems through studying the cellular response of the host to phage infection.

|

29

30

|

McAuliffe et al.

Biocontrol of plant pathogens Phages have been found in such plant-associated environments as leaves, root nodules, rotting fruit, seeds, stems, tomatoes, barley and beans (Goodridge, 2004) and phage impact on plants is mediated primarily through association with plant bacteria. Plant-associated symbiotic bacteria can range from mutuals to pathogens, examples of which are Bradyrhizobium japonicum and Erwinia amylovora respectively, and the phage impact on bacteria can also range from mutualistic to beneficial. Phage impact on plants can range from phageinduced bacterial lysis, to selection for phage resistance within bacterial communities, to phage conversion of bacterial phenotypes (Gill and Abedon, 2003). The use of phage in agriculture as bio-control agents against plant pathogens has been reviewed recently by Gill and Abedon (Gill and Abedon, 2003). Phage may lead to a reduction in the use of chemical pesticides, an outcome which should have widespread appeal among environmentalists, and also reduce the use of antibiotics, resistance to which may be transferred to humans via the food chain (Goodridge, 2004). Recent work performed by Jackson and co-workers utilized phage for the treatment of bacterial spot on tomatoes and blight on geraniums caused by Xanthomonas campestris pv. Vesicatoria (Flaherty et al., 2001; Flaherty et al., 2000). In both of these studies, host range mutant phage (phage which could lyse both the parent strain and bacteriophage insensitive mutants of the parent strain) were isolated and employed in a phage cocktail. In the first study, a phage cocktail containing four host range mutants reduced the incidence of bacterial spot from 40.5% to 0.9% in greenhouse grown seedlings (Flaherty et al., 2000). However, in field grown tomatoes the incidence of bacterial spot was only reduced from 17.5% to 16.8%. Daily application of a phage cocktail, containing four broad host range mutants controlled blight of geranium by 50%. More recently, this group have further reduced bacterial spot on tomatoes by increasing the longevity of the phage on the plant surface. Phages have also been used to control soft rot and fire blight on apple trees associated with Erwinia (Erskine, 1973), Xanthomonas pruni-associated bacterial spot disease of peaches, bacterial blotch in mushrooms caused by Pseudomonas tolaasii, Salmonella Enteritidis in experimentally contaminated broccoli and mustard seeds, Salmonella associated with freshcut fruit (Leverentz et al., 2001), and potato seed-tuber infected with Streptomyces scabies. However, the complex bacteriophage–host interactions that exist in plant environments must be further investigated before an efficient means of employing phages to control plant pathogens can be developed. Phage for biofilm eradication The ability of phage to control bacterial biofilms has also been recently reviewed (Suthereland et al., 2004). Biofilms have become a huge problem in food and medicine and are caused by the intrinsic interaction between bacterial cells, secreted polysaccharides and the inert matrix. Bacteria contained within the biofilm are less accessible to antibiotics and indeed, phage. However, phage K29 has the ability to penetrate the polysaccharide capsule (biofilm) of E. coli, by utilizing enzymes which can degrade polysaccharide (Bayer et al., 1979). More recent studies have demonstrated that E. coli biofilms on the surface of polyvinylchloride coupons can be lysed using the coliphage T4. The interactions between phage T4 and the biofilm were traced using fluorescent and chromogenic probes (Doolittle

Phage Genomics and Applications

et al., 1995). Tait et al. investigated the ability of phage and their associated polysaccharide depolymerases to control enteric biofilm formation. A cocktail of phages specific for Enterobacter, which had been isolated from sewage samples, successfully eradicated single species biofilms of Enterobacter clocae. However, similar attempts with a dual species biofilm were unsuccessful. Application of phage in wastewater treatment Phages are highly abundant in aquatic environments, ranging from 104 to 108 pfu/ml (Bergh et al., 1989). These viruses have a role in controlling the diversity of bacterial communities through control of selected species competing for resources (Hewson et al., 2003), which has raised interest in their use in aquatic environmental applications. A number of studies have examined the application of bacteriophages as indicators of the presence of bacteria in wastewater treatment processes (Withey et al., 2005). It has been suggested that phages are functionally important components of activated sludge systems (Ewert and Paynter, 1980a; Ewert and Paynter, 1980b; Hantula et al., 1991; Khan et al., 2002a; Khan et al., 2002b; Thomas et al., 2002) and phage lysis of bacterial components of these systems has the potential to influence treatment performance by controlling certain bacterial groups responsible for key sludge treatment processes, such as phosphorous removal and nitrification. One of the benefits of using phages in wastewater treatments is the control of specific bacterial pathogens in sludge that may be applied to agricultural land where there is a possibility of transmission of these pathogens to humans (Withey et al., 2005). Successful phage treatment of pathogens, including E. coli and Salmonella, is highly dependent on the prevalence and diversity of species within wastewater. There is potential to use a phage treatment in combination with biological sludge stabilization processes to reduce the pathogen load (Withey et al., 2005). Other benefits include delivering improvements in sludge dewaterability and digestibility by reducing biofilm and floc formation and the control of foaming. Concluding remarks: the future of phage research As the post-genomic era progresses, the complexity of biological systems becomes more and more apparent, so complex that they cannot be understood without simple “model systems” to analyze and conceptualize them. At present, we lack the tools to decipher the intricacies of the interactions of living systems, and returning to the use of phages as model systems may provide the answers. Therefore, it appears that phage research has come full circle in the past 80 years. From the advances in our understanding of complex biological mechanisms using phage in the 1960s and 1970s, to a move away from simple organisms to more complex systems in the 80s, the focus of phage research is changing again as the field re-embraces the simplicity of the virus as a “model system” for developing new types of complexity analyses. References

Anderson, T.F. (1966). Electron microscopy of phages. In: Phage and the Origins of Molecular Biology. Cold Spring Harbor Laboratory. Ashelford, K.E., Day, M.J., Bailey, M.J., Lilley, A.K., and Fry, J.C. (1999a). In situ population dynamics of bacterial viruses in a terrestrial environment. Appl. Environ. Microbiol. 65, 169–174.

|

31

32

|

McAuliffe et al.

Ashelford, K.E., Day, M.J., and Fry, J.C. (2003). Elevated abundance of bacteriophage infecting bacteria in soil. Appl. Environ. Microbiol. 69, 285–289. Ashelford, K.E., Fry, J.C., Bailey, M.J., and Day, M.J. (2002). Characterization of Serratia isolates from soil, ecological implications and transfer of Serratia proteamaculans subsp. quinovora Grimont et al., 1983 to Serratia quinivorans corrig., sp. nov. Int. J. Syst Evol. Microbiol. 52, 2281–2289. Ashelford, K.E., Fry, J.C., Bailey, M.J., Jeffries, A.R., and Day, M.J. (1999b). Characterization of six bacteriophages of serratia liquefaciens CP6 isolated from the sugar beet phytosphere. Appl. Environ. Microbiol. 65, 1959–1965. Ashelford, K.E., Norris, S.J., Fry, J.C., Bailey, M.J., and Day, M.J. (2000). Seasonal population dynamics and interactions of competing bacteriophages and their host in the rhizosphere. Appl. Environ. Microbiol. 66, 4193–4199. Atterbury, R.J., Dillon, E., Swift, C., Connerton, P.L., Frost, J.A., Dodd, C.E., Rees, C.E., and Connerton, I.F. (2005). Correlation of Campylobacter bacteriophage with reduced presence of hosts in broiler chicken ceca. Appl. Environ. Microbiol. 71, 4885–4887. Babalova, E.G. Katsitadze, K.T., Sakvarelidze, L.A., Imnaishvili, N., Sharashidze, T.G., Badashvili, V.A., Kiknadze, G.P., Meipariani, A.N., Gendzekhadze, N.D., Machavariani, E.V., et al. (1968). [Preventive value of dried dysentery bacteriophage]. Zh. Mikrobiol. Epidemiol. Immunobiol. 45, 143–145. Bachrach, G., Leizerovici-Zigmond, M., Zlotkin, A., Naor, R., and Steinberg, D. (2003). Bacteriophage isolation from human saliva. Lett. Appl. Microbiol. 36, 50–53. Bailey, S., Clokie, M.R., Millard, A., and Mann, N.H. (2004). Cyanophage infection and photoinhibition in marine cyanobacteria. Res. Microbiol. 155, 720–725. Barksdale, L., and Arden, S.B. (1974). Persisting bacteriophage infections, lysogeny, and phage conversions. Annu. Rev. Microbiol. 28, 265–299. Barrow, P., Lovell, M., and Berchieri, A., Jr. (1998). Use of lytic bacteriophage for control of experimental Escherichia coli septicemia and meningitis in chickens and calves. Clin. Diagn. Lab. Immunol. 5, 294–298. Bastien, N., Trudel, M., and Simard, C. (1999). Complete protection of mice from respiratory syncytial virus infection following mucosal delivery of synthetic peptide vaccines. Vaccine 17, 832–836. Bayer, M.E., Thurow, H., and Bayer, M.H. (1979). Penetration of the polysaccharide capsule of Escherichia coli (Bi161/42) by bacteriophage K29. Virology 94, 95–118. Bensing, B.A., Rubens, C.E., and Sullam, P.M. (2001). Genetic loci of Streptococcus mitis that mediate binding to human platelets. Infect. Immun. 69, 1373–1380. Berchieri, A., Jr., Lovell, M.A., and Barrow, P.A. (1991). The activity in the chicken alimentary tract of bacteriophages lytic for Salmonella Typhimurium. Res. Microbiol. 142, 541–549. Bergh, O., Borsheim, K.Y., Bratbak, G., and Heldal, M. (1989). High abundance of viruses found in aquatic environments. Nature 340, 467–468. Betley, M.J., and Mekalanos, J.J. (1985). Staphylococcal enterotoxin A is encoded by phage. Science 229, 185–187. Bina, J., Zhu, J., Dziejman, M., Faruque, S., Calderwood, S., and Mekalanos, J. (2003). ToxR regulon of Vibrio cholerae and its expression in vibrios shed by cholera patients. Proc. Natl. Acad. Sci. USA 100, 2801–2806. Binetti, A.G., and Reinheimer, J.A. (2000). Thermal and chemical inactivation of indigenous Streptococcus thermophilus bacteriophages isolated from Argentinian dairy plants. J. Food Prot. 63, 509–515. Bissonnette, F., Labrie, S., Deveau, H., Lamoureux, M., and Moineau, S. (2000). Characterization of mesophilic mixed starter cultures used for the manufacture of aged cheddar cheese. J. Dairy Sci. 83, 620–627. Biswas, B., Adhya, S., Washart, P., Paul, B., Trostel, A.N., Powell, B., Carlton, R., and Merril, C.R. (2002). Bacteriophage therapy rescues mice bacteremic from a clinical isolate of vancomycin-resistant Enterococcus faecium. Infect. Immun. 70, 204–210. Boyd, E.F. (2005). Bacteriophage and bacterial virulence. In: Bacteriophages: Biology and Applications, pp. 223–265. Breitbart, M., Felts, B., Kelley, S., Mahaffy, J.M., Nulton, J., Salamon, P., and Rohwer, F. (2004). Diversity and population structure of a near-shore marine-sediment viral community. Proc. Biol. Sci. 271, 565–574.

Phage Genomics and Applications

Breitbart, M., Hewson, I., Felts, B., Mahaffy, J.M., Nulton, J., Salamon, P., and Rohwer, F. (2003). Metagenomic analyses of an uncultured viral community from human feces. J. Bacteriol. 185, 6220– 6223. Breitbart, M., and Rohwer, F. (2005). Here a virus, there a virus, everywhere the same virus? Trends Microbiol. 13, 278–284. Breitbart, M., Salamon, P., Andresen, B., Mahaffy, J.M., Segall, A.M., Mead, D., Azam, F., and Rohwer, F. (2002). Genomic analysis of uncultured marine viral communities. Proc. Natl. Acad. Sci. USA 99, 14250–14255. Brussow, H., Canchaya, C., and Hardt, W.D. (2004). Phages and the evolution of bacterial pathogens: from genomic rearrangements to lysogenic conversion. Microbiol. Mol. Biol. Rev. 68, 560–602. Brussow, H., and Desiere, F. (2001). Comparative phage genomics and the evolution of Siphoviridae: insights from dairy phages. Mol. Microbiol. 39, 213–222. Brussow, H., Fremont, M., Bruttin, A., Sidoti, J., Constable, A., and Fryder, V. (1994). Detection and classification of Streptococcus thermophilus bacteriophages isolated from industrial milk fermentation. Appl. Environ. Microbiol. 60, 4537–4543. Brussow, H., and Hendrix, R.W. (2002). Phage genomics: small is beautiful. Cell 108, 13–16. Brussow, H., and Kutter, E. (2005). Genomics and evolution of tailed phages. In: Bacteriophages: Biology and Applications, pp. 91–128. Cairns, J., Stent, G.S., and Watson, J.D. (1966). Phage and the Origins of Molecular Biology. Cold Spring Harbor Laboratory. Calci, K.R., Burkhardt, W., 3rd, Watkins, W.D., and Rippey, S.R. (1998). Occurrence of male-specific bacteriophage in feral and domestic animal wastes, human feces, and human-associated wastewaters. Appl. Environ. Microbiol. 64, 5027–5029. Campbell, A. (2003). The future of bacteriophage biology. Nat. Rev. Genet. 4, 471–477. Canchaya, C., Fournous, G., Chibani-Chennoufi, S., Dillmann, M.L., and Brussow, H. (2003). Phage as agents of lateral gene transfer. Curr. Opin. Microbiol. 6, 417–424. Cann, A.J., Fandrich, S.E., and Heaphy, S. (2005). Analysis of the virus population present in equine faeces indicates the presence of hundreds of uncharacterized virus genomes. Virus Genes 30, 151–156. Casjens, S. (1998). The diverse and dynamic structure of bacterial genomes. Annu. Rev. Genet. 32, 339–377. Casjens, S. (2003). Prophages and bacterial genomics: what have we learned so far? Mol. Microbiol. 49, 277–300. Casjens, S.R. (2005). Comparative genomics and evolution of the tailed-bacteriophages. Curr. Opin. Microbiol. 8, 451–458. Chen, F., and Lu, J. (2002). Genomic sequence and evolution of marine cyanophage P60: a new insight on lytic and lysogenic phages. Appl. Environ. Microbiol. 68, 2589–2594. Chen, J., and Griffiths, M.W. (1996). Luminescent Salmonella strains as real time reporters of growth and recovery from sublethal injury in food. Int. J. Food Microbiol. 31, 27–43. Cheng, Q., Nelson, D., Zhu, S., and Fischetti, V.A. (2005). Removal of group B streptococci colonizing the vagina and oropharynx of mice with a bacteriophage lytic enzyme. Antimicrob. Agents Chemother. 49, 111–117. Chibani-Chennoufi, S., Canchaya, C., Bruttin, A., and Brussow, H. (2004a). Comparative genomics of the T4-Like Escherichia coli phage JS98: implications for the evolution of T4 phages. J. Bacteriol. 186, 8276–8286. Chibani-Chennoufi, S., Sidoti, J., Bruttin, A., Kutter, E., Sarker, S., and Brussow, H. (2004b). In vitro and in vivo bacteriolytic activities of Escherichia coli phages: implications for phage therapy. Antimicrob. Agents Chemother. 48, 2558–2569. Chopin, A., Bolotin, A., Sorokin, A., Ehrlich, S.D., and Chopin, M. (2001). Analysis of six prophages in Lactococcus lactis IL1403: different genetic structure of temperate and virulent phage populations. Nucleic Acids Res. 29, 644–651. Chow, M.S., and Rouf, M.A. (1983). Isolation and Partial Characterization of Two Aeromonas hydrophila Bacteriophages. Appl. Environ. Microbiol. 45, 1670–1676. Clark, A.J., Inwood, W., Cloutier, T., and Dhillon, T.S. (2001). Nucleotide sequence of coliphage HK620 and the evolution of lambdoid phages. J. Mol. Biol. 311, 657–679.

|

33

34

|

McAuliffe et al.

Clark, J.R., and March, J.B. (2004a). Bacterial viruses as human vaccines? Expert Rev. Vaccines 3, 463–476. Clark, J.R., and March, J.B. (2004b). Bacteriophage-mediated nucleic acid immunisation. FEMS Immunol. Med. Microbiol. 40, 21–26. Coffey, A., and Ross, R.P. (2002). Bacteriophage-resistance systems in dairy starter strains: molecular analysis to application. Antonie Van Leeuwenhoek 82, 303–321. Connerton, P.L., Loc Carrillo, C.M., Swift, C., Dillon, E., Scott, A., Rees, C.E., Dodd, C.E., Frost, J., and Connerton, I.F. (2004). Longitudinal study of Campylobacter jejuni bacteriophages and their hosts from broiler chickens. Appl. Environ. Microbiol. 70, 3877–3883. Cornax, R., Morinigo, M.A., Gonzalez-Jaen, F., Alonso, M.C., and Borrego, J.J. (1994). Bacteriophages presence in human faeces of healthy subjects and patients with gastrointestinal disturbances. Zentralbl. Bakteriol. 281, 214–224. d’Herelle, F. (1917). Sur un microbe invisible antagoniste des bacilles dysenteriques. Compt. Rend. Acad. Sci. 165, 373–375. Daly, C. (1983). The use of mesophilic cultures in the dairy industry. Antonie Van Leeuwenhoek 49, 297–312. Daubin, V., and Ochman, H. (2004). Bacterial genomes as new gene homes: the genealogy of ORFans in E. coli. Genome Res. 14, 1036–1042. Davis, B.M., Moyer, K.E., Boyd, E.F., and Waldor, M.K. (2000). CTX prophages in classical biotype Vibrio cholerae: functional phage genes but dysfunctional phage genomes. J. Bacteriol. 182, 6992–6998. Davis, C., Silveira, N.F., and Fleet, G.H. (1985). Occurrence and properties of bacteriophages of Leuconostoc oenos in Australian wines. Appl. Environ. Microbiol. 50, 872–876. Day, M. (2004). Bacterial sensitivity to bacteriophage in the aquatic environment. Sci. Prog. 87, 179– 191. Desiere, F., Lucchini, S., Bruttin, A., Zwahlen, M.C., and Brussow, H. (1997). A highly conserved DNA replication module from Streptococcus thermophilus phages is similar in sequence and topology to a module from Lactococcus lactis phages. Virology 234, 372–382. Desiere, F., Lucchini, S., Canchaya, C., Ventura, M., and Brussow, H. (2002). Comparative genomics of phages and prophages in lactic acid bacteria. Antonie Van Leeuwenhoek 82, 73–91. Desiere, F., Pridmore, R.D., and Brussow, H. (2000). Comparative genomics of the late gene cluster from Lactobacillus phages. Virology 275, 294–305. Dhillon, T.S., Dhillon, E.K., Chau, H.C., Li, W.K., and Tsang, A.H. (1976). Studies on bacteriophage distribution: virulent and temperate bacteriophage content of mammalian feces. Appl. Environ. Microbiol. 32, 68–74. Doolittle, M.M., Cooney, J.J., and Caldwell, D.E. (1995). Lytic infection of Escherichia coli biofilms by bacteriophage T4. Can J. Microbiol. 41, 12–18. Dykes, G.A., and Moorhead, S.M. (2002). Combined antimicrobial effect of nisin and a listeriophage against Listeria monocytogenes in broth but not in buffer or on raw beef. Int. J. Food Microbiol. 73, 71–81. Edwards, R.A., and Rohwer, F. (2005). Viral metagenomics. Nat. Rev. Microbiol. 3, 504–510. Entenza, J.M., Loeffler, J.M., Grandgirard, D., Fischetti, V.A., and Moreillon, P. (2005). Therapeutic effects of bacteriophage Cpl-1 lysin against Streptococcus pneumoniae endocarditis in rats. Antimicrob. Agents Chemother. 49, 4789–4792. Erskine, J.M. (1973). Characteristics of Erwinia amylovora bacteriophage and its possible role in the epidemology of fire blight. Can. J. Microbiol. 19, 837–845. Ewert, D.L., and Paynter, M.J. (1980a). Enumeration of bacteriophages and host bacteria in sewage and the activated-sludge treatment process. Appl. Environ. Microbiol. 39, 576–583. Ewert, D.L., and Paynter, M.J. (1980b). Technique for determining total bacterial virus counts in complex aqueous systems. Appl. Environ. Microbiol. 39, 253–260. Fernandez-Gacio, A., Uguen, M., and Fastrez, J. (2003). Phage display as a tool for the directed evolution of enzymes. Trends Biotechnol. 21, 408–414. Ferretti, J.J., McShan, W.M., Ajdic, D., Savic, D.J., Savic, G., Lyon, K., Primeaux, C., Sezate, S., Suvorov, A.N., Kenton, S., et al. (2001). Complete genome sequence of an M1 strain of Streptococcus pyogenes. Proc. Natl. Acad. Sci. USA 98, 4658–4663. Fiorentin, L., Vieira, N.D., and Barioni, W., Jr. (2005). Oral treatment with bacteriophages reduces the concentration of Salmonella Enteritidis PT4 in caecal contents of broilers. Avian Pathol. 34, 258–263.

Phage Genomics and Applications

Fischetti, V.A. (2005). Bacteriophage lytic enzymes: novel anti-infectives. Trends Microbiol. 13, 491– 496. Flaherty, J.E., Harbaugh, B.K., Jones, J.B., Somodi, G.C., and Jackson, D.A. (2001). H-mutnat bacteriophages as a potential biocontrol of bacterial blight of geranium. Hortscience 36, 98–100. Flaherty, J.E., Jones, J.B., Harbaugh, B.K., Somodi, G.C., and Jackson, L.E. (2000). Control of bacterial spot on tomato in the greenhouse and field with h-mutant bacteriophages. Hortscience 35, 882–884. Ford, M.E., Sarkis, G.J., Belanger, A.E., Hendrix, R.W., and Hatfull, G.F. (1998). Genome structure of mycobacteriophage D29: implications for phage evolution. J. Mol. Biol. 279, 143–164. Forde, A., and Fitzgerald, G.F. (1999). Bacteriophage defence systems in lactic acid bacteria. Antonie Van Leeuwenhoek 76, 89–113. Freeman, V.J. (1951). Studies on the virulence of bacteriophage-infected strains of Corynebacterium diphtheriae. J. Bacteriol. 61, 675–688. Fuhrman, J.A. (1999). Marine viruses and their biogeochemical and ecological effects. Nature 399, 541–548. Funatsu, T., Taniyama, T., Tajima, T., Tadakuma, H., and Namiki, H. (2002). Rapid and sensitive detection method of a bacterium by using a GFP reporter phage. Microbiol. Immunol. 46, 365–369. Furuse, K., Osawa, S., Kawashiro, J., Tanaka, R., Ozawa, A., Sawamura, S., Yanagawa, Y., Nagao, T., and Watanabe, I. (1983). Bacteriophage distribution in human faeces: continuous survey of healthy subjects and patients with internal and leukaemic diseases. J. Gen. Virol. 64 (Pt 9), 2039–2043. Gantzer, C., Henny, J., and Schwartzbrod, L. (2002). Bacteroides fragilis and Escherichia coli bacteriophages in human faeces. Int. J. Hyg. Environ. Health 205, 325–328. Gaubin, M., Fanutti, C., Mishal, Z., Durrbach, A., De Berardinis, P., Sartorius, R., Del Pozzo, G., Guardiola, J., Perham, R.N., and Piatier-Tonneau, D. (2003). Processing of filamentous bacteriophage virions in antigen-presenting cells targets both HLA class I and class II peptide loading compartments. DNA Cell Biol. 22, 11–18. Gill, J., and Abedon, S.T. (2003). Bacteriophage Ecology and Plants (APSnet). Goode, D., Allen, V.M., and Barrow, P.A. (2003). Reduction of experimental Salmonella and Campylobacter contamination of chicken skin by application of lytic bacteriophages. Appl. Environ. Microbiol. 69, 5032–5036. Goodridge, L., Chen, J., and Griffiths, M. (1999). The use of a fluorescent bacteriophage assay for detection of Escherichia coli O157:H7 in inoculated ground beef and raw milk. Int. J. Food Microbiol. 47, 43–50. Goodridge, L.D. (2004). Bacteriophage biocontrol of plant pathogens: fact or fiction? Trends Biotechnol. 22, 384–385. Gorski, A., Kniotek, M., Perkowska-Ptasinska, A., Mróz, A., Przerwa, A., Gorczyca, W., Dąbrowska, K., Weber-Dąbrowska, B., and Nowaczyk, M. (2006). Bacteriophages and transplantation tolerance. Transplant. Proc. 38, 331–333. Gorski, A., and Weber-Dąbrowska, B. (2005). The potential role of endogenous bacteriophages in controlling invading pathogens. Cell Mol. Life Sci. 62, 511–519. Greer, G.G. (1983). Psychrotrophic Brocothrix thermosphacta bacteriophages isolated from beef. Appl. Environ. Microbiol. 46, 245–251. Greer, G.G. (2005). Bacteriophage control of foodborne bacteriat. J. Food Prot. 68, 1102–1111. Greer, G.G., and Dilts, B.D. (2002). Control of Brochothrix thermosphacta spoilage of pork adipose tissue using bacteriophages. J. Food Prot. 65, 861–863. Groisman, E.A., and Casadesus, J. (2005). The origin and evolution of human pathogens. Mol. Microbiol. 56, 1–7. Guixa-Boixereu, N., Lysnes, K., and Pedros-Alio, C. (1999). Viral lysis and bacterivory during a phytoplankton bloom in a coastal water microcosm. Appl. Environ. Microbiol. 65, 1949–1958. Hambly, E., and Suttle, C.A. (2005). The viriosphere, diversity, and genetic exchange within phage communities. Curr. Opin. Microbiol. 8, 444–450. Hantula, J., Kurki, A., Vuoriranta, P., and Bamford, D.H. (1991). Ecology of bacteriophages infecting activated sludge bacteria. Appl. Environ. Microbiol. 57, 2147–2151. Havelaar, A.H., Furuse, K., and Hogeboom, W.M. (1986). Bacteriophages and indicator bacteria in human and animal faeces. J. Appl. Bacteriol. 60, 255–262. Hendrix, R.W. (2003). Bacteriophage genomics. Curr. Opin. Microbiol. 6, 506–511. Hendrix, R.W., Lawrence, J.G., Hatfull, G.F., and Casjens, S. (2000). The origins and ongoing evolution of viruses. Trends Microbiol. 8, 504–508.

|

35

36

|

McAuliffe et al.

Hendrix, R.W., Smith, M.C., Burns, R.N., Ford, M.E., and Hatfull, G.F. (1999). Evolutionary relationships among diverse bacteriophages and prophages: all the world’s a phage. Proc. Natl. Acad. Sci. USA 96, 2192–2197. Hershey, A.D., and Chase, M. (1952). Independent functions of viral protein and nucleic acid in growth of bacteriophage. J. Gen. Physiol. 36, 39–56. Hewson, I., Vargo, G.A., and Fuhrman, J.A. (2003). Bacterial diversity in shallow oligotrophic marine benthos and overlying waters: effects of virus infection, containment, and nutrient enrichment. Microb. Ecol. 46, 322–336. Higgins, J.P., Higgins, S.E., Guenther, K.L., Huff, W., Donoghue, A.M., Donoghue, D.J., and Hargis, B.M. (2005). Use of a specific bacteriophage treatment to reduce Salmonella in poultry products. Poult. Sci. 84, 1141–1145. Hitch, G., Pratten, J., and Taylor, P.W. (2004). Isolation of bacteriophages from the oral cavity. Lett. Appl. Microbiol. 39, 215–219. Hoogenboom, H.R., de Bruine, A.P., Hufton, S.E., Hoet, R.M., Arends, J.W., and Roovers, R.C. (1998). Antibody phage display technology and its applications. Immunotechnology 4, 1–20. Hudson, J.A., Billington, C., Carey-Smith, G., and Greening, G. (2005). Bacteriophages as biocontrol agents in food. J. Food Prot. 68, 426–437. Huff, W.E., Huff, G.R., Rath, N.C., Balog, J.M., and Donoghue, A.M. (2002a). Prevention of Escherichia coli infection in broiler chickens with a bacteriophage aerosol spray. Poult. Sci. 81, 1486–1491. Huff, W.E., Huff, G.R., Rath, N.C., Balog, J.M., Xie, H., Moore, P.A., Jr., and Donoghue, A.M. (2002b). Prevention of Escherichia coli respiratory infection in broiler chickens with bacteriophage (SPR02). Poult. Sci. 81, 437–441. Ioseliani, G.D., Meladze, G.D., Chkhetiia, N., Mebuke, M.G., and Kiknadze, N. (1980). [Use of bacteriophage and antibiotics for prevention of acute postoperative empyema in chronic suppurative lung diseases]. Grudn Khir, 63–67. Ishiguro, E.E., Ainsworth, T., Shaw, D.H., Kay, W.W., and Trust, T.J. (1983). A lipopolysaccharide-specific bacteriophage for Aeromonas salmonicida. Can. J. Microbiol. 29, 1458–1461. Jiang, S., Steward, G., Jellison, R., Chu, W., and Choi, S. (2004). Abundance, distribution, and diversity of viruses in alkaline, hypersaline Mono Lake, California. Microb. Ecol. 47, 9–17. Joerger, R.D. (2003). Alternatives to antibiotics: bacteriocins, antimicrobial peptides and bacteriophages. Poult. Sci. 82, 640–647. Josephsen, J., and Klaenhammer, T. (1990). Stacking of three different restriction and modification systems in Lactococcus lactis by cotransformation. Plasmid 23, 71–75. Juhala, R.J., Ford, M.E., Duda, R.L., Youlton, A., Hatfull, G.F., and Hendrix, R.W. (2000). Genomic sequences of bacteriophages HK97 and HK022: pervasive genetic mosaicism in the lambdoid bacteriophages. J. Mol. Biol. 299, 27–51. Kai, M., Watanabe, S., Furuse, K., and Ozawa, A. (1985). Bacteroides bacteriophages isolated from human feces. Microbiol. Immunol. 29, 895–899. Kalantri, S., Pai, M., Pascopella, L., Riley, L., and Reingold, A. (2005). Bacteriophage- based tests for the detection of Mycobacterium tuberculosis in clinical specimens: a systematic review and meta-analysis. BMC Infect. Dis. 5, 59. Karch, H., Tarr, P.I., and Bielaszewska, M. (2005). Enterohaemorrhagic Escherichia coli in human medicine. Int. J. Med. Microbiol. 295, 405–418. Keel, C., Ucurum, Z., Michaux, P., Adrian, M., and Haas, D. (2002). Deleterious impact of a virulent bacteriophage on survival and biocontrol activity of Pseudomonas fluorescens strain CHAO in natural soil. Mol. Plant Microbe Interact. 15, 567–576. Kennedy, J.E., Jr., and Bitton, G. (1988). Bacteriophages in foods. Phage Ecol. 289–316. Kennedy, J.E., Jr., Wei, C.I., and Oblinger, J.L. (1986). Methodology for enumeration of coliphages in foods. Appl. Environ. Microbiol. 51, 956–962. Kepner, R.L., Jr., Wharton, R.A., Jr., and Suttle, C.A. (1998). Viruses in Antarctic lakes. Limnol. Oceanogr. 43, 1754–1761. Khan, M.A., Satoh, H., Katayama, H., Kurisu, F., and Mino, T. (2002a). Bacteriophages isolated from activated sludge processes and their polyvalency. Water Res. 36, 3364–3370. Khan, M.A., Satoh, H., Mino, T., Katayama, H., Kurisu, F., and Matsuo, T. (2002b). Bacteriophage–host interaction in the enhanced biological phosphate removing activated sludge system. Water Sci. Technol. 46, 39–43.

Phage Genomics and Applications

Klaenhammer, T.R. (1991). Development of bacteriophage-resistant strains of lactic acid bacteria. Biochem. Soc. Trans 19, 675–681. Kochetkova, V.A., Mamontov, A.S., Moskovtseva, R.L., Erastova, E.I., Trofimov, E.I., Popov, M.I., and Dzhubalieva, S.K. (1989). [Phagotherapy of postoperative suppurative-inflammatory complications in patients with neoplasms]. Sov. Med., 23–26. Kodikara, C.P., Crew, H.H., and Stewart, G.S. (1991). Near on-line detection of enteric bacteria using lux recombinant bacteriophage. FEMS Microbiol. Lett. 67, 261–265. Koonin, E.V., Aravind, L., and Kondrashov, A.S. (2000). The impact of comparative genomics on our understanding of evolution. Cell 101, 573–576. Kovacikova, G., and Skorupski, K. (1999). A Vibrio cholerae LysR homolog, AphB, cooperates with AphA at the tcpPH promoter to activate expression of the ToxR virulence cascade. J. Bacteriol. 181, 4250–4256. Labrie, S., and Moineau, S. (2002). Complete genomic sequence of bacteriophage ul36: demonstration of phage heterogeneity within the P335 quasi-species of lactococcal phages. Virology 296, 308–320. Lasobras, J., Dellunde, J., Jofre, J., and Lucena, F. (1999). Occurrence and levels of phages proposed as surrogate indicators of enteric viruses in different types of sludges. J. Appl. Microbiol. 86, 723–729. Lawrence, J.G., Hatfull, G.F., and Hendrix, R.W. (2002). Imbroglios of viral taxonomy: genetic exchange and failings of phenetic approaches. J. Bacteriol. 184, 4891–4905. Leverentz, B., Conway, W.S., Alavidze, Z., Janisiewicz, W.J., Fuchs, Y., Camp, M.J., Chighladze, E., and Sulakvelidze, A. (2001). Examination of bacteriophage as a biocontrol method for salmonella on freshcut fruit: a model study. J. Food Prot. 64, 1116–1121. Leverentz, B., Conway, W.S., Camp, M.J., Janisiewicz, W.J., Abuladze, T., Yang, M., Saftner, R., and Sulakvelidze, A. (2003). Biocontrol of Listeria monocytogenes on fresh-cut produce by treatment with lytic bacteriophages and a bacteriocin. Appl. Environ. Microbiol. 69, 4519–4526. Leverentz, B., Conway, W.S., Janisiewicz, W., and Camp, M.J. (2004). Optimizing concentration and timing of a phage spray application to reduce Listeria monocytogenes on honeydew melon tissue. J. Food Prot. 67, 1682–1686. Lindell, D., Sullivan, M.B., Johnson, Z.I., Tolonen, A.C., Rohwer, F., and Chisholm, S.W. (2004). Transfer of photosynthesis genes to and from Prochlorococcus viruses. Proc. Natl. Acad. Sci. USA 101, 11013–11018. Loc Carrillo, C., Atterbury, R.J., el-Shibiny, A., Connerton, P.L., Dillon, E., Scott, A., and Connerton, I.F. (2005). Bacteriophage therapy to reduce Campylobacter jejuni colonization of broiler chickens. Appl. Environ. Microbiol. 71, 6554–6563. Loeffler, J.M., Djurkovic, S., and Fischetti, V.A. (2003). Phage lytic enzyme Cpl-1 as a novel antimicrobial for pneumococcal bacteremia. Infect. Immun. 71, 6199–6204. Loessner, M.J., Rees, C.E., Stewart, G.S., and Scherer, S. (1996). Construction of luciferase reporter bacteriophage A511::luxAB for rapid and sensitive detection of viable Listeria cells. Appl. Environ. Microbiol. 62, 1133–1140. Lopez, R., Garcia, E., Garcia, P., and Garcia, J.L. (1995). Architecture and domain interchange of the pneumococcal cell wall lytic enzymes. Dev. Biol. Stand 85, 273–281. Lopez, R., Garcia, E., Garcia, P., and Garcia, J.L. (1997). The pneumococcal cell wall degrading enzymes: a modular design to create new lysins? Microb. Drug Resist. 3, 199–211. Lu, Z., Breidt, F., Jr., Fleming, H.P., Altermann, E., and Klaenhammer, T.R. (2003). Isolation and characterization of a Lactobacillus plantarum bacteriophage, phiJL-1, from a cucumber fermentation. Int. J. Food Microbiol. 84, 225–235. Lucchini, S., Desiere, F., and Brussow, H. (1999). Comparative genomics of Streptococcus thermophilus phage species supports a modular evolution theory. J. Virol. 73, 8647–8656. Madera, C., Monjardin, C., and Suarez, J.E. (2004). Milk contamination and resistance to processing conditions determine the fate of Lactococcus lactis bacteriophages in dairies. Appl. Environ. Microbiol. 70, 7365–7371. Makita, S., Kanai, T., Oshima, S., Uraushihara, K., Totsuka, T., Sawada, T., Nakamura, T., Koganei, K., Fukushima, T., and Watanabe, M. (2004). CD4+CD25bright T cells in human intestinal lamina propria as regulatory cells. J. Immunol. 173, 3119–3130. Mann, N.H. (2003). Phages of the marine cyanobacterial picophytoplankton. FEMS Microbiol. Rev. 27, 17–34.

|

37

38

|

McAuliffe et al.

Mann, N.H., Cook, A., Millard, A., Bailey, S., and Clokie, M. (2003). Marine ecosystems: bacterial photosynthesis genes in a virus. Nature 424, 741. Manoutcharian, K., Terrazas, L.I., Gevorkian, G., Acero, G., Petrossian, P., Rodriguez, M., and Govezensky, T. (1999). Phage-displayed T-cell epitope grafted into immunoglobulin heavy-chain complementarity-determining regions: an effective vaccine design tested in murine cysticercosis. Infect. Immun. 67, 4764–4770. March, J.B., Clark, J.R., and Jepson, C.D. (2004). Genetic immunisation against hepatitis B using whole bacteriophage lambda particles. Vaccine 22, 1666–1671. March, J.B., Jepson, C.D., Clark, J.R., Totsika, M., and Calcutt, M.J. (2006). Phage library screening for the rapid identification and in vivo testing of candidate genes for a DNA vaccine against Mycoplasma mycoides subsp. mycoides small colony biotype. Infect. Immun. 74, 167–174. McNerney, R., Kambashi, B.S., Kinkese, J., Tembwe, R., and Godfrey-Faussett, P. (2004). Development of a bacteriophage phage replication assay for diagnosis of pulmonary tuberculosis. J. Clin. Microbiol. 42, 2115–2120. Meladze, G.D., Mebuke, M.G., Chkhetiia, N., Kiknadze, N., and Koguashvili, G.G. (1982). [Efficacy of staphylococcal bacteriophage in the treatment of purulent lung and pleural diseases]. Grudn Khir, 53–56. Meola, A., Delmastro, P., Monaci, P., Luzzago, A., Nicosia, A., Felici, F., Cortese, R., and Galfre, G. (1995). Derivation of vaccines from mimotopes. Immunologic properties of human hepatitis B virus surface antigen mimotopes displayed on filamentous phage. J. Immunol. 154, 3162–3172. Miller, E.S., Heidelberg, J.F., Eisen, J.A., Nelson, W.C., Durkin, A.S., Ciecko, A., Feldblyum, T.V., White, O., Paulsen, I.T., Nierman, W.C., et al. (2003a). Complete genome sequence of the broad-hostrange vibriophage KVP40: comparative genomics of a T4-related bacteriophage. J. Bacteriol. 185, 5220–5233. Miller, E.S., Kutter, E., Mosig, G., Arisaka, F., Kunisawa, T., and Ruger, W. (2003b). Bacteriophage T4 genome. Microbiol. Mol. Biol. Rev. 67, 86–156, table of contents. Mirold, S., Rabsch, W., Rohde, M., Stender, S., Tschape, H., Russmann, H., Igwe, E., and Hardt, W.D. (1999). Isolation of a temperate bacteriophage encoding the type III effector protein SopE from an epidemic Salmonella typhimurium strain. Proc. Natl. Acad. Sci. USA 96, 9845–9850. Moineau, S. (1999). Applications of phage resistance in lactic acid bacteria. Antonie Van Leeuwenhoek 76, 377–382. Moineau, S. (2005). Control of bacteriophages in industrial fermentations. In: Bacteriophages: Biology and Applications, pp. 285–296. Nakai, T., and Park, S.C. (2002). Bacteriophage therapy of infectious diseases in aquaculture. Res. Microbiol. 153, 13–18. Nakai, T., Sugimoto, R., Park, K.H., Matsuoka, S., Mori, K., Nishioka, T., and Maruyama, K. (1999). Protective effects of bacteriophage on experimental Lactococcus garvieae infection in yellowtail. Dis. Aquat. Organ. 37, 33–41. Nelson, D., Loomis, L., and Fischetti, V.A. (2001). Prevention and elimination of upper respiratory colonization of mice by group A streptococci by using a bacteriophage lytic enzyme. Proc. Natl. Acad. Sci. USA 98, 4107–4112. Nicholson, J.K., Holmes, E., and Wilson, I.D. (2005). Gut microorganisms, mammalian metabolism and personalized health care. Nat. Rev. Microbiol. 3, 431–438. O’Brien, A.D., Newland, J.W., Miller, S.F., Holmes, R.K., Smith, H.W., and Formal, S.B. (1984). Shigalike toxin-converting phages from Escherichia coli strains that cause hemorrhagic colitis or infantile diarrhea. Science 226, 694–696. O’Flaherty, S., Coffey, A., Meaney, W., Fitzgerald, G.F., and Ross, R.P. (2005). The recombinant phage lysin LysK has a broad spectrum of lytic activity against clinically relevant staphylococci, including methicillin-resistant Staphylococcus aureus. J. Bacteriol. 187, 7161–7164. O’Flynn, G., Ross, R.P., Fitzgerald, G.F., and Coffey, A. (2004). Evaluation of a cocktail of three bacteriophages for biocontrol of Escherichia coli O157:H7. Appl. Environ. Microbiol. 70, 3417–3424. Papke, R.T., and Doolittle, W.F. (2003). Phage evolution: new worlds of genomic diversity. Curr. Biol. 13, R606–607. Park, S.C., Shimamura, I., Fukunaga, M., Mori, K.I., and Nakai, T. (2000). Isolation of bacteriophages specific to a fish pathogen, Pseudomonas plecoglossicida, as a candidate for disease control. Appl. Environ. Microbiol. 66, 1416–1422.

Phage Genomics and Applications

Pedulla, M.L., Ford, M.E., Houtz, J.M., Karthikeyan, T., Wadsworth, C., Lewis, J.A., Jacobs-Sera, D., Falbo, J., Gross, J., Pannunzio, N.R., et al. (2003). Origins of highly mosaic mycobacteriophage genomes. Cell 113, 171–182. Perna, N.T., Plunkett, G., 3rd, Burland, V., Mau, B., Glasner, J.D., Rose, D.J., Mayhew, G.F., Evans, P.S., Gregor, J., Kirkpatrick, H.A., et al. (2001). Genome sequence of enterohaemorrhagic Escherichia coli O157:H7. Nature 409, 529–533. Prigent, M., Leroy, M., Confalonieri, F., Dutertre, M., and DuBow, M.S. (2005). A diversity of bacteriophage forms and genomes can be isolated from the surface sands of the Sahara Desert. Extremophiles 9, 289–296. Proctor, L.M. (1997). Advances in the study of marine viruses. Microsc. Res. Tech. 37, 136–161. Proux, C., van Sinderen, D., Suarez, J., Garcia, P., Ladero, V., Fitzgerald, G.F., Desiere, F., and Brussow, H. (2002). The dilemma of phage taxonomy illustrated by comparative genomics of Sfi21-like Siphoviridae in lactic acid bacteria. J. Bacteriol. 184, 6026–6036. Rice, G., Stedman, K., Snyder, J., Wiedenheft, B., Willits, D., Brumfield, S., McDermott, T., and Young, M.J. (2001). Viruses from extreme thermal environments. Proc. Natl. Acad. Sci. USA 98, 13341–13345. Sakandelidze, V.M. (1991). [The combined use of specific phages and antibiotics in different infectious allergoses]. Vrach Delo, 60–63. Sanger, F., Air, G.M., Barrell, B.G., Brown, N.L., Coulson, A.R., Fiddes, C.A., Hutchison, C.A., Slocombe, P.M., and Smith, M. (1977). Nucleotide sequence of bacteriophage phi X174 DNA. Nature 265, 687–695. Sanz, J.M., Garcia, P., and Garcia, J.L. (1996). Construction of a multifunctional pneumococcal murein hydrolase by module assembly. Eur. J. Biochem. 235, 601–605. Sarkar, B.L., Chakrabarti, A.K., Koley, H., Chakrabarti, M.K., and De, S.P. (1996). Biological activity and interaction of Vibrio cholerae bacteriophages in rabbit ileal loop. Indian J. Med. Res. 104, 139–141. Schuch, R., Nelson, D., and Fischetti, V.A. (2002). A bacteriolytic agent that detects and kills Bacillus anthracis. Nature 418, 884–889. Sing, W.D., and Klaenhammer, T.R. (1993). A Strategy for Rotation of Different Bacteriophage Defenses in a Lactococcal Single-Strain Starter Culture System. Appl. Environ. Microbiol. 59, 365–372. Ślopek, S., Durlakowa, I., Weber-Dąbrowska, B., Dąbrowski, M., and Kucharewicz-Krukowska, A. (1984). Results of bacteriophage treatment of suppurative bacterial infections. III. Detailed evaluation of the results obtained in further 150 cases. Arch. Immunol. Ther. Exp. (Warsz) 32, 317–335. Ślopek, S., Durlakowa, I., Weber-Dąbrowska, B., Kucharewicz-Krukowska, A., Dąbrowski, M., and Bisikiewicz, R. (1983a). Results of bacteriophage treatment of suppurative bacterial infections. I. General evaluation of the results. Arch. Immunol. Ther. Exp. (Warsz) 31, 267–291. Ślopek, S., Durlakowa, I., Weber-Dąbrowska, B., Kucharewicz-Krukowska, A., Dąbrowski, M., and Bisikiewicz, R. (1983b). Results of bacteriophage treatment of suppurative bacterial infections. II. Detailed evaluation of the results. Arch. Immunol. Ther. Exp. (Warsz) 31, 293–327. Ślopek, S., Kucharewicz-Krukowska, A., Weber-Dąbrowska, B., and Dąbrowski, M. (1985a). Results of bacteriophage treatment of suppurative bacterial infections. IV. Evaluation of the results obtained in 370 cases. Arch. Immunol. Ther. Exp. (Warsz) 33, 219–240. Ślopek, S., Kucharewicz-Krukowska, A., Weber-Dąbrowska, B., and Dąbrowski, M. (1985b). Results of bacteriophage treatment of suppurative bacterial infections. V. Evaluation of the results obtained in children. Arch. Immunol. Ther. Exp. (Warsz) 33, 241–259. Ślopek, S., Kucharewicz-Krukowska, A., Weber-Dąbrowska, B., and Dąbrowski, M. (1985c). Results of bacteriophage treatment of suppurative bacterial infections. VI. Analysis of treatment of suppurative staphylococcal infections. Arch. Immunol. Ther. Exp. (Warsz) 33, 261–273. Ślopek, S., Weber-Dąbrowska, B., Dąbrowski, M., and Kucharewicz-Krukowska, A. (1987). Results of bacteriophage treatment of suppurative bacterial infections in the years 1981–1986. Arch. Immunol. Ther. Exp. (Warsz) 35, 569–583. Smith, G.P. (1985). Filamentous fusion phage: novel expression vectors that display cloned antigens on the virion surface. Science 228, 1315–1317. Smith, H.W., and Huggins, M.B. (1982). Successful treatment of experimental Escherichia coli infections in mice using phage: its general superiority over antibiotics. J. Gen. Microbiol. 128, 307–318. Smith, H.W., and Huggins, M.B. (1983). Effectiveness of phages in treating experimental Escherichia coli diarrhoea in calves, piglets and lambs. J. Gen. Microbiol. 129, 2659–2675.

|

39

40

|

McAuliffe et al.

Smith, H.W., Huggins, M.B., and Shaw, K.M. (1987). The control of experimental Escherichia coli diarrhoea in calves by means of bacteriophages. J. Gen. Microbiol. 133, 1111–1126. Sternberg, N., and Hoess, R.H. (1995). Display of peptides and proteins on the surface of bacteriophage lambda. Proc. Natl. Acad. Sci. USA 92, 1609–1613. Stevenson, R.M., and Airdrie, D.W. (1984). Isolation of Yersinia ruckeri Bacteriophages. Appl. Environ. Microbiol. 47, 1201–1205. Streit, W.R., and Schmitz, R.A. (2004). Metagenomics—the key to the uncultured microbes. Curr. Opin. Microbiol. 7, 492–498. Suarez, V.B., Quiberoni, A., Binetti, A.G., and Reinheimer, J.A. (2002). Thermophilic lactic acid bacteria phages isolated from Argentinian dairy industries. J. Food Prot. 65, 1597–1604. Sulakvelidze, A., Alavidze, Z., and Morris, J.G., Jr. (2001). Bacteriophage therapy. Antimicrob. Agents Chemother. 45, 649–659. Sulakvelidze, A., Kutter, E. (2005). Bacteriophage therapy in humans. In: Bacteriophages: Biology and Applications, pp. 381–436. Sullivan, M.B., Coleman, M.L., Weigele, P., Rohwer, F., and Chisholm, S.W. (2005). Three Prochlorococcus cyanophage genomes: signature features and ecological interpretations. PLoS Biol. 3, e144. Suthereland, I.W., Hughes, K.A., Skillman, L.C., and Tait, K. (2004). The interaction of phage and biofilms. FEMS Microbiol. Lett. 232, 1–6. Suttle, C.A. (2005). Viruses in the sea. Nature 437, 356–361. Thomas, J.A., Soddell, J.A., and Kurtboke, D.I. (2002). Fighting foam with phages? Water Sci. Technol. 46, 511–518. Tolkacheva, T.V., Abakumov, E.M., Golosova, T.V., and Volodina, N.I. (1989). [Bacteriotherapy of intestinal dysbacteriosis in patients with acute leukemia]. Gematol. Transfuziol. 34, 11–14. Torrella, F., and Morita, R.Y. (1979). Evidence by electron micrographs for a high incidence of bacteriophage particles in the waters of Yaquina Bay, oregon: ecological and taxonomical implications. Appl. Environ. Microbiol. 37, 774–778. Trollip, A., Albert, H., and Maskell, T. (2001). Bacteriophage-based technologies for the rapid diagnosis and drug susceptibility testing of tuberculosis. Am. Clin. Lab. 20, 39–42. Trotter, M., McAuliffe, O., Callanan, M., Edwards, R., Fitzgerald, G.F., Coffey, A., and Ross, R.P. (2006). Genome analysis of the obligately lytic bacteriophage 4268 of Lactococcus lactis provides insight into its adaptable nature. Gene 366, 189–199. Twort, F.W. (1915). An investigation on the nature of ultra-microscopic viruses. Lancet II, 1241–1243. Ubukata, K., Konno, M., and Fujii, R. (1975). Transduction of drug resistance to tetracycline, chloramphenicol, macrolides, lincomycin and clindamycin with phages induced from Streptococcus pyogenes. J. Antibiot. (Tokyo) 28, 681–688. Ulitzur, S., and Kuhn, J. (2000). Construction of lux bacteriophages and the determination of specific bacteria and their antibiotic sensitivities. Methods Enzymol. 305, 543–557. van Regenmortel, M.H. V., C.M. Fauquet, D.H. L. Bishop, E.B. Carstens, M.K. Estes, S.M. Lemon, J. Maniloff, M.A. Mayo, D.J. McGeoch, C.R. Pringle, R.B. Wickner (2000). Virus taxonomy: classification and nomenclature of viruses. Seventh report of the International Committee on Taxonomy of Viruses. Academic Press, San Diego, California. van Wamel, W.J., Rooijakkers, S.H., Ruyken, M., van Kessel, K.P., and van Strijp, J.A. (2006). The innate immune modulators staphylococcal complement inhibitor and chemotaxis inhibitory protein of Staphylococcus aureus are located on {beta}-hemolysin-converting bacteriophages. J. Bacteriol. 188, 1310–1315. Ventura, M., Canchaya, C., Kleerebezem, M., de Vos, W.M., Siezen, R.J., and Brussow, H. (2003). The prophage sequences of Lactobacillus plantarum strain WCFS1. Virology 316, 245–255. Wagenaar, J.A., Van Bergen, M.A., Mueller, M.A., Wassenaar, T.M., and Carlton, R.M. (2005). Phage therapy reduces Campylobacter jejuni colonization in broilers. Vet. Microbiol. 109, 275–283. Wagner, P.L., Livny, J., Neely, M.N., Acheson, D.W., Friedman, D.I., and Waldor, M.K. (2002). Bacteriophage control of Shiga toxin 1 production and release by Escherichia coli. Mol. Microbiol. 44, 957–970. Wagner, P.L., Neely, M.N., Zhang, X., Acheson, D.W., Waldor, M.K., and Friedman, D.I. (2001). Role for a phage promoter in Shiga toxin 2 expression from a pathogenic Escherichia coli strain. J. Bacteriol. 183, 2081–2085.

Phage Genomics and Applications

Waldor, M.K., and Mekalanos, J.J. (1996). Lysogenic conversion by a filamentous phage encoding cholera toxin. Science 272, 1910–1914. Watson, B.B., and Eveland, W.C. (1965). The application of the phage-fluorescent antiphage staining system in the specific identification of Listeria monocytogenes. I. Species specificity and immunofluorescent sensitivity of Listeria monocytogenes phage observed in smear preparations. J. Infect. Dis. 115, 363–369. Weeks, C.R., and Ferretti, J.J. (1984). The gene for type A streptococcal exotoxin (erythrogenic toxin) is located in bacteriophage T12. Infect. Immun. 46, 531–536. Weinbauer, M.G. (2004). Ecology of prokaryotic viruses. FEMS Microbiol. Rev. 28, 127–181. Willats, W.G. (2002). Phage display: practicalities and prospects. Plant Mol. Biol. 50, 837–854. Williamson, K.E., Radosevich, M., and Wommack, K.E. (2005). Abundance and diversity of viruses in six Delaware soils. Appl. Environ. Microbiol. 71, 3119–3125. Winter, G., Griffiths, A.D., Hawkins, R.E., and Hoogenboom, H.R. (1994). Making antibodies by phage display technology. Annu. Rev. Immunol. 12, 433–455. Withey, S., Cartmell, E., Avery, L.M., and Stephenson, T. (2005). Bacteriophages—potential for application in wastewater treatment processes. Sci. Total Environ. 339, 1–18. Wolber, P.K., and Green, R.L. (1990). Detection of bacteria by transduction of ice nucleation genes. Trends Biotechnol. 8, 276–279. Yamamoto, M., Kominato, Y., and Yamamoto, F. (1999). Phage display cDNA cloning of protein with carbohydrate affinity. Biochem. Biophys. Res. Commun. 255, 194–199. Yoon, S.S., Barrangou-Poueys, R., Breidt, F., Jr., Klaenhammer, T.R., and Fleming, H.P. (2002). Isolation and characterization of bacteriophages from fermenting sauerkraut. Appl. Environ. Microbiol. 68, 973–976. Zhang, T., Breitbart, M., Lee, W.H., Run, J.Q., Wei, C.L., Soh, S.W., Hibberd, M.L., Liu, E.T., Rohwer, F., and Ruan, Y. (2005). RNA Viral Community in Human Feces: Prevalence of Plant Pathogenic Viruses. PLoS Biol. 4, e3. Zhukov-Verezhnikov, N.N., Peremitina, L.D., Berillo, E.A., Komissarov, V.P., and Bardymov, V.M. (1978). [Therapeutic effect of bacteriophage preparations in the complex treatment of suppurative surgical diseases]. Sov. Med. 64–66.

|

41

Bacteriophage Bioinformatics and Genomics

2

Carlos A. Canchaya, Marco Ventura, and Douwe van Sinderen

Abstract Bacteriophages have played and continue to play a key role in bacterial genetics and molecular biology. Historically, they were used to define gene structure and gene regulation. Also the first genome to be sequenced was a bacteriophage. However, bacteriophage research did not lead the genomics revolution, which is clearly dominated by bacterial genomics. Only very recently has the study of bacteriophage genomes become prominent, thereby enabling researchers to understand the mechanisms underlying phage evolution. Bacteriophage genome sequences can be obtained through direct sequencing of isolated bacteriophages, but can also be derived as part of microbial genomes. Analysis of bacterial genomes has shown that a substantial amount of microbial DNA consists of prophage sequences and prophage-like elements. A detailed database mining of these sequences offers insights into the role of prophages in shaping the bacterial genome. Introduction Bacteriophages are present in all environments that are colonized by bacteria. In fact, taking the paradigm “all the world’s a phage” (Brussow and Hendrix, 2002) it is clear that phages are one of the most prominent “organisms” on earth in terms of sheer numbers. However, phages are not a homologous group and their classification is currently based on the nucleic acid content of their genomes or on their morphology, the diversity of phages is perhaps best reflected by their differing genome sizes. Almost all bacteriophages identified to date possess a tail and belong to the order Caudovirales, which is further divided on the basis of tail morphology into Myoviridae, Podoviridae and Siphoviridae. Bacteriophages belonging to the Myoviridae possess a long contractile tail, whereas those that are classified among the Siphoviridae and Podoviridae have a long and short non-contractile tail, respectively (Maniloff and Ackermann, 1998). However, while this kind of classification is now widely accepted, it has no meaning in terms of the evolutionary development of bacteriophages. With the advent of genomics, the number of sequenced genomes of bacterial viruses has increased exponentially. The first genome to be published was that of phage FX174 in 1977 (Sanger et al., 1977), followed by that of one of the most studied phages, phage lambda (L), in 1982 (Sanger et al., 1982). Now, thanks to the help of powerful sequencing technologies, such as “shotgun sequencing”, more than 300 phage genomes are available and this figure can be expected to be in the thousands in a few years time. Phage genomics

44

|

Canchaya et al.

have been very informative with respect to the biology of the individual phages (Casjens et al., 1992). However, it was the relatively recent results of high throughput sequencing technologies coupled to an ever expanding bioinformatics tool box, which for the first time allowed scientists to study and compare multiple bacteriophage sequences at the whole genome resolution, thereby addressing fundamental biological questions relating to phage population genetics and evolution. A detailed analysis of the phage gene content of currently available phage genomes makes it possible to subdivide their genomes into functional units as first stated in the modular theory of phage evolution (Botstein, 1980). According to that theory, phage genomes are mosaics of modules (groups of functionally related genes) that are free to recombine in genetic exchanges between distinct phages infecting the same cell. The minimal genome of Caudovirales consists of discrete gene modules responsible for DNA packaging, morphogenesis, DNA replication, transcription, regulation, and lysis. With increasing genome size, virion morphology appears to become more sophisticated while the ability of the phage to interfere with the host’s cellular activities also increases. However, the function of many phage genes is still not understood. Even in one of the best studied bacterial viruses such as the Myovirus T4, only about 130 of the estimated 230 genes have assigned functions. However, analysis of phage genomes goes beyond the study of their individual genes, Phage genome sequences have also demonstrated their usefulness in the investigation of phage evolution, thereby providing insights into phage classification and taxonomy. This has been possible due to comparative genome analysis, which highlights differences and similarities between individual genes, gene clusters or complete genomes, thus providing a complete picture of relatedness between an individual phage and phage groups (Rohwer and Edwards, 2002; Lawrence et al., 2002; Proux et al., 2002). There are an estimated > 1030 tailed phages in the biosphere, and since phage particles largely outnumber prokaryotic cells in environmental samples, the tailed phages constitute an absolute majority of “organisms” on the planet, having impact in various environmental processes such as the movement of nutrients and energy within ecosystems primarily by lysing bacteria (Fuhrman, 1999). Significant efforts have been made to study phage populations present in environmental samples, e.g. sea water, through large scale isolation of phage particles followed by a high throughput genome sequencing. This has lead to the establishment of so-called phage metagenomics, a subdiscipline of bacteriophage ecology that is devoted to the analysis of vast amounts of phage sequences. A lot of effort is currently placed on sequencing of phage genomes (at the time of writing more than 300 phage genome sequences have been completed and entered into the corresponding NCBI web page of phage genomes (Figure 2.1). The analysis of this enormous amount of genomic information was facilitated by significant advances in computational biology or bioinformatics. Interestingly, bioinformatic analyses of bacterial genomes revealed the presence of a myriad of phage sequences (which are assumed to represent complete or defective prophages). Several of these prophage sequences are predicted to play an important role in increasing the ecological fitness of their bacterial host, e.g. through the expression of properties that prevents further phage infections, i.e. superinfection exclusion or immunity genes, as well as genes that contribute to bacterial virulence, i.e. toxin genes.

Bacteriophage Bioinformatics and Genomics

A

B

350 300 250 200 150 100 50 0 90 91 92 93 94 95 96 97 98 99 00 01 02 03 04 05

Figure 2.1 (A) Number of available complete phage genomes in the Genbank database by year. The black line indicates the exponential growth trend of the number of sequences annotated Fig. 1database entries. (B) Distribution of the number of phage sequences grouped by as individual bacterial host they interact, expressed as percentage.

Bioinformatics as a means for the identification and the analysis of phage genome sequences Bioinformatics derives knowledge from computer analysis of biological data; it is a rapidly developing branch of biology and is highly interdisciplinary, using techniques and concepts from informatics, statistics, mathematics, chemistry, biochemistry and physics. It has many practical applications in different areas of biology, including phage biology. The origin and development of bioinformatics proceeded in parallel with the genomics discipline, where the amount of data generated by genome sequence projects was simply too large to be analyzed manually. Although phages were the first target for complete genome determination efforts due to their small size, it was only after the sequencing of multiple bacterial genomes that a larger number of phage sequences became available for comparative analyses (Canchaya et al., 2003). New prophage sequences are brought into light by bacterial genome sequencing projects, these bacteria constitute a big reservoir of phage sequences which has not been yet fully exploited. Most of the sequenced bacterial genomes possess one or more prophages. However their bioinformatics analysis is more complicated when compared to phages isolated in the lytic phase since the exact position and extent of the prophage within the bacterial genome has to be established first. This computational effort involves identification of phage related sequences, prediction of the phage integration site, prediction of gene sequences, prediction of phage gene functions and functional annotation (Figure 2.2). Once each component and putative function has been identified, comparative genomics and database mining will contribute to obtain further insights into their evolution, which may allow a discussion about their classification and taxonomy. Finally, with increasing complexity and quantity of data, metagenomics will introduce the community and population variables to generate a holistic picture of phage impact not only with regards to the bacterial host but also to the environment.

|

45

46

|

Canchaya et al.

Phage Bioinformatic Analysis Comparative Phage Genomics

DNA sequence ATGGATATTTTAGAATTATTTAGTCAGAATAAGAAAGTCCAATCCTGGCACTCTGGATT A ACCACCTTAGGAAGACAACTGGTAATGGGGTTATCGGGTTCAAGTAAAACATTGGCTAT A GCTTCCGCTTATTTAGATGATCAAAAAAAAATAGTTGTGGTTACATCAACTCAAAATGA G GTTGAAAAATTAGCCAGCGATTTATCTAGTTTACTTGATGAAGAACTTGTTTTCCAATT T TTTGCAGACGATGTGGCTGCAGCGGAATTTATCTTTGCGTCAATGGATAAAGCTCTATC A ATGGATATTTTAGAATTATTTAGTCAGAATAAGAAAGTCCAATCCTGGCACTCTGGATT A ACCACCTTAGGAAGACAACTGGTAATGGGGTTATCGGGTTCAAGTAAAACATTGGCTAT A GCTTCCGCTTATTTAGATGATCAAAAAAAAATAGTTGTGGTTACATCAACTCAAAATGA G GTTGAAAAATTAGCCAGCGATTTATCTAGTTTACTTGATGAAGAACTTGTTTTCCAATT T TTTGCAGACGATGTGGCTGCAGCGGAATTTATCTTTGCGTCAATGGATAAAGCTCTATC A AGAATAGAAACCCTGCAATTTTTAAGGAATCCTAAATCTCAGGGCGTTTTAATTGTTAG T TTATCAGGCTTAAGAATTTTATTGCCAAACCCAGATGTTTTTACAAAGAGTCAGATTCA A CTAACAGTTGGAGAAGATTATGATAGTGATACTCTTACTAAACAACTGATGACAATTGG C TATCAGAAGGTCTCACAGGTTATTAGTCCAGGAGAATTTAGCCGTCGAGGGGATATTTT A

Orf Prediction Algorithms

10000

20000

30000

Orf prediction

Pairwise comparison

seq DB

40000

Lysogeny module

Replication

Regulation

Head morphogenesis

Head-tail joining

Tail genes

Tail fiber genes

Host Lysogeny lysis conversion

Function Prediction

Fig. 2 2.2 Figure

Diagram showing a schematic pipeline for the Bacteriophage bioinformatic analysis starting from the raw DNA sequence. Different levels of analysis depend on the use of predicting tools and their interaction with specialized sequence databases. Although some of these steps are done automatically, human annotation becomes more significant in each step than in the previous one. Arrows indicate open reading frames (orf), filled arrows indicate that they belong to the same functional module. Areas linking arrows at the comparative genomics level indicate the relatedness of linked orfs.

Phage genome annotation Identification of prophage sequences Some attempts have been reported in literature to automatically identify prophage sequences in bacterial genomes. Some of these rely on the assumption that the phage sequence and other regions coming from horizontal gene transfer possess nucleotide signatures that are different from the bacterial host (Dufraigne et al., 2005; Tsirigos and Rigoutsos, 2005). Other methods rely on the annotation and clustered organization of genes that are predicted to be phage associated (Fournous, 2003; Leplae et al., 2004). However, a single methodology that combines all previously proven approaches to identify phages would be ideal for the automatic localization of prophages. Currently, the identification and subsequent annotation and classification of prophage sequences is partially a manual task, since there are no publicly available automatic prediction tools to suit this task. To meet this computational shortcoming, one would first have to propose an exact bioinformatic definition of a prophage. Large and continuous sections of bacterial sequence DNA matching phage genes are a clear indication of the presence of prophages. However the size of these prophage regions varies. The smallest prophage regions containing few genes seem to be produced by a continuous deleterious bacterial random process where survival of bacteria is not affected by the loss of some prophage portions. It has been postulated that this progressive loss of prophage DNA does not equally affect all the genes, where those genes

Bacteriophage Bioinformatics and Genomics

that confer a selective advantage to the cell are presumably not removed (Lawrence and Ochman, 1998). The reduction of the number of phage genes renders their identification and localization difficult. This progressive reduction may of course harm the viability of the prophage although it may not always be clear whether apparent phage remnants represent non-functional relics of previously fully functional prophages or whether they may still be functional with the help of other lysogenic or lytic phages. Prophage identification in bacterial genomes is mainly performed through the similarity of individual open reading frames (ORFs) to typical phage genes. For example, genes that encode DNA-interacting proteins involved in the establishment of the lysogenic state constitute the lysogenic module in prophage sequences. As a result of the integration process, this module flanks the DNA target with which it interacts. One of these genes in lambdoid phages, the integrase, is typically located adjacent to the one of the phage attachment sites, determining consequently either the leftmost or rightmost border of the integrated phage into the bacterial chromosome. Besides integrases, other genes considered “cornerstones” for prophage identification in bacterial genomes include those that encode cI repressors, virion assembly proteins and lysins (Canchaya et al., 2004; Casjens, 2003). Some of these genes, such as those that specify proteins involved in building the virion, are unique to phages, as bacterial cells are not known to make similar structures. In this respect, some virion assembly and structural phage proteins are more highly conserved than others, and homology of these genes can often be recognized between phage types (Casjens, 2003). These include: (a) the larger of the two subunits of terminase, which is cleaving virionlength molecules from concatemeric phage DNA; (b) the portal protein, which constitutes the gateway through which DNA is packaged into the capsid; (c) the head maturation proteases, responsible for the proteolytic cleavage of specific structural proteins during head assembly; (d) the tail tape measure proteins, which is an essential component in the initiating complex for tail assembly, while also responsible for determining the length of the tail shaft; and (e) tail fibers and tail tip proteins which serve as contact points between the virion and bacteria. Moreover, a number of amino acid domains, such as the collagen Gly X-Y repeat, commonly present in tail fibers (Smith et al., 1998) can be useful for detecting prophage sequences. Gene prediction One of the first steps in phage sequence analysis is gene prediction. The accuracy of this process is crucial for correct functional annotation, which in turn will provide further information about modular organization and biology of phages. In the early days of sequencing gene prediction was performed manually, but in the last ten years this process has taken advantage of diverse bioinformatic tools. Prokaryotic methods of gene prediction are classified into three different groups according to the method they use. The first group corresponds to bioinformatic tools that use an intrinsic method (ab initio), usually based on Markov models, depending uniquely on the DNA sequence for the prediction. Glimmer (Salzberg et al., 1998) is a very popular software tool that uses an algorithm based on this intrinsic method. It can predict up to 99% of the genes present in a bacterial sequence in a fully automated fashion. Glimmer is based on a special kind of Markov chain called Interpolated Markov Models (IMM), to identify the coding regions and distinguish them from non-

|

47

48

|

Canchaya et al.

coding DNA. IMMs determine the probability a base belongs to a coding region based on the probabilities of preceding bases in a dynamic manner, by following an initial model built with known genes coming from the same organism Another important program belonging to this group is FrameD (Schiex et al., 2003), which predicts genes more accurately in high-GC DNA sequences. The second group corresponds to gene predictors based on extrinsic methods, where extensive comparisons against DNA and protein databases are the source to determine gene positions, among them ORPHEUS (Frishman et al., 1998) and CRITICA (Badger and Olsen, 1999). The third group corresponds to methods that use a combination of extrinsic and intrinsic methods to increase the accuracy of the gene prediction process as for example YACOP (Tech and Merkl, 2003). Bacterial gene prediction algorithms based on Markov models are “trained” with gene sequences belonging to the same bacterial species. Due to the dependency of the phage multiplication process on the host’s replication and protein synthesizing machinery, it is assumed that the gene prediction tools used for the host can also be used for their infecting phages, irrespective as to whether they are lysogenic or lytic. For example, Markov models trained with Lactobacillus plantarum were used to predict the genes of its lytic phage LP65. (Chibani-Chennoufi et al., 2004) Special care should be taken when considering the size of prophage genes. The size of phage genomes must be optimized to fit into the physically limited capsid, while the prophage genome should also not constitute a replication burden to the host since it will otherwise constitute a selective pressure that favors its removal. Due to these size constraints, the expected minimum phage gene size will be shorter than that of the host and this should be a parameter to be incorporated in the gene predicting algorithm. Two additional consequences of this selective pressure towards reduced gene size, i.e. gene overlap and translational frameshifting, are known to be present in phages. The phenomenon of gene overlap represents a subtle strategy to condense the maximum amount of genetic information into a given nucleotide sequence. Gene overlap seems to be more prevalent in small phages as those of the family Microviridae (Pavesi, 2006), although it has also been found in larger phages such as bacteriophage T4 (Zajanckauskaite et al., 1997) Programed translational frameshifting (Farabaugh, 1996) is a feature that is present in one of the best studied phages, L. It consists in the coding of two different proteins from the same gene sequence due to the tandem slippage of the tRNAs located at the ribosome during translation. There are some bioinformatic tools to predict a translational frameshift, and one specifically developed for bacteriophages has recently been developed (Xu et al., 2004) The presence of phage-encoded tRNA genes, although initially identified in lytic phages, have been increasingly found in lysogenic phages. Incidence and function of tRNA genes inside phage genomes has not been extensively studied; but their presence might indicate that prophages contribute in some way with the host translation process since it has been found that they are transcribed (Ventura et al., 2004). To predict the presence of tRNA genes in prophages the most used and robust tRNA prediction tool is tRNA-SE (Lowe and Eddy, 1997). Probably due to their conserved sequence, tRNA-encoding genes also represent the phage’s preferred insertion site into the bacterial chromosome, in some

Bacteriophage Bioinformatics and Genomics

cases disrupting the target tRNA gene sequence. Prophages are flanked by so-called attachment sites, two homologous sequences that are the targets of an integrase-mediated recombination event between the phage and the bacterial chromosome in order to allow integration (and excision). Identification of these repeat sequences at each border of the prophage thus indicates the exact position of prophage integration into the host genome. Functional annotation One of the fundamental steps in phage genome analysis is accurate function assignation to predicted genes. This function assignation will be very important in a subsequent step to extract further information relating to modular organization of specific phage functions (e.g. packaging, replication), which is determined based on the individual elements that together constitute its genomic gene repertoire and the comparison of these with other phages. Functional annotation can be defined as the process by which structural or functional information is inferred for genes or proteins, usually on the basis of observed similarities to previously characterized sequences in public databases (Ouzounis, 2002). A valid bioinformatics approach for functional gene assignment can start with the Basic Local Alignment Search Tool (BLAST) (Altschul et al., 1990) and FASTA (Pearson and Lipman, 1988). These two heuristic algorithms produce alignments that are not optimal as the ones produced by a dynamic programming algorithm such as the Smith-Waterman algorithm, however they succeed in quickly identifying relationships when scores are higher than 30%. As a rule of thumb for functional assignments based on BLAST alignments, the following assignment criteria can be used. When two sequences exhibit 40% or higher sequence identity, functional assignment is performed with high confidence, whereas alignments that exhibit identity levels between 35–20% cannot be relied upon, and this range is referred to as the twilight zone (Rost, 1999). However, in the case of “twilight zone” similarities found between particular phage gene and other phage sequences, the derived function can be assigned with high confidence if these sequences correspond to equivalent positions in a modularly organized phage genome. Despite the growing size of DNA and protein databases, the available sequences are probably not representative of the existing gene pool in nature, which is also true for phage sequences. However sequencing of new bacteria and phage and bacterial metagenomics analyses are producing databases with phage sequences more diverse than ever before, particularly when many phage sequences had originated from the dairy industry. Analysis of viral metagenomes from different environments have been feeding databases with new DNA and protein sequences. Such phage metagenomics projects combined with prophage sequences coming from bacterial genome projects will undoubtedly represent important sources of phage sequences for future discoveries in phage genomics. Further information can be extracted when no sequence based similarity is found. Searches against databases containing information based on secondary structure information (e.g. motifs and domains) will supply information on structure and possible function of phage proteins. The most frequently used database search tools for motifs are Pfam (Sonnhammer et al., 1998) PIR (George et al., 1996), PRINTS (Attwood and Beck, 1994) and Prosite (Bairoch, 1991). Most of these tools can be accessed through the program InterProScan (Quevillon et al., 2005).

|

49

50

|

Canchaya et al.

Genes in phages are not randomly distributed but they usually display a conserved functional order and cluster together into functional modules (see below). Thus, this may be helpful to predict gene function in cases where no clear database matches have been detected. In fact, the presence of anonymous genes within a specific functional phage module suggests that these genes may be involved in a similar function as its flanking genes. Genome analysis of prophages has shown that the order of these functional modules is widely conserved in bacteriophages, especially for those following the lambdoid structure. Recognized functional modules are those involved in lysogeny, replication, transcriptional regulation, head morphogenesis, head-to-tail joining, tail morphogenesis, tail fiber assembly and host lysis. In silico comparative analysis of the gene contents of various phage genomes reveal that this structure and order is well conserved among lambdoid phages. Only small differences, i.e. concerning the lysogeny-module, can be detected between phages with different host-lifestyle (e.g. virulent or temperate phages). Phage comparative genomics Analysis of phage genomes has highlighted the diversity existing between phages, but has on the other hand also revealed remarkable underlying similarities. A significant challenge is to interpret such observations and to use these to infer ancient and contemporary mechanisms of bacteriophage evolution. As mentioned above bacteriophage genomes are highly variable, in both gene size and content. Hence, Botstein (Botstein, 1980) proposed a model of phage evolution that could account for this surprising genetic variability of bacteriophages. According to this model, also known as the “modular theory of phage evolution,” new phage types are readily created by recombination processes. This theory was formulated on the basis of experimental data from lambda phage system, where homologous and heterologous genome segments are interspersed (Campbell, 1983). The essence of Botstein’s theory is that the product of evolution is not a given virus but a family of interchangeable genetic elements (modules), each of which carries out a particular biological function (Figure 2.3). Exchange of a given module for another occurs by recombination among viruses belonging to the same interbreeding population; and finally, these viruses and their hosts can differ widely in many characteristics. Genome comparison between lambdoid phages of enteric bacteria revealed that the genomes, which share a similar gene organization, are mosaic with respect to each other. Points of recombination, which are represented by short sequence of homology, have been identified at the extremities of gene modules in lambdoid coliphages (Clark et al., 2001). These linker sequences could promote genetic reassortment, i.e. modular exchanges, through homologous or site-specific recombination. An alternative manner of genetic recombination is represented by non-homologous recombination, which occurs indiscriminately across the phage genome, followed by selection for functional phages ( Juhala et al., 2000). In the majority of cases this will mean that the new hybrid phages that arose as a result of non-homologous recombination events within coding regions will be eliminated. A remarkable number of genomes have been sequenced from phages of different bacteria. However only few bacterial genera constitute about 70% of the hosts for those sequenced phages (Figure 2.1B). The bias towards these small set of bacterial genera will be reduced when a higher number of genome sequences of bacteriophages, isolated from

Bacteriophage Bioinformatics and Genomics

Primase

Helicase

Cro-like repressor

Holin Holin Lysin

Minor tail protein

Minor tail protein

Major tail protein

Major head protein

Portal protein ClpP protease

c os

Terminase

S. thermophilus Sfi19 phage

R

Packaging

Head Head-tail morphogenesis joining

Tail morphogenesis

Tail fiber proteins

Host lysis

Lysogeny replacement module

Ori

R

Replication

Transcriptional regulation

Packaging Head Head-tail morphogenesis joining

Tail morphogenesis 10 00 0

Tail fiber proteins

Lysogeny Host lysis Lysogenic module conversion 20000

Primase

Helicase

Integrase Sie cI repressor Cro repressor Antirepressor

Holin Holin Lysin

Minor tail protein

Minor tail protein

Major tail protein

Portal protein ClpP protease Major head protein

c os

Terminase

S. thermophilus Sfi21 phage

Ori

Replication

30000

Transcriptional regulation 37000

Fig. 3

Figure 2.3 Genome map of a typical virulent bacteriophage (S. thermophilus Sfi19) and a typical temperate bacteriophage (S. thermophilus Sfi21). Each arrow indicates an open reading frame (ORF). The modular structure of the genomes is indicated. Functions of the encoded proteins are noted below the ORFs.

a broader spectrum of environments, becomes available. Comparative genomic analyses within phages of mycobacteria and LAB demonstrate that they largely resemble the lambdoid group in their mosaic relationships, arguing for comparable mechanisms of gene exchange (Lucchini et al., 1999; Brussow, 2001; Ford et al., 1998). The conservation of the gene order in a genome or in a region within a genome, e.g. a phage-module, has been called synteny. The synteny argument has become a popular tool in phage genomics and suggests either very ancient divergence or rather impressive evolutionary convergence. However, the hypothesis of very ancient divergence, which implicates the existence of a common ancestor, has obtained significant support. This is based on more sensitive search algorithms that display previously invisible protein sequence relationships as well as the possibility of linking two very distantly related sequences by chains of similar sequences from other phages. On the other hand, related gene sequences are sometimes found between very distinct phages, suggesting that horizontal exchanges of sequences do also occur across the phage world, even if with reduced frequency (Hendrix et al., 1999). Examples of horizontal gene transfer across phages has been suggested for some early genes of P2-like phages and those of some lactic acid bacteriophages placed within the Myoand Siphoviridae family (Brussow and Hendrix, 2002). So the evolutionary development followed by phage seems to indicate that is not possible to represent phage history with a simple branching phylogeny.

|

51

52

|

Canchaya et al.

Prophage genomics Bacteriophages are strictly dependent on bacteria for replication and conservation of their genomes. The peculiar lifestyle of temperate phages, with alternating phases of lytic growth and prophage existence, subjects phage genomes to bacterial selective pressures. The sequencing of bacterial genomes has revealed a myriad of prophage sequences, which has become one of the major focal points for biologists interested in the evolution of parasitic DNA. Fully functional prophages are capable of initiating a round of lytic growth; however, not all prophage-like elements in bacterial genomes appear to represent functional bacteriophages. Other types of prophage-like elements have been characterized so far from bacterial genome screening. These include: (a) defective prophages (also mentioned as cryptic prophages or prophage remnants), which are prophages that are in a state of mutational decay and even if they still harbor functional genes, defective prophages are unable to program the full phage replication program (Campbell 1994); (b) satellite phages which are otherwise functional phages that do not carry their own virion structural encoding protein genes, with their chromosomes encapsidated by other virion proteins of specific helper phages. Finally the last type of prophage-like elements are the gene transfer agents (GTAs), which are tailed phage-like particles that package random fragments of the bacterial genome (Humphrey et al., 1997; Casjens, 2003). Analysis of prophage genomes has indicated the presence of both selfish and mutualistic properties with respect to their host (Canchaya et al., 2003; Casjens, 2003). It was found that prophage sequences may significantly contribute to the pathogenicity of their bacterial hosts. This is highlighted by the finding that a number of well-known bacterial toxins, such as the cholera toxin, botulinum toxin and salmonella toxin, are encoded by phages. Prophages are the major source of horizontally transferred DNA in bacteria, and thus contribute significantly to bacterial inter-strain variability. In bacterial genomes, 3% to 16% of the DNA is contributed by prophages. A striking example of this is the genome of the Streptococcus pyogenes strain SF370, in which prophages represent up to 16% of the DNA content of the bacterial chromosome (Ferretti et al., 2001). Another amazing example is represented by the emerging food pathogen Escherichia coli O157:H7, whose genome sequence contains 18 prophages or prophage remnants. Prophage DNA accounts for half of the 1.3 Mb of DNA found in O157:H7 strain but absent from the reference strain E. coli K-12 (Ohnishi et al., 2001). Unfortunately, the current database of bacterial genome sequences is biased towards genomes of pathogenic bacteria. Consequently, detailed analyses of prophages from sequenced non-pathogenic bacteria are still limited (Canchaya et al., 2003). Genome analysis of Lactococcus lactis IL1403 revealed three cos-site prophages and three prophage remnants (Chopin et al., 2001) (Figure 2.3). In Lactobacillus johnsonii NCC 533 DNA microarray analysis indicated that prophage DNA represents almost half of identified strain-specific DNA (Ventura et al., 2003b). Recently, it was shown that the Lactobacillus plantarum WCFS1 genome contains four prophage elements that share closely related genes encoding structural proteins (Ventura et al., 2003a). Finally, the genome sequence analysis of a human oral commensal such as Lactobacillus salivarius subsp. salivarius, revealed the presence of four prophages, two of which appear to represent com-

Bacteriophage Bioinformatics and Genomics

plete bacteriophage, whereas the other two seem to constitute phage remnants (Ventura et al., 2006). In view of the fact that prophages seem to be frequently present in genomes of this group of bacteria, one might suspect that prophages contribute to the evolutionary success of lactic acid bacteria living in strikingly distinct environments. Prophage-like elements and prophage remnants have been identified in almost all bacterial genomes sequenced so far (Canchaya et al., 2003), suggesting that this group of mobile elements are widespread in bacteria and may be considered to represent a useful tool in order to investigate bacterial evolution. Recently, the identification of prophage-like elements in genomes of bifidobacteria, which were not known to be infected by bacteriophages, indicated that prophage sequences could be used to investigate genome bacterial evolution (Ventura et al., 2005). The increasing number of available bacterial genome sequences has contributed to the understanding of prophage genome distribution and evolution. The mosaic pattern and localized diversity of many different prophage genomes is obvious from comparative analyses of prophage genome content and organization, as well as similarities of orthologous gene products encoded by these elements. Functional analyses performed on many prophages revealed that most of them have lost the capacity to produce active phage particles. Most prophages have suffered massive deletion events, which inactivated or eliminated structural protein-encoding genes or genes that are required to effect bacterial lysis. A high deletion rate may be a bacterial defense against a high rate of DNA influx by potentially dangerous foreign DNA elements. This “cleaning” activity could account for the relatively compact and pseudogene-free nature of many bacterial genomes (Lawrence et al., 2001). Moreover, homologous recombination between prophages sharing some DNA homology and residing in the same host would give rise to deleterious host genome rearrangements or even deletions. Phage phylogeny and taxonomy For cellular organisms, phylogenetic and taxonomic relationships can be derived from universal conserved genes, i.e. molecular markers, such as rRNA genes, recA, tufA, and groEL genes. However, this technique cannot be applied to phage sequences, because there is no single sequence that is ubiquitously present in all phage genomes (Rohwer and Edwards, 2002). Official viral classification is based on characteristics of the virions and host range, but no sequence data have been considered or endorsed (Büchen-Osmond, 2006). However, a sequence-based taxonomic system for inferring phylogeny among phages and prophages by means of proteomic tree constructions, has been recently proposed (Rohwer and Edwards, 2002). This system is based on an algorithm that uses every gene in every phage or prophage genome in order to determine an average distance between pairs of phages. The proteomic tree analysis may thus be a useful means to infer phage evolution despite the fact that as yet it is not a molecular taxonomical tool that has been fully accepted by the International Committee on the Taxonomy of Viruses (ICTV). Furthermore, the additional genomes that were incorporated from the time of the development of this proteomic tree approach have not changed the overall phylogenetic image that was proposed in

|

53

54

|

Canchaya et al.

the first version of the tree (Rohwer and Edwards, 2002). This suggests that the proteomic tree is providing a reliable mean to infer phylogeny among bacteriophages. Another very common sequence-based approach for phage taxonomy is the use of a single gene locus, such as capsid or DNA polymerase gene, to characterize a specific group. PCR-primers can be designed to specifically amplify these genes, and the diversity of specific genes in a particular environment can be assessed by cloning and sequencing of PCR products amplified directly from environmental samples. Metagenomics and bacteriophages The identification and evaluation of viral community dynamics in the environment is a very complicated task. In fact, less than one percent of microbial hosts have been cultivated. Research in metagenomics, the study of environmental population genomics (Streit and Schmitz, 2004), of uncultivated viral communities circumvent this limitation and can provide insights into their composition. The metagenomics of viruses started in 2002 with the publication of two uncultured marine viral communities, i.e. a marine water sample and a marine sediment sample (Steward, 2000; Wommack et al., 1999), and has recently been reinforced with the publication of two other viral metagenomic libraries from human fecal samples (Breitbart et al., 2003) and equine fecal samples (Cann et al., 2005). One of the crucial steps in the preparation of viral community libraries for metagenomic analyses is the isolation of phage DNA. In fact, the presence of free and cellular DNA, which is about 50 times bigger than the average of viral DNA may overwhelm the viral signal. Generally a combination of differential filtration, DNase treatment and density centrifugation in cesium chloride is used to separate intact phage particles from bacteria and free DNA. Once intact virions have been isolated, the viral DNA is extracted and cloned. Thus the achieved viral library is then subjected to DNA sequence analysis. Genomic analyses of the previously mentioned viral metagenomic libraries show that around 60% of the new phage sequences have no similarity to any sequence in GenBank. This finding highlights the fact that a large bacteriophage variability exists on the Earth and that only a minor part of the existing bacteriophage have been so far characterized. In addition to cataloging the phage composition, viral metagenomics makes it possible to reconstruct the structure of uncultured viral communities. The fecal and marine-water viral communities each contained a different dominant viral genotype that constituted at least 1% of total community, whereas the most dominant virus in the marine-sediment community made up less than 0.01% of total community. This suggests that marine sediment communities are the most diverse biological systems characterized to date (Breitbart et al., 2004). Comparison of the available viral metagenomes using the phage proteomic tree approach (see above) suggests that Siphophage fragments are the most common DNA sequences observed in such environments (Edwards and Rohwer, 2005). Notably, Siphophage represent 44% of phage sequences of the marine-sediment metagenome. Remarkably, oceanic sediments contain the largest microbial biomass existing on the planet (Whitman et al., 1998). Viruses are abundant also in these environments (Danovaro and Serresi, 2000), which suggests that Siphophages represent one of the most abundant genome arrangement on Earth (Edwards and Rohwer, 2005). The application of wholegenome taxonomy systems to metagenome libraries enable investigators to determine if any

Bacteriophage Bioinformatics and Genomics

particular bacteriophage taxonomic group is preferentially associated with one environment or another (Martin, 2002). Analysis of metagenomes requires intensive bioinformatic efforts. In particular, the inclusion of large numbers of unrelated genomes in the sample and the presence of repeated sequences, such as insertion elements and transposons, require extensive computational inputs. These problems may be overcome with different assembly algorithms, longer sequence reads and higher sequence coverage of such environmental samples. An existing limitation of bioinformatics for metagenomics is represented by the fact that all comparisons between metagenomic libraries are currently performed using sequence-similarity algorithms such as BLAST and FASTA. However, these metagenomic analyses are affected by the fact that most of the phage sequences have no recognizable similarity to any GenBank database entitles. Future analyses of metagenomic sequences should therefore also include G+C/A+T content, codon usage and oligomer skews using different-sized sequence strings such as dinucleotide, trinucleotide. Sequence skews will be useful to associate viral sequences to their bacterial host or for tracing phage-cluster of phylogeny. Outlook The origin and development of bioinformatics proceeded in parallel with the genomics discipline, to cope with the analysis of the enormous amount of data derived from genome projects. Furthermore, going to a higher level of complexity, the number of phage sequences is going to be dramatically increased with sequences resulting from metagenomic analysis of phages and bacterial communities isolated from different environments. Phage biologists are also beginning to explore methods to sample phage sequences from environmental sources without the necessity to grow them on culturable bacteria. These surveys will provide the raw data necessary for understanding the size of viral metagenome and community structure. The initial bias of the pool of phage sequences toward sequences coming from pathogens and those important for the food industry will be reduced producing a more robust picture of the phage gene reservoir. In addition, new developments in internet technologies, software, algorithms, and computer cluster technology will continue to enable bioinformatics to make great leaps in terms of the amount of data which can be efficiently analyzed. This computer power together with the availability of larger databases will constitute the groundwork for wet biologists. The synergistic work between “wet” and in silico research will bring to us a better understanding of the role of phages in nature and their impact to the bacterial world. Apart from direct phage genome sequencing and metagenomic investigations, another way to explore phage genome variability is represented by bacterial genomes. Bacterial genome sequencing projects represent an important source of bacteriophage sequences. They have already proved their usefulness in identifying novel phage sequences also in bacterial group such as bifidobacteria, which have been so far considered to be free from phage infection. The increasing number of phage sequences coming from phage sequencing projects and those of prophages coming from bacterial genomes will increase the understanding of phage biology and will provide a deeper insight into the relationship of phages with their hosts. Prophages are an important source of gene variability; for some bacteria they

|

55

56

|

Canchaya et al.

are the main contributors of intraspecific variability. Furthermore, phages can play a role as hotspots of recombination altering the bacterial chromosome architecture. Phage sequences coming from bacteria of diverse environments will increase the knowledge of the prophage diversity, clarifying or challenging the definition of prophages and their impact in the bacterial chromosome. The importance of phages as vehicles of horizontal gene transfer in bacteria has already been studied in bacterial pathogens, where phages seem to contribute to the bacterial shortterm evolution bringing genes that play an important role in their pathogenicity. The nature and relevance of this contribution will be determined for non-pathogenic bacteria when new sequences coming from diverse ecological niches become available. This will enrich the pool of known phage genes not only in number but it is also expected to contribute with novel functions. Novel bioinformatic tools adapted to analyze phage sequences will be needed to automatically identify new genes and annotate those with no phage-related functions. In vitro and in vivo experiments with these genes will probably shed light on the manner in which prophages modify the fitness of their hosts to their corresponding environments. The investigation of bacteriophage genome variability represents a crucial step in understanding the evolutionary development of phages. Moreover, through comparative genomics as well as genome sequence-based methods, novel schemes of bacteriophage taxonomy will be introduced that will be based on genome sequence relationship rather then the classical phage morphology schemes. References

Altschul, S.F., Gish, W., Miller, W., Myers, E.W., and Lipman, D.J. (1990). Basic local alignment search tool. J. Mol. Biol. 215, 403–410. Attwood, T.K., and Beck, M.E. (1994). PRINTS—a protein motif fingerprint database. Protein Eng. 7, 841–848. Badger, J.H., and Olsen, G.J. (1999). CRITICA: coding region identification tool invoking comparative analysis. Mol. Biol. Evol. 16, 512–524. Bairoch, A. (1991). PROSITE: a dictionary of sites and patterns in proteins. Nucleic Acids Res. 19 Suppl, 2241–2245. Botstein, D. (1980). A theory of modular evolution for bacteriophages. Ann. NY Acad. Sci. 354, 484– 490. Breitbart, M., Felts, B., Kelley, S., Mahaffy, J.M., Nulton, J., Salamon, P., and Rohwer, F. (2004). Diversity and population structure of a near-shore marine-sediment viral community. Proc. Biol. Sci. 271, 565–574. Breitbart, M., Hewson, I., Felts, B., Mahaffy, J.M., Nulton, J., Salamon, P., and Rohwer, F. (2003). Metagenomic analyses of an uncultured viral community from human feces. J. Bacteriol. 185, 6220– 6223. Brussow, H. (2001). Phages of dairy bacteria. Annu. Rev. Microbiol. 55, 283–303. Brussow, H., and Hendrix, R.W. (2002). Phage genomics: small is beautiful. Cell 108, 13–16. Büchen-Osmond, C. (2006). Index to ICTVdB virus descriptions. (ICTVdB Management, Mailman School of Public Health, Columbia University, New York, NY, USA). Campbell, A.M. (1996). Cryptic prophages. In Escherichia coli and Salmonella: cellular and molecular biology, 2nd ed, F.C. Neidhardt, R. Curtis III, J.L. Ingraham, E.C.C. Lin, K.B. Low, B. Magasarik, W.S. Reznikoff, M. Riley, M. Schaechter, and H.E. Umbarger, eds. Washington, DC: ASM Press, pp. 2041–2046. Canchaya, C., Fournous, G., and Brussow, H. (2004). The impact of prophages on bacterial chromosomes. Mol. Microbiol. 53, 9–18.

Bacteriophage Bioinformatics and Genomics

Canchaya, C., Proux, C., Fournous, G., Bruttin, A., and Brussow, H. (2003). Prophage genomics. Microbiol. Mol. Biol. Rev. 67, 238–276. Cann, A.J., Fandrich, S.E., and Heaphy, S. (2005). Analysis of the virus population present in equine faeces indicates the presence of hundreds of uncharacterized virus genomes. Virus Genes 30, 151–156. Casjens, S. (2003). Prophages and bacterial genomics: what have we learned so far? Mol. Microbiol. 49, 277–300. Casjens, S., Wyckoff, E., Hayden, M., Sampson, L., Eppler, K., Randall, S., Moreno, E.T., and Serwer, P. (1992). Bacteriophage P22 portal protein is part of the gauge that regulates packing density of intravirion DNA. J. Mol. Biol. 224, 1055–1074. Chibani-Chennoufi, S., Dillmann, M.L., Marvin-Guy, L., Rami-Shojaei, S., and Brussow, H. (2004). Lactobacillus plantarum bacteriophage LP65: a new member of the SPO1-like genus of the family Myoviridae. J. Bacteriol. 186, 7069–7083. Chopin, A., Bolotin, A., Sorokin, A., Ehrlich, S.D., and Chopin, M. (2001). Analysis of six prophages in Lactococcus lactis IL1403: different genetic structure of temperate and virulent phage populations. Nucleic Acids Res. 29, 644–651. Clark, A.J., Inwood, W., Cloutier, T., and Dhillon, T.S. (2001). Nucleotide sequence of coliphage HK620 and the evolution of lambdoid phages. J. Mol. Biol. 311, 657–679. Danovaro, R., and Serresi, M. (2000). Viral density and virus-to-bacterium ratio in deep-sea sediments of the Eastern Mediterranean. Appl. Environ. Microbiol. 66, 1857–1861. Dufraigne, C., Fertil, B., Lespinats, S., Giron, A., and Deschavanne, P. (2005). Detection and characterization of horizontal transfers in prokaryotes using genomic signature. Nucleic Acids Res. 33, e6. Edwards, R.A., and Rohwer, F. (2005). Viral metagenomics. Nat. Rev. Microbiol. 3, 504–510. Farabaugh, P.J. (1996). Programmed translational frameshifting. Annu. Rev. Genet. 30, 507–528. Ferretti, J.J., McShan, W.M., Ajdic, D., Savic, D.J., Savic, G., Lyon, K., Primeaux, C., Sezate, S., Suvorov, A.N., Kenton, S., et al. (2001). Complete genome sequence of an M1 strain of Streptococcus pyogenes. Proc. Natl Acad. Sci. USA 98, 4658–4663. Ford, M.E., Sarkis, G.J., Belanger, A.E., Hendrix, R.W., and Hatfull, G.F. (1998). Genome structure of mycobacteriophage D29: implications for phage evolution. J. Mol. Biol. 279, 143–164. Fournous, G. (2003). Automatisation de la recherche et de l’analyse de prophages dans les génomes bactériens, Travail de DEA, Université Henri Poincaré, Nancy. Frishman, D., Mironov, A., Mewes, H.W., and Gelfand, M. (1998). Combining diverse evidence for gene recognition in completely sequenced bacterial genomes. Nucleic Acids Res. 26, 2941–2947. Fuhrman, J.A. (1999). Marine viruses and their biogeochemical and ecological effects. Nature 399, 541–548. George, D.G., Barker, W.C., Mewes, H.W., Pfeiffer, F., and Tsugita, A. (1996). The PIR-International Protein Sequence Database. Nucleic Acids Res. 24, 17–20. Hendrix, R.W., Smith, M.C., Burns, R.N., Ford, M.E., and Hatfull, G.F. (1999). Evolutionary relationships among diverse bacteriophages and prophages: all the world’s a phage. Proc. Natl Acad. Sci. USA 96, 2192–2197. Humphrey, S.B., Stanton, T.B., Jensen, N.S., and Zuerner, R.L. (1997). Purification and characterization of VSH-1, a generalized transducing bacteriophage of Serpulina hyodysenteriae. J. Bacteriol. 179, 323–329. Juhala, R.J., Ford, M.E., Duda, R.L., Youlton, A., Hatfull, G.F., and Hendrix, R.W. (2000). Genomic sequences of bacteriophages HK97 and HK022: pervasive genetic mosaicism in the lambdoid bacteriophages. J. Mol. Biol. 299, 27–51. Lawrence, J.G., Hatfull, G.F., and Hendrix, R.W. (2002). Imbroglios of viral taxonomy: genetic exchange and failings of phenetic approaches. J. Bacteriol. 184, 4891–4905. Lawrence, J.G., Hendrix, R.W., and Casjens, S. (2001). Where are the pseudogenes in bacterial genomes? Trends Microbiol. 9, 535–540. Lawrence, J.G., and Ochman, H. (1998). Molecular archaeology of the Escherichia coli genome. Proc. Natl Acad. Sci. USA 95, 9413–9417. Leplae, R., Hebrant, A., Wodak, S.J., and Toussaint, A. (2004). ACLAME: a CLAssification of Mobile genetic Elements. Nucleic Acids Res. 32, D45–49. Lowe, T.M., and Eddy, S.R. (1997). tRNAscan-SE: a program for improved detection of transfer RNA genes in genomic sequence. Nucleic Acids Res. 25, 955–964.

|

57

58

|

Canchaya et al.

Lucchini, S., Desiere, F., and Brussow, H. (1999). Comparative genomics of Streptococcus thermophilus phage species supports a modular evolution theory. J. Virol. 73, 8647–8656. Maniloff, J., and Ackermann, H.W. (1998). Taxonomy of bacterial viruses: establishment of tailed virus genera and the order Caudovirales. Arch. Virol. 143, 2051–2063. Martin, A.P. (2002). Phylogenetic approaches for describing and comparing the diversity of microbial communities. Appl. Environ. Microbiol. 68, 3673–3682. Ohnishi, M., Kurokawa, K., and Hayashi, T. (2001). Diversification of Escherichia coli genomes: are bacteriophages the major contributors? Trends Microbiol. 9, 481–485. Ouzounis, C. (2002). Bioinformatics and the theoretical foundations of molecular biology. Bioinformatics 18, 377–378. Pavesi, A. (2006). Origin and evolution of overlapping genes in the family Microviridae. J. Gen. Virol. 87, 1013–1017. Pearson, W.R., and Lipman, D.J. (1988). Improved tools for biological sequence comparison. Proc. Natl Acad. Sci. USA 85, 2444–2448. Proux, C., van Sinderen, D., Suarez, J., Garcia, P., Ladero, V., Fitzgerald, G.F., Desiere, F., and Brussow, H. (2002). The dilemma of phage taxonomy illustrated by comparative genomics of Sfi21-like Siphoviridae in lactic acid bacteria. J. Bacteriol. 184, 6026–6036. Quevillon, E., Silventoinen, V., Pillai, S., Harte, N., Mulder, N., Apweiler, R., and Lopez, R. (2005). InterProScan: protein domains identifier. Nucleic Acids Res. 33, W116–120. Rohwer, F., and Edwards, R. (2002). The Phage Proteomic Tree: a genome-based taxonomy for phage. J. Bacteriol. 184, 4529–4535. Rost, B. (1999). Twilight zone of protein sequence alignments. Protein Eng. 12, 85–94. Salzberg, S.L., Delcher, A.L., Kasif, S., and White, O. (1998). Microbial gene identification using interpolated Markov models. Nucleic Acids Res. 26, 544–548. Sanger, F., Air, G.M., Barrell, B.G., Brown, N.L., Coulson, A.R., Fiddes, C.A., Hutchison, C.A., Slocombe, P.M., and Smith, M. (1977). Nucleotide sequence of bacteriophage phi X174 DNA. Nature 265, 687–695. Sanger, F., Coulson, A.R., Hong, G.F., Hill, D.F., and Petersen, G.B. (1982). Nucleotide sequence of bacteriophage lambda DNA. J. Mol. Biol. 162, 729–773. Schiex, T., Gouzy, J., Moisan, A., and de Oliveira, Y. (2003). FrameD: A flexible program for quality check and gene prediction in prokaryotic genomes and noisy matured eukaryotic sequences. Nucleic Acids Res. 31, 3738–3741. Smith, M.C., Burns, N., Sayers, J.R., Sorrell, J.A., Casjens, S.R., and Hendrix, R.W. (1998). Bacteriophage collagen. Science 279, 1834. Sonnhammer, E.L., Eddy, S.R., Birney, E., Bateman, A., and Durbin, R. (1998). Pfam: multiple sequence alignments and HMM-profiles of protein domains. Nucleic Acids Res. 26, 320–322. Steward, G., Montiel, J., and Azam, F (2000). Genome size distributions indicate variability and similarities among marine viral assemblages from diverse environments. Limnol. Oceanogr. 45, 1697−1706 Streit, W.R., and Schmitz, R.A. (2004). Metagenomics—the key to the uncultured microbes. Curr. Opin. Microbiol. 7, 492–498. Tech, M., and Merkl, R. (2003). YACOP: Enhanced gene prediction obtained by a combination of existing methods. In Silico Biol. 3, 441–451. Tsirigos, A., and Rigoutsos, I. (2005). A new computational method for the detection of horizontal gene transfer events. Nucleic Acids Res. 33, 922–933. Ventura, M., Canchaya, C., Bernini, V., Altermann, E., Barrangou, R., Mc Grath, S., Claesson, M.J., Li, Y., Leahy, S., Walker, C.D., et al. (2006). Comparative genomics and transcriptional analysis of prophages identified in the genomes of Lactobacillus gasseri, Lactobacillus salivarius, and Lactobacillus casei. Appl. Environ. Microbiol. 72, 3130–3146. Ventura, M., Canchaya, C., Kleerebezem, M., de Vos, W.M., Siezen, R.J., and Brussow, H. (2003a). The prophage sequences of Lactobacillus plantarum strain WCFS1. Virology 316, 245–255. Ventura, M., Canchaya, C., Pridmore, D., Berger, B., and Brussow, H. (2003b). Integration and distribution of Lactobacillus johnsonii prophages. J. Bacteriol. 185, 4603–4608. Ventura, M., Canchaya, C., Pridmore, R.D., and Brussow, H. (2004). The prophages of Lactobacillus johnsonii NCC 533: comparative genomics and transcription analysis. Virology 320, 229–242. Ventura, M., Lee, J.H., Canchaya, C., Zink, R., Leahy, S., Moreno-Munoz, J.A., O’Connell-Motherway, M., Higgins, D., Fitzgerald, G.F., O’Sullivan, D.J., and van Sinderen, D. (2005). Prophage-like elements

Bacteriophage Bioinformatics and Genomics

in bifidobacteria: insights from genomics, transcription, integration, distribution, and phylogenetic analysis. Appl. Environ. Microbiol. 71, 8692–8705. Whitman, W.B., Coleman, D.C., and Wiebe, W.J. (1998). Prokaryotes: the unseen majority. Proc. Natl Acad. Sci. USA 95, 6578–6583. Wommack, K.E., Ravel, J., Hill, R.T., Chun, J., and Colwell, R.R. (1999). Population dynamics of chesapeake bay virioplankton: total-community analysis by pulsed-field gel electrophoresis. Appl. Environ. Microbiol. 65, 231–240. Xu, J., Hendrix, R.W., and Duda, R.L. (2004). Conserved translational frameshift in dsDNA bacteriophage tail assembly genes. Mol. Cell 16, 11–21. Zajanckauskaite, A., Malys, N., and Nivinskas, R. (1997). A rare type of overlapping genes in bacteriophage T4: gene 30.3a is completely embedded within gene 30.3 by one position downstream. Gene 194, 157–162.

|

59

Bacteriophage in the Environment

3

Markus G. Weinbauer, Martin Agis, Osana Bonilla-Findji, Andrea Malits, and Christian Winter

Abstract One and a half decades ago, it was detected that phages are much more abundant in the water column of freshwater and marine habitats than previously thought and that they can cause significant mortality of bacterioplankton. Methods in phage community ecology have been developed to assess phage-induced mortality of bacterioplankton and its role for food web process and biogeochemical cycles, to genetically fingerprint phage communities or populations and estimate viral biodiversity by metagenomics. The release of lysis products by phages converts organic carbon from particulate (cells) to dissolved forms (lysis products), which makes organic carbon more bio-available and thus acts as a catalyst of geochemical nutrient cycles. Phages are not only the most abundant biological entities but probably also the most diverse ones. The majority of the sequence data obtained from phage communities has no equivalent in data bases. These data and other detailed analyses indicate that phage-specific genes and ecological traits are much more frequent than previously thought. In order to reveal the meaning of this genetic and ecological versatility, studies have to be performed with communities and at spatiotemporal scales relevant for microorganisms. Introduction In 1989/1990 it was shown using techniques such as transmission electron microscopy (TEM) and epifluorescence microscopy that viruses are much more abundant in pelagic systems (i.e. the water column of the ocean and lakes) than previously thought (Bergh et al., 1989; Proctor and Fuhrman, 1990; Suttle et al., 1990). It is now well known that viral abundance exceeds the abundance of their most likely host, the prokaryotes (domains Bacteria and Archaea), by 10–100 fold. A ca. 10-fold higher viral than prokaryotic abundance was even found for the deep ocean, where host abundance is as low as in tap or drinking water. Also, high viral infection frequencies where found in prokaryotic and cyanobacterial communities using TEM (Proctor and Fuhrman, 1990). These findings mark the beginning of the research field of virus community ecology. Previous research had suggested high viral abundance in aquatic systems, however, this did not attract the attention of microbial ecologists.

62

|

Weinbauer et al.

Much of the research on viruses ever since has been linked to the community ecology of bacterioplankton and the role of bacterioplankton in biogeochemical cycles. Indeed, the research groups popularizing the significance of the ecological and biogeochemical role of phages (and viruses infecting phytoplankton) (Fuhrman, 1999), have a background in community microbial ecology and biogeochemistry rather than in microbiology, virology, genetics or molecular biology. For example, the ecological and biogeochemical role of viruses was compared to that of small eukaryotic grazers, i.e. flagellates and ciliates, since causes of mortality for bacterioplankton were missing in budgets of carbon flow and phage were promising suspects. This is one of the reasons why the role of lytic phages in microbial food webs has been described as plankton predation (Proctor, 1997) rather than as cellular parasitism. The advent of molecular tools to study microbial diversity, gene expression, metagenomics and proteomics has advanced both ecology and microbiology. There is hope that the increasing use of molecular tools will help to create more common ground for the various disciplines. This chapter deals with bacteriophages and archaeal viruses or “archaephages.” In the following both types of viruses will be called phages. Most of the environmental examples are derived from pelagic marine and freshwater systems. This reflects not only the expertise of the authors but also the majority of the work done in viral community ecology. However, some research from sediments and soils will be discussed as well. Traditional and emerging methodologies Total viral abundance, phage production, phage-induced mortality and lysogeny Total estimates of viruses allow for characterizing environments, compare the number of viruses to that of the potential prokaryotic hosts and are essential for measuring phage production and prophage induction in natural communities. TEM (Bergh et al., 1989) and epifluorescence microscopy (Proctor and Fuhrman, 1990; Suttle et al., 1990; Hennes et al., 1995; Noble and Fuhrman 1998) have been used to count viruses (and prokaryotyes), however, these methods are progressively being replaced by flow cytometry (FCM) (Marie et al., 1999). Using FCM a larger number of samples can be processed—mainly because sample preparation is facilitated—and a higher accuracy can be achieved compared to other methods. Also, FCM allows for distinguishing distinct viral populations based on differences in nucleic acid fluorescence and side scatter (Figure 3.1). In order to obtain consistent results, the samples need to be preserved and stored properly (Brussaard, 2004; Wen et al., 2004). The majority of viruses found in aquatic environments are infecting prokaryotes (including cyanobacteria), however, viruses infecting algae can become dominant during phytoplankton blooms. Since most of the data discussed in this section were not from bloom situations, we will consider the entire viral community as phage community. The first methods developed to assess phage-induced mortality at the community level were based on TEM inspection of cells. Either thin sectioning (Proctor and Fuhrman, 1990) or whole cell approaches (Heldal and Bratbak, 1991; Weinbauer et al., 1993) were used to estimate the frequency of visibly infected cells (FVIC). As phage particles are only visible during the last phase of the latent period, conversion factors have been applied to

Bacteriophage in the Environment

beads HNAls HNAhs LNA

0

10

101 SSC-H

102

Figure 3.1 Cytograms for prokaryotic and viral settings showing different prokaryotic and viral populations. Note that large viral populations can also be detected with bacterial settings.

convert FVIC to the frequency of infected cells (FIC) (Proctor et al., 1993; Weinbauer et al., 2002). Models have been developed to assess phage-induced mortality of bacterioplankton from FIC (Proctor et al., 1993; Binder, 1999). More recently a virus-dilution approach was developed to assess FIC (Weinbauer et al., 2002). The principle of the virusdilution approach is to numerically reduce viruses in the sample (e.g. by ultrafiltration) in order to prevent new infection. Thus, an increase in viral abundance during the subsequent incubation can be attributed to viral lysis resulting from previous infections in the environment. Dividing the number of produced phage by a burst size yields an estimate of the percentage of infected cells. Burst size estimates of bacterioplankton can be obtained from TEM (Figure 3.2). The same dilution method can also be used to estimate phage production (Wilhelm et al., 2002) and this allows for assessing the fraction of bacterial production converted into phage lysis products. Decay rates of viral communities have been measured either after stopping new production with cyanide (Heldal and Bratbak, 1991) or by adding fluorescently labeled viruses (Noble and Fuhrman, 2000). Assuming steady state conditions, the decay rates correspond to production rates. In addition, an approach has been developed using radioactively labeled DNA precursors, which are incorporated into phage DNA (Steward et al., 1992). All these methods have their pros and cons and none has made it into a standard method. Lysogeny and prophage induction was also studied at the community level in bacterioplankton. Usually, mitomycin C was used as inducing agent, but sunlight, hydrogen peroxide, polyaromatic hydrocarbons, Bunker C fuel oil, trichloroethylene, polychlorinated biphenyl mixtures, Arochlor 1248, pesticide mixtures and sunscreen can also cause detectable prophage induction in prokaryotic communities ( Jiang and Paul 1996; Cochran et al., 1998, Weinbauer and Suttle 1999, Danovaro and Coridaldesi 2003). Mitomycin C does not induce all lysogens and thus its efficiency as an inducing agent for natural communities remains unknown. In the community ecology of phages, ultra- and tangential flow filtrations are important tools to concentrate phages (Suttle et al., 1991). These viral or phage concentrates

|

63

64

|

Weinbauer et al.

Figure 3.2 Transmission electron microscopy pictures of visibly infected cells. Electron micrographs were made at 50 000s magnification. Phage have a capsid diameter of ca. 63 nm in the upper panel and of ca. 81 nm in the lower panel.

constitute a library of phages, which can be used for isolation and genetic fingerprinting. In addition, phage concentrates can be used in experimental studies to increase the phage pressure on bacterioplankton. Moreover, phage-free water can be obtained and combined with bacterioplankton as phage-free controls in experimental studies to assess effects of phages. Isolation-based techniques to identify phage in the environments Isolation of phage–host systems is still a paramount task in phage ecology. This also means that many hosts remain to be isolated for which it is known from culture-independent methods that they are abundant in situ. PFUs and MPNs are traditional techniques to titer phage infecting specific isolated hosts. Phage typing is another classic technique, which has only rarely been applied in the environment. A modification of phage typing has been used to identify hosts in situ by adding a fluorescently labeled phage and detection of the phage on the target cells using epifluorescence microscopy (Hennes and Simon, 1995). Several methods have been used to detect phage or phage infection in the environment. For example, monoclonal antibodies have been applied to detect lactococcal c2 like bacteriophages (Azaiez et al., 1998) and DNA probes to detect phages of Bacterioides fragilis HSP40 (Puig et al., 2000a). Gene probes were also developed to detect human enteric viruses in environmental samples including coastal marine waters (Gerba et al., 1989; Ogunseitan et al., 1992). However, most phage-detection methods are PCR (Lopez-Pila et al., 1993; Sobsey, 1993; Straub et al., 1994) and RT-PCR -based technology (Tsai et al., 1993; Tsai

Bacteriophage in the Environment

et al., 1994; Tsai et al., 1995). In the years to come quantitative PCR is expected to become a powerful technique to detect and enumerate phages in the environment (Gruber et al., 2001). Phage infection has been studied by identifying the release of phage-borne enzymes following phage-induced cell lysis (Blasco et al., 1998; Puig et al., 1998) and probes have been used to identify phage genomes in cells (Labrie and Moineau, 2000; Ogunseitan et al., 1992; Puig et al., 2000a; Puig et al., 2000b). In single cells in situ PCR has been used in the medical field targeting viral DNA and mRNA (Haase et al., 1990; Mehta et al., 1995; Nuovo et al., 1992b; Patterson et al., 1993). This technique has a huge potential for studying phage in the environment, however, it will be a challenge to apply it to heterogeneous communities in aquatic and terrestrial systems. Isolation-independent methods to identify phages in the environment and assess diversity Sequences of viral isolates have been used to develop primers for specific groups such as cyanophages (Fuller et al., 1998) and temperate phage infecting Enterobacteriaceae (Balding et al., 2005). Using techniques such as denaturing gradient gel electrophoresis (DGGE) and terminal restriction fragment length polymorphisms (T-RFLP) genetic fingerprints for specific environments can be established (Short and Suttle, 1999; Wilson et al., 2000). For DGGE the bands can be excised and sequenced and sequence information can also be obtained by combining these primers with clone libraries (Zhong et al., 2002). Many environmental sequences could be collected in this manner without cultivation and used for phylogenetic affiliation. In addition, these fingerprints can be used to assess the diversity of different phage groups and thus applied to ecological studies. Viral genome size distributions can be studied using pulsed field gel electrophoresis (Klieve and Swain, 1993; Swain et al., 1996) and this method has been used as a rough proxy for the diversity of (dominant) viruses in the environment. Moreover, environmental probes can be made from excised bands and used in hybridization studies to investigate the distribution of specific uncultured phages in the environment (Wommack et al., 1999a; Wommack et al., 1999b). These probes can also be sequenced to assess the phylogenetic position of the phage. Culture independent approaches such as fingerprinting methods to assess viral diversity are still impossible since there is no universal bacteriophage gene analogous to the 16S rDNA of prokaryotes (Hendrix et al., 1999). Shotgun sequencing from environmental metagenomes can bypass this dilemma (Breitbart et al., 2003; Breitbart et al., 2002; Breitbart et al., 2004c; Edwards and Rohwer, 2005). It provides the qualitative as well as quantitative information of every genotype. Moreover, the online computational tool PHACCS (Angly et al., 2005) to analyze environmental viral communities from shotgun sequence data has recently been created to build models of possible community structure as a basis for making estimates of uncultured viral community richness, evenness, diversity index and abundance of the most abundant genotype. The metagenomics approach, however, harbors drawbacks such as free DNA in the environment, viral genes that kill the host in the cloning approach and unclonable viral DNA (e.g. Edwards and Rohwer, 2005). Nevertheless, viral metagenomics is a very powerful tool for understanding phage in the environment. Recently, TEM and in situ hybridization (ISH) has been used to identify specific bacterial groups natural prokaryotic communities (Gérard 2005). This method could be

|

65

66

|

Weinbauer et al.

combined with TEM studies on visibly infected cells to assess which groups of bacterioplankton are infected at which frequencies. Large prokaryotic groups such as Bacteria and Archaea, or Alpha-, Beta- and Gammaproteobacteria can be investigated, however, probes are also available for smaller taxonomic groups such as different genera. Abundance and diversity of bacteriophages in the environment Total viral abundance Viral abundance generally increases with abundances of prokaryotes and concentrations of Chlorophyll a in benthic (Glud and Middelboe, 2004) and pelagic systems (Maranger and Bird, 1995). In coastal waters there are more phages (107–8 ml–1) than in offshore water (106–7 ml–1). The interior of the ocean contains about 105–6 ml–1. Viral abundance is often higher in freshwater than marine systems and can be as high as 108 ml–1 in productive lakes. In sediments, viral abundance varies strongly between 107 and 1010 per gram with lower abundances in deep water sediments (Danovaro et al., 2005). Viral abundance decreases often with sediment depth. Some viruses can remain infectious in sediments from decades to a hundred years or more (Lawrence et al., 2002). Thus, sediments might constitute a reservoir of phage survival. Overall, phage counts from sediments are orders of magnitude higher than in the water column of systems with comparable trophic status. After improving the method for extraction of bacteriophages from soils (Williamson et al., 2003), measurement yielded higher viral abundances in forest soils (1.31 s 109 to 4.17 s 109 g–1 dry weight) than in agricultural soils (8.7 s 108 to 1.1 s 109 g–1 dry weight) (Williamson et al., 2005). Estimates of phage abundance are also available for so-called extreme habitats. In hot springs, above the upper temperature limit for eukaryotic life, phage abundance ranged from 0.07 to 7 s 106 ml–1 (on average 2 s 106) and showed a strong temporal variation (Breitbart et al., 2004b). Abundances from active deep-sea hydrothermal vents ranged from 1.45 s 105 to 9.90 s 107 ml–1 and averaged 3.5 s 106 ml–1 in the neutrally buoyant plume associated with the Endeavour Ridge system (Ortmann and Suttle, 2005). In the alcaline hypersaline Mono Lake, phage abundance ranged from 1 s 108 to 1 s 109 ml–1, which is the highest observed value in any natural pelagic system examined so far ( Jiang et al., 2004). With abundances of 106–7 ml–1, sea ice contains a significant number of viruses as well (Maranger et al., 1994). Phages are the most abundant biological entities in the biosphere. It has been estimated that there are ca. 1030 phages in the ocean and 1031 in the entire viriosphere (Suttle 2005). The amount of carbon bound in phages exceed that bound to whales by far and is ca. 10% of the organic carbon in prokaryotes (Figure 3.3). However, there is still considerable uncertainty about viral abundance and biomass in the subsurface, i.e. the crust of the Earth, which might be the biome containing the largest amount of prokaryotes. Dynamics of specific phages and phage diversity The great majority of isolated phages (at least 96%) are the tailed phages (Caudovirales). Tailed phages appear as monophyletic and as the oldest known virus group (Ackermann, 1999). They probably evolved before the separation of life into the domains Bacteria, Ar-

Bacteriophage in the Environment

Figure 3.3 Food web model in pelagic systems and carbon content in different compartments. The dotted lines show virus-mediated pathways. Carbon data are for marine system and humans serve as comparison.

chaea and Eukarya. An extensive TEM study of isolated bacteriophages from marine waters gives a glimpse of their enormous morphological diversity (Frank and Moebus, 1987). A careful TEM study in a lake also suggested that tailed phages dominate in the environment. However, morphology is not sufficient to describe phages. Viral species have been defined as “…a polythetic class that constitutes a replicating lineage and occupies a particular ecological niche.” The parameters for species description can only be obtained after isolation. An approach based on viral sequences has been proposed (Nelson, 2004), which is based on single gene loci, for example a DNA polymerase gene. In contrast, Edwards and Rohwer (2002) proposed a phage proteomic tree for viral taxonomy. This system is based on an algorithm, which uses every gene in every phage to calculate an average distance between pairs of phages. Species definitions for phages have been criticized. For example, the extensive gene transfer in phages could make such definitions futile. However, viruses might possess conserved genes at least for specific groups (CIESM, 2003), i.e. signature genes (Rohwer and Edwards, 2002). The latter approach

|

67

68

|

Weinbauer et al.

is pragmatically used in the following paragraphs to describe what is known about phage diversity in the environment. Diversity has been studied for phages infecting specific hosts. Among the best studied are vibriophages and cyanophages. In a series of studies, Moebus and coworkers (Frank and Moebus, 1987; Moebus, 1983; Moebus and Nattkemper, 1981; Moebus and Nattkemper, 1983) have found a high diversity of phages in the Atlantic and the North Sea. In a seasonal study on phages infecting Vibrio parahaemolyticus Paul and coworkers (Kellogg et al., 1995; Paul, 2000) detected seven groups of genetically different types and one was dominant. Several studies suggest that some phages are cosmopolitan in marine systems. A wide geographic distribution has been shown for bacteriophages that lyse the same indigenous freshwater isolate Sphingomonas sp. Strain B18 (Wolf et al., 2003). A wide occurrence of phage genes has also been reported using culture-independent approaches (Breitbart et al., 2004b; Short and Suttle, 2005). Phage production could be demonstrated when marine microbial communities were exposed to viral communities from soil, freshwater and sediment indicating that (1) phages can move between biomes and/or (2) there are cosmopolitan microbial hosts (Sano et al., 2004). However, the known bacteriophages are mostly host specific (Wommack and Colwell, 2000). Recent studies suggest that the host range might be broader than previously assumed (see below). It is also possible that the across biome transplantation has induced prophages. Sequences of structural proteins of cyanophages revealed a tremendous genetic variation not represented in cultures (Short and Suttle, 2005). Furthermore, sequences with > 99% identities were recovered from the Gulf of Mexico, the Southern Pacific Ocean, an Arctic freshwater cyanobacterial mat, and Lake Constance, Germany. It has also been suggested that phage-encoded sequences might move between biomes, e.g. the HECTOR sequence was found in salterns, marine water, corals and rumen fluid (Breitbart et al., 2004b). These results imply that closely related hosts and the phages infecting them are distributed widely across environments or that horizontal gene exchange occurs among phage communities from very different environments (Short and Suttle, 2005). Much of what we know about phage diversity comes from methods developed in genetics and molecular biology, which were adapted for studying phage in the environment. Some data are available from clone libraries, T-RFLP, DGGE and PFGE. Using PFGE, between 10 and 40 genome size classes are typically found, which are often distributed in 2–3 size class groups. The larger size classes might be due to algal viruses, however, this has not been tested rigorously. Clearly, this number is only a conservative estimate of phage diversity, since several phage types can have the same or a very similar genome size. Nevertheless, even with this crude technique, it could be shown that phage communities differ at seasonal and spatial scales, along plumes and between water masses (Riemann et al., 2000; Steward et al., 2000). Furthermore, diurnal and day-to-day changes in the rumen of individual sheep, as well as differences between sheep specimen could also be detected (Klieve and Swain 1993). The most powerful application of this approach is probably that the genomes excised from gels can be used in combination with hybridization to assess the dynamics of uncultured phages (Wommack et al., 1999a,b). Cyanophages are the best studied group of phages in marine and freshwater systems. For cyanophages, primers have been developed, which are targeted against the g20 capsid

Bacteriophage in the Environment

protein. For example, Wilson and Mann (2000) studied cyanophage diversity along a south-north transect from Falkland to UK with DGGE and found dramatic changes in cyanophage population structure in surface waters across the transect and also over depths. Using genetic fingerprints, it has been shown during an annual cycle, that the diversity of cyanophage and cyanobacteria co-varies (Mühling et al., 2005). This observation and multivariate tests are consistent with cyanophage infection being a major controlling factor in picophytoplankton succession. Using clone libraries, Zhong et al. (2002) found distinct cyanophage populations along transects from estuarine to offshore waters. Since cyanophage sequences have been also found in the deep-sea, the primers might not only detect cyanophages. This idea is also supported by the finding that identical sequences have been found in marine and freshwater systems and this implies a strong potential for gene transfer (Short and Suttle 2005). However, cyanobacteria can be transported rapidly by organic particles (marine snow) into deep water and cyanophages could be transported in the same way and released into ambient water during the decay of particles. Thus, cyanophage signatures from the deep sea could be due to sinking phages. Using a metagenomics approach from two marine environments, it has been stated that there are between 370 and 7100 different virus types present (Breitbart et al., 2002). This corresponds to a genome complexity of 1.5 s 107 to 3.5 s 108 bp per sample (Weinbauer and Rassoulzadegan, 2004). Assuming unique endemic phage populations in each environment, about a 100 million distinct viral genotypes can be predicted when metagenomic data are extrapolated to assess global viral diversity and less than 0.0002% have been sampled so fare (Rohwer, 2003). It is well known that microorganisms are associated with multicellular organisms. However, only recently has the diversity of these microorganisms been studied. It seems that invertebrate species such as corals or marine sponges harbor specific prokaryotic communities. Although phage communities have not been studied, it is known that e.g. oysters contain up to 105 phages cm–1, which infect Vibrio parahaemolyticus, and are seasonally dynamic (e.g. Comeau et al., 2005). This represents a potentially large but unexplored reservoir of phage genetic diversity. Phage genomics and metagenomics The first phage genome sequence was already published in 1977 for F X174 infecting Escherichia coli (Sanger et al., 1977). Sequencing of phages requires the growth of their microbial hosts on culture plates. However, from aquatic biomes typically less than one percent of microbes have been cultivated and no archaeal isolates have been obtained from pelagic marine systems. Thus, only few phage genomes have been sequenced (Paul et al., 2002), although a rapid progress is made for isolated phages. A metagenomics approach tries to circumvent this. Five viral metagenomic libraries have been published so far, two from marine coastal waters (Breitbart et al., 2002), and one each from marine sediments (Breitbart et al., 2004a), human feces (Breitbart et al., 2003) and equine feces (Cann et al., 2005). 75% of the sequences had no significant similarity to any sequence in Genbank. Thus, compared to metagenomes of microorganisms, for which only about 10% of unknown sequences were found (Venter et al., 2004) the viral metagenome remains essentially unknown. This has

|

69

70

|

Weinbauer et al.

led to the speculation that the viral metagenome could serve as a gene stock for microbial innovation (Daubin and Ochman, 2004). Algorithm-based analyses of viral metagenomes indicate that coastal marine sediments are the most diverse biological systems characterized so far (Breitbart et al., 2004a). Moreover, the first exploration of a metagenome library generated from a viral community in a Delaware agricultural soil uncovered that 78% of soil viral metagenome sequences were novel to GenBank and that 4% of the unknown soil sequences had significant homologs in marine environments (Wommack et al., 2004). This would imply that soils have even a larger reservoir of unknown genes than in marine sediments. The phage proteomic tree is a potentially appropriate tool for metagenomic analysis since it uses all of the genomic sequences. For example, this whole-genome taxonomy system has shown that Siphoviridae might be the most abundant genome arrangement on earth. According to the bank model, only few phages are abundant while the rest of phages are rare thus forming a potential recruitment bank capable of responding to environmental changes along with changes in the microbial host community (Breitbart and Rohwer, 2005). Power-law functions describe best the shape of rank-abundance curves of phage communities as assessed by metagenomics. This stems from the fact that each lytic cycle produces about 30 new phages which increases the encounter probability dramatically for the phage genotype which has randomly hit more hosts than others or which is specific to a more competitive host (Edwards and Rohwer, 2005). Clearly, metagenomics is not only a powerful tool for the study of the genetics and molecular biology but also for the ecology of phage. Prophages and diversity About 60% of the published microbial genomes contain prophages, which can constitute as much as 10–20% of the host’s genome. Prophages may exist in a dormant stage for perhaps millions of years and remain functional (Casjens, 2003). Edwards and Rohwer (2005) extrapolated from similarities between the Sargasso Sea metagenome and known phages to unknown phages and estimated that about 1% of microbial metagenome encode phage protein. Whole genome sequencing has shown that the occurrence of prophages and phage-mediated lateral gene transfer is much more frequent than previously thought. Thus, phage DNA in cells is an important component of diversity. However, it is not known whether the phage diversity in situ is due to lytic or temperate phages. Phage in the environment—principles of phage ecology The ecological conditions and resources, which are two major parameters defining the niche, are not well known for most phages in pelagic environments. However, fragmentary information is available from studies with phage–host systems and in situ studies (Goyal et al., 1987). Goyal and co-authors (Goyal et al., 1987) present data and methods in phage ecology, which are mainly based on phage–hosts systems. A large data set on niche parameters has been published since the detection of phages in 1915, which awaits to be discussed in terms of ecological theory. Some attempts towards such an integration have been made, for example for cyanophages (Suttle, 2000a; Suttle, 2000b) and lytic phages (Lenski, 1988). The host provides not only the metabolism but also the chemical elements (“nutrients”) for

Bacteriophage in the Environment

phage replication and assembly. Thus, the hosts can be considered as habitable patches or island in the environment (Weinbauer 2004). In the following, we discuss some aspects of phage proliferation and survival for phages, which have been studied only recently or applied to natural communities. Nutrient limitation Phages can propagate in recently killed cells (Anderson, 1948) and can infect starving cells and propagate in them, although cells lysis might be prevented (Kokjohn et al., 1991; Schrader et al., 1995; Schrader et al., 1997). Low nutrient availability often results in prolonged latent period and a smaller burst size and suggests a strong control of the metabolic activity on phage production. However, it has also been argued that a short latent period and a small burst size are typical for productive, high host density environments (Abedon, 1989; Abedon et al., 2001). In addition, the type of nutrient limitation might be crucial as the limitation of the chemical elements necessary for phage assembly. For example, phage have a high phosphorus content (ca. 50% is DNA) and there is experimental evidence that phage production in the environment can be P-limited. Support for this notion stems from whole genome sequencing of the marine cyanophage P60 and the roseophage SIO1 showing that the two phages have obtained DNA metabolism genes from the hosts (Chen and Lu, 2002; Rohwer et al., 2000). Moreover, marine phages can use host DNA (Wikner et al., 1993) and in the roseophage genome a host protein was detected, which is induced in cells during phosphorus limitation. From an ecological point of view and in light of the fact that the host is the major nutritional resource for phage, a state of nutrient limitation may also apply when a receptor is masked or not expressed on the cell surface. DNA damage and repair Many phages in the environment are exposed to sunlight and thus have to deal with sunlight-induced DNA damage. Many studies have been performed with isolates and UVC radiation, however, in situ studies with sunlight are still sparse. In a study conducted in the Gulf of Mexico, cyclobutane pyrimidine dimers were ca. one order of magnitude more frequently detected in phage communities than pyrimidone-pyrimidine (6–4) photoproducts (Weinbauer et al., 1999; Wilhelm et al., 1998b). Different phage isolates exposed along a depth profile in the water column showed a strong variability in the response of DNA damage formation to sunlight. In addition, phage communities were more resistant against DNA damage formation than phage isolates. It could also be shown that in situ phage populations infecting specific hosts developed resistance against sunlight-induced DNA damage during periods when solar radiation was strong (Garza and Suttle 1998). Increasing the GC content of the phage DNA is one of the mechanisms to reduce loss of infectivity (Kellogg and Paul, 2002). Phages can exploit the host dark repair systems and the enzyme photolyase to repair their DNA. In addition, a sunlight-inducible recA-independent repair system was found for the phage UNL-1 infecting Pseudomonas aeruginosa (Shaffer et al., 1999). Studies with natural phage and host communities in the Gulf of Mexico suggested that 20–50% of the phages which could not be repaired in the dark, could restore their infectivity in UVA and visible light (Weinbauer et al., 1997; Wilhelm et al., 1998a). Since the percentage of phages

|

71

72

|

Weinbauer et al.

repaired by photoreactivation was higher in non-productive offshore than in more productive coastal environments, the ATP independent mechanisms photoreactivation might be preferable in nutrient poor environments. Viruses can also carry their own repair genes, which have no equivalents in their host cells and these genes can be genetically wide-spread. For example the den V genes were found in the T4 phage infecting the prokaryote E. coli and viruses infecting the eukaryote Chlorella, a photosynthetic algae. Host range and resistance The majority of phage is considered to be species- or even strain-specific: phage typically do not trespass the genus barrier (Ackermann and DuBow, 1987). Nevertheless, broad-host range viruses traversing the genus or even domain level do exist, although it has been estimated that they typically do not exceed 0.5% of the total virus community (Chiura, 2002; Chiura and Tagaki, 1994; Chiura and Umitsu, 2004; Chiura et al., 2000; Chiura, 1997; Wichels et al., 2002). “Broad-host range vector particles” can originate from spontaneous induction and are able to infect recipient cells not phylogenetically related to the original host organism. In addition, it has been argued that isolation techniques select for narrowhost range phage and that host ranges are typically broad ( Jensen et al., 1998). Findings from genomics, from the wide-spread occurrence of similar genes and the potential of phages to cross biomes are further indications that the host range might be broader than previously thought. The ability of bacterial isolates to rapidly develop immunity to viral infection upon exposure to phage in batch and chemostat cultures (e.g. Lenski 1988) is in stark contrast to the high viral abundance commonly found in the environment and has been termed the infectivity paradox (Weinbauer, 2004). Resistance to viral infection can result from a reduction in the number and/or structural modification of cell surface components such as nutrient receptors, which also serve as recognition and docking sites for phage. It is reasonable to assume that under the conditions of nutrient rich culture media commonly employed for isolating and growing bacterial strains, acquiring immunity to viral infection has no obvious adverse effect. In contrast, the growth of bacteria in many environments might be nutrient limited and the costs of acquiring immunity to viral infection might be substantial resulting in a competitive disadvantage over sensitive cells. Indeed, it has been shown that resistance can have physiological costs (Chao et al., 1977; Lenski, 1988; Levin et al., 1977). Since even dormant cells can be consumed by grazers, bacteria in nutrient-poor environments might have no other choice than expressing receptors and try to grow and divide. The finding that cyanophage resistance decreased towards oligotrophic offshore environments (Garza and Suttle, 1998; Suttle and Chan, 1994) supports this idea. Thus, there might be a trade-off between the chance to get infected and the chance to acquire nutrients. Fuhrman (1999) even suggested that bacteria might develop decoy receptors to invite infection in order to assess nutrient rich phage protein and DNA. As discussed above, phosphorus might often be the limiting element. Modeling the development of resistance in phage–host isolates suggested that when a more complex situation was assumed by including grazers in the model, susceptible cells recovered and high phage infection was sustained (Middelboe et al., 2001). Clearly, the phenomenon of immunity to phage infection and its consequences remain poorly understood at the level of bacterioplankton communities in situ.

Bacteriophage in the Environment

Phage life styles Several hypotheses have been developed to describe the environmental conditions when lysogenization and prophage induction occurs (Weinbauer, 2004). Lysogeny should occur in environments, where the contact rate between infective phage and hosts is too low to sustain the lytic life style. This can be due to low host abundances (Echols, 1972, Steward and Levin, 1984) or high phage decay rates (Lenski, 1988). An across system study showed that lysogeny was highest in the nutrient limited deep sea (Weinbauer et al., 2003a) and this would support the contact rates hypothesis. The lysogenic decision can depend on the multiplicity of infection, nutrient and divalent cation concentrations. It has been shown that lysogenization occurs in starving cells and prophage induction in rapidly growing cells. The strategy behind this could be to feast when hosts grow well and are abundant while remain in the cell when environmental conditions are not favorable. Induction can also occur upon DNA damage. It makes sense to abandon the sinking ship to save some of the progeny. Sunlight does not seem to be a potent inducing agent for bacterioplankton ( Jiang and Paul 1996, Weinbauer and Suttle 1999), however, it cannot be excluded that the majority of the lysogens was induced already when these studies were performed. Some studies with phage communities suggest that prophage induction typically does not contribute strongly to phage production (Bratbak et al., 1994; Wilcox and Fuhrman, 1994; Weinbauer and Suttle, 1996;) and that lytic phages might out-compete prophages (Weinbauer 2004). However, it has been pointed out that in many situations temperate phages may not lysogenize host cells but retain a lytic life style (Lenski 1988). Overall, we do not know whether the majority of phages in the environment are lytic or temperate and it may well be that this differs between biomes. It is also possible that generalizations across systems are not possible because individual phage–host systems differ strongly (Schrader et al., 1995). This constitutes a problem, since the major phage–host systems in situ remain unknown. Clearly, new methods to detect temperate phages and their hosts in situ are necessary to tackle these questions. Phage are obligate cell parasites. However, their ecological role is more complex (Table 3.1). Lytic phage infect cells, lyse them and the cell content and cell wall fragments are released, which has similarities to a grazer attacking a cell, digesting it and releasing cell remnants and nutrients. Thus, lytic phage might be best described as predators. Temperate phage can have mutualistic interactions with the host (Edlin et al., 1975), whereas chronic infection is a proper parasitism. Other life style concepts popular in ecology might be applicable to phage as well (CIESM, 2003). A compilation of life strategy traits (Table 3.1) suggests similarities to concepts such the r/K continuum with lytic phage being rather r-strategists or opportunistic species and lysogenic phage rather K-strategists or equilibrium species. The concept of evolutionary trade-offs assumes that resource scarcity and abundance are major determining factors for life styles. Using this concept, lytic phage might be favored when the resource is abundant, whereas lysogeny should prevail when the resource is scarce. However, this concept should be extended to “resource-independent” survival mechanisms such as resistance against destruction (such as sunlight-induced nucleic acid damage), repair capacity, and temperature and salinity tolerance, which differ between phage species and thus strongly influence their life styles. To further tackle the question, whether or not different types of life styles have evolved as distinct ecological

|

73

74

|

Weinbauer et al.

Table 3.1 Viral life styles in potential relation to types of interaction with host, evolutionary trade-offs and some life style traits. NA, not applicable

Life styles

Type of interaction with host

Tradeoff by resource

Lytic

Predator

Lysogenic Chronic

Life style traits Resource abundance

Production rate

Latent period

Burst size

Abundance Short

Low

High

High

Parasite/ mutualist

Scarcity

Long

High

Low

Low

Parasite

Scarcity

NA

NA

Low?

Low?

strategies to optimize resource allocation and survival, more information on parameters such as infection efficiency, carrying capacity, burst size, intrinsic rate of increase, and viral survival has to be obtained from in situ studies or in experiments relevant to in situ conditions. It is also possible that phage morphotypes are linked to the life strategy of the phage, since Siphoviridae are typically temperate, while most Myoviridae are lytic (Suttle, 2005). Whole genome sequencing and metagenomics might also allow the development of testable hypotheses about ecological strategies of phages. Role of phage for biogeochemical processes and food web processes Phage-induced mortality of prokaryotic communities Phage can cause significant mortality in prokaryotic communities. Trends of viral infection along trophic gradients (Weinbauer et al., 1993) or during diurnal cycles (Bettarel et al., 2002; Winter et al., 2004a) could be detected indicating that these methods are useful, although they are not precise enough to make accurate measurements on the effect of viruses in biogeochemical cycles. Viruses and protists (flagellates and ciliates) can cause similar mortality of bacterioplankton (Fuhrman and Noble, 1995) and in specific environments or situations virus-induced mortality can be even higher than grazing-induced mortality (Wommack and Colwell, 2000; Weinbauer, 2004). For example, in anoxic marine and freshwater environments high infection frequencies were high (Weinbauer and Höfle, 1998; Weinbauer, 2003), maybe because the other type of predators of the plankton, the protists, showed low abundances. High infection frequencies have also been found in solar saltern and very cold marine waters; in both studies grazing rates were low (Guixa-Boixareu et al., 1996; Guixa-Boixereu et al., 2001). These situations may have some similarities to the scenario before the evolutionary appearance of eukaryotic grazers when phage were probably the only large group of predators. Overall, there is a tendency that viral infection frequencies are higher in productive than in non-productive environments (Weinbauer, 2004). On average, ca 25% of virus-induced bacterial mortality was found in offshore marine systems compared to > 50% in coastal marine systems. These data are based on steady-state assumptions, where 100% mortality means that offspring and death are balanced, i.e. the population numbers remains stable.

Bacteriophage in the Environment

Phages can also influence prokaryotes in other ways. Phage ghosts, i.e. phage without DNA, and phage tails can kill hosts or at least impair their growth. Phages can also be used as bacteriocins in the chemical warfare of prokaryotes. Bacteriocins can kill potential hosts as well as infected cells and thus reduce viral production. It is well known that phage lysates can inhibit bacterial growth without phage production (Moebus, 1983; Moebus and Nattkemper, 1981). This lysis inhibition might be due to phage-borne bacteriolytic enzymes released during lysis. However, the significance of these mechanisms in situ remains unknown. Lysogeny Published estimates on the percentage of lysogenic bacterioplankton show a wide range from ca. 0% to 100%. Some reports suggest that the percentage of lysogenic bacterioplankton increases towards nutrient poorer environments and in habitats where the loss of phage infectivity is high. However, uncertainty remains, e.g. due to the fact mitomycin C does not induce all lysogens. The potential of lysogeny for food web processes and biogeochemical cycles is poorly investigated. A mass lysis event of filamentous cyanobacteria caused changes in the bacterial community in a pond (Van Hannen et al., 1999). Such phenomena could locally change the dissolved organic matter (DOM) composition and availability. However, since the small number of available studies suggest that viral lysis typically comes from lytic infection and not from prophage induction (Bratbak et al., 1994; Wilcox and Fuhrman, 1994; Weinbauer and Suttle, 1996;), induction does probably not strongly affect the functions of the bacterioplankton ecosystem and it’s associated biogeochemical cycles. However, as discussed below, lysogeny might influence the ecology and the species composition of bacterial hosts and this could affect food web processes and biogeochemical cycles. Food web processes and biogeochemical cycles Prokaryotes play a crucial role for the flow of carbon and nutrients in aquatic systems. They are the only group of organisms, which can convert organic carbon, either in the form of dissolved or particulate organic carbon (DOC, POC; nominally separated by 0.2-µm pore-size filters), into biomass at a significant rate. This carbon pool, which approximates the amount of carbon bound in atmospheric CO2, would otherwise be lost from the food web. Via grazing by protists (flagellates and ciliates), carbon bound in prokaryotes can be transferred back to the grazing food chain. This microbial part of the food web is known as microbial loop (Azam et al., 1983) (Figure 3.3). Phage cause mortality and thus influence the standing stock and production of prokaryotes. From a food web perspective, viral lysis of cells has counter-intuitive consequences (Fuhrman, 1999). Bacterial production that is lysed, i.e. converted in to DOC and cell wall fragments, cannot be taken up efficiently by grazers. This represents a short circuit to or semi-enclosed trophic loop that has been called “viral loop” or “viral shunt.” Therefore, lysis reduces the transfer of organic carbon to higher trophic levels via grazing. For this reason and since lysis products enter mainly the DOC pool, bacterial production and respiration should be stimulated. Evidence for this has been found in experimental studies (Middelboe et al., 1996; Middelboe and Lyck, 2002; Noble and Fuhrman, 1999).

|

75

76

|

Weinbauer et al.

In a food web model, Wilhelm and Suttle (1999) estimated that 6–26% of the photosynthetically fixed carbon enters the DOC pool via viral lysis at different trophic levels. This suggests that the viral shunt is a significant source of DOC and should have a strong influence on biogeochemical cycles. Prokaryote-derived capsular polysaccharides, cell wall components and membrane porins are important components of recalcitrant DOC, which is the largest DOC pool in the ocean. The cell content such as proteins and nucleic acids should be rich in N and P, whereas cell wall fragments should be rich in C. However, it is not known in detail how viral lysis of cells contributes to the recalcitrant and labile DOC pool. Moreover, released host and phage-borne enzymes might affect the DOM pool composition and reactivity. Since lysis probably results in a disintegration of cells rather than in the production of ghost cells (Riemann and Middelboe, 2002), the accessibility of lysis products might be high and this could stimulate biogeochemical cycles. One of the most significant biologically driven mechanisms involved in the interaction between global change and the ocean is the biological pump, i.e. the transport of carbon from the light flooded euphotic zone to the dark interior of the ocean (Legendre and Le Fèvre, 1995). The rate of this transport, the residence time of the aggregates in the dark ocean and the burial rates into deep sea sediments are major determinants of the amount of CO2 that can be stored in the ocean. These aggregates have different origins, but come mainly from different species of phytoplankton and appendicularians (gelatinous zooplankton) and are modified by bacterioplankton. Due to their appearance the larger aggregates are sometimes called marine or lake snow. There is evidence that viruses are able to influence the rate of the biological pump. By converting POC (i.e. cells) into DOC, the retention time of organic carbon (and other elements) should increase due to lysis (Fuhrman 1999), however, other mechanisms are possible as well. For example, sticky lysis products can coagulate and form larger colloids (Shibata et al., 1997). Also, it has been shown that increasing the concentration of the natural viral community in seawater increases the abundance, size and stability of algal flocs (Peduzzi and Weinbauer, 1993). It is not clear yet, whether viruses increase or reduce the retention time of aggregates, however, it seems likely that this also depends on the origin of the aggregates. Prokaryotes inhabiting aggregates are infected by phage and thus lysis products might increase the stickiness of aggregates, whereas enzymes released during lysis might contribute to the dissolution of aggregates (Proctor and Fuhrman, 1991). There is also an abundant phage community embedded in the organic matrix of marine snow (Peduzzi and Weinbauer, 1993). Phage can also cause mortality of photosynthetic bacteria, i.e. cyanobacteria (Suttle, 2000a,b). Cyanobacteria can contribute significantly to photosynthesis especially in nutrient-poor offshore environments and cause nuisance blooms in freshwater. Thus, one can assume that viral lysis reduces the rate or magnitude of the ecosystem functions mediated by these organisms such as photosynthetic carbon fixation or nitrogen fixation. However, more complex patterns are possible in situations where phage–hosts systems have to deal with physical, chemical and biological interactions. For example, it has been shown that in the absence of phages, cyanobacteria do not grow well (Suttle et al., 1996). The reason for this could be that cyanobacteria either need lysis products from heterotrophic bacteria or that the enhanced respiration rates in the presence of phage (as discussed above) increase remineralization of inorganic nutrients. It has also been argued that lysed photosynthesis

Bacteriophage in the Environment

is lost to grazers but available to bacteria, which might increase remineralization and thus photosynthesis (Weinbauer, 2004). Overall, phages might acts as catalysts of geochemical nutrient cycles (Suttle, 2005). Role of phage for the diversity of prokaryotic communities From “killing the winner” to “kill the killer of the winner,” an overview of mathematical models dealing with the viral influence on bacterial communities A central question in aquatic microbial ecology is: what determines the high number of different bacterial types commonly found in aquatic environments? Originally this question was described as the “paradox of the plankton” to explain why there are so many phytoplankton species when the types of limiting nutrients are so few (Hutchinson, 1957; Hutchinson, 1961) and subsequently extended to bacterioplankton. Generally, bacterial community composition appears to be regulated by three factors: size-specific grazing by protists, nutrient supply, and viral infection. The following discussion focuses exclusively on the influence of phages on bacterioplankton communities, whereas a detailed overview of the other two factors can be found elsewhere (Torsvik et al., 2002, Pernthaler 2005). Viral infection rates depend on the abundance of hosts and their phage (Murray and Jackson, 1992). Taking this into consideration, Thingstad and Lignell (1997) described a simple model of the interactions of heterotrophic bacteria with their environment. The authors considered three types of growth rate limiting factors (availability and degradability of organic carbon, inorganic phosphorus, organic and inorganic nitrogen) and two sources of bacterial mortality (non-selective grazing by flagellates and viral lysis). The model assumed steady state conditions and strict species-specificity of phages. The results suggested that in all cases bacterivorous flagellates control bacterial abundance whereas viral lysis affects bacterial richness by selectively killing the most abundant members of the community. This regulating effect of viruses on bacterial richness is often referred to as the “killing the winner” hypothesis and several authors found evidence that viruses are indeed capable of influencing bacterial community composition (e.g. Hewson et al., 2003; Schwalbach et al., 2004; Winter et al., 2004b). The model is a mathematical formulation of the idea that due to relative host specificity and dependence on host abundance, phage keep competitive dominants in check (Fuhrman and Suttle, 1993). Despite the model’s simplicity, its predictions appear to be intuitively plausible and as a consequence “killing the winner” enjoys great popularity. In a refined version of the original model Thingstad (2000) also considered controlling mechanisms of viral abundance. This model predicted that viruses with fast growing hosts are the most abundant ones and that the number of different bacterial types (in the sense of different substrate affinities and, thus, growth rates) controls viral abundance. Thus, the model suggested a reciprocal relationship between bacterial richness and the abundance of viruses. These predictions are supported by the findings of Winter et al. (2005), who detected a strong positive relationship between bacterial production and viral abundance concomitant with a negative relationship between viral abundance and bacterial richness in the North Sea. These empirical results are in contrast to the original “killing the winner”

|

77

78

|

Weinbauer et al.

hypothesis that suggested a positive relationship between viral abundance and bacterial richness. However, both models (Thingstad, 2000; Thingstad and Lignell, 1997) have not been tested frequently. One of the few exceptions is the study of Hennes et al. (1995), who added the bacterial isolate Vibrio natriegens PWH3A to seawater (Hennes et al., 1995). This bacterium grew rapidly in an initial phase as determined by phage typing but at the end of the experiment, the abundance dropped rapidly concomitant with a strong increase of phage infective for this isolate. Clearly, the natural phage community could control the winner in this experiment. In order to scrutinize the model predictions and relate them to situations found in natural environments it will be necessary to follow the abundance of individual phage populations and their hosts and to monitor total bacterial and phage community composition over time. To do this remains a methodological challenge. Middelboe et al. (2001) performed chemostat experiments to determine the influence of viruses on the population dynamics of four strains of marine planktonic bacteria. Their results showed that viral lysis had only temporary effects on the species composition of the host cells and that the growth of resistant cells resulted in a very similar distribution of the four strains when compared to the control culture without phage. The authors concluded that inter-species competition was the driving force in determining the bacterial community composition after the initial lysis phase. However, the clonal composition was strongly influenced by viral lysis, since the simple model community changed completely from sensitive to resistant cells within 5–10 generations. The authors also provided a mathematical model accompanying their experimental results based on carbon flow with fixed C:N and C:P ratios. The model was composed of bacteria, phage, protozoa, and DOC. These four compartments were interacting with each other via several processes (predation, uptake, respiration, infection, basic metabolism, nutrient addition, and dilution) to build the model. In the model, resistance to viral infection resulted in a competitive disadvantage over sensitive cells. The model predicted a strong influence of phages on community and clonal composition (sensitive and resistant cells) and no influence on bacterial abundance, since resistant clones complement the losses caused by viral lysis. The latest stage in the development of models aiming to understand the effects of viruses on bacterial communities is the study by Miki and Yamamura (2005). These authors modeled and analyzed the effects of a viral latent period. The basic assumptions are similar to Thingstad’s model (Thingstad, 2000), however, these authors considered an open system and allowed immigration of bacteria and phages as well as nutrients into their theoretical chemostat. Miki and Yamamura’s model (2005) predicted that non-selective protistan grazing (infected and uninfected bacterial cells are equally vulnerable to grazing) reduces bacterial richness by eliminating phages within infected bacterial cells. Thus, these indirectly grazed phages cannot act in the “killing the winner” way anymore and so their hosts may become more abundant and eventually out-compete other bacterial types (“kill the killer of the winner”). Furthermore, selective grazing on infected bacterial cells and/or a long latent period would intensifies this negative effect on bacterial species richness. However, previous studies in a meso-eutrophic reservoir (Simek et al., 2003; Simek et al., 2001; Weinbauer et al., 2003b) demonstrated that protistan grazing on bacterioplankton increases standing stocks of viruses and virus-induced mortality of bacteria and these find-

Bacteriophage in the Environment

ings are in contrast to the predictions of the model by Miki and Yamamura (2005). Such synergism might kill the “killer of the killing of the winner” hypothesis. Protistan grazing also increases bacterial growth rates by recycling nutrients and releasing easily degradable substances such as amino acids (Pernthaler, 2005). Weinbauer et al. (2003b) suggested that the increase in bacterial growth rates caused by protistan grazing might induce lysogens and/or force pseudolysogens into the lytic cycle thus increasing phage production. This might explain the discrepancy between the model and the empirical results, since Miki and Yamamura (2005) did not consider lysogenic infection in their model. However, an increased growth rate might also decrease the latent period, since growth rate and latent period are roughly similar (Proctor et al., 1993), and this would violate some assumptions of the “kill the killer of the winner” model. Finally, it has been shown that infected cells express genes, which might act as defense against grazing (Clokie et al., 2006). Thus, phages might have developed strategies to circumvent or reduce grazing by protists and this might also explain, why a higher infection rate sometimes concurs with grazing pressure. Each of the presented models advanced our understanding of the viral influence on bacterial community composition by focusing on the effects of one or a few of the known interactions between phage and their hosts in the context of a simplified ecosystem. However, there are a number of phage–host interactions that were so far not incorporated into a mathematical model. One example is lysogenic infection. Another missing feature in mathematical models is that they do not incorporate more than one phage population per host. Non-selective protistan grazing, i.e. grazing affects all bacterioplankton cells equally, was assumed for all but one model (Miki and Yamamura, 2005). A more realistic approach for protistan grazing would be to include the effect of protists on the size structure, species composition and activity of bacterioplankton. The models are based on simplifying assumptions that are both their strength and weakness. The strength lies in the ability to make emerging trends and patterns more visible, while the weakness lies clearly in an oversimplification of nature. Now that an increasing number of viral effects on their hosts have been modeled and analyzed in detail, it should be a worthwhile endeavor to gradually combine these models to obtain a more unifying picture of how resource availability, protistan grazing, and the effects of phages are shaping the composition of bacterioplankton communities. Virus-mediated gene transfer Our knowledge on the significance of phage-mediated gene transfer in large ecosystems such as marine and freshwater pelagic systems is extremely limited. Most studies have been performed in lakes using Pseudomonas aeruginosa (e.g. Ripp and Miller, 1995). However, an increasing number of studies has been performed in marine systems (Paul, 1999). Estimated transduction rates from marine systems are up to several fold higher than for freshwater systems. A critical determinant for phage-mediated gene shuffling is the host range, which might be (as discussed above) broader than previously thought. Broad-range vector particles exhibit mild lethality while having an unusually high transduction frequency in the order of 10–3 per particle. Other estimates from freshwater and marine systems are typically much lower ranging from 10–5 to 10–8. Jiang and Paul (1998) estimated that 1014 transduction

|

79

80

|

Weinbauer et al.

events could occur per year in Tampa Bay, whereas Weinbauer and Rassoulzadegan (2004) estimated 1013 transduction events for the Mediterranean Sea. The water column is less homogenous than previously thought and consists of a spectrum of aggregates ranging from small colloids to large (mm to cm) marine or lake snow particles. It is believed that such particles represent hot spots of microbial abundance and activity. This might also enhance the chance of gene transfer by bringing together hosts and phage in a similar way as this has been shown for suspended matter in freshwater systems (Ripp and Miller, 1995; Vettori et al., 1999). Based on a detailed analysis of the increasing number of completely sequenced prokaryotic genomes many sequences of apparent viral origin were detected, which were scattered throughout these genomes. This suggests that viruses had a significant influence on the evolutionary development of prokaryotes. Moreover, this should also strongly influence the diversity of host communities, however, almost nothing is known for natural prokaryotic communities (Weinbauer and Rassoulzadegan 2004). Another mechanism by which viruses can act as vehicles of gene transfer is lysogenic conversion (phage conversion). For example, cells infected with temperate phage acquire immunity to super-infection by homologous phage. This could represent a mechanism for temperate phage to influence bacterial community composition. However, the consequences of such a mechanism currently remain unexplored. Phage conversion sensu stricto can also transfer morphological and metabolic changes to the lysogens and this can change the host’s fitness and thus, bacterial community composition. For particularly well-studied pathogenic bacteria several examples have been discovered where phage conversion plays a crucial role in the acquisition of pathogenicity (Brussow et al., 2004). To our best knowledge, the effect of phage conversion for bacterioplankton diversity has never been studied. Viral influence on Archaea Most of the studies described above deal with the domain Bacteria, however, it is now well known that prokaryotic communities can comprise a significant fraction of Archaea such as in the deep ocean (Karner et al., 2001). Because of the enormous volume of the deep ocean, marine planktonic Archaea comprise an abundant cell-type on Earth. There are a number of studies addressing archaeal viruses and hosts at the community level in so-called “extreme” environments such as solar salterns, the Dead Sea or hot springs (Guixa-Boixareu et al., 1996; Zillig et al., 1996; Oren et al., 1997; Häring et al., 2005a; Häring et al., 2005b). In general, marine planktonic Archaea are poorly understood and only a single experimental study has investigated the potential role of archaephages. Winter et al. (2004b) followed the development of marine bacterial and archaeal communities over time exposed to different concentrations of viruses in batch-culture experiments. Although differences between treatments were small when looking at total bacterial and archaeal communities, clear differences in the response of individual bacterial and archaeal types to virus additions could be detected. For example, the intensity of some archaeal peaks on T-RFLP fingerprints was inversely related to the number of viruses present. Thus, phages seem to lyse marine planktonic Archaea, at least under experimental conditions. However, the question whether or not this viral influence on Archaea is important in the ocean and under which conditions this occurs remains currently unexplored.

Bacteriophage in the Environment

Release of lysis products The potential role of lysis products on bacterial productivity has been described above. Changes in DOM composition and reactivity due to lysis might also affect bacterial species composition, since changes in dissolved organic carbon concentration as occurring during phytoplankton blooms have been shown to influence bacterioplankton community composition (Arrieta and Herndl, 2002; Riemann et al., 1999; Riemann et al., 2000). Van Hannen et al. (1999) reported on a virus-driven mass lysis event of a cyanobacterial bloom in a mesocosm study. These authors found pronounced changes in bacterial community composition during the periods before and after the crash of the bloom. Although the authors interpret their findings as evidence for a direct influence of viruses on the members of the community in the sense of the “killing the winner” hypothesis, it is possible that the changes found in bacterial community composition are a result of the release of lysis products into the medium. Phage oddities It is becoming more and more apparent that phage have more unique genes and ecological traits than previously thought. Here, we describe some of these phage oddities for bacterio-, cyano- and archaephages. Grazers of bacteria probably ingest not only non-infected cells but also infected cells and this might reduce the spread of phage as discussed above. However, it was shown that induction of bacterial prophages occurred in the food vacuole of a ciliate (Clarke, 1998). Some of these phage were released along with remnants of the host cells, thus, grazing directly produced phage. A release of infective cyanophage from food vacuoles was also reported for amoebae grazing on filamentous cyanobacteria (Cannon, 1987). This suggests that induction can be a strategy to “abandon the sinking ship.” Using TEM, virus particles were also detected in food vacuoles of radiolarians (Gowing, 1993) and amoebae (Corpe and Jensen, 1996). Although the origin of the phage has not been confirmed in the two latter studies, phage seem to have some unorthodox strategies do deal with specific ecological conditions they encounter. Prophage induction in the vacuole of the grazer could also increase the chance of gene transfer. Thus, protists might be what they eat. Recently, using phage genomics, it was shown that cyanophage carry photosynthesis genes (Mann et al., 2003; Lindell et al., 2004). This finding was at odds with the general perception that phage have no metabolism and that their genome is too small to carry junk DNA. Subsequent studies on quantifying gene expression in infected cells (Clokie et al., 2006) clearly showed that these phage photosynthesis genes were expressed to prevent a shut-off of photosynthesis either due to photoinhibition or due to a host antiphage defense system. Thus, specific phage genes manipulate the host metabolism to maximize the viral progeny production, i.e. burst size. A paradigm in viral biology states that viruses are completely assembled by the metabolism of the host within the cell. However, in a recent study, it could be demonstrated that a archaephage can develop a tail at the capsid outside the cell (Häring et al., 2005a; Häring et al., 2005b). This shows that phage development outside a cell is possible. Maybe even more intriguing is the idea that these “tails” could be used as paddles or at least as passive transport mechanisms.

|

81

82

|

Weinbauer et al.

Phage communities in situ—microhabitat of phage and host Research with phage–host systems has been successfully applied to study specific aspects of phage biology and ecology. However, it is difficult to extrapolate from phage–host systems to environmental conditions. For example, phage to host ratios in chemostats are typically well below 1, since hosts develop resistance rapidly under these conditions (Lenski, 1988). From such studies one could not predict the high virus abundance and virus infection of bacterioplankton found in situ. This is one of the most important lessons of phage community ecology. The problem is amplified by the difficulty to isolate bacteria, which are abundant in situ. However, recent studies have suggested that cultivation is not impossible (Hahn et al., 2003; Rappé et al., 2002). This might also allow for the isolation of representative phage hosts systems and boost research as has been shown for cyanophages. Cyanobacteria are comparatively easy to culture and this has facilitated isolation of cyanophages. Whole genome sequences of cyanophages and their host allowed Clokie et al. (2006) for example to study gene expression during infection and assess the role of phage photosynthesis genes. Using cyanobacterial isolates the distribution of cyanophages in the ocean could be studied (Suttle and Chan, 1994) and it was shown that cyanophages are selected for UV resistance in summer (Garza and Suttle, 1998). In situ, phage do not only have to find a susceptible host cell which in turn does not have to deal with just a single phage, but both phage and host cell are exposed to natural communities of a multitude of species and a variable and complex array of physicochemical conditions. The seemingly homogeneous water column is more heterogeneous than previously thought and some TEM pictures of that are shown in Figure 3.4. For example,

Figure 3.4 Transmission electron microscopy pictures of the habitat of natural phage communities. In the upper panel, several bacteria are visible, whereas in the lower panel, the large structure is the skeleton of a diatom (single-celled alga). The cobweblike structures are scales from different species of protists (protozoa). Amorphous aggregations of organic matter can also be seen.

Bacteriophage in the Environment

decaying phytoplankton cells are hot spots of local nutrient enrichment thus chemotactically attracting prokaryotes (Blackburn et al., 1998). Also, in addition to cells, there is a continuum of organic matter particles from small colloids to larger (mm to cm) size aggregates, which are colonized by prokaryotes. This microheterogeneity not only creates microniches for prokaryotes but also concentrates cells around the hot spots and thus increases the probability of phage encounter. Lysing cells are part of the organic matter continuum and create a plume of organic molecules thus attracting motile prokaryotes (Riemann and Middelboe, 2002). The formation of such a nutrient plume strongly depends on whether a cell disrupts during lysis or whether phage leave a cell though ruptures and leave a ghost cell behind. Optimization of organic matter leakage could be a strategy of phage to attract and concentrate potential hosts. One of the major aims in the years to come will be the identification and study of the in situ dynamics of the major phage host systems in different environments and the assessment of how these dynamics are influenced by factors such as host range and resistance. The microenvironment for prokaryotes and thus for phage host systems, i.e. processes that matter for microorganisms at the femto-, nano- and micrometer scale, are hardly known in situ but might be crucial for understanding phage in the environment. This will require methodological progress, not only in terms of molecular tools but also in terms of isolation techniques and physicochemical measurements at the microscale. References

Abedon, S.T. (1989). Selection for bacteriophage latent period length by bacterial density: a theoretical examination. Microb. Ecol. 18, 79–88. Abedon, S.T., Herschler, T.D., and Stopar, D. (2001). Bacteriophage latent-period evolution as a response to resource availability. Appl. Environ. Microbiol. 67, 4233–4241. Ackermann, H.-W. (1999). Tailed bacteriophages: the order Caudovirales. Adv. Virus Res. 51, 135–201. Ackermann, H.-W., and DuBow, M.S. (1987). Viruses of Prokaryotes. Volume I. General Properties of Bacteriophages (Boca Raton, CRC Press). Anderson, T.F. (1948). The growth of T2 virus on ultraviolet-killed host cells. J. Bacteriol. 56, 403–410. Angly, F., Rodriguez-Brito, B., Bangor, D., McNairnie, P., Breitbart, M., Salamon, P., Felts, B., Nulton, J., Mahaffy, J., and Rohwer, F. (2005). PHACCS, an online tool for estimating the structure and diversity of uncultured viral communities using metagenomic information. BMC Bioinformatics 6, 41. Arrieta, J.M., and Herndl, G.J. (2002). Changes in bacterial B-glucosidase diversity during a coastal phytoplankton bloom. Limnol. Oceanogr. 47, 594–599. Azaiez, S.R. C., Fliss, I., Simard, R.E., and Moineau, S. (1998). Monoclonal antibodies raised against native major capsid proteins of lactococcal c2-like bacteriophages. Appl. Environ. Microbiol. 64, 4255–4259. Azam, F., Fenchel, T., Field, J.G., Gray, J.S., Meyer-Reil, L.A., and Thingstad, F. (1983). The ecological role of water-column microbes in the sea. Mar. Ecol. Prog. Ser. 10, 257–263. Balding, C., Bromley, S.A., Pickup, R.W., and Suanders, J.R. (2005). Diversity of phage integrases in Enterobacteriaceae: development of markers for environmental analysis of temperate phages. Environ. Microbiol. 7, 1558–1567 Bergh, Ø., Børsheim, K.Y., Bratbak, G., and Heldal, M. (1989). High abundance of viruses found in aquatic environments. Nature 340, 467–468. Bettarel, Y., Dolan, J., Hornak, K., Lemée, R., Masin, M., Pedrotti, M.-L., Rochelle-Newall, E., Simek, K., and Sime-Ngando, T. (2002). Strong, weak and missing links in a microbial community of the N.W. Mediterranean Sea. FEMS Microbiol. Ecol. 42, 451–462. Binder, B. (1999). Reconsidering the relationship between virally induced bacterial mortality and frequency of infected cells. Aquat. Microb. Ecol. 18, 207–215.

|

83

84

|

Weinbauer et al.

Blackburn, N., Fenchel, T., and Mitchell, J. (1998). Microscale nutrient patches in planktonic habitats shown by chemotactic bacteria. Science 282, 2254–2256. Blasco, R., Murphy, M.J., Sanders, M.F., and Squirrell, D.J. (1998). Specific assays for bacteria using phage mediated release of adenylate kinase. J. Appl. Microbiol. 84, 661–666. Bratbak, G., Thingstad, F., and Heldal, M. (1994). Viruses and the microbial loop. Microb. Ecol. 28, 209–221. Breitbart, M., Felts, B., Kelley, S., Mahaffy, J.M., Nulton, J., Salamon, P., and Rohwer, F. (2004a). Diversity and population structure of a near-shore marine-sediment viral community. Proc. Biol. Sci. 271, 565–574. Breitbart, M., Hewson, I., Felts, B., Mahaffy, J.M., Nulton, J., Salamon, P., and Rohwer, F. (2003). Metagenomic analyses of an uncultured viral community from human feces. J. Bacteriol. 185, 6220– 6223. Breitbart, M., Miyake, J.H., and Rohwer, F. (2004b). Global distribution of nearly identical phage-encoded DNA sequences. FEMS Microbiol. Lett. 236, 249–256. Breitbart, M., and Rohwer, F. (2005). Here a virus, there a virus, everywhere the same virus? Trends Microbiol. 13, 278–284. Breitbart, M., Salamon, P., Andresen, B., Mahaffy, J.M., Segall, A.M., Mead, D., Azam, F., and Rohwer, F. (2002). Genomic analysis of uncultured marine viral communities. Proc. Natl. Acad. Sci. USA 99, 14250–14255. Breitbart, M., Wegley, L., Leeds, S., Schoenfeld, T., and Rohwer, F. (2004c). Phage community dynamics in hot springs. Appl. Environ. Microbiol. 70, 1633–1640. Brussaard, C.P. (2004). Optimization of procedures for counting viruses by flow cytometry. Appl. Environ. Microbiol. 70, 1506–1513. Brussow, H., Canchaya, C., and Hardt, W.D. (2004). Phages and the evolution of bacterial pathogens: from genomic rearrangements to lysogenic conversion. Microbiol. Mol. Biol. Rev. 68, 560–602, table of contents. Cann, A.J., Fandrich, S.E., and Heaphy, S. (2005). Analysis of the virus population present in equine faeces indicates the presence of hundreds of uncharacterized virus genomes. Virus Genes 30, 151–156. Cannon, R.E., ed. (1987). Cyanophage Ecology (New York, Wiley-Interscience). Casjens, S. (2003). Prophages and bacterial genomics: what have we learned so far? Mol. Microbiol. 49, 277–300. Chao, L., Levin, B.R., and Stewart, F.M. (1977). A complex community in a simple habitat: an experimental study with bacteria and phage. Ecology 58, 369–378. Chen, F., and Lu, J. (2002). Genomic sequence and evolution of marine cyanophage P60: a new insight on lytic and lysogenic phage. Appl. Environ. Microbiol. 68, 2589–2594. Chiura, H. (2002). Broad host range xenotrophic gene transfer by virus-like particles from a hot spring. Microbes Environ. 1, 53–58. Chiura, H., and Tagaki, J. (1994). Phage-like particles production and gene transfer by marine bacteria. Bull Jap. Soc. Microb. Ecol. 9, 75–90. Chiura, H., and Umitsu, M. (2004). Isolation and characterization of broad-host range gene transporter particles from geo-thermal vent of the Toyoha mine. Microbes Environ. 19, 20–30. Chiura, H., Velimirov, B., and Kogure, K. (2000). Virus-like particles in microbial population control and horizontal gene transfer in the aquatic environment. In: Microbial Biosystems, C. Bell, M. Brylinsky, and P. Johnson-Green, eds. (Halifax, Atlantic Canada Society for Microbial Ecology), pp. 145–151. Chiura, H.X. (1997). Generalized gene transfer by virus-like particles from marine bacteria. Aquat. Microb. Ecol. 13, 75–83. CIESM (2003). Ecology of Marine Viruses, vol. 21 (Monaco). Clarke, K. (1998). Virus particle production in lysogenic bacteria exposed to protozoan grazing. FEMS Microbiol. Lett. 166, 1777–1180. Clokie, M.R.J., Shan, J., Bailey, S., Jia, Y., Krisch, H.M., West, S., and Mann, N.H. (2006). Transciption of a “photosynthetic” T4-type phage during infection of a marine cyanobacterium. Environ. Microbiol. 8, 827–835. Cochran, P.K., Kellogg, C.A., and Paul, J.H. (1998). Prophage induction of indigenous marine lysogenic bacteria by environmental pollutants. Mar. Ecol. Prog. Ser. 164, 125–133 Comeau, A.M., Buenaventura, E., and Suttle, C.A. (2005). A persistent, productive, and seasonally dynamic vibriophage population within Pacific oysters (Crassostrea gigas). Appl. Environ. Microbiol. 71, 5324–5331.

Bacteriophage in the Environment

Corpe, W., and Jensen, T. (1996). The diversity of bacteria, eukaryotic cells and viruses in an oligotrophic lake. Appl. Microbiol. Biotechnol. 46, 622–630. Danovaro, R., and Corinaldesi, C. (2003). Sunscreen products increase virus production through prophage induction in marine bacterioplankton. Microb. Ecol. 45, 109–118 Danovaro, R., Corinaldesi, C., Dell’Anno, A., Fabiano, M., and Corselli, C. (2005). Viruses, prokaryotes and DNA in the sediments of a deep-hypersaline anoxic basin (DHAB) of the Mediterranean sea. Environ. Microbiol. 7, 586–592. Daubin, V., and Ochman, H. (2004). Bacterial genomes as new gene homes: the genealogy of ORFans in E. coli. Genome Res. 14, 1036–1042. Edwards, R.A., and Rohwer, F. (2005). Viral metagenomics. Nat. Rev. Microbiol. 3, 504–510. Frank, H., and Moebus, K. (1987). An electron microscopic study of bacteriophages from marine waters. Helgoländer Meeresunters 41, 385–414. Fuhrman, J.A. (1999). Marine viruses and their biogeochemical and ecological effects. Nature 399, 541–548. Fuhrman, J.A., and Noble, R.T. (1995). Viruses and protists cause similar bacterial mortality in coastal seawater. Limnol. Oceanogr. 40, 1236–1242. Fuhrman, J.A., and Suttle, C.A. (1993). Viruses in marine planktonic systems. Oceanography 6, 51–63. Fuller, N.J., Wilson, W.H., Joint, I.R., and Mann, N.H. (1998). Occurrence of a sequence in marine cyanophages similar to that of T4 g20 and its application to PCR-based detection and quantification techniques. Appl. Environ. Microbiol. 64, 2051–2060. Garza, D.R., and Suttle, C.A. (1998). The effect of cyanophages on the mortality of Synechococcus spp., and selection for UV resistant viral communities. Microb. Ecol. 36, 281–292. Gérard, E., Guyot, F., Philippot, P., and Lopez-Garcia, P. (2005). Fluorescence in situ hybridisation coupled to ultra small immunogold detection to identify prokaryotic cells using transmission and scanning electron microscopy. J. Microbiol. Methods 63, 20–28. Glud, R.N., and Middelboe, M. (2004). Virus and bacteria dynamics of a coastal sediment: Implication for benthic carbon cycling. Limnol. Oceanog. 49, 2073–2081. Gowing, M.M. (1993). Large virus-like particles from vacuoles of phaeodarian radiolarians and from other marine samples. Mar. Ecol. Prog. Ser. 101, 33–43. Goyal, S., Gerba, C., and Bitton, G., eds. (1987). Phage Ecology (New York, Wiley & Sons). Gruber, F., Falkner, F.G., Dorner, F., and Hammerle, T. (2001). Quantitation of viral DNA by real-time pcr applying duplex amplification, internal standardization, and two-color fluorescence detection. Appl. Environ. Microbiol. 67, 2837–2839. Guixa-Boixareu, N., Calderón-Paz, J.I., Heldal, M., Bratbak, G., and Pedrós-Alió, C. (1996). Viral lysis and bacterivory as prokaryotic loss factors along a salinity gradient. Aquat. Microb. Ecol. 11, 215–227. Guixa-Boixereu, N., Vaqué, D., Gasol, J., Sánchez-Cámara, J., and Pedrós-Alió, C. (2001). Viral distribution and activity in Antarctic waters. Deep Sea Res. II 49, 827–845. Haase, A.T., Retzel, E.F., and Staskus, K.A. (1990). Amplification and detection of lentiviral DNA inside cells. Proc. Natl. Acad. Sci. USA 7, 1874–1878. Hahn, M.W., Lunsdorf, H., Wu, Q., Schauer, M., Hofle, M.G., Boenigk, J., and Stadler, P. (2003). Isolation of novel ultramicrobacteria classified as Actinobacteria from five freshwater habitats in Europe and Asia. Appl. Environ. Microbiol. 69, 1442–1451. Häring, M., Rachel, R., Peng, X., Garrett, R., and Prangishvili, D. (2005a). Viral diversity in hot springs in Pozzuoli, Italy, and characterization of a unique archaeal virus, Acidianus bottle-shaped virus, from a new family, the Ampullaviridae. Appl. Environ. Microbiol. 79, 9904–9911. Häring, M., Vestergaard, G., Rachel, R., Chen, L., Garrett, R., and Prangishvili, D. (2005b). Independant virus development outside a host. Nature 436, 1101–1102. Heldal, M., and Bratbak, G. (1991). Production and decay of viruses in aquatic environments. Mar. Ecol. Prog. Ser. 72, 205–212. Hendrix, R., Smith, M., Burns, R., Ford, M., and Hatfull, G. (1999). Evolutionary relationships among diverse bacteriophages and prophages: all the world’s a phage. Proc. Natl. Acad. Sci. USA 96, 2192– 2197. Hennes, K.P., and Simon, M. (1995). Significance of bacteriophages for controlling bacterioplankton growth in a mesotrophic lake. Appl. Environ. Microbiol. 61, 333–340. Hennes, K.P., Suttle, C.A., and Chan, A.M. (1995). Fluorescently labeled virus probes show that natural virus populations can control the structure of marine microbial communities. Appl. Environ. Microbiol. 61, 3623–3627.

|

85

86

|

Weinbauer et al.

Hewson, I., Vargo, G.A., and Fuhrman, J.A. (2003). Bacterial diversity in shallow oligotrophic marine benthos and overlying waters: effects of virus infection, containment, and nutrient enrichment. Microb. Ecol. 46, 322–336. Hutchinson, G. (1957). Concluding remarks. Paper presented at: Cold Spring Harbor Symposium on Quantitatve Biology. Hutchinson, G. (1961). The paradox of the plankton. Am. Nature 95, 137–145. Jensen, E., Schrader, H., Rieland, B., Thompson, T., Lee, K., Nickerson, K., and Kokjohn, T. (1998). Prevalence of broad-host-range lytic bacteriophages of Spherotilus natans, Escherichia coli, and Pseudomonas aeruginosa. Appl. Environ. Microbiol. 64, 575–580. Jiang, S., Steward, G., Jellison, R., Chu, W., and Choi, S. (2004). Abundance, distribution, and diversity of viruses in alkaline, hypersaline Mono Lake, California. Microb. Ecol. 47, 9–17. Jiang, S.C., and Paul, J.H. (1996). Occurrence of lysogenic bacteria in marine microbial communities as determined by prophage induction. Mar. Ecol. Prog. Ser. 142, 27–38 Jiang, S.C., and Paul, J.H. (1998). Gene transfer by transduction in the marine environment. Appl. Environ. Microbiol. 64, 2780–2787. Karner, M.B., DeLong, E.F., and Karl, D.M. (2001). Archaeal dominance in the mesopelagic zone of the Pacific Ocean. Nature 409, 507–510. Kellogg, C.A., Rose, J.B., Jiang, S.C., Thurmond, J.M., and Paul, J.H. (1995). Genetic diversity of related vibriophages isolated from marine environments around Florida and Hawaii, USA. Mar. Ecol. Prog. Ser. 120, 89–98. Kellogg, C.A., and Paul, J.H. (2002). Degree of ultraviolet radiation damage and repair capabilities are related to G+C content in marine vibriophages. Aquat. Microb. Ecol. 27, 13–20. Klieve, A., and Swain, R. (1993). Estimation of ruminal bacteriophage numbers by pulsed-field gel electrophoresis and laser densitometry. Appl. Environ. Microbiol. 59, 2299–2303. Kokjohn, T.A., Sayler, G.S., and Miller, R.V. (1991). Attachment and replication of Pseudomonas aeruginosa bacteriophages under conditions simulating aquatic environments. J. Gen. Microbiol. 137, 661–666. Labrie, S., and Moineau, S. (2000). Multiplex PCR for detection and identification of lactococcal bacteriophages. Appl. Environ. Microbiol. 66, 987–994. Lawrence, J., Chan, A., and Suttle, C. (2002). Viruses causing lysis of the toxic bloom-forming Heterosigma akashiwo (Raphidophyceae) are widespread in coastal sediments of British Columbia, Canada. Limnol. Oceanogr. 47, 545–550. Legendre, L., and Le Fèvre, J. (1995). Microbial food webs and the export of biogenic carbon in oceans. Aquat. Microb. Ecol. 9, 69–77. Lenski, R.E. (1988). Dynamics of interactions between bacteria and virulent bacteriophage. Adv. Microb. Ecol. 10, 1–44. Levin, B.R., Steward, F.M., and Chao, L. (1977). Resource-limited growth, competition, and predation: a model and experimental studies with bacteria and bacteriophage. Am. Nature 111, 3–24. Lindell, D., Sullivan, M.B., Johnson, Z.I., Tolonen, A.C., Rohwer, F., and Chisholm, S.W. (2004). Transfer of photosynthesis genes to and from Prochlorococcus viruses. Proc. Natl. Acad. Sci. USA 101, 11013–11018. Lopez-Pila, J.M., Krawelitzki, B., Hachula, G., Sheler, C., and Kopecka, H. (1993). Detection of viruses in water and soil by molecular hybridisation techniques and by PCR: pitfalls and limitations. In: Trends in Microbial Ecology, R. Guerrero, and C. Pedros-Alio, eds. (Barcelona, Spanish Society for Microbiology), pp. 595–600. Mann, N.H., Cook, A., Millard, A., Bailey, S., and Clokie, M. (2003). Marine ecosystems: bacterial photosynthesis genes in a virus. Nature 424, 741. Maranger, R., Bird, D.F., and Juniper, S.K. (1994). Viral and bacterial dynamics in Arctic sea ice during the spring algal bloom near resolute, N.W.T., Canada. Mar. Ecol. Prog. Ser. 111, 121–127. Maranger, R., and Bird, D.F. (1995). Viral abundance in aquatic systems: a comparison between marine and fresh waters. Mar. Ecol. Prog. Ser. 121, 217–226. Marie, D., Brussaard, C., Thyrhaug, R., Bratbak, G., and Vaulot, D. (1999). Enumeration of marine viruses in culture and natural samples by flow cytometry. Appl. Environ. Microbiol. 65, 45–52. Mehta, A., Maggioncalda, J., Bagasra, O., Thikkavarapu, S., Saikumari, P., Valyi-Nagy, T., Fraser, N.W., and Block, T.M. (1995). In situ DNA PCR and RNA hybridization of herpes simplex virus sequences in trigeminal ganglia of latently infected mice. Virology 206, 633–640.

Bacteriophage in the Environment

Middelboe, M., Hagström, A., Blackburn, N., Sinn, B., Fischer, U., Borch, N., Pinhassi, J., Simu, K., and Lorenz, M. (2001). Effects of bacteriophages on the population dynamics of four strains of pelagic marine bacteria. Microb. Ecol. 42, 395–406. Middelboe, M., Jørgensen, N.O. G., and Kroer, N. (1996). Effects of viruses on nutrient turnover and growth efficiency of noninfected marine bacterioplankton. Appl. Environ. Microbiol. 62, 1991–1997. Middelboe, M., and Lyck, P. (2002). Regeneration of dissolved organic matter by viral lysis in marine microbial communities. Aquat. Microb. Ecol. 27, 187–194. Moebus, K. (1983). Lytic and inhibition responses to bacteriophages among marine bacteria, with special reference to the origin of phage–host systems. Helgoländer Meeresunters 36, 375–391. Moebus, K., and Nattkemper, H. (1981). Bacteriophage sensitivity patterns among bacteria isolated from marine waters. Helgoländer Meeresunters 34, 375–385. Moebus, K., and Nattkemper, H. (1983). Taxonomic investigations of bacteriophage sensitive bacteria isolated from marine waters. Helgoländer Meeresunters 36, 357–373. Mühling, M., Fuller, N.J., Millard, A., Somerfield, P.J., Marie, D., Wilson, W.H., Scanlan, D.J., Post, A.F., Joint, I., and Mann, N.H. (2005). Genetic diversity of marine Synechococcus and co-occurring cyanophage communities: evidence for viral control of phytoplankton. Environ. Microbiol. 7, 499–508 Murray, A.G., and Jackson, G.A. (1992). Viral dynamics: a model of the effects of size, shape, motion and abundance of single-celled planktonic organisms and other particles. Mar. Ecol. Prog. Ser. 89, 103–116. Nelson, D. (2004). Phage taxonomy: we agree to disagree. J. Bacteriol. 186, 7029–7031. Noble, R., and Fuhrman J. (1998). Use of SYBR Green I for rapid epifluorescence counts of marine viruses and bacteria. Aquat. Microb. Ecol. 14, 113–118 Noble, R., and Fuhrman, J. (1999). Breakdown and microbial uptake of marine viruses and other lysis products. Aquat. Microb. Ecol. 20, 1–11. Noble, R.T., and Fuhrman, J.A. (2000). Rapid virus production and removal as measured with fluorescently labeled viruses as tracers. Appl. Environ. Microbiol. 66, 3790–3797. Nuovo, G.J., Margiotta, M., MacConnell, P., and Becker, J. (1992b). Rapid in situ detection of PCR-amplified HiV-1 DNA. Diagnostic Mol. Pathol. 1, 98–102. Ogunseitan, O.A., Sayler, G.S., and Miller, R.V. (1992). Application of DNA probes to analysis of bacteriophage distribution patterns in the environment. Appl. Environ. Microbiol. 58, 2046–2052. Oren, A., Bratbak, G., and Heldal, M. (1997). Occurrence of virus-like particles in the Dead Sea. Extremophiles 1, 143–149. Ortmann, A.C., and Suttle, C.A. (2005). High abundances of viruses in a deep-sea hydrothermal vent system indicates viral mediated microbial mortality. Deep Sea Research Part I: Oceanographic Research Papers 52, 1515–1527. Patterson, B.K., Till, M., Otto, P., Goolsby, C., Furtado, M.R., McBride, L.J., and Wolinsky, S.M. (1993). Detection HiV-1 DNA and messenger RNA in individual cells by PCR-driven in situ hybridization and flow cytometry. Science 260, 976–979. Paul, J.H. (1999). Microbial gene transfer. J. Mol. Microbiol. Biotechnol. 1, 45–50. Paul, J. (2000). Ecology of bacteriophages in nature. In: Viral Ecology, C. Hurst, ed. (San Diego, Academic Press), pp. 211–246. Paul, J.H., Sullivan, M.B., Segall, A.M., and Rohwer, F. (2002). Marine phage genomics. Comp. Biochem. Physiol. B Biochem. Mol. Biol. 133, 463–476. Peduzzi, P., and Weinbauer, M.G. (1993). The submicron size fraction of seawater containing high numbers of virus particles as bioactive agent in unicellular plankton community successions. J. Plankton Res. 15, 1375–1386. Pernthaler, J. (2005). Predation on prokaryotes in the water column and its ecological implications. Nature Rev. 3, 537–546 Proctor, L. (1997). Advances in the study of marine viruses. Microsc. Res. Tech. 37, 136–161. Proctor, L.M., and Fuhrman, J.A. (1990). Viral mortality of marine bacteria and cyanobacteria. Nature 343, 60–62. Proctor, L.M., and Fuhrman, J.A. (1991). Roles of viral infection in organic particle flux. Mar. Ecol. Prog. Ser. 69, 133–142. Proctor, L.M., Okubo, A., and Fuhrman, J.A. (1993). Calibrating estimates of phage-induced mortality in marine bacteria: ultrastructural studies of marine bacteriophage development from one-step growth experiments. Microb. Ecol. 25, 161–182.

|

87

88

|

Weinbauer et al.

Puig, A., Jofre, J., and Araujo, R. (1998). Use of oxyrase enzyme (Oxyrase®) for the detection of bacteriophages of Bacteroides fragilis in aerobic incubation conditions. J. Microbiol. Methods 31, 205–207. Puig, M., Jofre, J., and Girones, R. (2000a). Detection of phages infecting Bacteroides fragilis HSP40 using a specific DNA probe. J. Virol. Methods 88, 163–173. Puig, M., Pina, S., Lucena, F., Jofre, J., and Girones, R. (2000b). Description of a DNA amplification procedure for the detection of bacteriophages of Bacteroides fragilis HSP40 in environmental samples. J. Virol. Methods 89, 159–166. Rappé, M.S., Connon, S.A., Vergin, K.L., and Giovannoni, S.J. (2002). Cultivation of the ubiquitous SAR11 marine bacterioplankton clade. Nature 418, 630–633. Riemann, L., and Middelboe, M. (2002). Viral lysis of marine bacterioplankton: implications for organic matter cycling and bacterial clonal composition. Ophelia 56, 57–68. Riemann, L., Steward, G., Fandino, L., Campbell, L., Landry, M., and Azam, F. (1999). Bacterial community composition during two consecutive NE Monsoon periods in the Arabian Sea studied by denaturing gradient gel electrophoresis (DGGE) of rRNA genes. Deep-Ses Res. Part II 46, 1791–1811. Riemann, L., Steward, G.F., and Azam, F. (2000). Dynamics of bacterial community composition and activity during a mesocosm diatom bloom. Appl. Environ. Microbiol. 66, 578–587. Ripp, S., and Miller, R.V. (1995). Effects of suspended particles on the frequency of transduction among Pseudomonas aerugonosa in a freshwater environment. Appl. Environ. Microbiol. 61, 1214–1219. Rohwer, F. (2003). Global phage diversity. Cell 113, 141. Rohwer, F., and Edwards, R. (2002). The Phage Proteomic Tree: a genome-based taxonomy for phage. J. Bacteriol. 184, 4529–4535. Rohwer, F., Segall, A., Steward, G., Seguritan, V., Breitbart, M., Wolven, F., and Azam, F. (2000). The complete genomic sequence of marine phage Roseophage SIO1 shares homology with nonmarine phages. Limnol. Oceanogr. 45, 408–418. Sanger, F., Air, G.M., Barrell, B.G., Brown, N.L., Coulson, A.R., Fiddes, C.A., Hutchison, C.A., Slocombe, P.M., and Smith, M. (1977). Nucliotide sequence of bacteriophage phi X174 DNA. Nature 265, 687–695. Sano, E., Carlson, S., Wegley, L., and Rohwer, F. (2004). Movement of viruses between biomes. Appl. Environ. Microbiol. 70, 5842–5846. Schrader, H.S., Schrader, J.O., Walker, J.J., Bruggeman, N., B., Vanderloop, J.M., Shaffer, J.J., and Kokjohn, T.A. (1995). Effects of host starvation on bacteriophage dynamics. In: Bacteria in Oligotrophic Environments, R.Y. Morita, ed. (New York, Chapman & Hall), pp. 368–385. Schrader, H.S., Schrader, J.O., Walker, J.J., Wolf, T.A., Nickerson, K.W., and Kokjohn, T.A. (1997). Bacteriophage infection and multiplication occur in Pseudomonas aerigunosa starved for 5 years. Can J. Microbiol. 43, 1157–1163. Schwalbach, M., Hewson, I., and Fuhrman, J. (2004). Viral effects on bacterial community composition in marine plankton microcosms. Aquat. Microb. Ecol. 34, 117–127. Shaffer, J.S., Jacobson, L.M., Schrader, J.O., Lee, K.W., Martin, E.L., and Kokjohn, T.A. (1999). Characterization of Pseudomonas aeruginosa bacteriophage UNL-1, a bacterial virus with a novel UVA-inducible DNA damage reactivation phenotype. Appl. Environ. Microbiol. 65, 2606–2613. Shibata, A., Kogure, K., Koike, I., and Ohwada, K. (1997). Formation of submicron colloidal particles from marine bacteria by viral infection. Mar. Ecol. Prog. Ser. 155, 303–307. Short, C.M., and Suttle, C.A. (2005). Nearly identical bacteriophage structural gene sequences are widely distributed in both marine and freshwater environments. Appl. Environ. Microbiol. 71, 480–486. Short, S., and Suttle, C. (1999). Use of the polymerase chain reaction and denaturing gradient gel electrophoresis to study diversity in natural virus communities. Hydrobiologia 401, 19–33. Simek, K., Horak, K., Mazin, M., Christaki, U., Nedoma, J., Weinbauer, M., and Dolan, J. (2003). Comparing the effects of resource enrichment and grazing on a bacterioplankton community of a meso-eutrophic reservoir. Aquat. Microb. Ecol. 31, 123–135. Simek, K., Pernthaler, J., Weinbauer, M., Hornák, K., Dolan, J., Nedoma, J., Masin, M., and Amann, R. (2001). Changes in bacterial community composition, dynamics and viral mortality rates associated with enhanced flagellate grazing in a meso-eutrophic reservoir. Appl. Environ. Microbiol. 67, 1723–2733. Sobsey, M.D. (1993). The use of molecular techniques for the detection of viruses in environmental samples. In: Trends in Microbial Ecology, R. Guerrero, and C. Pedros-Alio, eds. (Barcelona, Spanish Society for Microbiology), pp. 601–606.

Bacteriophage in the Environment

Steward, G.F., Wikner, J., Smith, D.C., Cochlan, W.P., and Azam, F. (1992). Estimation of virus production in the sea: I. method development. Mar. Microb. Food Webs 6, 57–78. Straub, T.M., Pepper, I.L., Abbaszadegan, M., and Gerba, C.P. (1994). A method to detect enteroviruses in sewage sludge-amended soil using the PCR. Appl. Environ. Microbiol. 60, 1014–1017. Suttle, C. (2000a). Cyanophages and their role in the ecology of cyanobacteria. In: The Ecology of Cyanobacteria, B. Whitton, and M. Potts, eds. (The Netherlands, Kluwer Acadamic Publishers), pp. 563–589. Suttle, C. (2000b). Ecological, evolutionary, and geochemical consequences of viral infection of cyanobacteria and eukaryotic algae. In: Viral Ecology, C. Hurst, ed. (San Diego, Academic Press), pp. 247–296. Suttle, C.A. (2005). Viruses in the sea. Nature 437, 356–361. Suttle, C.A. (2005). The viriosphere: the greatest biological diversity on Earth and driver of global processes. Environ. Microbiol. 7, 481–482. Suttle, C.A., Chan, A.M., and M.T. Cottrell, M.T. (1991). Use of ultrafiltration to isolate viruses from seawater which are pathogens of marine phytoplankton. Appl. Environ. Microbiol. 57, 721–726. Suttle, C.A., and Chan, A.M. (1994). Dynamics and distribution of cyanophages and their effect on marine Synechococcus spp. Appl. Environ. Microbiol. 60, 3167–3174. Suttle, C.A., Chan, A.M., and Cottrell, M.T. (1990). Infection of phytoplankton by viruses and reduction of primary productivity. Nature 347, 467–469. Suttle, C.A., Chan, A.M., Short, S.M., Weinbauer, M.G., and Wilhelm, S.W. (1996). The effect of cyanophages on Synechococcus during a bloom in the Western Gulf of Mexico. Ocean Sciences MeetingAmerican Geophysical Union, San Diego CA. Eos Transaction 76, 0S207–0S208 Swain, R.S., Nolan, J.V., and Klieve, A.V. (1996). Natural variability and diurnal fluctuations within the bacteriophage population of the rumen. Appl. Environ. Microbiol. 62, 994–997. Thingstad, T. (2000). Elements of a theory for the mechanisms controlling abundance, diversity, and biogeochemical role of lytic bacterial viruses in aquatic systems. Limnol. Oceanogr. 45, 1320–1328. Thingstad, T.F., and Lignell, R. (1997). Theoretical models for the control of bacterial growth rate, abundance, diversity and carbon demand. Aquat. Microb. Ecol. 13, 19–27. Torsvik, V., Ovreas, L., and Thingstad, T.F. (2002). Prokaryotic diversity-magnitude, dynamics, and controlling factors. Science 296, 1064–1066. Tsai, Y.L., Sobsey, M.D., Sangermano, L.R., and Palmer, C.J. (1993). Simple method of concentrating enteroviruses and hepatitis A from sewage and ocean water for rapid detection by reverse transcriptase-polymerase chain reaction. Appl. Environ. Microbiol. 59, 3488–3491. Tsai, Y.L., Tran, B., and Palmer, C.J. (1995). Analysis of viral RNA persistence in seawater by transcriptase PCR. Appl. Environ. Microbiol. 61, 363–366. Tsai, Y.L., Tran, B., Sangermano, L.R., and Palmer, C.J. (1994). Detection of poliovirus, hepatitis A virus and rotavirus from sewage and ocean water by triplex reverse transcriptase PCR. Appl. Environ. Microbiol. 60, 2400–2407. Van Hannen, E.J., Zwart, G., van Agterveld, M.P., Gons, H.J., Ebert, J., and Laanbroek, H.J. (1999). Changes in bacterial and eukaryotic community structure after mass lysis of filamentous cyanobacteria associated with viruses. Appl. Environ. Microbiol. 65, 795–801. Venter, J.C., Remington, K., Heidelberg, J.F., Halpern, A.L., Rusch, D., Eisen, JA., Wu, D., Paulsen, I., Nelson, K.E., Nelson, W., Fouts, D.E., Levy, S., Knap, A.H., Lomas, M.W., Nealson, K., White, O., Peterson, J., Hoffman, J., Parsons, R., Baden-Tillson, H., Pfannkoch, C., Rogers, Y.H., and Smith, H.O. (2004). Environmental genome shotgun sequencing of the Sargasso Sea. Science 304, 66–74. Vettori, C., Stotzky, G., Yoder, M., and Gallori, E. (1999). Interaction between bacteriophage PBS1 and clay minerals and transduction of Bacillus subtilis by clay-phage complexes. Environ. Microbiol. 1, 347–355. Weinbauer, M. (2004). Ecology of prokaryotic viruses. FEMS Microbiol. Rev. 28, 127–181. Weinbauer, M., Brettar, I., and Höfle, M. (2003a). Lysogeny and virus-induced mortality of bacterioplankton in surface, deep, and anoxic waters. Limnol. Oceanogr. 48, 1457–1465. Weinbauer, M., Nedoma, J., Christaki, U., and Simek, K. (2003b). Comparing the effects of resource enrichment and grazing on viral production in a meso-eutrophic reservoir. Aquat. Microb. Ecol. 31, 137–144. Weinbauer, M., Winter, C., and Höfle, M. (2002). Reconsidering transmission electron microscopy based estimates of viral infection of bacterioplankton using conversion factors derived from natural communities. Aquat. Microb. Ecol. 27, 103–110.

|

89

90

|

Weinbauer et al.

Weinbauer, M.G., Fuks, D., and Peduzzi, P. (1993). Distribution of viruses and dissolved DNA along a coastal trophic gradient in the northern Adriatic Sea. Appl. Environ. Microbiol. 59, 4074–4082. Weinbauer, M.G., and Höfle, M.G. (1998). Significance of viral lysis and flagellate grazing as factors controlling bacterioplankton production in a eutrophic lake. Appl. Environ. Microbiol. 64, 431–438. Weinbauer, M.G., and Rassoulzadegan, F. (2004). Are viruses driving microbial diversification and diversity? Environ. Microbiol. 6, 1–11. Weinbauer, M.G., and Suttle, C.A. (1996). Potential significance of lysogeny to bacteriophage production and bacterial mortality in coastal waters of the Gulf of Mexico. Appl. Environ. Microbiol. 62, 4374–4380. Weinbauer, M.G., Wilhelm, S.W., Garza, D.R., and Suttle, C.A. (1997). Photoreactivation compensates for UV damage and restores infectivity to natural marine viral communities. Appl. Environ. Microbiol. 63, 2200–2205. Weinbauer, M.G., Wilhelm, S.W., Pledger, R., Mitchell, D., and Suttle, C.A. (1999). Sunlight-induced DNA damage and resistance in natural virus communities. Aquat. Microb. Ecol. 17, 111–120. Wen, K., Ortmann, A.C., and Suttle, C.A. (2004). Accurate estimation of viral abundance by epifluorescence microscopy. Appl. Environ. Microbiol. 70, 3862–3867. Wichels, A., Gerdts, G., and Schütt, C. (2002). Pseudoalteromonas spp. phages, a significant group of marine bacteriophages in the North Sea. Aquat. Microb. Ecol. 27, 233–239. Wikner, J., Vallini, J.J., Steward, G.F., Smith, D.C., and Azam, F. (1993). Nucleic acids from the host bacterium as amajor source of nucleotides for three marine bacteriophages. FEMS Microbiol. Ecol. 12, 237–248. Wilcox, R.M., and Fuhrman, J.A. (1994). Bacterial viruses in coastal seawater: lytic rather than lysogenic production. Mar. Ecol. Prog. Ser. 114, 35–45. Wilhelm, S., Brigden, S., and Suttle, C. (2002). A dilution technique for the direct measurement of viral production: a comparison in stratified and tidally mixed coastal waters. Microb. Ecol. 43, 168–173. Wilhelm, S., and Suttle, C. (1999). Viruses and nutrient cycling in the sea. Bioscience 49, 781–788. Wilhelm, S.W., Weinbauer, M.G., Suttle, C.A., and Jeffrey, W.H. (1998a). The role of sunlight in the removal and repair of viruses in the sea. Limnol. Oceanogr. 43, 586–592. Wilhelm, S.W., Weinbauer, M.G., Suttle, C.A., Pledger, R.J., and Mitchell, D.L. (1998b). Measurements of DNA damage and photoreactivation imply that most viruses in marine surface waters are infective. Aquat. Microb. Ecol. 14, 215–222. Williamson, K.E., Radosevich, M., and Wommack, K.E. (2005). Abundance and diversity of viruses in six Delaware soils. Appl. Environ. Microbiol. 71, 3119–3125. Williamson, K.E., Wommack, K.E., and Radosevich, M. (2003). Sampling natural viral communities from soil for culture-independent analyses. Appl. Environ. Microbiol. 69, 6628–6633. Wilson, W.H., Fuller, N.J., Joint, I.R., and Mann, N.H. (2000). Analysis of cyanophage diversity in the marine environment using denaturing gradient gel electrophoresis. In: Microbial Biosystems, C. Bell, M. Brylinsky, and P. Johnson-Green, eds. (Halifax, Atlantic Canada Society for Microbial Ecology), pp. 565–570. Winter, C., Herndl, G., and Weinbauer, M.G. (2004a). Diel cycles in viral infection of bacterioplankton in the North Sea. Aquat. Microb. Ecol. 35, 207–216. Winter, C., Smit, A., Herndl, G., and Weinbauer, M. (2004b). Impact of virioplankton on archaeal and bacterial community richness in seawater batch cultures. Appl. Environ. Microbiol. 70, 804–813. Winter, C., Smit, A., Herndl, G.J., and Weinbauer, M.G. (2005). Linking bacterial richness with viral abundance and prokaryotic activity. Limnol. Oceanogr. 50, 968–977. Wolf, A., Wiese, J., Jost, G., and Witzel, K.P. (2003). Wide geographic distribution of bacteriophages that lyse the same indigenous freshwater isolate (Sphingomonas sp. strain B18). Appl. Environ. Microbiol. 69, 2395–2398. Wommack, K.E., Bench, S.R., Williamson, K.E., and Radosevich, M. (2004). Viruses in soils: The first terrestrial viral metagenome. 10th International Symposium in Microbial Ecology, Cancun, Mexico. Wommack, K.E., and Colwell, R.R. (2000). Virioplankton: Viruses in aquatic ecosystems. Microbiol. Mol. Biol. Rev. 64, 69–114. Wommack, K.E., Ravel, J., Hill, R.T., and Colwell, R.R. (1999a). Hybridization analysis of Chesapeake Bay virioplankton. Appl. Environ. Microbiol. 65, 241–250.

Bacteriophage in the Environment

Wommack, K.E., Ravel, J., Holl, R.T., Chun, J., and Colwell, R.R. (1999b). Population dynamics of Chesapeake Bay virioplankton: total-community analysis by pulse-field gel electrophoresis. Appl. Environ. Microbiol. 65, 231–240. Zhong, Y., Chen, F., Wilhelm, S., Poorvin, L., and Hodson, R. (2002). Phylogenetic diversity of marine cyanophage isolates and natural virus communities as revealed by sequences of viral capsid assembly protein gene g20. Appl. Environ. Microbiol. 68, 1576–1584. Zillig, W., Prangishvilli, D., Schleper, C., Elferink, M., Holz, I., Albers, S., Janekovic, D., and Götz, D. (1996). Viruses, plasmids and other genetic elements of thermophilic and hyperthermophilic Archaea. FEMS Microbiol. Rev. 18, 225–236.

|

91

Bacteriophages and Food Fermentations

4

Eric Emond and Sylvain Moineau

Abstract A broad number of food products, commodity chemicals, and biotechnology products are manufactured industrially by large-scale bacterial fermentation of various organic substrates. Because enormous amounts of bacteria are being cultivated each day in large fermentation vats, the risk that bacteriophage contamination rapidly brings fermentations to a halt and cause economical setbacks is a serious threat in these industries. This chapter describes the relationship between bacteriophages and their bacterial hosts in the context of the food fermentation industry. Sources of phage contamination, measures to control their propagation and dissemination, and biotechnological defense strategies developed to restrain phages are discussed. The primary focus will be given to the dairy fermentation industry because it has openly acknowledged the problem of phage and has been working with academia and starter culture companies to develop defense strategies and systems to curtail the propagation and evolution of phages for decades. Other industries will be discussed where appropriate to highlight their similarities and specificities. Introduction Looking at the microbial world is always a fascinating and mind-blowing experience. The observer rapidly realizes that bacteria are among the most prolific organisms on earth. Their incredible biodiversity and metabolic capabilities allow them to grow and prosper in every ecosystem where life forms have been found, including extreme environments such as deepsea thermal vents, and sulfur-rich volcanic geysers. In favorable ecological niches where nutrients and co-factors are available and abundant, such as the gut of mammals, the bacterial populations are so dense that their numbers exceed (over a 10-fold) the total number of cells constituting the body of their mammalian hosts (Whitman et al., 1998; Wommack and Colwell, 2000; Breitbart et al., 2005). Given their ubiquity, one can ask why bacteria are not out-competing all other life forms on earth? Part of the answer resides in the existence of bacteriophages. Bacteriophages or “phages” are viruses that infect bacteria (Eubacteria and Archaea). These prokaryotic viruses are present in ecosystems where bacteria have been found, including man-made ecological niches such as food fermentation vats. They may not have been found in all ecosystems yet, but it is simply a matter of looking for them using the appropriate methods. It is believed that they represent the most abundant biological

94

|

Emond and Moineau

entities in the world with estimates ranging from 1030 to 1032 total phage particles on earth. It is generally assumed that they outnumber bacteria about 10-fold (Whitman et al., 1998; Breitbart et al., 2005). To illustrate the astronomical proportion of these numbers, one can imagine that if phages (approx. 250 nm in length) would be laid end to end, they would form a bridge of such proportions that it would be possible to reach Alpha Centauri, the star nearest to our solar system (4.35 light years) 46 million times (Hendrix, 2005)! Without a doubt, bacteriophages play a key role in regulating bacterial populations in all habitats. Phage particles disseminate freely in environmental fluids where they make contact with susceptible bacterial hosts and initiate an infection cycle. If a phage does not meet an appropriate host, it eventually becomes inactivated by exposure to environmental stresses such as temperature, UV light, pH, etc. Phages that undergo a lytic cycle will generally release a progeny ranging from a few to hundreds of phage particles per infected cell within a timeframe ranging from minutes to a day. With such explosive proliferations, phages could rapidly eradicate entire populations of bacteria. But in nature, a large pool of bacterial species and their respective phages are involved in continuous cycles of co-evolution where phage-resistant mutants help to preserve bacterial cell lineages from extinction and counter-resistant mutant phages threaten new bacterial strains. These co-evolution cycles, as well as the balance between phage production and phage decay, maintain these microorganisms in a constant state of flux where the rapid growth of one group is followed by the burst of the other, thus establishing a relative homeostasis and quick turnover of the populations, often modulated by environmental factors. Nowadays, many bio-industries rely on bacteria for the large-scale fermentation of various substrates. The conditions prevailing in these industrial environments represent unique ecological niches that favor phage proliferation. Fermentation processes usually involve large volumes (up to millions of liters) and high-density cultures containing one or a few selected bacterial strains. In other cases, mixed starters, composed of undefined numbers and ratios of bacterial strains, are used to drive the fermentation process (Stahouders and Leenders, 1984; Bissonnette et al., 2000). These cultures are maintained in exponential phase of growth for extended periods of time and the manufacturing process is regularly repeated over and over in successive batches with the same strains. These conditions are ideal for the propagation of bacteriophage, and contamination of fermentation vessels by virulent phages has dramatic consequences. It is therefore not surprising that fermentation industries live in fear of phage infection and deploy major efforts to prevent and limit contamination. Despite all the control points, bacteriophages eventually find their way into the fermentation vats and their explosive multiplication rapidly leads to delays or complete arrest of the fermentation process. The economical setbacks can be substantial and include product loss, raw material spoilage, non-productive operation costs, and plant shutdown for decontamination purposes. Most industry sectors relying on fermentation of bacteria have experienced, at various extents, problems with bacteriophages and are constantly waging war against these viruses to keep them under control (Bogosian, 2005). The objective of this chapter is to provide the reader with a view of the dynamics of phage–host interactions in the fermentation industry, and to provide comprehensive information on the methods and strategies that were developed to help control phage problems in industrial settings.

Bacteriophages and Food Fermentations

Phages biology Phage structure Bacteriophages are composed of a nucleic acid—mostly double-stranded DNA but also single-stranded DNA, single-stranded RNA, and double-stranded RNA—enclosed in a protein or lipoprotein shell named “the capsid.” Some phages have small genomes of a few thousand base pairs (bp) encoding 12 or fewer genes (Calendar and Inman, 2005), and others have a genome as large as 480 000 bp. Phages are heterogeneous in their morphology and they can be polyhedral, filamentous, or pleomorphic. Phages with a polyhedral capsid often carry a tail “the Caudovirales order,” to which many appendages such as whiskers, base plate, spikes, and tail fibers can be attached (Figure 4.1). The tails can be contractile “the Myoviridae family,” long and non-contractile “the Siphoviridae family,” or short and non-contractile “the Podoviridae family.” The virulent phages of the Caudovirales order are by far the largest and most predominant group of bacteriophages (Ackermann, 2005). Accordingly, they are also relevant to food fermentations and will be the main focus of this chapter. For

Figure 4.1 Electron micrograph (upper panel) and schematic representation (lower panel) of representative phages of the Caudovirales order. The phage T4 of E. coli has a prolate head and a long contractile tail (Myoviridae); the temperate phage TP901-1 (P335 species) of L. lactis has an isometric head and a long non-contractile tail (Siphoviridae); and the virulent phage KSY1 of L. lactis has a rare morphology characterized by an elongated head and a short non-contractile tail (Podoviridae).

|

95

96

|

Emond and Moineau

a more detailed description of phage structures and their classification, the reader should refer to Ackermann (1999). Phage life cycle A common characteristic of phages is that, although their genome carries the information required to drive their own multiplication, they rely on the energy and the protein biosynthetic machinery of their bacterial hosts to complete their lytic cycle. Hence, phages are obligatory parasites. The capsid protects the genetic material of the phage against environmental damage while the distal part of the tail of Caudovirales phages is providing the specific mechanism required to recognize susceptible hosts. Once, by chance, a phage meets an appropriate host, the infection starts by the specific binding of receptor-binding protein (the key) and the receptor (the lock), located on the host’s surface. Hosts receptors include structures such as proteins [Pip/(Valyasevi et al., 1991); LamB/(Randall-Hazelbauer and Schwartz, 1973); OmpC/(Yu and Mizushima, 1982); FhuA/(Braun et al., 1973)], lipopolysaccharides (Heller and Braun, 1979) or sugars anchored to the cell membrane or part of the cell wall structure (Dupont et al., 2004; Geller et al., 2005; Spinelli et al., 2006; Tremblay et al., 2006). Ensues a rapid reorganization of the bacterial cell wall and phage structures, with concomitant translocation of the phage genome into the bacterial cytoplasm. Once the nucleic acid has entered into the cytoplasm, the phage becomes an active intracellular parasite. Three different scenarios are then possible that depend on the genetic information encoded by the phage and the physiological status of the host at the time of infection. They include: (1) the lytic cycle that results in host lysis and concomitant release of newly assembled phage particles (2) the intracellular chronic cycle that results in processive release of progeny phages without killing the host bacterium (Maniloff et al., 1981; Calendar and Inman, 2005), and (3) the lysogenic cycle, where infections result in the integration of the phage genome into host DNA or maintenance of the phage nucleic acid as a self-replicating circular plasmid in the cytoplasm (Little, 2005). The lytic cycle Virulent phages are a group of bacterial viruses that can only undergo lytic infection cycles. In conditions where susceptible hosts are abundant and grow actively, like those found in industrial fermentation vessels, the populations of virulent phages proliferate at an incredible speed. It is thus not surprising that most, if not all of the phages causing food fermentation problems are virulent. Upon entry of the virulent phage DNA into the cell, the host undergoes major physiological changes. A cascade of finely tuned sequential events orchestrated by the phage and driven towards the production of phage progeny takes place. The majority of the host cellular processes are interrupted with the exception of those required for the production of phage particles. Phage DNA is replicated massively, genes encoded by the phage are being orderly transcribed and translated into regulatory and structural proteins. Thereafter, the structural constituents of the phage are being irreversibly assembled into precursor polymers and gradually processed into mature phage particles. The newly assembled phage particles are eventually released as the cell wall is being disrupted by the dual action of the phage-encoded lysis system, often made of a holin and an endolysin. In most

Bacteriophages and Food Fermentations

cases of industrial fermentation infection, the lytic cycle takes 30 to 60 min to complete and each infected cell release a phage progeny ranging from 10 to 500 virions. The chronic infection cycle The intracellular chronic cycle has been best described for filamentous phages of Escherichia coli (Maniloff et al., 1981; Calendar and Inman, 2005). Chronic infection is particular in that the production of phages particles does not result in cell lysis and death. Instead, phage precursors consisting of a single-stranded DNA molecule coated with a phage singlestranded DNA binding protein are being packaged into mature phage particles while they are being extruded from the cell (Rakonjac et al., 1999). The extrusion process, driven by phage-encoded proteins located in the cell envelope, preserves the integrity of the cell membranes of the host (Snyder and Champness, 2003). Extrusion of phages continues for extended periods while cells grow, albeit slowly. Phages with filamentous morphology were isolated during a survey of phages active on Propionibacterium freudenreichii (Chopin et al., 2002), a bacterium widely used in Swiss-type cheese manufacture. However at this time, filamentous phages are not significant in industrial food fermentations processes and will not be further described here. The readers are referred to the following reference to learn more about this type in infection (Russel and Model, 2005). The lysogenic cycle Another group of bacterial viruses named “temperate phages” are capable of two alternate lifestyles: lytic and lysogenic. Upon infection and DNA translocation into the host cytoplasm, a decision is quickly made to undertake one or the other route. If the phage undergoes the lytic path, the cascade of events described above for virulent phages takes place. If the phage embarks into the lysogenic mode, the expression of the lytic genes is repressed, and the phage establishes, and more importantly, maintains a stable relationship with its host. The bacterium is then referred to as a “lysogen.” In most cases, formation of a lysogen involves the integration of the phage genome into the host chromosome. The genetic code of the integrated phage, also called “a prophage,” is replicated along with the host chromosome and transmitted to daughter cells with every cell division. Alternatively, some temperate phages are maintained as self-replicating plasmids in the host. Prophages are extremely stable and can be maintained indefinitely in this state. However, they can switch to the lytic cycle if triggered by appropriate environmental changes. For different viruses, this switching occurs by different mechanisms, but many, including the majority of food-related prophages, respond to stress-induced physiological changes of the host such as starvation, heat shock, and UV-induced DNA damage. In temperate phages, the establishment of lytic or lysogenic lifestyle, and the conversion from each other, depend on a molecular switch that involves phage regulatory proteins expressed early upon infection and a DNA region coding a set of operators; the so-called genetic switch (Ptashne, 2004). Competitive binding of the operators by these regulatory proteins is the key that modulates the expression of specific sets of genes needed to establish lysogenic or lytic cycles. Because the intracellular environment of the host is determinant in controlling the molar ratio of these regulators, this genetic switch is in fact a molecular sensor that determines the best

|

97

98

|

Emond and Moineau

course of action to undertake under specific conditions. This sensor constitutes a strong adaptive tool to determine the best survival strategy for temperate phages. From an applied perspective, bacterial strains used in biotechnological processes should preferably be devoid of inducible prophages. Industrial incidence and significance of phage The bioconversion of substrates into value-added compounds by fermentation is the basic process used by a wide variety of industries to manufacture products such as fermented foods, fermented feeds (silages), organic acids, alcohols, solvents, amino acids, biopolymers, pesticides, antimicrobials, and many other drugs (Table 4.1). Altogether, the bacterial fermentation industry represents a multibillion dollar a year market. As any bacterial population is susceptible to bacteriophage infection, it is not surprising that most fermentation industries have experienced problems with phage contamination (Table 4.1). The frequency and severity of phage problems vary for different industrial sectors, but in all cases, losses are considerable and the risk associated with phages cannot be neglected. The presence of bacteriophage in foods is not a safety concern because the phages of bacteria used in food fermentation, mostly lactic acid bacteria, do not represent a health hazard to consumers. However, these bacteriophages are a serious concern for the manufacturers of fermented foods because they can slow down or arrest the fermentation process and adversely impact product quality. In the dairy industry, this usually happens when a defined culture is used and phage titers in the fermentation vessels rise above a threshold of about 106–107 plaque-forming units per mL (PFU/mL). At this concentration and beyond, the rate of lactic acid production is reduced and the organoleptic properties of the product are altered.

Table 4.1 Examples of fermentation industry having experienced phage contamination Industry sector

Bacterial species

Product

Bacillus subtilis

Fermented coco beans

Bacillus subtilis var. natto

Fermented soy beans

Lactobacillus acidophilus

Fermented milk

Lactobacillus casei

Fermented milk

Lactobacillus brevis

Sauerkraut

Lactobacillus delbrueckii subsp. bulgaricus

Yogurt

Food

Lactobacillus delbrueckii subsp. lactis Cheese Lactobacillus fermentum

Sourdough bread

Lactobacillus helveticus

Cheese

Lactobacillus plantarum

Sauerkraut

Bacteriophages and Food Fermentations

Industry sector

Bacterial species

Product

Lactococcus lactis

Buttermilk, cheese, sour cream

Leuconostoc mesenteroides

Sauerkraut, buttermilk, sourcream

Leuconostoc fallax

Sauerkraut

Oenococcus oeni

Wine (malolactic fermentation)

Propionibacterium freudenreichii

Cheese

Streptococcus thermophilus

Cheese, yogurt

Tetragenococcus halophila

Soy sauce

Lactobacillus plantarum

Silage

Acetobacter sp.

Vinegar

Bacillus subtilis

Amylases, proteases

Brevibacterium lactofermentum

L-glutamic acid

Clostridium sp.

Acetone, butanol

Corynebacterium sp.

L-glutamic acid

Gluconobacter sp.

Gluconic acid

Lactobacillus brevis

Lactic acid

Pseudomonas aeruginosa

2-Ketogluconic acid

Xanthomonas campestris

Xantham

Feed

Commodity chemical

Biotechnology/Pharmaceutical Bacillus colistinus

Colistin

Bacillus polymyxa

Polymycin

Escherichia coli

Various biotechnology products

Streptomyces aureofaciens

Tetracycline

Streptomyces endus

Endomycin

Streptomyces griseus

Streptomycin

Streptomyces kanamycetus

Kanamycin

Streptomyces venezuelae

Chloramphenicol

Bacillus thuringensis

Insecticide (BT)

Pesticide

|

99

100

|

Emond and Moineau

As dairy phages are safe for human consumption, failed batches can be used to manufacture derivative products, although this implies inherent losses in product value and increases in manufacturing costs. As documented below, it is difficult to prevent the emergence of new phages in the dairy industry because of limitations in the stringency of the treatments that can be applied to reduce phage contamination in fermentation media. Thus, the number of dairy fermentation batches that are compromised by bacteriophage is relatively high, with estimates ranging from 0.1% to 10% depending on the factory and the products being made (Bogosian, 2005; Moineau and Levesque, 2005). In the biotechnology and pharmaceutical industries, this rate is much lower (0.1% of the batches lost to bacteriophages contamination) because stricter containment measures are employed, including the use of sterile fermentation substrates (Bogosian, 2005). However, the application of stringent manufacturing standards (cGMP) and strict product definitions in terms of purity, strength, and quality, imply that no level of phage contamination is acceptable due to the potential for process or product adulteration. Therefore, the losses in product rejection are also costly for this industry. For all industrial sectors, severe phage outbreaks are costly and time-consuming because investigations must be done to identify the source of the phage and to implement corrective actions. A shut down of production is often required to thoroughly clean up the facilities, which adds up to the losses in raw material and valuable products, as well as a reduction in productivity. Finally, the bacterial strains used to make the product may be replaced with a non-sensitive derivative, which add up the costs and delays due to the fine-tuning of the processes with the new strain(s). Phage monitoring methods The proliferation of virulent phages is particularly explosive in food fermentation factories and a good hygiene and sanitation program is required to control them and prevent outbreaks. But a sanitation program would not remain effective without constant monitoring and dependable tests that enable swift identification of new threats and quick implementation of corrective actions. Phage detection methods belong to three categories: (1) the microbiological assays, which are based on the phage ability to infect and lyse susceptible bacterial hosts; (2) the biochemical assays, which detect components of the virion using various biochemical procedures, and (3) electron microscopy, which permits direct visualization of the phage particles. Microbiological methods Plaque assay is a conventional and direct microbiological method for quantitative and sensitive detection of phages; but it is not a quick test. Briefly, samples are obtained by swabbing solid surface or by taking samples from a liquid. Swabs are dipped into diluents to suspend the collected material. The suspensions or liquid samples are then centrifuged at low speed to pellet cells and debris, and filtered (0.45 micrometer) to remove residual bacterial cells. Serial 10-fold dilutions of the filtrates in an appropriate buffer may be prepared if quantification is desirable. A drop containing about 50 microliters is then spotted on a plate seeded (100 to 500 microliters of an overnight culture) with an indicator bacterial strain, which is generally prepared in a top agar. To ensure detection of relevant bacteriophages,

Bacteriophages and Food Fermentations

the indicator bacterium should be the same as the one used in the fermentation process. If several known bacterial strains are used in production, then each of them should be separately tested with the filtrates. Plates are incubated in conditions that are adequate for the growth of the indicator bacteria, and then the areas where the samples were spotted are checked for zones of clearing, or plaques within the bacterial growth. It is possible, for samples that contain a higher number of phages that the spots turn out completely clear because of high phage concentrations. In such cases, it may be difficult to determine if the clearing is caused by phages or by chemical agents that may inhibit bacterial growth (e.g. antibiotics, residues of disinfectants). To distinguish between these possibilities, it is necessary to proceed with serial dilutions. Phages would then form discrete plaques on diluted samples, while the inhibition zone due to chemicals would gradually disappear as dilution factor increases. Results can be obtained in a matter of hours to a couple of days, depending on the bacterial host/phage combination. Plaques formed by some phages may be difficult to observe on the bacterial lawn. It is possible to alleviate some of these problems by modifying the incubation temperature (higher or lower), by adding cofactors such as Ca2+ or Mg2+, or by adding agents that weaken the bacterial cell (e.g. 0.25% glycine) and help the development of larger plaques (Lillehaug, 1997). The use of agarose instead of agar can also lead to increased plaque size. It should be noted that plaque assays does not work for all bacterial species. Strains of the Gram-negative bacteria Acetobacter europaeus, responsible for the efficient oxidation of ethanol into acetic acid in submerged vinegar fermentations, grow poorly on plates and do not produce visible plaques (Sellmer et al., 1992). Plaque assays are not appropriate for all occurrences in the food industry. They work better with pure bacterial cultures because otherwise plaques could be masked by the growth of additional unsusceptible strains present in the bacterial lawn. For instance, plaque assays are appropriate for the majority of the starter cultures currently used in large-scale cheese productions because they contain a small number of strains that can be assayed individually and in parallel. However, plaque assays are not always practical for specialty cheeses because they are often manufactured with traditional undefined bacterial cultures comprising a large number of strains for their superior flavoring properties. Similar limitations apply to fermentations that develop from epiphytic microflora (vinegar, malolactic fermentation, sauerkraut, cucumber), because they also contain large number of uncharacterized strains. In addition, quantitative detection of plaques does not necessarily provide a clear indication of the imminence of a phage outbreak. Depending on the virulence of a given bacteriophage, it may represent a very serious threat at low titers while others would remain relatively innocuous at higher titers. It should be noted that a high phage count could be sometimes observed in fermentations conducted with undefined or mixed cultures without significantly altering the fermentation process. In many instances, it is more convenient to monitor the impact of phages on the biological activity of the biocatalysts used in a fermentation process. These indirect microbiological methods do not detect phages but instead measure their impact on the activity of the bacterial culture. A classical test consists in measuring phage-induced lysis of a bacterial culture. Two tubes containing culture media are inoculated with a bacterial culture, and one of the tubes is further inoculated with a small amount of a filtrated-sample suspected

|

101

102

|

Emond and Moineau

to contain phages. The cultures are incubated under growth conditions appropriate for the indicator strain and regularly checked to determine if the tube co-inoculated with the filtered-sample becomes clear (or less dense), which would indicate the presence of phages. A variation of this method consists in measuring specific metabolic activities of the culture that are relevant to the fermentation process. Probably the most common and classical example for this type of assay is the milk activity test used by the dairy industry. Two tubes containing milk inoculated with an indicator lactic acid bacterium are incubated under appropriate conditions after one of the tubes has been seeded with a sample suspected to contain phages. The tubes are then monitored to determine their acidification rates. If the rate of acidification of the tube seeded with the suspected sample is slower than the other, phage contamination is suspected. A pH indicator such as bromocresol blue can be added to avoid direct pH measurements. In this assay, the indicator turns from purple to yellow when the pH drops below 5.4. Thus, the absence of a color change indicates loss of acidifying activity, which suggests the presence of phages. This technique is not particularly sensitive because the initial phage counts in the inoculum might not be enough to produce a detectable impact on cell activity. It is, however, a good tool to measure phage populations that represent an immediate threat to a fermentation process. Increasing the sensitivity of the assay is achievable by repeating the growth cycles successively using filtrates from the previous batch as inoculum, but this usually takes too much time for routine testing in the industry. For many years, the large starter culture companies have been providing a free service to help monitor phages in North American dairy fermentation plants, though at considerable costs to the service provider. Fermented dairy products manufacturers collect samples from their production lots and ship them to the technical service of the starter culture companies who perform one or both of the microbiological tests described above. Test results are being sent back as a report that specifies the risk that a phage might lead to fermentation failure and include recommendation to implement corrective actions. Performing these tests requires considerable efforts and the usefulness of this practice is certainly questionable. Firstly, a long period of time elapses between sample collection and the time results become available (3 to 5 working days). Within this timeframe, phages have more than enough time to multiply and lead to fermentation failure. Secondly, results may overestimate the phage titers at time of sampling because samples contain both phages and bacteria and there is no guarantee that phage will not multiply during shipping. In-house testing (within the dairy factories) would be faster and would provide a higher degree of control, but this is not always possible because starter culture companies do not usually provide isolates of the starter culture to their customers. Some larger dairies are performing such in-house testing, particularly those using multiple culture suppliers. On the other hand, the availability of a broad and large number of phage-containing samples provides the culture companies with a remarkable biological material that can be used to develop phage-resistant strains and to improve their starter rotation scheme (see below). In an effort to reduce cost associated with this type of analyses, it is possible to miniaturize this milk activity assay to allow automation. The scaled-down version of the assay is performed in 96-well microtiter plates for higher throughput. The pH drop associated with the acidifying activity of the cultures is measured by a plate reader or through the analysis

Bacteriophages and Food Fermentations

of digital images of the wells obtained at different time points during incubation. The key to the method is a composition of pH indicators that provide an array of colors that change with the acidity level (Houlberg et al., 2005). An image of the wells can be captured with a scanner or a CCD camera, and digital data representing the color of the wells are translated into pH values by reference to a standard curve. The dynamic range of pH values that can be inferred by this method and the accuracy of the pH estimates are adequate for the needs of milk activity tests. Because the method is non-invasive and non-destructive, there is virtually no risk of contamination and samples can be scanned as many times as required. Also, the assay only requires visualizing the light reflected by the surface of the sample, thus this colorimetric assay is not affected by sample turbidity. Another advantage of this technology is that it is relatively easy to set up a custom system with low cost scanners, standard incubators and a computer. Moreover, such computerized data can be stored and retrieved as needed. Biochemical methods Biochemical assays have been developed for direct detection of phage components. They include immunochemical detection of bacteriophages with antibodies raised against specific structural protein components of the virion (Schouler et al., 1992; Moineau et al., 1993b; Chibani Azaiez et al., 1998; Ledeboer et al., 2002), and molecular DNA detection techniques (e.g. DNA hybridization, PCR) with DNA primers and probes that are specific for nucleic acid sequences of a given phage, or group of phages (Moineau et al., 1992; Labrie and Moineau, 2000; Binetti et al., 2005; Dupont et al., 2005). These techniques are generally well suited for rapid detection, can be quite sensitive (e.g. PCR), and can be adapted to provide quantitative data (e.g. Q-PCR, ELISA). Although popular for research activities, these analytical tools are mainly used for phage characterization and they are seldom used by the food industry. This is because the development of molecular probes requires that a large number of phage isolates and phage data be available to develop probes that have the specificity and the range entailed by the fermentation industry. In that regard, the dairy industry is far ahead of other industries, as hundreds of phages of dairy lactic acid bacteria have been characterized worldwide (Ackermann, 2001; Brüssow, 2001; Brüssow and Desiere, 2001). Molecular probes do not distinguish phages that are infectious from those that have been inactivated, and thus, positive results would not be an appropriate indicator of relevant phage contamination, which would limit prediction of the imminence of a phage outbreak. Moreover, the presence of these phages does not necessary indicate that they will infect the forthcoming strains to be used in a fermentation process. Nonetheless, these probes are useful tools to rapidly determine the species of phage isolates (Labrie and Moineau, 2000) and to identify potential sources of contamination in the fermentation facilities. Electron microscopy Reliable microbiological methods are not always available to isolate and characterize phages and some fermentation processes involve so many undefined bacterial strains that phage monitoring can be difficult to do. In these cases, observation of a sample with an electron microscope can be used to confirm that faulty fermentations are caused by phages. However,

|

103

104

|

Emond and Moineau

these techniques are labor intensive and could hardly be used as a routine phage-monitoring tool. In all likelihood, the samples will have to be sent to microscopy service, subsequently centrifuged to remove large debris, properly stained (uranyl acetate or potassium phosphotungstate), and finally, observed at the right magnification (usually 300 000s). Since phages are heterogeneous in their morphology and their biology, samples should be observed by an experienced phage biologist, particularly for outbreaks occurring in industries experiencing rare phage problems and for which limited literature are available. Sources of phage contamination The establishment of a good strategy to control phage contamination requires that the external sources of the new emerging phage be recognized. The locations within the fermentation facilities that represent internal phage reservoirs should also be identified. The latter is important because new phages can remain undetected until their populations build up to levels where they become a threat to the fermentation process. These aspects are discussed in the following sections. External sources Phages are ubiquitous and any fluid entering the fermentation facility should be considered a risk factor. Among these, the fermentation substrates are by far the most important of all external sources of phages (Teuber et al., 1987; Stamm et al., 1989; McIntyre et al., 1991; Yoon et al., 2002). Thus, a way of looking at the risk associated with new incoming phages is to divide the industries into three risk groups based upon the type of fermentation substrates used: (1) industries that utilize sterile fermentation material, (2) industries that treat fermentation material to reduce microbial contamination, and (3) industries that use raw substrates. Sterile fermentation substrates Sterilization should be used whenever possible because, as most microorganisms, phages are inactivated by this treatment. Consequently, industries that use this process have lower incidences of phages contamination. Sterilization is usually done through a “heat-in-place” treatment of the substrate in the fermenter, or by running the fermentation substrate through an ultra-high temperature (UHT) unit. But many substrates including most foods cannot sustain such heat treatment without altering essential properties. Heat-sterilization is generally appropriate for fermentations that use classical culture media. This would include many products made by the pharmaceutical/biotechnology industry, several commodity chemicals (Table 4.1), as well as companies that manufacture starter culture and other bacterial cultures used in foods, animal health (silage inoculants, direct-fed microbials), and human health (probiotics). Pasteurized fermentation substrates In the dairy industry, the raw milk used to make cheese, yoghurt, and fermented milks is an important natural reservoir of lactic acid bacteria and their phages [101 to 104 pfu/mL of raw milk; (McIntyre et al., 1991)]. The fact that raw milk is pooled from several dairy farms increases phage biodiversity and raises the risk that a specific phage capable of infecting

Bacteriophages and Food Fermentations

a starter culture is present in the raw milk tanks. Thus, storage conditions and treatment aiming at reducing the indigenous microflora of raw milk are critical to control the emergence of new bacteriophages in the fermentation facilities. As mentioned before, not all fermentation substrates can endure treatments that effectively inactivate phages. While the milk used for yoghurt production undergoes a treatment at 90oC, which usually kills high levels of phages (Quiberoni A. et al., 1999), the milk used for cheese is treated under less stringent conditions to prevent defects in product texture and flavor. Consequently, phages can remain undamaged after pasteurization (Chopin, 1980; Madera et al., 2004). Raw fermentation substrates The use of raw materials is seldom warranted, as risk of phage contamination is very high. However, substrates such as cabbage, coco beans, soybeans, wine (malolactic fermentation), and the mash used in vinegar production cannot sustain treatments to reduce microflora without seriously damaging the products. Many of these fermentations are actually driven by the epiphytic lactic acid bacteria associated with the fermentation substrate. Bacterial strains developing by spontaneous fermentation are in equilibrium with their phages and usually survive the fermentation process, albeit fermentation progress at reduced speed and is more prone to inconsistencies compared to a defined starter culture system. Starter cultures have been, or are being developed for some of these products to speed up the fermentation process and improve production consistency. As these fermentation materials carry an assortment of virulent bacteriophages, the risk that the inoculated strains become inactivated by these phages still remains very high (Davis et al., 1985; Henick-Kling et al., 1986; Yoon et al., 2002; Lu et al., 2003; Doi et al., 2003). Lysogenic starter cultures Another external source of phage contamination that must be considered is the now-established fact that many bacteria carry prophages. The analysis of bacterial genomes initiated with the genomic era in the mid-nineties revealed that prophages are more widespread than previously thought. Numerous lactic acid bacteria carry one or several inducible prophages and/or pseudo-phages integrated into their genomes (Canchaya et al., 2003). Pseudophages are mutated prophages that have lost part or all of their functionality and are not capable of completing infectious cycles upon induction. The risk for prophage induction and its consequences on the final product must be carefully evaluated when developing industrial fermentation processes and selecting the appropriate bacterial strains for fermentation. Maintaining lysogenic bacteria in a fermentation vat for extended periods of time and manipulating them through stressful downstream processing may induce prophages into lytic cycle and result in fermentation failure or product inconsistencies. Therefore, the candidate bacterial strain should be tested in laboratory conditions in presence of inducible agents (mitomycin C, hydrogen peroxide, or UV exposure) as well as under industrial manufacturing conditions. Indeed, starter culture companies thoroughly test their strains in conditions that mimic industrial fermentation processes to ensure that prophages (if present) will not be induced under processing conditions. The long and successful use of strains carrying prophages shows that many lysogenic strains can be used in the dairy industry without noticeable problems. In fact, some believed that the presence of prophages

|

105

106

|

Emond and Moineau

or pseudo-phages may contribute favorably to the development of characteristic flavors by enhancing the releases of ripening enzymes in the cheese matrix through the expression of prophage-encoded endolysin and holin genes (Lepeuple et al., 1998). Recent reports suggest that prophages and pseudo-phages might also represent a risk factor for the emergence of new lytic phages because they constitute a pool of genes that can be transferred to incoming lytic phages by homologous recombination, thus expanding their host range (Bouchard and Moineau, 2000; Durmaz and Klaenhammer, 2000). Moreover, it has been shown in laboratory that serially replicating a temperate phage on an indictor (sensitive) strain may lead to a virulent derivative (Bruttin and Brüssow, 1996). The industrial significance of these phenomena, however, remains to be demonstrated. Ingredients derived from dairy fermentation In their efforts to maximize productivity and increase yields, many industries utilize dairy ingredients such as whey components to standardize the milk used to manufacture fermented dairy products. Being by-products of dairy fermentation, these ingredients potentially carry important loads of phages and should be regarded as important risk factors for bacteriophage contamination. Thus, even if some of these ingredients may have to go through some harsh phage-inactivating treatments, proper quality control experiments should be performed on these ingredients by testing for the presence of phages against the starter cultures used in the manufacturing process. The means by which these ingredients are delivered to the vat (pipelines, mixers) should also be closely monitored to avoid cross contamination. Other external sources Because bacteriophages can become airborne, contamination of the air supply is a critical weak point. Fermentation facilities located near to a natural reservoir of phages against the bacterial strains used in their fermentation process (e.g. cheese factory located close to dairy farms; biotech industry using E. coli as biocatalyst and located near to a sewage treatment plant) should be particularly concerned by airborne contamination. An HVAC system equipped with HEPA filters appropriately installed and maintained is usually efficacious to control airborne phage contamination. People (and their lab coats), equipment, carts and vehicles moving in and out of the fermentation facilities should also be considered an important risk factor for phage contamination, and special efforts should be done to restrict access of non-essential personnel and to provide adequate decontamination devices. Utility water supply is another potential reservoir of phages, particularly for coliphages associated with fecal contamination. This is particularly important in biotechnology and pharmaceutical industries, which use, most of the time, recombinant E. coli strains as fermentation biocatalysts. Food industries should also take measures to assess phage contamination of their water supply, especially when located in rural environments where phages of lactic acid bacteria are more likely to be found. Internal phage reservoirs We have seen earlier that new virulent bacteriophages are constantly introduced into food fermentation facilities despite efforts to control external sources of contamination. But very seldom would raw materials contain phage at levels that immediately threaten food fermen-

Bacteriophages and Food Fermentations

tations. Phages must first proliferate to critical levels and disseminate into the fermentation facilities; thus controlling phage proliferation and dissemination within the production area is the most critical step of a phage control strategy. As a general rule, phages are found where bacteria are in high numbers. The likely internal sources of phages are specific for each industrial set up, but clearly, the materials and equipments used in the fermentation process, as well as the fermentation products and byproducts, are the most obvious reservoirs. This implies that phage numbers likely increase as they progress further downstream into food fermentation manufacturing equipments and environments. Thus, a well-adapted manufacturing design would include a unidirectional flow of the fermentation materials through the processing equipment, with a parallel and positively pressurized flow of air. The products and by-products would be maintained into closed compartments as much as possible throughout the process. Each step of the process would be performed in a separate room and designed to minimize circulation of objects and personnel between the production units. As raw materials are potential sources of phages, they should be kept in separate warehouses and quickly processed upon entry into production facilities. Bacteriophages survive in humid environments, disseminate in all fluids as well as air. However, most bacteriophages have decimal reduction times ranging from a few hours to several days on dry surfaces and thus, they should eventually disappear from a contaminated facility (Bogosian, 2005). Liquids should be confined as much as possible to the production line, and whenever possible, humidity levels in the factory should be kept at minimum. Airborne phages are particularly important to control, albeit very few analysis have been performed in this area. Concentrations of up to 105 pfu/m3 of air have been reported in some dairy plants (Neve et al., 1994). We feel that control of airborne phages is particularly important for fermentations that operate in open vat systems or for any process that is prone to aerosol generation (e.g. fermentation involving gas production, aerated fermentation vessel, stirring, centrifugation of fermentation product, spray drying of whey, etc.). We have already mentioned the importance of HVAC air handling systems. Although HEPA filters are not certified to eliminate all phages, these systems are very effective at reducing airborne phages when the air is changed at an adequate rate. Any measure that contributes to keep the air as dry as possible will also reduce the number of infective airborne phages. It is also imperative to ensure that the HEPA filters stay dry at all time, as humid air makes filters less efficient. Many of the large-scale cheese plants in North America use starter culture rooms to prepare inoculums. To prevent contamination from the outside environment, bacterial starter cultures should be located upstream of the process and prepared in a positively pressurized room. More importantly, the fermentation by-products, which represent a very potent source of phages, should be handled in closed system and quickly moved downstream of the fermentation room. Also important are wet and damp areas within the production facilities where bacteria and their phages can develop and become a source of cross-contamination. Spills and leaks from the production line should be immediately decontaminated. Wet and damp areas such as drains should be regarded as phages reservoirs and be regularly and thoroughly disinfected. Sanitation process Special care should be taken to set up thorough and efficacious sanitation procedures that preserve the integrity of the equipment and are not hazardous to personnel. The efficacy

|

107

108

|

Emond and Moineau

of numerous sanitizing agents has been tested on bacteriophages, and several reviews have listed chemicals to which bacteriophages are susceptible (Hall et al., 1951; Murata et al., 1972; Rudolph, 1978; Quiberoni A. et al., 1999; Quiberoni et al., 2003). Phage-killing agents that are appropriate to the food industry include heat, alkaline (pH 11 and higher) and acidic (pH 4 and lower) environments, and disinfecting agents such as sodium hypochlorite (100 ppm) and 0.15% (w/v) peracetic acid (Adams, 1959). Alcohols such as ethanol and isopropanol (75–100%) exhibit suboptimal biocidal activity, with decimal reduction time ranging from 12 hours to 2 days (Bogosian, 2005). Heat treatments are also convenient for decontamination of fixed equipment because it diffuses readily in stainless steel and reach areas that may be difficult to contact with chemical sanitizers where biofilms are likely to develop. Ascorbic acid (vitamin C) with trace amounts of copper has also been reported to be an effective phage-killing agent that appears to be effective against all phages (Murata et al., 1971; Bogosian, 2005). It has the advantage of being safer for the health of the personnel and not as corrosive as other chemical agents. The use of Clean In Place (CIP) and Sterilize In Place (SIP) systems, which are designed for automatic cleaning and disinfecting without major disassembly and assembly work, should be used whenever possible because they are less prone to human error and they enable to clean one part of the plant while other areas continue to make products. That said, it might be useful to make sure that the chemical agents such as nitric and phosphoric acids (at the concentration recommended by the manufacturer) as well as the temperature used in these systems are effective against phages. Moreover, the used cleaning and disinfecting solutions should be discarded according to safe and standardized practices as well as never be re-used within the facilities. Phage control strategies The phage control measures described above such as factory designs, manufacturing processes, and sanitation procedures contribute largely to keep phages under control; and in fact, these are often the only countermeasures available to minimize the impact of phages in many food fermentation industries. But the dairy industry has developed additional defense strategies, built on effective starter cultures and designs that combine high industrial performance with reduced phage susceptibilities. These strategies will be described in the following sections and references to further reading will be provided where appropriate. Focus will be given to describing the advantages and limitations with regard to the application of these strategies in the food industry. Strain selection and culture rotation A simple and efficacious strategy to prevent bacteriophage-induced fermentation failures consists in using a set of bacterial strains with distinct phage susceptibilities. Starter cultures with such properties may be of defined or undefined strain composition. Undefined starter cultures have a long history of traditional use in the manufacture of fermented dairy products. The exact bacterial composition of these cultures is unknown but it contains several phage-related and unrelated strains. Undefined-strain compositions reduce the negative impact of a phage infection because phage-insensitive cells are readily available to carry on the milk acidification process in the advent of a phage attack. But the fermentation process may be delayed if predominant and fast-acid producing components of the culture

Bacteriophages and Food Fermentations

are being destroyed by virulent phages. Such fermentation delays may not be a significant concern in small-scale craftwork productions, as time may not be the prime consideration; but in large-scale fermentations where tight production schedules are the norm, these fermentation delays are a serious concern. At any rate, severe phage infections will negatively affect the final composition of the fermented products and it may be difficult to salvage the fermentations made with these undefined starters because there is no phage–host relationship data available to orient the choice towards an alternative culture. Defined starter cultures are a lot more prevalent nowadays. They are particularly well suited for large-scale production industries because they have been optimized for fast-acid production, they provide a higher degree of control and reproducibility on the fermentation process, and they are amenable to culture rotation systems. In general, defined starter cultures are composed of 2 to 5 thoroughly characterized strains with well-established microbiological, biochemical, and phage-susceptibility profiles. The strains making a defined starter are carefully selected to meet the specific requirements of the product. Selection criteria may vary, but they usually include acidification rates, flavor profiles, strain compatibility, and phage susceptibility profiles. Because the development of starter cultures takes time and requires considerable amount of resources, starter companies are highly concerned that a culture might suffer phage hits that could compromise its effectiveness and lifespan on the market. Consequently, new starter cultures are seldom released individually, but instead are integrated into a defense strategy that consists of a culture rotation program (Sing and Klaenhammer, 1993; Neve et al., 1994; Durmaz and Klaenhammer, 1995). A culture rotation program is designed to serve as a phage control strategy whereby a series of starter cultures (usually 3 to 5) can be swapped with another one upon emergence of new lytic phages. The cultures included in a rotation scheme are specifically formulated to be phage-unrelated (i.e. distinct phage susceptibility patterns). When used properly, culture rotation can be very effective at maintaining phage titers at levels that do not harm the fermentation process for extended periods of time. The cultures included in a rotation scheme are carefully designed to provide similar industrial performances. However, dairy fermentation industries commonly develop a preference towards specific cultures within a rotation scheme and are inclined to use them more extensively than others. After a while, this practice puts a lot of pressure on a single component of the culture sets and may endanger the effectiveness of the whole rotation program. The fermentation plants must also be careful not to constantly switch one culture program to another because the phage relatedness of cultures from different programs is not usually known—particularly true when switching between starter culture manufacturers—and frequent swapping may put phage-related cultures one after another in the rotation scheme. Another strategy to further control phage consists in rotating the types of fermented products manufactured on a given production line to dilute out phage population that built up during previous production cycles. This strategy can be effective for products manufactured with phage-unrelated cultures, such as Cheddar cheese (mesophilic cultures) and Mozzarella cheese (thermophilic cultures). However, the application of this phage control strategy needs to be carefully analyzed since the recent introduction on the market of new starter cultures that contain both mesophilic and thermophilic strains to maximize acidification rates in some cheesemaking processes. Undoubtedly, these ready-to-use cultures break the natural phage barrier that existed between products manufactured with

|

109

110

|

Emond and Moineau

mesophilic and thermophilic starter cultures. Several procedures need to be reassessed to identify new putative contamination channels. On the other hand, these high performance cultures alleviate the requirement of growing bulky volumes of starter cultures in largescale fermentation factories, which eliminates an important risk of phage contamination. Selection of bacteriophage-insensitive mutants A starter culture that is repeatedly hit by phages must be removed from the starter culture rotation program, or it can be salvaged if an appropriate substitute strain is available to replace the phage-susceptible component in the starter culture. Ideally, the faulty bacterial strain would be replaced with a phage-resistant isogenic variant to preserve integrally (or as much as possible) the functional properties of the original culture. Bacteriophage-insensitive mutants (BIMs) are spontaneous phage-resistant derivatives that survive long exposure to one or several lytic phages in a classical phage challenge experiment. This random mutational method is one of the first and among the most successful approaches used to develop phage-resistant strains (Forde and Fitzgerald, 1999b; Coffey and Ross, 2002). The benefit of this approach is that it can be applied to any bacterial strain including Gram-positive and Gram-negative bacteria. Therefore, it can be used to generate phage-resistant strains for almost all fermentation processes. The mutations underlying the resistance phenotype in BIMs are generally unknown. However if the wild-type phages are no longer capable to adsorb to the BIMs, then the mutation(s) likely occurred in the phage receptors. Mutations affecting components located in the cell envelope often have negative impact on the cell performance because essential cellular functions (e.g. membrane transport, signal transduction, cell wall structure) may be altered. Thus, it is not surprising that many BIMs grow slowly, have reduced capacity to produce lactic acid, or form aggregates under stressful conditions. Another problem with BIMs is that the rate of reversion to phage susceptibility phenotype can be high. It is also possible that phage mutants infecting these BIMs be easily found in manufacturing facilities. But all these problems can be circumvented if an appropriate number of candidate BIMs are put through a carefully designed screening process. If fact based on the authors knowledge, many BIMs have been isolated and are being used successfully under industrial conditions. BIMs have been particularly useful to salvage key strains for which there were no other phage-insensitive alternatives available. It is common to find strains that have been going through 4 or 5 successive rounds of BIM selection (using different phages each time) over the years. The downside of BIMs is that, inevitably their industrial performance gradually deteriorates with increasing rounds of phage challenges, which sets a limit to the utilization of this process based on natural selection. Native defense mechanisms Since the pioneer work by Max Delbrück and his collaborators, phages have been used as a prime model to elucidate DNA replication and establish the foundations of modern molecular biology (Summers, 2005). Delbrück and his collaborators chose to focus on a restricted group of phages to make rapid progress and hasten the discovery process. This turned out to be a very successful approach, and tremendous knowledge has been acquired on the molecular mechanisms underlying the life cycle of E. coli phages. Although phages

Bacteriophages and Food Fermentations

infecting other bacterial species have received less attention, the industrial challenge represented by dairy phages fostered the isolation of a myriad of host-encoded natural defense mechanisms. Over the years, culture companies have been collecting phages that caused industrial fermentation failures and used them to build up phage cocktails employed to screen for new bacterial strains capable of sustaining these phage challenges. These repeated cycles of screening led to the selection of highly resilient bacteria, among which, many possess one or several natural defense mechanisms. To these days, over 50 such phage resistance mechanisms have been characterized only in the dairy mesophilic bacterium Lactococcus lactis (Coffey and Ross, 2002). The majority of these natural defense mechanisms are encoded on transmissible genetic elements, typically consisting of conjugative or mobilizable plasmids. This useful feature enables the transfer of the defense mechanisms to other strains by conjugation, and permits to salvage industrial strains (including BIMs) that would otherwise be discarded because of repeated phage infections. All of these defense mechanisms interfere with an essential step of the phage lytic cycle. For classification purpose, they have been divided into four groups based upon the phase of the phage life cycle with which they interfere: (1) adsorption blocking, (2) DNA ejection blocking (3) restriction/modification (R/M) systems, and (4) abortive infection mechanisms (Abi). Each of these functional classes will be discussed with a focus on practical aspects regarding their application into the fermentation industry. Adsorption blocking This class of natural defense mechanisms has features that are reminiscent of some of the BIMs described earlier, because they also prevent adsorption of the phages to the cell surface. Two distinct mechanisms of action have been suggested to explain adsorption inhibition (Klaenhammer and Fitzgerald, 1994). The first phage inhibitory system produces substances that sterically hinder the receptor in such a way that the phage is no longer able to bind. The second mechanism would produce an alteration that prevents receptor binding by the phage, or would result in complete loss of the receptor. Although adsorption blocking has been characterized to some extent in L. lactis, the genetic determinants responsible for this natural defense mechanism have not been always identified. A number of authors have suggested that the production of exopolysaccharides or a capsule may protect lactic acid bacteria against phages (Forde and Fitzgerald, 1999a; Looijesteijn et al., 2001) while others argued that it is not providing a sufficient protection (Deveau et al., 2002; Broadbent et al., 2003). For example, it has also been shown that some phages possess polysaccharidedegrading enzyme (Sutherland et al., 2004). Adsorption blocking mechanisms can be very effective at controlling phage infection because they make the cells impervious to the phage DNA (phage genome does not enter into the cell), which prevent, to a large extent, the possibility of genetic exchanges. As a result, mutant phages capable of circumventing this type of defense mechanism are rarely observed. It should also be pointed out that because the phages stay outside the cell, the intracellular biochemical activities are not altered. Blockage of phage DNA ejection DNA ejection blocking has not been thoroughly characterized (Garvey et al., 1996). Phages adsorb to the cell envelope of cells carrying this defense mechanism, but DNA is prevented

|

111

112

|

Emond and Moineau

from being ejected into the host’s cytosol. The sie2009 gene of the temperate lactococcal bacteriophage Tuc2009, was shown to mediate a phage-resistance phenotype in Lactococcus lactis against a number of bacteriophages. The Sie2009 protein is associated with the cell membrane and its expression leaves phage adsorption, transfection and plasmid transformation unaffected, but interferes with plasmid transduction, as well as phage replication. These observations indicate that this resistance is a result of DNA injection blocking, thus representing a novel superinfection exclusion system (Mc Grath et al., 2002). Restriction and modification Once a phage has successfully adsorbed to a susceptible bacterium and translocated its DNA into the host’s cytosol, the first intracellular defense mechanism that it is likely to face is a R/M system. These R/M systems are distinguished from other intracellular phage hurdles in that they act early upon phage DNA entry, and prevent damage that would otherwise cause cell death. R/M systems consist of two enzymatic activities that function as a complementary system. One is a restriction endonuclease that cleaves DNA at specific recognition sites; the other is a modification enzyme that methylates DNA at specific loci. Methylated DNA, usually referred as modified, is protected against cleavage by the restriction endonuclease. Modification of the DNA is required to protect the bacterial DNA against its own restriction endonuclease. Newly synthesized bacterial DNA is quickly modified upon replication and remains fully protected against cleavage by its restriction enzyme (Tock and Dryden, 2005). Upon entry into the bacterial host, the phage genome is not modified, which gives the restriction enzyme a chance to cleave this DNA before modifying enzymes completes the methylation of the recognition sites. In this competition between modification and restriction, it happens, although at low frequency, that a phage successfully escapes restriction, because modification completes before the restriction enzyme had a chance to cut at recognition sites. The chances of evading restriction via modification increase as the number of restriction sites in a phage genome diminishes (Moineau et al., 1993a). If the phage genome contains a functional methylase gene, it may also limit the efficacy of a specific R/M system (Hill et al., 1991). A phage that escaped restriction becomes totally immune to the R/M system and can pursue its lytic cycle unconstrained. Moreover, the genome of the phage progeny will also be modified, and be impervious to R/M in subsequent infection cycles. This immunity will be maintained for as long as the host use for propagation expresses the R/M system. Conversely, propagation of the modified phage on a strain that does not harbor the R/M system will produce a progeny that do not carry methylation; and the susceptibility to R/M will be restored. Thus, incorporation of R/M defense mechanisms into rotation programs is advisable to ensure that immune phages due to modification will be purged upon culture rotation, which will extend the useful lifespan of strains in industrial fermentations (Sing and Klaenhammer, 1993; Durmaz and Klaenhammer, 1995). Abortive infection Abortive infection mechanisms, also designated phage exclusion systems, constitute the last known line of natural defense against phages. Abi systems intervene after phages had adsorbed and ejected their nucleic acid into the host’s cytosol. They cause an interruption

Bacteriophages and Food Fermentations

of phage development, resulting in the release of few or no progeny particles and to the death of the infected cells. Cells harboring Abi systems are like a trap that captures free phage particles and strongly limits their spread to other cells; thus allowing the bacterial population to survive (altruistic suicide system). Abi systems have been identified in several bacterial species, but most have been isolated from lactococci (22 Abis published in L. lactis alone), probably as a result of the large efforts devoted to the study of lactococcal phages and to the analysis of lactococcal phage defense mechanisms (reviewed in Coffey and Ross, 2002; Chopin et al., 2005). Although several groups of lactococcal phage were reported based upon DNA homology ( Jarvis et al., 1991), only phages of the 936, c2, and P335 species are repeatedly isolated from industrial dairy fermentations. Abi proteins have generally been tested against phages representative of these three phage groups, and have shown effective against one, two, or all three of them (Chopin et al., 2005). The efficacy of a given Abi, however, may vary considerably from one phage species to another, and also on different representatives within a species. Abi systems may represent a very strong barrier against dairy phages as evidenced by efficiencies of plaquing (EOPs) in the range of 10–7 to < 10–9 for some of them. Others are considered to have a medium potency if EOPs are in the range of 10–4 to 10–6, whereas a weak Abi would produce an EOP in the range of 10–1 to 10–3 (Moineau, 1999). EOP is a method that compares the phage titer obtained on susceptible and resistant variants of a bacterial strain (Sanders and Klaenhammer, 1980). It is calculated by dividing the titer obtained on the phage-resistant derivative by the titer of the phage-susceptible derivative. This method is relatively simple to perform and is widely used to evaluate the efficiency of Abi defense systems. However, one has to be careful when interpreting EOP data because abortive systems do not prevent all infected cells from releasing phages, and hosts carrying an Abi system may produce plaques of reduced size but in numbers that are comparable to that of the susceptible host. Thus, high EOP values do not warrant that an Abi would not perform appropriately in the industry. Another way to evaluate Abi’s potential is to perform efficiency of centers of infection (ECOI) and one-step growth experiments that measure the percentage of infected cells that do release a phage progeny, and the size of the released progeny, respectively. The most reliable method to evaluate the industrial potential of a strain harboring a natural defense system is to run it through a starter culture activity assay (SAT) as described by Heap and Lawrence (1977). This test is similar to the milk activity assay described earlier, with the exception that the culture is deliberately contaminated with a high number of a phage or a phage cocktail (106 pfu/mL) to challenge the culture. The milk is fermented in conditions that mimic the industrial process, and the fermentation is repeated in several successive batches with a supplement of whey/milk from the previous batch as a phage inoculum. The first fermentation cycle may suffer a delay because phage-infected cells die despite the presence of an Abi system in the cells. However, the rate of acidification should resume to full speed in the second and subsequent fermentations if the Abi is effective. Abi systems interfere with intracellular steps of the phage lytic cycle. Attempts have been made to identify these steps by comparing phage DNA replication, transcription, synthesis of structural proteins, and lysis in the presence or absence of Abis. This approach, however, did not provide much insight into their mechanisms of action because most of the downstream steps in the cascade of events that normally takes place upon phage infec-

|

113

114

|

Emond and Moineau

tion are also disrupted. This makes it difficult to separate the primary target of Abi from subsequent effects. At best, it may point towards interference with earlier or later phases of the phage lytic cycle. At the molecular level, it has been shown that seventeen of the lactococcal Abis require a single gene to express their phenotype, four require two genes, and one requires four genes (Chopin et al., 2005; Yang et al., 2006). Abi systems have little or no sequence similarity to one another, and though some homologs exist in databases, no homology with proteins of known function has been found, which does not allow any prediction to be made on their mode of action. However, a reverse transcriptase motif was recently found in AbiA and AbiK (Fortier et al., 2005). Mutant phages able to circumvent Abi defenses have provided more insight on the possible mechanisms of action of Abi systems (reviewed in Chopin et al., 2005). It was suggested that phage-triggered host suicide may result from the deregulation, upon phage infection, of the expression of Abi proteins at the transcriptional and translational levels. The over-expressed Abi would then interfere with a function of the host, or of the phage, that is essential for phage multiplication and shut down the lytic cycle and kill the host. Another plausible explanation for some of these systems would be that, by the time the Abi interrupts the lytic cycle, the host has already endured irreversible phage-induced damages and dies. Combinations of natural phage defense mechanisms The vast majority of all known natural defense mechanisms are encoded on plasmids. For example, in many lactococci, several phage barriers are carried by a single strain, which can be encoded on single or distinct plasmids. These combinations of natural phage defense systems provide expanded protection against phages. The availability of numerous phage resistance plasmids encoding R/M and Abi systems enabled the construction of designer phage-resistant systems where isogenic variants of L. lactis carrying single or paired natural defense mechanisms were incorporated into rotation programs (Sing and Klaenhammer, 1993; Durmaz and Klaenhammer, 1995). In these combinations, a phage escaping an antiphage barrier is prevented from completing its infective cycle by the other antiviral systems. The culture rotation successfully controlled phages that were resistant to any one of the individual defense systems by presenting a different set of defenses in the next rotation. These combinations of culture rotation and natural anti-phage system pairing improved the protection of valuable starter cultures from phage attack during SAT assay and may extend the lifespan of these starters in the industry. Another advantage of this approach relies on the use of isogenic variants of a single strain, which provide a higher degree of control and reproducibility over the industrial process. However, the use of recombinant DNA technology to construct some of these isogenic variants may be an issue for industrial applications given the current consumer reluctance with genetically modified organisms (GMO) and strict ruling regarding use of GMO, especially in Europe. To circumvent regulatory issues, combinations of phage defense system could be based only on conjugation experiments with native plasmids, but this would strongly limit the array of possible combinations. Engineered defense systems Two major developments in recent years have enabled the engineering of a wide variety of designer phage resistance systems. Firstly, several model phages of L. lactis and S. ther-

Bacteriophages and Food Fermentations

mophilus have been characterized to some extent at the molecular level, which identified phage components that have become a source for engineering resistance traits. Secondly, comparative phage genomics have become possible due to the availability of more than 400 complete phage genome sequences and the development of powerful bioinformatics tools, which allowed identification of well-conserved genetic regulation elements among groups of phages susceptible to cause damage in dairy factories. These engineered phage-resistance systems are briefly described here (for a review, see Moineau, 1999; Coffey and Ross, 2002; Sturino and Klaenhammer, 2004b). Superinfection immunity Phage-resistance based upon superinfection exclusion has been described in L. lactis (Durmaz et al., 2002) and in S. thermophilus (Bruttin et al., 1997). Superinfection immunity is a well-characterized phage resistance system that relies on a genetic switch encoded by a temperate phage, which is required to establish and maintain the lysogenic state (see above, Chapter 9 of this book, and Ptashne, 2004). When a temperate phage infects a bacterium carrying a prophage with a functionally identical genetic switch, the expression of the genes required to initiate the lytic cycle are turned off by the prophage and the infection aborts. Phages that are inhibited by a cognate genetic switch are said homoimmune. Typically, virulent phages do not carry the so-called genetic switch and are impervious to this type of defense mechanism. However, some virulent phages from specific groups or species (e.g. P335 species of L. lactis, phages of S. thermophilus) have maintained part or all of the functional properties of the genetic switch harbored by their temperate ancestors and are inhibited when specific elements of the switch are expressed in-trans in the host. The immunity phenotype was observed for several but not all virulent phages belonging to the P335 species of L. lactis [12 out of 16 phages; (Durmaz et al., 2002)] and phages of S. thermophilus (Bruttin et al., 1997). Although effective, this type of engineered defense mechanism only inhibits a relatively narrow range of phages. Nonetheless, these engineered systems could become valuable against phages for which there are no alternative resistance mechanisms available. Origin-derived phage-encoded resistance Soon after a dairy phage has injected its DNA into a susceptible host, early phage products and host-encoded factors rapidly bind to the phage origin (ori) to initiate the process of massive phage DNA replication. The formation of an initiation complex, which is one of the rate-limiting steps in DNA replication, relies on a delicate balance between molar concentration of ori sequences and both host- and phage-encoded initiation factors. Origin-derived phage-encoded resistance (PER) is a system whereby multicopies of a phage ori sequences are supplied in-trans on a plasmid carried by the host (Hill et al., 1990; O’Sullivan et al., 1993). Upon phage infection, the plasmid-based copies of the phage origin presumably segregate most of the phage- and host-encoded replication initiation factors, which reduce the rate of initiation complex formation at the phage origin. As a result, the replication of the phage genome is severely reduced while the replication events initiated at the phage ori region located on plasmid dramatically increase its copy numbers, which further contribute to limit phage genome replication. To efficiently sequester away replication factors, PER systems must provide multicopies of the plasmid-encoded phage origin. This is usually

|

115

116

|

Emond and Moineau

achieved by cloning the phage ori region on a high-copy-number plasmid (O’Sullivan et al., 1993) and/or by cloning multicopies of the ori region on the plasmid (McGrath et al., 2001). PER systems are only effective against the phages that have a functionally similar replication origin, which is usually limited to members of a common group or species of phages. However, there are few natural defense systems available to control phages in thermophilic starter cultures and PER may represent one of very few recombinant alternatives to protect these bacteria (Foley et al., 1998; Stanley et al., 2000; Lamothe et al., 2005). Gene silencing Gene silencing is a type of engineered defense mechanism that express anti-sense RNAs to high levels to block the translation of essential genes required for phage development. The formation of double-stranded RNA (dsRNA) between the target mRNAs and the anti-sense molecules interfere with several components of the translation machinery. Firstly, anti-sense RNAs prevent formation of an initiation complex with the 30S ribosomal subunit by masking the ribosome-binding site (RBS). Secondly, the formation of dsRNA downstream of the RBS site sterically hinders ribosome translocation and retards translation during the elongation phase. Thirdly, the formation of dsRNA may destabilize the mRNA by promoting the action of dsRNA-specific ribonucleases. Finally, polar inhibitory effect may be observed on translation of genes located downstream on a polycistronic mRNA (Sturino and Klaenhammer, 2004b). Many anti-sense RNAs developed against dairy phages of L. lactis and S. thermophilus showed various levels of efficacy (reviewed by Sturino and Klaenhammer, 2004b). In general, anti-sense RNAs that target widely distributed and highly conserved genes are more likely to be effective against a broad range of industrial phages. In addition, anti-sense RNAs that target genes expressed early upon phage infection are more effective. Because the phage-encoded genes involved with DNA replication are highly conserved and expressed early upon infection, it is not surprising that they proved to be effective targets for engineering gene silencing defense systems (Kim and Batt, 1991; McGrath et al., 2001; Sturino and Klaenhammer, 2002; Sturino and Klaenhammer, 2004a). Triggered suicide Triggered suicide is another engineered phage defense mechanism where a toxic gene is cloned under the control of a phage-inducible promoter. When a phage infects a bacterial cell, the expression of the lethal gene is induced, which kills the host and aborts the phage lytic cycle. A triggered suicide system was engineered by cloning the restriction endonuclease gene cassette LlaIR+ under the control of the phage promoter that normally controls expression of late genes in F31 (Djordjevic and Klaenhammer, 1997; Djordjevic et al., 1997). Once induced by an infecting phage, the restriction enzyme cuts the unmethylated DNA from the host and phage—the cognate modifying enzyme is not present in the cell to modify DNA—and the phage lytic cycle aborts. This system proved to work properly when harbored by a high-copy number plasmid and requires a tight promoter that does not leak to prevent any toxic effect on the host in absence of phage infection. Unfortunately, this system only works with closely related phages able to effectively induce expression from the F31 promoter, which limits industrial applications. However, such a system could

Bacteriophages and Food Fermentations

potentially be used to salvage an important industrial strain for which there are not many replacement alternatives. Subunit poisoning This new type of engineered defense mechanism has been proposed and appears to be very effective in S. thermophilus particularly when combined with a PER system (reviewed in Sturino and Klaenhammer, 2004b). Subunit poisoning targets an essential phage enzyme that requires oligomerization to become activated. This approach is based on the identification of conserved amino acids in positions that are critical for the catalytic activity of the enzyme and regions of the polypeptide involved in oligomerization of the subunits. A mutated allele of the gene is generated that substitute amino acids in the catalytic domain of the enzyme, but keep intact the regions required for oligomerization. When this is expressed in high copy number in a PER-containing vector in the bacterial host, phage infection is severely reduced as evidenced by a dramatic reduction of the EOP (< 10–9). The authors present evidences that supports a mechanism whereby the mutant allele would associate with wild-type subunits to form dysfunctional heteropolymers that lost activity (Sturino and Klaenhammer, 2004b). The activity of subunit poisoning systems has not yet been reported on heterologous phages, but the range of phage inhibition could resemble that of other engineered defense systems. Phage structural proteins and antibodies A novel anti-phage strategy was recently designed based on the use of a phage neutralizing heavy-chain antibody fragment obtained from Llama. Immunization of a llama with lactococcal phage p2 (936 species) produced high titers of neutralizing heavy-chain antibodies (i.e. devoid of light chains) (de Haard et al., 2005). A panel of p2 specific single-domain antibody fragments was obtained using phage display technology, from which a group of potent neutralizing antibodies were identified. The antigen bound by these antibodies was identified as the receptor-binding protein (RBP), which was located at the distal part of the phage tail (de Haard et al., 2005). It was also shown that the addition of purified RBP protein to a bacterial culture suppressed phage infection. The neutralizing heavy-chain antibody fragment was also produced at a large scale in the GRAS microorganism Saccharomyces cerevisiae and shown to impede phage infection during a cheese process (Ledeboer et al., 2002). The infection of the cheese starter culture by 105 pfu/ml cheese-milk of the phage p2 was prevented by the addition of only 0.1 µg/ml (7 nM) of the neutralizing antibody fragment. The low amount of antibody fragments needed and the fact that these antibodies are produced by a food grade production organism and can easily be isolated from the fermentation liquid in a pure form and free of the DNA of the production organism (not GMO by European standards; see below), makes the use of such antibody fragments a surprising possibility. However as with the other engineered antiviral strategies, the antibody fragment are likely to be effective against only a closely related group of phages. Bioengineered organisms and the food industry Bioengineered defense systems provide useful alternatives to enhance phage resistance in bacteria used in food fermentations. Unfortunately, only a restricted range of phages

|

117

118

|

Emond and Moineau

belonging to the group from which the systems were derived are inhibited, which limits their usefulness in the factory environment. Expanding their range is required to make them more attractive from a commercial viewpoint. More importantly, consumers are concerned with bioengineered organisms, particularly in Europe where the regulatory agencies have set strict rules to control the distribution of organisms that were developed using recombinant DNA technologies. Currently, the only gene transfer technique accepted in the European Community is natural conjugation. The situation is different in the USA, where it is possible to commercialize bioengineered foods after informing the Food and Drug Administration through the procedure for Premarket Notice Concerning Bioengineered Foods. But still, large starter culture companies operate internationally, and the manufacture of bioengineered cultures for a specific market may be difficult to integrate into their production logistics. Application of recombinant DNA technology in food fermentations will progress slowly until consumer’s perception changes through proper and non-biased information. Concluding remarks Bacteriophages are natural constituents of most microbiota and are found in all ecosystems where bacteria flourish. Because of their ubiquity in nature, phages eventually find their way into fermentation facilities and threaten the industrial process used to manufacture fermented foods. Over the years, the fermentation industries have developed a series of control measures to keep phage in check, including adapted factory design and processes, cleaning and sanitation procedures, and tailored air control systems. Strategies based upon culture rotation of phage-unrelated strains or strains harboring natural phage-resistance mechanisms tremendously helped in the containment of phages in these industries. More recently, genetic engineering has enabled the design of new strains that combine excellent industrial performance with enhanced phage resistance (GMOs); however their application should remain marginal until they receive general acceptance in the public. Although complete eradication of bacteriophages in industrial fermentation settings is not possible, the implementation of an integrated multihurdle program and appropriate staff education should provide the means to contain phage contamination in the food fermentation industry. Acknowledgments The authors are grateful to Hélène Deveau for providing the electronic microscope images of the phages. The authors are very thankful to the following agencies and organizations for their continued and long term support of our research program on bacteriophages: Natural Sciences and Engineering Research Council (NSERC) of Canada, Fonds de recherche sur la nature et les technologies du Québec, Agropur, and Danisco. References

Ackermann, H.W. (1999). Tailed bacteriophages. The order Caudovirales. Adv. Virus Res. 51, 135–201. Ackermann, H.W. (2001). Frequency of morphological phage descriptions in the year 2000. Arch. Virol. 146, 843–857. Ackermann, H.W. (2005). Bacteriophage classification. In: Bacteriophages: biology and applications, E. Kutter and A. Sulakvelidze, eds. (Boca Raton: CRC Press), pp. 67–89. Adams, M.H. (1959). Bacteriophages. (New York: Interscience Publishers).

Bacteriophages and Food Fermentations

Binetti, A.G., Del Rio, B., Martin, M.C., and Alvarez, M.A. (2005). Detection and characterization of Streptococcus thermophilus bacteriophages by use of the antireceptor gene sequence. Appl. Environ. Microbiol. 71, 6096–6103. Bissonnette, F., Labrie, S., Deveau, H., Lamoureux, M., and Moineau, S. (2000). Characterization of mesophilic mixed starter cultures used for the manufacture of aged cheddar cheese. J. Dairy Sci. 83, 620–627. Bogosian, G. (2005). Control of bacteriophage in commercial microbiology and fermentation facilities. In: The bacteriophages, 2nd ed., R. Calendar, ed. (New York, NY: Oxford University Press). Bouchard, J.D., and Moineau, S. (2000). Homologous recombination between a lactococcal bacteriophage and the chromosome of its host strain. Virology 270, 65–75. Braun, V., Schaller, K., and Wolff, H. (1973). A common receptor protein for phage T5 and colicin M in the outer membrane of Escherichia coli B. Biochim. Biophys. Acta 323, 87–97. Breitbart, M., Rohwer, F., and Abedon, S.T. (2005). Phage ecology and bacterial pathogenesis. In: Phages: Their Role in Bacterial Pathogenesis and Biotechnology, M.K. Waldor, D.I. Friedman, and S.L. Adhya, eds. (Washington: ASM Press), pp. 66–91. Broadbent, J.R., McMahon, D.J., Welker, D.L., Oberg, C.J., and Moineau, S. (2003). Biochemistry, genetics, and applications of exopolysaccharide production in Streptococcus thermophilus: a review. J. Dairy Sci. 86, 407–423. Bruttin, A., and Brüssow, H. (1996). Site-specific spontaneous deletions in three genome regions of a temperate Streptococcus thermophilus phage. Virology 219, 96–104. Bruttin, A., Desiere, F., Lucchini, S., Foley, S., and Brüssow, H. (1997). Characterization of the lysogeny DNA module from the temperate Streptococcus thermophilus bacteriophage phi Sfi21. Virology 233, 136–48. Brüssow, H. (2001). Phages of dairy bacteria. Annu. Rev. Microbiol. 55, 283–303. Brüssow, H., and Desiere, F. (2001). Comparative phage genomics and the evolution of Siphoviridae: insights from dairy phages. Mol. Microbiol. 39, 213–223. Calendar, R., and Inman, R. (2005). Phage biology. In: Phages: Their role in bacterial pathogenesis and biotechnology, M.K. Waldor, Friedman D.I., and Adhya S.L., eds. (Washington: ASM Press), pp. 18–36. Canchaya, C., Proux, C., Fournous, G., Bruttin, A., and Brüssow, H. (2003). Prophage genomics. Microbiol. Mol. Biol. Rev. 67, 238–276. Chibani-Azaiez, S.R., Fliss, I., Simard, R.E., and Moineau, S. (1998). Monoclonal antibodies raised against native major capsid proteins of lactococcal c2-Like bacteriophages. Appl. Environ. Microbiol. 64, 4255–4259. Chopin, M.C. (1980). Resistance of 17 mesophilic lactic Streptococcus bacteriophages to pasteurization and spray-drying. J. Dairy Res. 47, 131–139. Chopin, M.C., Chopin, A., and Bidnenko, E. (2005). Phage abortive infection in lactococci: variations on a theme. Curr. Opin. Microbiol. 8, 473–479. Chopin, M.C., Rouault, A., Ehrlich, S.D., and Gautier, M. (2002). Filamentous phage active on the Grampositive bacterium Propionibacterium freudenreichii. J. Bacteriol. 184, 2030–2033. Coffey, A., and Ross, R.P. (2002). Bacteriophage-resistance systems in dairy starter strains: molecular analysis to application. Antonie van Leeuwenhoek 82, 303–321. Davis, C., Silveira, N.F., and Fleet, G.H. (1985). Occurrence and properties of bacteriophages of Leuconostoc oenos in Australian wines. Appl. Environ. Microbiol. 50, 872–876. de Haard, H.J.W., Bezemer, S., Ledeboer, A.M., Müller, W.H., Boender, P.J., Moineau, S., Coppelmans, M.-C., Verkleij, .A.J., Frenken, L.G.J., and Verrips, C.T. (2005). Llama antibodies against a lactococcal protein located at the tip of the phage tail prevent phage infection. J. Bacteriol. 187, 4531–4541. Deveau, H., van Carsteren, M.-R., and Moineau, S. (2002). The effect of exopolysaccharides on phage–host interactions in Lactococcus lactis. Appl. Environ. Microbiol. 68, 4364–4369. Djordjevic, G.M., and Klaenhammer, T.R. (1997). Bacteriophage-triggered defense systems: phage adaptation and design improvements. Appl. Environ. Microbiol. 63, 4370–4376. Djordjevic, G.M., O’Sullivan, D.J., Walker, S.A., Conkling, M.A., and Klaenhammer, T.R. (1997). A triggered-suicide system designed as a defense against bacteriophages. J. Bacteriol. 179, 6741–6748. Doi, K., Zhang, Y., Nishizaki, Y., Umeda, A., Ohmomo, S., and Ogata, S. (2003). A comparative study and phage typing of silage-making Lactobacillus bacteriophages. J. Biosci. Bioeng. 95, 518–525.

|

119

120

|

Emond and Moineau

Dupont, K., Vogensen, F.K., and Josephsen, J. (2005). Detection of lactococcal 936-species bacteriophages in whey by magnetic capture hybridization PCR targeting a variable region of receptor-binding protein genes. J. Appl. Bacteriol. 98, 1001–1009. Dupont, K., Janzen, T., Vogensen, F.K., Josephsen, J., and Stuer-Lauridsen, B. (2004). Identification of Lactococcus lactis genes required for bacteriophage adsorption. Appl. Environ. Microbiol. 70, 5825– 5832. Durmaz, E., and Klaenhammer, T.R. (1995). A starter culture rotation strategy incorporating paired restriction/modification and abortive infection bacteriophage defenses in a single Lactococcus lactis strain. Appl. Environ. Microbiol. 61, 1266–1273. Durmaz, E., Madsen, S.M., Israelsen, H., and Klaenhammer, T.R. (2002). Lactococcus lactis lytic bacteriophages of the P335 group are inhibited by overexpression of a truncated CI repressor. J. Bacteriol. 184, 6532–6544. Durmaz, E., and Klaenhammer, T.R. (2000). Genetic analysis of chromosomal regions of Lactococcus lactis acquired by recombinant lytic phages. Appl. Environ. Microbiol. 66, 895–903. Foley, S., Lucchini, S., Zwahlen, M.C., and Brüssow, H. (1998). A short noncoding viral DNA element showing characteristics of a replication origin confers bacteriophage resistance to Streptococcus thermophilus. Virology 250, 377–387. Forde, A., and Fitzgerald, G.F. (1999a). Analysis of exopolysaccharide (EPS) production mediated by the bacteriophage adsorption blocking plasmid, pCI658, isolated from Lactococcus lactis ssp. cremoris HO2. Int. Dairy J. 9, 465–472. Forde, A., and Fitzgerald, G.F. (1999b). Bacteriophage defence systems in lactic acid bacteria. Antonie van Leeuwenhoek 76, 89–113. Fortier, J.-C., Bouchard, J.D., and Moineau, S. (2005). Expression and site-directed mutagenesis of the lactococcal abortive phage infection protein AbiK. J. Bacteriol. 187, 3721–3730. Garvey, P., Hill, C., and Fitzgerald, G.F. (1996). The lactococcal plasmid pNP40 encodes a third bacteriophage resistance mechanism, one which affects phage DNA penetration. Appl. Environ. Microbiol. 62, 676–679. Geller, B.L., Ngo, H.T., Mooney, D.T., Su, P., and Dunn, N. (2005). Lactococcal 936-species phage attachment to surface of Lactococcus lactis. J. Dairy Sci. 88, 900–907. Hall, E.A., Kavanagh, F., and Asheshov, I.N. (1951). Action of forty-five antibacterial substances on bacterial viruses. Antibiot. Chemother. 1, 233–265. Heller, K., and Braun, V. (1979). Accelerated adsorption of bacteriophage T5 to Escherichia coli F, resulting from reversible tail fiber-lipopolysaccharide binding. J. Bacteriol. 139, 32–38. Hendrix, R.W. (2005). Bacteriophage evolution and the role of phages in host evolution. In: Phages: Their Role in Bacterial Pathogenesis and Biotechnology, M.K. Waldor, D.I. Friedman, and S.L. Adhya, eds. (Washington, D.C.: ASM Press), pp. 55–65. Henick-Kling, T., Lee, T.H., and Nicholas, D.J.D. (1986). Inhibition of bacterial growth and malolactic fermentation in wine by bacteriophage. J. Appl. Bacteriol. 61, 287–294. Hill, C., Miller, L.A., and Klaenhammer, T.R. (1990). Cloning, expression, and sequence determination of a bacteriophage fragment encoding bacteriophage resistance in Lactococcus lactis. J. Bacteriol. 172, 6419–26. Hill, C., Miller, L.A., and Klaenhammer, T.R. (1991). In vivo genetic exchange of a functional domain from a type II A methylase between lactococcal plasmid pTR2030 and a virulent bacteriophage. J. Bacteriol. 173, 4363–4370. Houlberg, U., Herbsleb, P., and Sturino, J. (2005). Method and system for colorimetric determination of a chemical or physical property of a turbid medium. Patent: WO2005068982. Jarvis, A.W., Fitzgerald, G.F., Mata, M., Mercenier, A., Neve, H., Powell, I.B., Ronda, C., Saxelin, M., and Teuber, M. (1991). Species and type phages of lactococcal bacteriophages. Intervirology 32, 2–9. Kim, S.G., and Batt, C.A. (1991). Antisense mRNA-mediated bacteriophage resistance in Lactococcus lactis subsp. lactis. Appl. Environ. Microbiol. 57, 1109–1113. Klaenhammer, T.R., and Fitzgerald, G.F. (1994). Bacteriophage and bacteriophage resistance. In: Genetics and Biotechnology of Lactic Acid Bacteria, M.J. Gasson and W.M. de Vos, eds. (London: Chapman and Hall), pp. 106–168. Labrie, S., and Moineau, S. (2000). Multiplex PCR for detection and identification of lactococcal bacteriophages. Appl. Environ. Microbiol. 66, 987–994.

Bacteriophages and Food Fermentations

Lamothe, G., Levesque, C., Bissonnette, F., Cochu, A., Vadeboncoeur, C., Frenette, M., Duplessis, M., Tremblay, D., and Moineau, S. (2005). Characterization of the cro-ori region of the Streptococcus thermophilus virulent bacteriophage DT1. Appl. Environ. Microbiol. 71, 1237–1246. Ledeboer, A.M., Bezemer, S., de Haard, J.J.W., Schaffers, I.M., Verrips, C.T., van Vliet, C., Düsterhöft, E.-M., Zoon, P., Moineau, S., and Frenken, L.G.J. (2002). Preventing phage lysis of Lactococcus lactis in cheese production using a neutralizing heavy-chain antibody fragment. J. Dairy Sci. 85, 1376–1382. Lepeuple, A.S., Van Gemert, E., and Chapot-Chartier, M.P. (1998). Analysis of the bacteriolytic enzymes of the autolytic Lactococcus lactis subsp. cremoris strain AM2 by renaturing polyacrylamide gel electrophoresis: identification of a prophage-encoded enzyme. Appl. Environ. Microbiol. 64, 4142–4148. Lillehaug, D. (1997). An improved plaque assay for poor plaque-producing temperate lactococcal bacteriophages. J. Appl. Microbiol. 83, 85–90. Little, J.W. (2005). Lysogeny, prophage induction, and lysogenic conversion. In: Phages: Their Role in Bacterial Pathogenesis and Biotechnology, M.K. Waldor, D.I. Friedman, and S.L. Adhya, eds. (Washington: ASM Press), pp. 37–54. Looijesteijn, P.J., Trapet, L., de Vries, E., Abee, T., and Hugenholtz, J. (2001). Physiological function of exopolysaccharides produced by Lactococcus lactis. Int. J. Food Sci. 64, 71–80. Lu, Z., Breidt, F., Plengvidhya, V., and Fleming, H.P. (2003). Bacteriophage ecology in commercial sauerkraut fermentations. Appl. Environ. Microbiol. 69, 3192–3202. Madera, C., Monjardin, C., and Suarez, J.E. (2004). Milk contamination and resistance to processing conditions determine the fate of Lactococcus lactis bacteriophages in dairies. Appl. Environ. Microbiol. 70, 7365–7371. Maniloff, J., Cadden, S.P., and Putzrath, R.M. (1981). Maturation of an enveloped budding phage: mycoplasmavirus L2. Prog. Clin. Biol. Res. 64, 503–513. McGrath, S., Fitzgerald, G.F., and van Sinderen, D. (2001). Improvement and optimization of two engineered phage resistance mechanisms in Lactococcus lactis. Appl. Environ. Microbiol. 67, 608–616. McGrath, S., Fitzgerald, G.F., and van Sinderen, D. (2002). Identification and characterization of phageresistance genes in temperate lactococcal bacteriophages. Mol. Microbiol. 43, 509–520. McIntyre, K., Heap, H.A., Davey, G.P., and Limsowtin, G.K.Y. (1991). The distribution of lactococcal bacteriophage in the environment of a cheese manufacturing plant. Int. Dairy J. 1, 183–197. Moineau, S., Fortier, J., and Pandian, S. (1992). Direct detection of lactococcal bacteriophages in cheese whey using DNA probes. FEMS Microbiol. Lett. 92, 169–174. Moineau, S., Pandian, S., and Klaenhammer, T.R. (1993a). R/M systems and restriction endonucleases are more effective on lactococcal bacteriophages that have emerged recently in the industry. Appl. Environ. Microbiol. 59, 197–202. Moineau, S. (1999). Applications of phage resistance in lactic acid bacteria. Antonie van Leeuwenhoek 76, 377–382. Moineau, S., Bernier, D., Jobin, M., Hebert, J., Klaenhammer, T.R., and Pandian, S. (1993b). Production of monoclonal antibodies against the major capsid protein of the Lactococcus bacteriophage ul36 and development of an enzyme-linked immunosorbent assay for direct phage detection in whey and milk. Appl. Environ. Microbiol. 59, 2034–2040. Moineau, S., and Levesque, C. (2005). Control of bacteriophages in industrial fermentations. In: Bacteriophages: Biology and Applications, E. Kutter and A. Sulakvelidze, eds. (Boca Raton: CRC Press), pp. 285–296. Murata, A., Kitagawa, K., and Saruno, R. (1971). Inactivation of bacteriophages by ascorbic acid. Agr. Biol. Chem. 35, 294–296. Murata, A., Kitawaga, K., and Saruno, R. (1972). Inactivation of bacteriophage by thiol reducing agents. Agr. Biol. Chem. 36, 1065–1067. Neve, H., Kemper, U., Geis, A., and Heller, K. (1994). Monitoring and characterization of lactococcal bacteriophages in a dairy plant. Kieler Milchwirtschaftliche Forschungsberichte 46, 167–178. O’Sullivan, D.J., Hill, C., and Klaenhammer, T.R. (1993). Effect of increasing the copy number of bacteriophage origins of replication, in trans, on incoming-phage proliferation. Appl. Environ. Microbiol. 59, 2449–2456. Ptashne, M. (2004). A Genetic Switch: Phage Lambda Revisited. (New York: Cold Spring Harbor Laboratory Press). Quiberoni, A., Suarez, V.B., and Reinheimer, J.A. (1999). Inactivation of Lactobacillus helveticus bacteriophages by thermal and chemical treatments. J. Food Prot. 62, 894–908.

|

121

122

|

Emond and Moineau

Quiberoni, A., Guglielmotti, D.M., and Reinheimer, J.A. (2003). Inactivation of Lactobacillus delbrueckii bacteriophages by heat and biocides. Int. J. Food Microbiol. 84, 51–62. Rakonjac, J., Feng, J.-N., and Model, P. (1999). Filamentous phages are released from the bacterial membrane by a two-step mechanism involving a short C-terminal fragment of pIII. J. Mol. Biol. 289, 1253–1265. Randall-Hazelbauer, L., and Schwartz, M. (1973). Isolation of the bacteriophage lambda receptor from Escherichia coli. J. Bacteriol. 116, 1436–1446. Rudolph, V. (1978). Bacteriophages in fermentation. Process Biochem. ( Jan), 16–26. Russel, M., and Model, P. (2005). Filamentous bacteriophages, 2nd ed. In: The Bacteriophages, R. Calendar, ed. (New York, NY: Oxford University Press). Sanders, M.E., and Klaenhammer, T.R. (1980). Restriction and modification in group N Streptococci: effect of heat on development of modified lytic bacteriophage. Appl. Environ. Microbiol. 40, 500–506. Schouler, C., Bouet, C., Ritzenthaler, P., Rouet, X., and Mata, M. (1992). Characterization of Lactococcus lactis phage antigens. Appl. Environ. Microbiol. 58, 2479–2484. Sellmer, S., Sievers, M., and Teuber, M. (1992). Morphology, virulence and epidemiology of bacteriophage particles isolated from industrial vinegar fermentations. Syst. Appl. Microbiol. 14, 610–616. Sing, W.D., and Klaenhammer, T.R. (1993). A strategy for rotation of different bacteriophage defenses in a lactococcal single-strain starter culture system. Appl. Environ. Microbiol. 59, 365–372. Snyder, M., and Champness, W. (2003). Molecular Genetics of Bacteria. (Washington DC: ASM Press). Spinelli, S., Desmyter, A., Verrips, C.T., de Haard, H.J.W., Moineau, S., and Cambillau, C. (2006). Lactococcal bacteriophage p2 receptor-binding protein structure suggests a common ancestor gene with bacterial and mammalian viruses. Nat. Struct. Mol. Biol. 13, 85–89. Stahouders, J., and Leenders, G.J.M. (1984). Spontaneously developed mixed-strain cheese starters. Their behavior towards phages and their use in the Dutch cheese industry. Neth. Milk Dairy J. 38, 157. Stamm, W.W., Kittelmann, M., Follmann, H., and Truper, H.G. (1989). The occurrence of bacteriophages in spirit vinegar fermentation. Appl. Microbiol. Biotech. 30, 41–46. Stanley, E., Walsh, L., van der Zwet, A., Fitzgerald, G.F., and van Sinderen, D. (2000). Identification of four loci isolated from two Streptococcus thermophilus phage genomes responsible for mediating bacteriophage resistance. FEMS Microbiol. Lett. 182, 271–277. Sturino, J.M., and Klaenhammer, T.R. (2002). Expression of antisense RNA targeted against Streptococcus thermophilus bacteriophages. Appl. Environ. Microbiol. 68, 588–596. Sturino, J.M., and Klaenhammer, T.R. (2004a). Antisense RNA targeting of primase interferes with bacteriophage replication in Streptococcus thermophilus. Appl. Environ. Microbiol. 70, 1735–1743. Sturino, J.M., and Klaenhammer, T.R. (2004b). Bacteriophage defense systems and strategies for lactic acid bacteria. Adv. Appl. Microbiol. 56, 331–378. Summers, W.C. (2005). History of phage research and phage therapy. In: Phages: Their Role in Bacterial Pathogenesis and Biotechnology, M.K. Waldor, D.I. Friedman, and S.L. Adhya, eds. (Washington: ASM Press), pp. 3–17. Sutherland, I.W., Hughes, K.A., Killman, L.C., and Tait, K. (2004). The interaction of phage and biofilms. FEMS Microbiol. Lett. 232, 1–6. Teuber, M., Andresen, A., and Sievers, M. (1987). Bacteriophage problems in vinegar fermentations. Biotechnol. Lett. 9, 37–38. Tock, M.R., and Dryden, D.T. (2005). The biology of restriction and anti-restriction. Curr. Opin. Microbiol. 8, 466–472. Tremblay, D.M., Tegoni, M., Spinelli, S., Campanacci, V., Blangy, S., Huyghe, C., Desmyter, A., Labrie, S., Moineau, S., and Cambillau.C. (2006). Receptor binding protein of Lactococcus lactis phages: identification and characterization of the saccharide receptor binding site. J. Bacteriol. 188, 2400–2410. Valyasevi, R., Sandine, W.E., and Geller, B.L. (1991). A membrane protein is required for bacteriophage c2 infection of Lactococcus lactis subsp. lactis C2. J. Bacteriol. 173, 6095–6100. Whitman, W.B., Coleman, D.C., and Wiebe, W.J. (1998). Prokaryotes: The unseen majority. Proc. Natl. Acad. Sci. USA 95, 6578–6583. Wommack, K.E., and Colwell, R.R. (2000). Virioplankton: viruses in aquatic ecosystems. Microbiol. Mol. Biol. Rev. 64, 69–114. Yang, J.M., Deurraza, P.J., Matvienko, N., and O’Sullivan DJ. (2006). Involvement of the LlaKR2I methylase in expression of the AbiR bacteriophage defense system in Lactococcus lactis subsp. lactis biovar diacetylactis KR2. J. Bacteriol. 188, 1920–1928.

Bacteriophages and Food Fermentations

Yoon, S.S., Barrangou-Poueys, R., Breidt, F., Jr., Klaenhammer, T.R., and Fleming, H.P. (2002). Isolation and characterization of bacteriophages from fermenting sauerkraut. Appl. Environ. Microbiol. 68, 973–976. Yu, F., and Mizushima, S. (1982). Roles of lipopolysaccharide and outer membrane protein OmpC of Escherichia coli K-12 in the receptor function for bacteriophage T4. J. Bacteriol. 151, 718–722.

|

123

Bacteriophages in Medicine* Andrzej Górski, Jan Borysowski, Ryszard Mię dzybrodzki, and Beata-Weber-Dą browska

5

Abstract Bacteriophages, or phages, are viruses of bacteria. Thus, by their very nature, they can be considered as potential antibacterial agents. Over the past decade or two, the idea of phage therapy, i.e. the use of lytic bacteriophages for both the prophylaxis and the treatment of bacterial infections, has gained special significance in view of a dramatic rise in the prevalence of highly antibiotic-resistant bacterial strains paralleled by the withdrawal of the pharmaceutical industry from research into new antibiotics. As an alternative to “classic” phage therapy, in which whole viable phage particles are used, one can also employ bacteriophage-encoded lysis-inducing proteins, either as recombinant proteins or as lead structures for the development of novel antibiotics. Two additional, rather less-recognized potential medical applications of phages are the treatment of viral infections and their use as immunizing agents in diagnosing and monitoring patients with immunodeficiences. We also discuss very interesting novel findings demonstrating the immunomodulatory activity of bacteriophages, suggestive of a potential role of endogenous phages in maintaining the homeostasis of the immune system. Introduction Bacteriophages, or phages, are viruses that infect solely bacterial cells. The rich history of research into bacteriophages began with their discovery by Frederick Twort in 1915 and, independently, by Felix d’Herelle in 1917. Although at the beginning of the twentieth century the knowledge of bacteriophages was very scant and it took several decades to formulate its basic principles, investigators even then realized phages’ antibacterial activity and tried to exploit it for therapeutic purposes. Indeed, in view of their very nature, i.e. the capability to infect and, in the case of lytic phages, also kill bacteria, bacteriophages seem to constitute apparent potential antibacterial agents, with numerous studies (both preclinical

* “Polen habe in den vergangengen Jahrzehnten international die Fuehrung in klinischer Phagen-Therapie uebernommen”—U.Bettge: Viren als Verbuendete gegen Infektionen. Die Welt, 6 December, 2004. “Poland has taken on the international leadership in clinical phage therapy in the last decades”—U. Bettge: Viruses as allies against infections.

126

|

Górski et al.

and those performed in humans) confirming their high efficacy in this regard. However, we believe that phages should not be viewed merely as “bacterial viruses,” as they are also capable of exerting other interesting and sometimes unexpected activities, at least some of which apparently can be exploited in medicine. This chapter has two major goals: (1) to present medical applications of bacteriophages and (2) to highlight the potential role of endogenous phages in the regulation of different (patho)physiological processes in the human organism. Although the latter line of research into bacteriophages is a rather new field of study, it obviously provides new interesting insights into phage biology and holds promise of introducing new practical applications of phages. Phage therapy Background Over the past years, antibiotic resistance has emerged as a major public-health crisis. The consequences of the appearance and spread of antimicrobial resistance have included increasing morbidity, mortality, and the cost of health care. Suffice to say that rates of methicillin-resistant Staphylococcus aureus bacteremia in children in the UK increased 19fold between 1990 and 2001 (Editorial Lancet, 2004). At the same time, the “antibiotic development pipeline runs dry” and “drug companies snub antibiotics,” to cite only a few headlines taken from articles recently published in leading biomedical journals (Clarke, 2003; Nelson, 2003). For example, of 89 drugs newly approved by the FDA in 2002, none were antibiotics, and only nine new antibiotics have been approved since 1998 (Remuzzi et al., 2004). Thus, increasing levels of antibiotic resistance, an insecure pipeline, and a dwindling number of companies dealing with anti-infective agents are causing an unsettling impasse in medicine (Norrby et al., 2005; Wenzel, 2004), which is further aggravated by recent reports indicating that antibiotic treatment may increase the risk of developing breast cancer in women (Velicer et al., 2004). In light of the above menace, many experts believe that there is sufficient scientific justification to study and develop the use of bacteriophages for therapy and prophylaxis. Although phages were discovered almost a century ago, only recently has their true potential in medicine been recognized, not only as classical “phage therapy,” but also as a tool in diagnostics and, possibly, immunotherapy. Preclinical studies Well-controlled preclinical studies have revealed several important features of bacteriophages that may have great significance in the context of the use of phages as antibacterial agents in humans. These include: (1) a potent antibacterial activity against both Grampositive and Gram-negative bacteria regardless of antibiotic sensitivity in vivo, (2) a mode of action associated with direct killing of bacteria, (3) apparent safety, (4) good systemic distribution, including the central nervous system, and (5) a relatively low probability of developing resistance. The most thorough research into phage antibacterial activity in vivo was conducted by Smith and Huggins (1982). In this classic study, one dose of phage was shown to be actually more efficacious than multiple doses of four different antibiotics, i.e. tetracycline, ampicillin, chloramphenicol, and trimethoprim plus sulphafurazole, and at least as effective as streptomycin in curing mice of a potentially lethal E. coli infection

Bacteriophages in Medicine

(Smith and Huggins, 1982). The higher therapeutic effectiveness of bacteriophages was seemingly determined by a phage-unique capacity for exponential growth, a phenomenon that is dependent upon replication within bacterial cells (Levin and Bull, 1996). In this regard, the phage titer actually grows over the course of the treatment, rendering therapy far more efficacious. Apparently, it was Smith and Huggins’ research that prompted a resurgence of interest in experimental phage therapy. Interestingly, several studies conducted since then have consistently confirmed that just single doses of phage exert a significant antibacterial activity in vivo in infections caused by different species of both Gram-positive (Matsuzaki et al., 2003) and Gram-negative bacteria (Soothill, 1992; Barrow et al., 1998). Particularly impressive results were obtained by Soothill, who showed that as few as 102 (!) phage virions are sufficient to protect mice challenged with 108 A. baumanii cells (Soothill, 1992). Several studies have also been conducted to assess the potential of phages for the treatment of alimentary tract infections of bacterial etiology. One of the most painstaking studies in this regard was conducted by Chibani-Chennoufi et al., who found that four orally applied T4-like coliphages were capable of transiting the entire alimentary canal of mice without loss of viability. The phages were not absorbed in the gut, and no virions were detectable in the mesenteric lymph nodes or the liver. Following oral administration, viable phages were consistently recovered from the feces in a dose-dependent manner, with the minimal dose required for recovery being merely 103 pfu per milliliter of drinking water. Importantly, oral administration of phages did not bring about any histopathological changes in the gut mucous membrane. Another fundamental conclusion arising from this research was that only bacteria recently introduced into the murine gut could be successfully killed by phage, whereas the resident intestinal microflora was generally resistant, even though it is composed of bacteria that are sensitive to the lytic activity of phages in vitro. Apparently, either the metabolic state of the microflora or its localization within the mucin layer protects the bacteria from phage lytic activity. However, for no apparent reason, the lytic activity of phages against bacteria recently introduced into the gut was only transient and failed to completely eliminate them (Chibani-Chennoufi et al., 2004). In another study, coliphage Esc-A administered orally was shown to be superior to the antibiotic chloromycetin with respect to decreasing both the incidence of E. coli diarrhea and the associated death rate in chickens. Interestingly, phage therapy using the coliphage reduced the chickens’ susceptibility to other intestinal infectious diseases. Another beneficial effect of this treatment was an enhancement of the conversion rate of feed and an increase in the weight of animals compared with antibiotic therapy (Xie et al., 2005). Taken together, the above findings indicate that bacteriophages may provide a safe and effective means of treating diarrhea of bacterial etiology without disturbing the balance of the natural gut microflora. Bacteriophages are also considered as a potential means of treating tuberculosis and other mycobacterial diseases. Basically, due to a very low number of relevant original papers with generally inconsistent results, at present no general conclusions can be drawn as to the real therapeutic potential of mycobacteriophages. However, in view of the prevalence of tuberculosis and other diseases of mycobacterial etiology, especially those caused by multidrug resistant mycobacteria, further research into the therapeutic use of mycobacteriophages appears to be needed (McNerney et al., 2005).

|

127

128

|

Górski et al.

Potential phage applications also include stomatology: the addition of phages to root canals experimentally infected with bacteria led to a substantial reduction in bacteria or their complete eradication even when the root dentin was infected (Paisano et al., 2004). One of the most fundamental questions regarding phage therapy is that of the major mode of antibacterial action of bacteriophages. Theoretically, two basic mechanisms could be considered in this regard, i.e. direct killing of bacterial cells by phage virions and the antibacterial immune response induced by some component(s) of phage preparation. Although phage lysates are indeed highly immunogenic, and therefore potentially capable of exerting an antibacterial immune response in vivo (Sulakvelidze et al., 2005), a considerable body of experimental findings clearly points to direct killing of bacteria by phage particles as the major mode of antibacterial action of bacteriophages. For example, only functional phage particles, i.e. those capable of lysing bacteria in vitro, also have the capacity to rescue mice challenged with a lethal dose of bacteria, whereas the therapeutic effect of both non-functional phages (Biswas et al., 2002; Wang et al., 2006a; Wang et al., 2006b) and a mechanical lysate of bacterial cells (an analog of phage lysate) (Matsuzaki et al., 2003) is drastically diminished. Furthermore, there is a correlation between phages’ antibacterial activity found in vitro and their therapeutic effect observed in vivo, i.e. only phages lysing bacteria in vitro are capable of curing infection, and those acting more potently in vitro are also more efficient in vivo (Smith and Huggins, 1982). And finally, upon administration of phages to infected experimental animals, a decrease in the number of bacterial cells is paralleled by an increase in the quantity of phage virions, suggestive of a replication of bacteriophages combined with the killing of host bacterial cells (Smith and Huggins, 1982). Thus, it is direct killing of bacteria that is certainly the major and, apparently, also the solely significant mode of the antibacterial action of bacteriophages. In addition to lytic bacteriophages, i.e. those killing their host bacterial cells during replication (Sulakvelidze et al., 2001), one can use for therapeutic purposes also non-lytic filamentous phages (e.g. M13), which do not kill their host cells during infection (Russel, 1995). In this approach, genetically engineered filamentous phages themselves are not agents directly killing bacteria, but rather constitute vehicles delivering phagemids encoding some bactericidal proteins to bacterial cells. This novel method, still at a very early stage of development, was shown to be effective both in vitro and in vivo (in a murine model of bacteremia) (Westwater et al., 2003). As phages kill bacteria by a mode of action that is completely different from those employed by traditional antibiotics, one could expect them to exhibit activity also against antibiotic-resistant bacterial strains. Indeed, four recent studies have highlighted their high efficacy in this regard. In the first, one i.p. dose of phage was sufficient to cure 100% of the mice that received a potentially lethal dose of a clinical isolate of vancomycin-resistant E. faecium (Biswas et al., 2002). Likewise, in a second study just a single dose of bacteriophage administered to mice upon challenge with four different clinical MRSA strains significantly increased survival rates (Matsuzaki et al., 2003). Infections caused by antibiotic-resistant strains of clinically relevant species of Gram-negative bacteria can also be successfully cured with single doses of phage, as reported for extended-spectrum beta-lactamase-producing E. coli (Wang et al., 2006a) and imipenem-resistant P. aeruginosa (Wang et al., 2006b).

Bacteriophages in Medicine

Lysis of bacterial cells leads to the release of different pro-inflammatory membrane and cell wall components, e.g. endotoxin, which may result in life-threatening complications, including septic shock and multiple organ failure (Nau et al., 2002). Recently, the use of lysis-deficient bacteriophages was proposed as an interesting alternative to decrease the risk of rapid endotoxin release during phage therapy (Hagens et al., 2004; Matsuda et al., 2005). Antimicrobial-induced endotoxemia can also be elicited by some antibiotics such as beta-lactams to a level which could even be associated with clinical deterioration (Lepper et al., 2002). In this regard it should be pointed out that in animal experiments, phage therapy is apparently safe and no significant bacteriolysis-induced side effects have been observed. Moreover, in contrast to many antibiotics and their well known adverse events, phage particles themselves are also not likely to exert any detrimental effect on mammalian tissues, as phages, being viruses specific to bacteria, generally display no tropism to eukaryotic cells (although, as will be pointed out later, phage-eukaryotic cell interactions can take place). Another significant feature of bacteriophages having clear implications for phage therapy is their good systemic distribution. Following i.p. administration, for example, phage particles could be recovered in remarkable titers from different tissues and organs of mice, including the liver, spleen, kidneys, skeletal muscles, as well as the brain (Matsuzaki et al., 2003). The capability of phages to cross the blood-brain barrier was also reported by other authors (Dubos et al., 1943; Smith and Huggins, 1982) (more detailed data are presented in our recent review (Dąbrowska et al., 2005b). One of the earlier reproaches against phage therapy was that bacteria can rapidly develop resistance to bacteriophages (Smith and Huggins, 1982). Indeed, phage-resistant bacterial mutants rapidly emerge following exposure to bacteriophages in vitro (Smith and Huggins, 1982; Tanji et al., 2005). However, the appearance of resistant bacteria can be markedly delayed by using a phage “cocktail,” i.e. a mixture of a few different phages (Tanji et al., 2005). Furthermore, the development of resistance to bacteriophages was found to be accompanied by a decrease in bacterial virulence (Smith and Huggins, 1982). Levin and Bull postulated that phages are effective in treating acute infections because they reduce the densities and dissemination rates of the infecting populations of bacteria to levels which can be controlled by the immune system of the host. In contrast, in chronic infections the immune system is unable to clear the infection efficiently, and bacterial populations may remain at substantial densities for years. At the present time it is unclear whether phages could provide an efficient means of eradicating such infections without the evolution of bacterial resistance to phages. The development of such resistance resembles a race between the ability of bacteria to evolve and the counter-evolution of phages to thereby, overcome bacterial resistance. In theory, the evolution of phage resistance by bacteria could also be beneficial for an infected organism, as the development of resistance can also reduce the fitness of the bacteria, change their receptors, and lower their virulence (Levin and Bull, 2004). Phage therapy in humans The history of using phages for the clinical treatment of bacterial infectious diseases and its perspectives have been subject of recent detailed reviews (Sulakvelidze et al., 2001; Stone,

|

129

130

|

Górski et al.

2002; Duckworth and Gulig, 2002; Clewley, 2003; Inal, 2003; Merril et al., 2003; Thacker, 2003; Thiel, 2004; Bradbury, 2005; Brussow, 2005; Sulakvelidze, 2005; Matsuzaki et al., 2005; Parfitt, 2005; Skurnik and Strauch, 2006). Probably the most complete description of efforts aiming at the eradication of bacterial infection using phages was published recently (Sulakvelidze et al., 2005). Therefore it is not the intention of our review to return again to issues already adequately addressed by other authors; rather, we would like to emphasize other aspects of the therapy. As pointed out in reviews on the therapeutic application of phages, after the discovery and introduction of antibiotics, phage therapy was continued only in Eastern Europe and the Soviet Union (the countries most active and experienced in this field have been Georgia, Russia, Poland, and Romania, at least judging from available publications). In fact, the first documented Polish use of phages in human therapy dates back to 1925, when it was applied in patients of the surgery clinic of the Jagiellonian University in Cracow (traditionally, this clinic has had the highest rank among university surgical departments in Poland). The author treated 40 patients with suppurative staphylococcal infections using sterile phage lysates of staphylococci obtained from the patients. While subcutaneous administration of phages was usually not effective, very good results were obtained with topical phage application (e.g. intra-articular injections in cases of joint abscesses and direct phage application to the skin in cases of furunculi). Although this study did not provide definite proof of phage efficacy, we believe it is nevertheless important in that it showed unequivocally the safety of phage therapy, including their administration by parenteral routes. Notably, the author did not observe any local reactions to subcutaneous phage lysate injections (even though he made approx. 100 injections). Similarly, no reactions were seen when phages were applied in compresses soaked with the lysates. Although in some cases there were general reactions, they were weak and subsided within 24 hours after phage administration (elevated temperature). Importantly, there were no significant side effects when phages were given intrapleurally, including the case of a 3-year-old child ( Jasieński, 1927). In 1942, a famous Polish writer was cured of a urinary tract infection; this incident was later described in her memoirs (Dąbrowska, 1988). During the war, phages were successfully used in the Polish army, following the expertise of Red Army doctors (Szarecki, 1942). Likewise, Lityński used phages to treat severe cases of peritonitis caused by gunshot wounds; he later described a case of prompt, phage-induced recovery from postpartum peritonitis in a 22-year-old woman who was too ill to undergo surgical intervention and did not respond to antimicrobial treatment (Lityński, 1949). In the 1960s, phage therapy was successfully used at the Department of Infectious Diseases of the Medical University of Warsaw to treat S. aureus-induced suppurative skin infections (Olczak and Strumiłło, 1961). When Prof. Ludwik Hirszfeld moved to Wroclaw after the World War II, bringing his collection of phages from the State Institute for Hygiene in Warsaw, the Lower Silesian capital became a major center of phage research and therapy, which was further enhanced when our Institute (now bearing his name) was created in 1954. In this way, his idea of the clinical application of the phenomenon referred to as “the battle of the invisible with the imperceptible” could slowly materialize, especially as he justly believed that the efficacy of phages exceeds the strength of the defense mechanisms of the human body (Hirszfeld,

Bacteriophages in Medicine

1948a). Hirszfeld also introduced the application of phages for typing bacterial strains in Poland (Hirszfeld, 1948b), a technique which still has its value and continues to be used today (Sola, 2006). Since then, our Institute’s phages have been used to treat more than 2000 patients with antibiotic-resistant bacterial infections, as presented on our phage website http://www.iitd.pan.wroc.pl/phages/phages.html, as well as in our publications (Ślopek et al., 1983a, Ślopek et al., 1983b; Ślopek et al., 1984; Ślopek et al., 1985a; Ślopek et al., 1985b; Ślopek et al., 1985c; Ślopek et al., 1987; Weber-Dąbrowska et al., 2000a; Weber-Dąbrowska et al., 2003). Our phage laboratory has been isolating, characterizing, and producing phage preparations for therapy performed at clinics and hospitals, mostly in the Lower Silesian region, who are supposed to feed results back to our Institute. While continuing this activity, we opened our own phage therapy center in 2005 at our Institute (an outpatient clinic) which accepts patients from Poland and abroad; further data are available on our website: http:// www.iitd.pan.wroc.pl/phages/phages.html According to Polish law, phage therapy is considered an experimental treatment which is carried out on the basis of the respective legislation (pharmacological law, regulations of the Minister of Health). Experimental treatment (or, translated literally, a therapeutic experiment) occurs when a physician introduces new or only partially tested diagnostic, therapeutic, or prophylactic methods for the direct benefit of the person being treated. In contrast, an investigational experiment has the primary purpose of broadening medical science (and is tantamount to clinical research). To satisfy the existing requirements, two basic items are prerequisites for experimental therapy: (a) the written informed consent of the patient and (b) approval by an institutional review board (bioethics commission). Furthermore, it may be implemented only by a qualified doctor and when available treatment has failed (arts. 29/1, 21/2, and 21/3 of the law on the physician’s profession). Therefore, our current therapy involves cases in which prior antibiotic treatment did not lead to the eradication of infection. The opening of our phage outpatient clinic, the first such unit in a country of the European Union, operating in accordance with currently prescribed administrative, legal, and ethical standards, constitutes, according to Nature Medicine, “a fundamental step” (Hausler, 2006). According to our protocol, 10 ml of phages are administered orally three times daily before eating and after neutralization of the gastric juices. Phages have also been applied directly to wounds, as ear and nose drops, infusions to fistulas, washing of the nasal cavity, intraperitoneally during washing of the peritoneal cavity, and topically in cases of multiple skin abscesses. Phage treatment has been highly effective in infections caused by different species of bacteria: Escherichia, Klebsiella, Proteus, Enterobacter, Pseudomonas, and Staphylococcus aureus, with an average success rate of 85%. Importantly, our sets of phages have been highly efficient against such dangerous pathogens as S. aureus, methicillin-resistant strains (MRSA), and Pseudomonas aeruginosa. There have been only minor side effects in patients, such as gastric discomfort, nausea, or skin irritation (approx. 2% of cases). The safety of the therapy was also confirmed in renal transplant recipients, in whom improvement of allograft function was noted after completion of phage therapy as a result of eradication of urinary tract infection; importantly, no other changes in laboratory parameters were seen in the patients receiving phages (Boratyńska et al., 1994). The potential use

|

131

132

|

Górski et al.

of phages in renal allograft recipients was recently confirmed by a Russian report extracted from Pubmed: Bacteriophages use in the treatment of pyoseptic complications in a female patient with renal allotransplant. Urologiia, 2005, 6, 43–46. An important clinical question is whether phage treatment could be safely applied in patients with primary and secondary immunodeficiencies (e.g. those receiving chemotherapy, immunosuppressive therapy, etc.). Several lines of evidence confirm the safety of phage administration in these clinical settings: (a) antiphage humoral responses following their subcutaneous injection have been used to evaluate immunocompetence in patients with various forms of inborn and acquired immunodeficiency syndromes, as described in detail in one of the next chapters, (b) our preliminary results in transplant patients cited above as well as the successful use of phages in treating bacterial infections in cancer patients (Weber-Dąbrowska et al., 2001) are encouraging, (c) patients treated with phages have been previously receiving antibiotics, sometimes for prolonged periods, and therefore many of them may have immunodeficiency anyway, and (d) treatment with phage lysates appears to produce immunostimulatory effects (see the chapter on phage-mediated immunomodulation). Therefore, immunodeficiency and transplantation do not appear to be contraindications for phage therapy. Recently, a very clear response to phage therapy was seen in a female patient who had undergone removal of a pituitary tumor by the transphenoid approach. Her postoperative course was complicated by a staphylococcal infection at the site of surgical intervention. Several weeks of antibiotic therapy did not clear the infection and she was subsequently treated with a course of oral anti-staphylococcal phage, again without a clear response. Eventually, a phage preparation was applied using a cannula communicating with the site of infection, this time with a prompt and dramatic response, normalization of temperature, and general well-being. The striking effectiveness of this route of phage administration highlights the potential future application of phages in cocaine esterase therapeutics. Carrera has shown the therapeutic potential of a phage-displayed cocaine binding antibody using a rat model of cocaine addiction (Carrera et al., 2004). A further study by this group includes the use of cocaine esterase displayed on a phage that is able to penetrate the central nervous system after intranasal application (Rogers et al., 2005). Colonization with MRSA has been described as a risk factor for subsequent MRSA infection, and the presence of MRSA in the gastrointestinal tract of patients in intensive care units and in prospective liver graft recipients poses a difficult clinical dilemma (Barret, 2005; Woeste et al., 2005). In addition, controlling the MRSA carrier status among healthcare workers poses a significant logistic challenge. A nurse from a university hospital of the Warsaw Medical University had an episode of MRSA-dependent urinary tract infection, and the intestinal tract was identified as a colonization source for the genitourinary tract. MRSA sensitivity to specific phage strains was confirmed at our Institute, and a phage preparation containing three of the most efficient anti-MRSA phage strains was prepared. The preparation contained polyvalent, virulent S. aureus phages, causing complete lysis of the MRSA strain isolated from the patient. Rectal swabs were obtained during the standard treatment with the phage preparation. Single MRSA colonies were detectable during the first week of phage treatment, the rectal swabs becoming negative thereafter. Decolonization was confirmed by negative results of rectal swabs taken at monthly intervals during the next

Bacteriophages in Medicine

half year as well as of fecal and vaginal cultures. No side effects of phage therapy, nor any significant changes in laboratory parameters were noted (Leszczyński et al., 2006). These findings suggest that oral phages could be used for the decolonization of MRSA carriers, which is usually a difficult and costly procedure. The value of a specific bacteriophage in eradicating carrier status for another bacterium (Shigella) was already described by our Institute’s team some 40 years ago (Mulczyk et al., 1967). An important recent contribution by an Irish group indicates that washing human fingers with anti-Staphyloccocus phage K (a polyvalent phage able to lyse MRSA strains) reduces the number of staphylococci remaining on skin 100-fold in contrast to Ringer’s solution, which causes only slight effects. Furthermore, even though the MRSA strains may be weakly sensitive to the phage, they can still be lysed by phage variants obtained by passage of the phage through these initially insensitive bacteria. This polyvalent anti-Staphylococcus phage was also highly active as a cream. These results suggest that anti-MRSA phages can control hospital S. aureus infections when used in hand wash or a cream (O’Flaherty et al., 2005b). Our experience with phage therapy is similar to descriptions of its effectiveness and safety provided by the Eliava Institute in Tbilisi, Georgia, Russian centers, as well as a small number of reports originating from other Eastern European centers (Bucharest) (Sulakvelidze et al., 2005). Taken together, these are quite extensive studies which probably involve several thousand patients. The most important conclusion derived from these studies is the safety of phage therapy: virtually no side effects were encountered, and no significant alterations in laboratory parameters were noted in patients treated. Since no therapeutic study has been performed in accordance with current rigorous requirements for clinical trials (appropriate controls, randomization, etc.), their scientific value is limited, but the results strongly suggest that phages can combat bacterial infections and that clinical trials performed according to the currently required standards of clinical investigation are urgently needed. Our protocols of phage therapy involve patients with antibiotic-resistant infections in whom antimicrobials should normally be discontinued. However, some Russian and Georgian centers use combinations of antibiotics and phages, claiming the superiority of such protocols (Sulakvelidze et al., 2005). On the other hand, antibiotics may impair bacterial ability to support phage replication and thereby the effectiveness of phage therapy (Merril et al., 2003). Lack of controlled studies on this important issue leaves the problem open for further investigation. Probably the first safety test done according to current standards that also included evaluation of the bioavailability of phages was performed recently by the Swiss scientists (Bruttin and Brussow, 2005). Using oral administration of T4 phages, no adverse events related to the study intervention were reported, including normal aminotransferase levels prior to and after the study. Although the value of this report is limited (T4 has a narrow host range on E. coli and would therefore not be expected to be used in therapy), it confirms earlier reports indicating the safety of phage treatment. Those reports, confirming that the administration of phages does not produce harmful effects and is well tolerated, are not unexpected when one takes into consideration the fact that phages are omnipresent, i.e. are also present in the bodies of animals and man, and their presence has been documented in

|

133

134

|

Górski et al.

feces, saliva, sputum, as well as bovine and calf serum. Of particular interest are two reports in the literature describing natural phage occurrence in human sera. There, Mankiewicz and Liivak demonstrated mycobacteriophages in the sera of the majority of patients with sarcoidosis (Mankiewicz and Liivak, 1967), and Parent and Wilson found these phages in the sera of some patients with Crohn’s disease. Interestingly, some normal subjects’ sera were also positive (Parent and Wilson, 1973). Endogenous phage circulation in normal and diseased individuals is an exciting phenomenon which, regretfully, has not been explored further. We believe that such phage circulation may confirm our recent hypothesis on phage translocation (Górski et al., 2006b), further supported by observations made already 80 years ago showing that blood and tissues of domestic fowl may contain phages causing lysis of various strains of Salmonella pullora (Pyle, 1926). Furthermore, it was shown that phages may penetrate mouse skin and migrate to the bloodstream (Keller, 1958). The possible phage occurrence in urine is also very interesting: such phages could, in theory, play an important role in protecting the urinary tract against invading bacteria. Again, the literature addressing this important problem is incredibly scarce: we have found only one paper on the issue of phage presence in human urine that addresses the problem in some detail (from 1928!) and another short report on the same subject based on a single case (Caldwell, 1928; Caroli et al., 1980). In contrast, a combination of the terms “bacteria + urine” yields more than 14 000 Pubmed records. Evidently, the subject of possible phage circulation in human blood and their secretion in urine requires more in-depth studies. The occurrence of phages in the human body, especially in the intestinal tract, raises the possibility that phages can translocate in a process similar to the well-known bacterial translocation (Górski et al., 2006b). The problem of whether phages can pass the gut barrier, reaching the peripheral lymph and blood, is still open, but there are studies suggesting that phages administered orally in mice can be later found in the peripheral blood, although other authors have not confirmed that such passage can indeed take place (for more details, see our recent review on phage penetration in vertebrates by Dąbrowska et al., 2005b). Similarly, authors from our Institute (Weber-Dąbrowska et al., 1987) as well as Babalova et al. (cited by Sulakvelidze et al., 2005) have determined that phages can migrate to the blood during oral treatment of patients; on the other hand, Bruttin and Brussow were unable to detect T4 coliphage following its oral administration to healthy volunteers (Bruttin and Brussow, 2005). It appears that phage translocation may be determined by a variety of factors, including phage type, dosage (phages given in higher amounts are more likely to pass the gut barrier), neutralization of gastric juice, as well as the individual treated (it may well be that phage translocation can occur more easily in patients than in normal donors: the gut barrier may be more permeable in disease, as occurs in bacterial translocation). Similar mechanisms may be responsible for phage transit to the urinary tract; for example, it has been demonstrated in mice that phages appear in urine when a threshold dose of 109/mouse is applied and serum phagemia exceeds 105 pfu/ml. Phages can be concentrated by the kidneys, and urine does not appear to affect their antibacterial action, a fact with great practical implications (Keller and Engley, 1958; Keller and Zatzman, 1959; Schultz and Neva, 1965; Uhr and Weissman, 1965). Recently we summarized available data on phage penetration and bioavailability (Dąbrowska et al., 2005b). As pointed out by Meril et al., pharmacokinetics data are still

Bacteriophages in Medicine

rudimentary and their veracity is complicated by the fact that phages can replicate exponentially. Formal mathematical models have been developed to study the complex interactions of phages with bacteria and the host; nevertheless, only detailed clinical studies can provide the necessary data (Merril et al., 2003). Natural and exogenous phage circulation in the human body may also be important in light of our hypothesis implying that endogenous phages may have protective potential not only against invading bacteria, but also viruses and, perhaps, the development of some forms of cancer (Górski et al., 2005). Phages may play an important role in the epidemiology of disease. For example, it has been suggested that phage amplification by cholera victims leads to an increase in the phage concentration in the environment, with subsequent predation decreasing the load of viable V. cholerae and collapse of the epidemic (Faruque et al., 2005). Bacteriophage lytic proteins Background One of the essential functions of the cell wall is protection of the bacterial cell against osmotic rupture (lysis), which could be effected by the high intracellular osmotic pressure typical of bacteria. This function is determined largely by peptidoglycan (murein), a heteropolymer of alternating N-acetylglucosamine and N-acetylmuramic acid connected by glycoside bonds, whose adjacent chains are cross-linked by short peptides (Todar’s Online Textbook of Bacteriology). Thus, upon damage to the proper structure of peptidoglycan, bacterial cells rapidly undergo lysis (Bernhardt et al., 2002). Over the course of evolution, bacteriophages have developed some very efficient means of inducing lysis, which enables progeny virions to be successfully liberated from the host cell. Exploitation of phageencoded lysis-inducing proteins for both the prophylaxis and the treatment of bacterial infections appears to be a viable alternative to classic phage therapy, in which whole viable phage particles are administered. Some of these proteins can be used in a natural form (as recombinant proteins), whereas others can provide lead compounds for the development of novel antibiotics. Endolysins Endolysins, or lysins, are enzymes encoded by bacteriophages with large dsDNA genomes (Bernhardt et al., 2002). These are produced in phage-infected bacterial cells during the later stages of the lytic cycle, their major function being the cleavage of peptidoglycan covalent bonds, which leads to lysis of the host bacterial cell and enables progeny virions to be liberated (Loessner, 2005). Toward the end of the lytic cycle, endolysin molecules reach their substrate, peptidoglycan, through lesions in the inner membrane which are formed by phage-encoded hydrophobic proteins termed holins (Young et al., 2000). However, lysins are capable of digesting peptidoglycan also when applied externally to bacterial cells (as recombinant purified proteins). This phenomenon, known as “lysis from without,” occurs especially in Gram-positive bacteria, whereas Gram-negatives are considered to be generally not amenable owing to the presence of the outer membrane (see below) (Nelson et

|

135

136

|

Górski et al.

al., 2001; Loessner, 2005). Whether reaching peptidoglycan from inside the cell or acting from outside, lysins rapidly cleave it, thereby inducing lysis of the bacterial cell (Young et al., 2000, Nelson et al., 2001). Depending on the enzymatic specificity, endolysins fall into five major classes: N-acetylmuramoyl-L-alanine amidases, N-acetylmuramidases (lysozymes), endo-B-N-acetylglucosaminidases, lytic transglycosylases, and endopeptidases. With very few exceptions, endolysins display only one kind of muralytic activity (Borysowski et al., 2006). Many lysins possess a modular structure, i.e. they are made up of at least two major functional domains or modules, with the enzymatic module located typically at the N-terminus and the cell wall-binding module at the C-terminus (Loessner et al., 2002; Hermoso et al., 2003). This modular structure gains special significance in the context of using endolysins as antibacterial agents, as it enables one to alter either of their basic functions independently. To that end, either of the domains can be replaced with the corresponding module from another lysin, thereby yielding an enzyme with a new enzymatic or cell wall-binding specificity (Fischetti, 2005a; Fischetti, 2005b). Furthermore, in some endolysins the C-terminal domain, though responsible for high-affinity binding of the bacterial cell wall, is not essential for antibacterial activity. Interestingly, C-truncated forms of those lysins actually exhibit bacteriolytic activity superior to that of the full-length enzyme molecules. Thus, a viable method of improving the antibacterial activity of such endolysins could be deletion of a fragment of the C-terminus (Low et al., 2005). Other strategies for engineering enzymes might also be used in that regard. One of the most fundamental features of lysins is their rapid and potent antibacterial activity both in vitro and in vivo, especially against Gram-positive bacteria (see below). For example, the release of intracellular ATP from bacterial cells can be observed within 10 seconds upon the addition of enzyme (Schuch et al., 2002). In another study, complete cell death of ~107 group A streptococci was achieved within 5 seconds upon exposure to 1000 U of lysin (Nelson et al., 2001). In a rat endocarditis model, endolysin was also found to exert a more rapid antibacterial activity than vancomycin both in blood and in aortic vegetations (Entenza et al., 2005). Importantly, lysins kill bacteria by a mechanism that is completely different from those employed by antibiotics, i.e. the cleavage of peptidoglycan covalent bonds. Thus, predictably, they are also capable of killing antibiotic-resistant bacteria, including penicillin-resistant S. pneumoniae (Loeffler et al., 2001; Loeffler et al., 2003a; Loeffler et al., 2003b; Jado et al., 2003; Entenza et al., 2005), vancomycin-resistant E. faecium and E. faecalis (Yoong et al., 2004), as well as methicillin-resistant S. aureus (O’Flaherty et al., 2005a). Another typical feature of bacteriophage lytic enzymes is their narrow antibacterial range, limited basically to the host species (Loeffler et al., 2001; Schuch et al., 2002) or genus (O’Flaherty et al., 2005a) of the phage that encodes a given enzyme. In view of this lysins, unlike traditional antibiotics, provide an opportunity for the targeted killing of single pathogenic bacterial species without disturbing the balance of natural microflora (Fischetti, 2005a). Endolysins were originally developed as a novel means of targeted killing of potentially pathogenic bacteria colonizing mucous membranes (Nelson et al., 2001; Fischetti, 2003). Not only does mucosal colonization provide a potential starting point for infection, but it also results in the horizontal spread of pathogenic bacteria within the community. For

Bacteriophages in Medicine

that reason its elimination leads to a decrease in both the incidence of infection and the community spread of bacteria (Fischetti, 2001; Fischetti, 2005a). Thus far, four endolysins have been tested in different murine models of mucosal colonization. These include a lysin of group C streptococci C1 phage (Nelson et al., 2001), phage Dp-1 Pal amidase (Loeffler et al., 2001), Cp-1 phage Cpl-1 lysozyme (Loeffler et al., 2003b), and phage NCTC 11261 PlyGBS (Cheng et al., 2005). All were found highly effective, with just single doses being sufficient for exerting a significant antibacterial activity in vivo, and in some cases for clearing of bacteria from the surface of mucosa. Thus, owing to the capacity for rapid targeted killing of single pathogenic bacterial species, endolysins constitute a novel and unique means of selective, i.e. generally species-specific, prophylaxis of bacterial infections (Fischetti, 2005a). In view of the prevalence and significance of mucosal colonization in humans, endolysins are quite likely to find wide application in medicine in the future. Another obvious application of lysins is the treatment of invasive bacterial infections. Thus far, these have been investigated in murine models of bacteremia (Loeffler et al., 2003b; Jado et al., 2003), B. anthracis infection (Schuch et al., 2002) and a rat model of endocarditis (Entenza et al., 2005). In essence, all of these studies revealed the high potential of endolysins, with just single doses being sufficient for exerting a significant therapeutic effect. It should also be stressed that in none of the studies have any serious side effects been observed. Although endolysin therapy leads to the release of different cytokines, including pro-inflammatory ones (Entenza et al., 2005), it was not found to result in any serious complications, e.g. shock as a potential consequence of massive bacteriolysis. One of the major potential obstacles to lysin treatment could be the development of resistance. However, it has been argued that the emergence of resistance to endolysins is rather unlikely, as these enzymes target some essential constituents of the bacterial cell wall. A classic example is provided by lytic enzymes encoded by the pneumococcal phages that bind choline, a component of the pneumococcal cell wall that is essential for bacterial viability (Fischetti, 2005b). Experimental findings appear to confirm this assumption. For example, even following repeated exposure of bacteria to lower doses of enzyme on both agar plates and in liquid cultures, no resistant mutants could be recovered (Loeffler et al., 2001; Schuch et al., 2002). However, it should be noted that some changes occurred in the cell wall during the stationary phase that decreased bacterial susceptibility to lysins (Loeffler et al., 2001; Pritchard et al., 2004). Furthermore, one of the biggest challenges to future endolysin treatment is Gram-negative bacteria that, due to the presence of the outer membrane, are generally considered not amenable to lysin activity. Nevertheless, the results of some studies suggest that Gram-negative bacteria may also be successfully killed by lysins (Borysowski et al., 2006). One of the major shortcomings of endolysins is their apparent immunogenicity, i.e. the capacity to induce a specific humoral immune response following both mucosal and systemic administration (Fischetti, 2005b). However, a considerable body of experimental findings strongly suggests that “neutralizing” antibodies actually do not neutralize, but rather slightly reduce endolysin’s antibacterial activity both in vitro and in vivo (Loeffler et al., 2003b). Moreover, there are some effective means of improving the proteins’ pharmacokinetics, e.g. conjugation to polyethylene glycol (Walsh et al., 2003). Thus, at the current, rather preliminary state of research into the therapeutic applications of lysins, it appears that im-

|

137

138

|

Górski et al.

munogenicity in fact might not be a very serious obstacle to their use in the treatment of systemic bacterial infections. Other lytic proteins Bacteriophages with small ssRNA or DNA genomes have developed an entirely different mechanism of releasing their progeny from bacterial cells. These induce lysis of a bacterial cell by means of a single lytic protein which displays no enzymatic activity but rather inhibits peptidoglycan synthesis (Bernhardt et al., 2002). Thus far, three prototypical lytic proteins have been reported, i.e. the QB phage A2 protein (Bernhardt et al., 2001b), the Ms2 phage L protein (Bernhardt et al., 2002), and the F X174 phage E protein (Bernhardt et al., 2001a). Both the A2 protein and the E protein indeed have been shown to constitute cell wall-synthesis inhibitors: the former specifically blocks the activity of MurA (Bernhardt et al., 2001b), whereas the latter inhibits MraY, or translocase I (Bernhardt et al., 2001a). In the case of the L protein, an analogous mechanism (i.e. inhibition of one of the enzymes of the pathway for peptidoglycan synthesis) is also very likely, though not proven as yet (Bernhardt et al., 2002). Owing to their major mode of action, i.e. murein-synthesis inhibition, lytic proteins encoded by small phages were collectively termed amurins (Bernhardt et al., 2002). As opposed to endolysins, these are quite unlikely to be used as antibacterial agents in their natural form. Their major application appears rather for use as lead structures for the synthesis of novel non-peptide small-molecule peptidoglycan-synthesis inhibitors. Due to the conserved character of their target enzymes, such inhibitors should feature a broad spectrum of antibacterial activity (Fischetti, 2001). Immunomodulatory effects of phages Antimicrobial agents are biological response modifiers which can exert potent systemic immunomodulatory effects (van der Meer, 2003). It is also well known that bacteria and their products may cause immunomodulation, their prevailing effect being to enhance immune responses, a phenomenon which found practical implications in attempts to correct depressed immunity in patients with immune abnormalities (Brockstedt et al., 2005). Furthermore, Leiva et al. recently demonstrated that antimicrobial agents may cause immunomodulatory effects through the release of active fractions from intestinal bacteria; polymyxin and teicoplanin lowered the counts of some bacteria of the intestinal flora and induced significant modifications in spleen cells’ ability to proliferate in response to mitogens (Leiva et al., 2005). In view of these facts, it is not surprising that phage lysates have also been found to exert immunoenhancing action. Staphage lysate (SPL) was found to induce pronounced enhancement of the hemagglutinin response in mice; moreover, when administered alone, SPL evoked a heightened immunoglobulin level (Esber et al., 1985). SPL added in vitro enhanced mitogenic responses of human lymphocytes and may induce immunoglobulin production (Lee et al., 1985). SPL has also been used for the in vitro measurement of the functional reactivity of lymphocytes from normal and immunodeficient patients (Wanebo et al., 1988). Phage lysates administered to patients were shown to cause normalization of TNF-alpha and IL-6 production by blood cell cultures (WeberDąbrowska et al., 2000b). Furthermore, we have shown that lymphocytes from patients treated with staphylococcal lysates show enhanced immunoglobulin production in vitro

Bacteriophages in Medicine

in response to a polyclonal activator pokeweed mitogen (PWM) (Kniotek et al., 2004a). Those data, taken together, strongly suggest that phage lysates exert immunostimulatory action on the immune system, a phenomenon which may contribute to the beneficial effects of phage treatment. In contrast to lysates, purified phage preparations may have immunosuppressive rather than immunostimulatory activities. Our recent studies performed using purified T4 phages in vitro indicate that they inhibit human T cell proliferation triggered via the CD3-TCR complex and downregulate in vivo antibody production to a specific antigen in mice. Moreover, phages extend skin allograft survival in mice in both naïve and sensitized recipients and inhibit mononuclear cell migration to the site of an allograft (Górski et al., 2006a). Furthermore, phages inhibit activation of NF-kappa B (Górski et al., 2006a), a key regulator of transcription after TCR and costimulatory receptor ligation regulating the expressions of almost 400 different genes, including inflammatory cytokines, whose activation has been linked to a variety of human diseases (Ahn and Aggarwall, 2005). Our findings indicating that phages themselves may have immunosuppressive effects in vitro and in vivo are not unexpected, since human viruses have been shown to have similar effects, e.g. live measles virus may inhibit human T cell proliferative responses in vitro, and immune deficiency is a frequent syndrome in patients with viral infections (Schneider-Schaulies and Meulen, 2002). Świtała-Jeleń of our Institute demonstrated that platelets can bind phages using B3 integrins; this was confirmed by a lack of reactivity of platelets from a patient with Glanzmann’s thrombastenia (an inborn deficiency of the B3 integrin). Furthermore, phages can inhibit platelet aggregation induced by collagen and thrombin (Świtała-Jeleń, 2005), as well as platelet adhesion to fibrinogen (Kniotek et al., 2004b). Immunomodulatory influences of phages on the immune system are also emphasized by the presence of the Hoc protein within the bacteriophage T4 head. The exact function of Hoc is unknown: it is necessary neither for T4 viability nor for its structure. Interestingly, Bateman et al. revealed Hoc’s relatedness to the eukaryotic immunoglobulin superfamily of proteins, which are engaged in cell adhesion, antigen presentation, antigen recognition (T-cell receptor), and other processes (Bateman et al., 1997). Our studies suggest that Hoc could participate in phage interactions with the immune system and platelets, mediating some immunomodulatory effects of T4 phages (e.g. inhibition of T cell and platelet adhesion to fibrinogen, inhibitory effects on T cell activation and antibody synthesis). Lack of Hoc may be responsible for the observed differences in cancer cell binding, antitumor activity, and clearance/removal from the murine organism between T4 and its Hoc-lacking mutant HAP1 (Dąbrowska et al., 2005a, 2007). It is also possible that other phage proteins may have effects on the immune system. Evidently, further research in this field should yield data which may have very important implications for human disease and therapy. Normal human serum may show coliphage-neutralizing activity, which appears to possess characteristics of the properdin system ( Jerne, 1956; Barlow et al., 1958; Cowan, 1962). Of particular interest is the phenomenon of the possible occurrence of anti-phage antibodies in the sera of normal individuals. A group at our Institute detected antibodies against S. aureus phage in 11% of healthy controls and 23% of patients prior to phage therapy (Kucharewicz-Krukowska et al., 1987). Interestingly, Jerne described the pres-

|

139

140

|

Górski et al.

ence of specific antibody against bacteriophage T4 in normal mouse serum already in the 1950s ( Jerne, 1956). Smith described anti-coliphage antibodies in human, cattle, pig, and bovine sera (Smith and Huggins, 1982). Furthermore, Kamme and Hedstrom and Kamme reported anti-S. aureus phage antibodies in normal sera, whose titers increased in staphylococcal infections; this might have been caused by a immune response to phages released from lysogenic microorganisms (Kamme, 1973; Hedstrom et al., 1973). As shown by our group, oral phage therapy does not induce significant anti-phage antibody production, a finding confirmed by a recent report (Bruttin and Brussow, 2005). On the other hand, other authors observed significant systemic (IgG/IgA subclass) and local mucosal (IgA subclass) immune responses against phage coat protein following oral delivery in rodents (Clark et al., 2004). As already pointed out, parenteral phage administration induces humoral responses in animals and man, although it is unclear whether this could lead to a significant decrease in phage anti-bacterial action. As emphasized by Sulakvelidze and Kutter, the immunogenicity of phages may vary, with some phages being only very weak immunogens, requiring the use of adjuvant and repeated administration in order to elicit detectable antibody responses (Sulakvelidze et al., 2005). In addition, the significance of anti-phage antibodies induced by virus administration is unclear. Earlier reports emphasized the role of reticuloendothelial system in phage elimination with liver being the most active in this process followed by the spleen; in these experiments anti-phage antibodies played no role (experiments were performed in germ-free mice) (Geier et al., 1973). However, Srivastava demonstrated an important role of B cells and their products in T7 phage elimination; no role for NK, T cells and macrophages was shown (Srivastava et al., 2004) Apparently, the issue of phage interactions with human serum, the presence of “natural” as well as therapyinduced anti-phage antibodies, constitutes a problem of paramount practical implication which should be further explored in more detail. Reactive oxygen species (ROS) play an essential role in phagocyte defenses against microbial pathogens. However, their excess may overwhelm the body’s endogenous antioxidant defense mechanisms and lead to oxidative stress, which causes tissue damage and contributes to the high mortality rates in sepsis and endotoxin shock (Victor et al., 2004). ROS-suppressive effects may be beneficial in controlling inflammation and may constitute an effective therapeutic approach to sepsis (Victor et al., 2005). Our recent data indicate that homologous, but not heterologous phages inhibit ROS production by human granulocytes and monocytes triggered by bacteria (Przerwa et al., 2006). In addition, phages may inhibit the stimulation of ROS production by endotoxin (Międzybrodzki et al., manuscript submitted for publication). Interestingly, in contrast to bacteria and pathogenic viruses, phages induce only minimal ROS production from phagocytes. Phage-mediated inhibition of ROS production by phagocytes may be a very important phenomenon contributing to the beneficial effects of phage therapy which sheds more light on our preliminary data suggesting that phages may exert beneficial effects in patients with sepsis (Weber-Dąbrowska et al., 2003), a clinical setting where excessive production of ROS appears to play an important role and agents interfering with ROS have been advocated for treatment (Victor et al., 2005). While phages could affect some functions of phagocytes, activated (but not resting) polymorphonuclear neutrophils can inactivate phages, probably by releasing extracellular

Bacteriophages in Medicine

factors (e.g. hypochloric acid) (Ferrini et al., 1989). If confirmed, this phenomenon could explain why phages are generally more efficient in killing bacteria in vitro than in vivo. Thus we postulate that, aside from their well-known antibacterial actions, phages may play an important role in clinical immunology as immunoregulatory agents. By their ability to exert immunosuppressive and anti-inflammatory actions, endogenous phages could contribute to maintaining immune homeostasis. This phenomenon should be especially relevant in the gut, where phages may be present in high concentrations. At the same time, phage-dependent bacterial destruction may release bacterial products with immunostimulatory activities. Therefore, intestinal phages may help GALT to respond correctly to a variety of antigens and maintain an appropriate balance between inducing a state of unresponsiveness (mucosal tolerance) and eliciting a strong immune response when confronted with pathogenic bacteria (Smith et al., 2005). It is known that orally administered antigens are presented in GALT by dendritic cells, macrophages, and B cells. Furthermore, antigen-specific T cells in GALT upregulate activation markers and proliferate after oral administration of soluble protein antigens. Intestinal dendritic cells can extend their dendrites between intestinal epithelial cells and sample intestinal flora from the gut lumen. Moreover, intestinal dendritic cells can engulf commensal bacteria and induce the production of protective IgA (Smith et al., 2005). Interestingly, dendritic cells abundantly phagocytize T4 coliphage (Barfoot et al., 1989) which, as we have shown, may have immunosuppressive activities. These data lend further support to our hypothesis suggesting that intestinal phages can downregulate local immune reactions and help maintain mucosal tolerance within the gut lumen. Anti-phage humoral response as a measure of immunocompetence Bacteriophages replicate solely in bacterial cells. Thus when no potential host cells are available in the human organism, they behave as inert antigens (Clark et al., 2004b). This feature, coupled with its ease of quantification and apparent lack of toxicity, has enabled the use of one coliphage, &X 174, for the assessment of humoral immunity in diagnostics and the monitoring of both primary and secondary immunodeficiency diseases (Ochs et al., 1971; Bernstein et al., 1985; Wedgwood et al., 1975). In fact, over 30 years after the first reports on the topic were published, &X 174 is considered to be one of the standard antigens for the evaluation of humoral immunity in clinical medicine (Bearden et al., 2005). In terms of immunogenicity, it is a potent T cell-dependent neoantigen that induces typical primary, secondary, and tertiary humoral responses following intravenous administration (Wedgwood et al., 1975; Pyun et al., 1989). Basically, while the primary response comprises largely IgM antibody, the secondary and tertiary responses feature memory, amplification, and isotype switching from IgM to IgG (Wedgwood et al., 1975; Pyun et al., 1989). Based on abnormalities in both the rate of antibody-dependent phage particle clearance and the quantitative and qualitative features of the generated primary and secondary responses, patients with immunodeficiencies fall into six major groups (types 0-V) (Wedgwood et al., 1975). The most profound immunodeficiency is found in patients having type 0 immune response, i.e. those who produce no antibody and therefore do not clear phage particles in the typically accelerated way. On the other hand, subjects with type V, in whom im-

|

141

142

|

Górski et al.

munodeficiency is least marked, show only some minor quantitative impairment of their immune response (Wedgwood et al., 1975). Generation of the humoral response to &X 174 is a complex process which requires the proper functioning of antigen-presenting cells, T lymphocytes, and B lymphocytes. Thus, its abnormalities following immunization may be indicative not only of dysfunction in B cells, but in T cells and complement and of adhesive molecule defects as well (Andrews et al., 1997). For example, in a recent study, immunization with &X 174 was successfully used for the assessment of CD4+ T cell function in vivo in HIV-infected patients (Fogelman et al., 2000). Thus far, immunization with &X 174 has been reportedly used for the diagnosing and monitoring of patients with adenosine deaminase deficiency (Ochs et al., 1992), X-linked agammaglobulinemia (Ochs et al., 1971), X-linked hyper-IgM syndrome (Seyama et al., 1998), major histocompatibility complex class II deficiency (Nonoyama et al., 1998), the Wiskott-Aldrich syndrome (Ochs et al., 1980), AIDS (Fogelman et al., 2000), and patients after bone marrow transplantation (Witherspoon et al., 1982). T2 phage given subcutaneously in repeated doses was used to immunize patients with immune abnormalities whose urine concentrates were subsequently assayed for anti-phage antibody. Urine concentrates from patients immunized with T2 phages all showed antiphage activity, in contrast to urine samples obtained prior to immunization or obtained from non-immunized individuals (Hanson and Tan, 1965). Studies on the phage application in the treatment of pathogenic virus infection It seems that many commercial bacteriophage preparations, although they were introduced shortly after the discovery of bacteriophages, were used improperly for conditions against which phages could not be effective. Enterophagos is cited as one such preparation that was allegedly effective against many diseases, including herpes infections (Barrow and Soothill., 1997). In fact, there are already data in the literature indicating that the effect of phages seems to be broader than the mere elimination of bacteria (Górski et al., 2003; Dąbrowska et al., 2004a; Dąbrowska et al., 2004b; Górski et al., 2005). Some suggest the use of phages to treat infections caused by pathogenic viruses in humans (Międzybrodzki et al., 2005). In 1965 Centifanto published her observations on the antiviral activity of the L phage lysate of E. coli K-12 (Centifanto, 1965). This activity was shown against herpes simplex (HSV) and vaccinia viruses in a plaque inhibition assay on chick embryo monolayer cultures infected with and in a herpetic corneal ulcer model in rabbits. The lysate did not block the absorption of the virus, but rather inhibited intracellular replication. Control extract prepared from the E. coli K-12 that did not contain phages was inactive in vitro, and the phage lysates did not induce interferon (IFN) in the cells, suggesting that this cytokine as well as endotoxin were not involved in the mechanism of the observed phenomenon. Moreover, Centifanto showed that the antiviral activity of the lysate remained after removing the infective lambda particles. Because the putative soluble antiviral agent was not found in the uninfected E. coli culture and because its production was closely associated with phage replication (its concentration increased during a one-step growth experiment), she called it “phagicin” (Centifanto, 1968). Later she found that this substance could also be obtained by disruption of purified L phage particles by incubation with LiCl at 46oC for 15 min or by

Bacteriophages in Medicine

sonic treatment (Centifanto, 1968). Phagicin activity was rather specific: it was not active against poliovirus, Newcastle disease virus, influenza virus, Semliki Forest virus, vesicular stomatitis virus (VSV), or rhinovirus. It was susceptible to proteolytic enzymes such as trypsin and pepsin, but not to deoxyribonuclease, ribonuclease, and ultraviolet irradiation. On this basis she hypothesized that this substance might be a phage-internal protein which interferes with the intracellular replication of viral DNA. The antiviral activity of phagicin (obtained by centrifugation and gel filtration on Sephadex columns of L phage lysate of E. coli K-12) was confirmed by Meek and Takahashi (Meek et al., 1968). They also showed that phagicin inhibited the synthesis of vaccinia virus DNA and did not affect host-DNA synthesis. There are also a few other data in the literature on the antiviral action of purified phages. Merril observed inhibition of plaque formation on chick embryo fibroblasts infected with HSV or vesicular stomatitis virus (VSV) by L phage as well as anti-VSV activity of F80 bacteriophage (when added 2–4 hours before VSV infection of cell culture) (Merril, 1977). In 1970, Kleinschmidt reported in Nature the inhibition of VSV in primary rabbit kidney cells by T4 phage applied to monolayers 24 hours before challenge with the virus (Kleinschmidt et al., 1970). However, in 1973 Brown and Cohen published their unsuccessful results on the lack of specific inhibition of vaccinia plaque formation by L phage lysates of E. coli, opposite to Centifanto’s earlier studies (Brown and Cohen, 1973). A protective experiment in vivo was done in 1969 by Keyhani. He administrated the lysate of the Brucella bacteriophage type abortus strain 3 (1014 pfu per dose) to chickens weekly for 12 consecutive weeks and then inoculated the animals with virulent Newcastle disease virus. While animal survival in the control group (no bacteriophage given) was only 12%, it was 69% in the animals receiving the phage lysate orally and 83% in the group that received oral doses of phage during the first week and then the preparation intramuscularly. Very interesting novel data showed that a virus not known to be pathogenic in humans and a pathogenic virus can interact on the level of cellular receptors and even alter the course of disease. GB virus C (GBV-C), for example, induces chemokines in peripheral blood mononuclear cells that reduce the expression of the human immunodeficiency virus (HIV) coreceptor CCR5 in vitro, thus decreasing its replication in target cells (Xiang et al., 2004). This effect could be responsible for the improved survival of HIV-positive patients co-infected with GBV-C. A hypothesis of bacteriophage-virus interactions through specific cellular receptors, e.g. integrins, was presented by Górski (Górski et al., 2003). It is based on the finding that some phages may express integrin-binding sequences on their capsids, as found in the case of T4 phage (Dąbrowska et al., 2004a; Dąbrowska et al., 2004b). Using a library of phage-displayed RANTES mutants, Hartley isolated a phage clone that could block in vitro infection of peripheral blood plasma mononuclear cells by the R5-tropic strain of HIV type 1 (HIV-1) (Hartley et al., 2003). Recently, Krichevsky and colleagues demonstrated that filamentous bacteriophages that are used to construct phage-display peptide libraries recognize the nuclear localization signal (NLS) sequence of the HIV-1 Tat protein (Krichevsky et al., 2005). This protein is essential for the HIV life cycle and pathogenesis, and the NLS domain mediates Tat activa-

|

143

144

|

Górski et al.

tion and internalization. The N-terminus of the bacteriophage coat protein p8 promotes interaction between fd phage and Tat-NLS. Fd phage specifically inhibited Tat-NLS-mediated nuclear-import, and a peptide bearing the first 20 amino acids of the p8 protein (NTP8) inhibited HIV-1 propagation in cultured cells. The influence of whole phage particles on interferon induction was reported only in the previously mentioned study by Kleinschmidt (Kleinschmidt, 1970). Whole T4 phage particles administrated intravenously to mice were capable of inducing IFN in the blood. Interestingly, T4 ghosts and extracted T4 DNA failed. The kinetics of the IFN production by mice injected with a purified T4 preparation was characteristic of viral IFN inducers. In a control group treated with E. coli endotoxin at a dose comparable to that of the T4 preparation, an increase in IFN serum level was observed during the first 6 hours after endotoxin injection, while IFN stimulation was several times higher in the blood of the T4-treated mice, and was prolonged to more than 24 hours. The conclusion was drawn that DNA is the T4 molecule responsible for the induction of interferon and that the encapsulation of the DNA in the intact phage enables the delivery of the DNA to the inducer recognition site. The phenomenon of inhibition of viral infection by bacteriophage-derived nucleic acids was shown in vitro and in vivo in a several studies. In all of them, induction of IFN may be part of the mechanism of that phenomenon. In a model of duck hepatitis B virus (DHBV) infection, Iizuka et al. presented specific antiviral activity of single-stranded DNA (ssDNA) from M13 phage (Iizuka et al., 1994). Daily intravenous injections of M13 ssDNA (but not S1 nuclease-digested M13 DNA) decreased serum level of DHBV DNA during the first 10 days after infection, but continuous DNA administration was required for prolongation of the antiviral effect, which was superior to that observed for acyclovir. It also reduced the viral DNA level in the livers of infected ducks. Treatment of ducks with M13 ssDNA increased the serum level of 2–5 A synthetase, which is known to be induced by endogenous IFN, suggesting its possible involvement in the observed antiviral effect. This M13 ssDNA activity was confirmed by Mori by showing an M13 ssDNA-induced reduction of the formation of tail lesions caused by vaccinia virus (Mori et al., 1995). A single intravenous dose of the nucleic acid administered one day before virus infection was sufficient to cause a significant clinical effect. It also induced IFN (mainly IFN-B) in mice that reached a peak in serum 4 hours after injection and was detectable for up to 12 hours. Bacteriophage double-stranded RNA (dsRNA) is also effective. As shown by Vales, dsRNA extracted from F6 phage inhibited in vitro HSV type 1 (HSV-1) and HSV type 2 (HSV-2) infection in fetal rhesus monkey kidney cells (MA104) (Vales et al., 1991). This inhibition was cell specific: F6 dsRNA did not alter the cytopathic effect of viruses in an African green monkey kidney (Vero) cell culture. When applied in vivo on day five after induction of HSV-2 genital infection in guinea pigs, single intravaginal administration of 600 µg/kg of dsRNA completely healed animals in 21 days post infection with few recurrences and no death was observed, while in the untreated group the mortality ratio was 39%. The intraperitoneal route of dsRNA administration seemed to be less effective in this model. An acute toxicity study revealed that F6 dsRNA was rather toxic: LD50 estimated in mice was 4.62 mg/kg and 5.37 mg/kg, respectively, for the intravenous and

Bacteriophages in Medicine

the intraperitoneal route of administration. In another study, dsRNA obtained from the amB11 mutant of the R17 E. coli phage protected mice infected with VSV, encephalomyocarditis, and Moloney mice sarcoma viruses against their lethality (Feldmane et al., 1977). Administration of this nucleic acid, similarly as in Mori’s study, induced production of IFN, its maximum level in blood being reached at 4 hours after intravenous injection. In the 1970s the application of native bacteriophage dsRNA in the treatment of viral skin infection was studied in five hospitals in Slovakia (Borecky, 1977). Over two hundred patients suffering from viral dermatoses and viral eye disease (herpes simplex recidivans, herpes zoster, male genitokeratoconjunctivitis herpetica, verrucae vulgares, and conjunctivitis lignosa) were treated topically with dsRNA derived from a suppressorsensitive (amber)mutant f2 coliphage (Borecky et al., 1978). The ointment used contained 100–1000 µg of f2 RNA per g of base and was administer for a period shorter than 20 days. Evident failure of the therapy (a need for substitution therapy) was observed in about 10% of the patients infected with herpes simplex, 15% of herpes zoster cases, and 30% of verrucae cases, depending on the disease. 79.3% of 29 former patients suffering from repeated herpes simplex attacks considered the f2 RNA therapy effective. The results of a small double-blind study (12 treated patients and another 25 in the placebo group) indicated that 10-day topical f2-RNA treatment of HSV-infected patients proved significantly effective in 94.1% of cases compared with 25% receiving placebo. Due to great variation in the severity of the dermatoses, difficult control of the therapeutic regimen, and the insufficient number of patients for statistical evaluation in the open study, the total study could not provide a clear answer as to the therapeutic value of f2-RNA treatment. The author’s conclusion was that treatment with f2-RNA was harmless to the patients, brought a rapid feeling of relief in the majority of cases, and shortened the duration of the disease. Phages are also being considered as a modern platform for vaccines (Clark and March, 2006). They are known to be inducers of IgM, IgG, and IgE class antibody formation (Pyun et al., 1989, Zuercher et al., 2000). Some of them, such as FX 174, have been used in purified form to monitor the immunity of immunocompromised patients (Ochs et al., 1971; Price et al., 1994; Fogelman et al., 2000). Viable bacteriophages readily accumulate in the spleen and may persist there for a longer time (Keller and Zatzman, 1958; Geier et al., 1973, Reynaud et al., 1992). Their virions can be taken up by antigen-presenting cells and processed efficiently by MHC class I and II pathways (Gaubin et al., 2003). They elicit immunological response not only after intravenous, but also after oral and even intranasal administration (Manoutcharian et al., 2004; Zuercher et al., 2000; Delmastro et al., 1997). This response can be directed against wild type proteins of the phage, peptides displayed on their coats, and DNA packed in phage particles. A phage displaying a human respiratory syncytial virus protective epitope (viral glycoprotein G) was constructed by Bastien, who showed that it has protective activity as a vaccine in a mouse model (Bastien et al., 1997). HIV peptides may also be displayed on the surface of filamentous phages (closely resembling the natural viral epitopes); vaccination with such phages causes the production of high antibody titers capable of neutralizing viruses (Di Marzo Veronese et al., 1994; De Berardinis et al., 2003). Delmastro and colleagues induced an immunological response in mice to mimotopes of human HBSAg

|

145

146

|

Górski et al.

displayed on a phage (Delmastro et al., 1997). The stability of vaccines of this type may be relatively high, as demonstrated for L phage ( Jepson and March, 2004). Bacteriophage DNA can reach peripheral leukocytes, spleen, and liver via the intestinal wall mucosa and can be covalently linked to mouse DNA even when administered orally, as was shown for M13mp18 phage (Schubbert et al., 1997). Clark and March proved that effective antibody production can be elicited by genetically modified phages carrying eukaryotic DNA-expression cassettes coding the proper antigen (Clark and March, 2004). In their thorough study, whole bacteriophage L-gt11 particles that contained the gene for hepatitis B surface antigen under the control of the cytomegalovirus (CMV) promoter (phage L-HBsAg) were used as delivery vehicles for nucleic acid immunization. Intramuscular injection of mice at weeks 0 and 3 with L-HbsAg induced significantly higher anti-HBsAg antibody response than that observed in control mice injected with unmodified phage or in a group injected with control plasmid DNA at a dose several dozen times higher. Phage typing Bacterial viruses have been commonly acknowledged for bacterial identification, and in the last few decades phage typing has been an essential method in the epidemiological investigation of infections to determine their source and their way of spread (Sechter et al., 2000; Demczuk et al., 2003; Wildemauwe et al., 2004; Rybniker et al., 2006). Phage typing allows the characterization of heterogeneous bacterial strains belonging to the same species. This is a simple, rapid, economical, and reproducible method with an internationally standardized procedure for the discrimination of pathogenic bacteria. It provides information about human infections, animals reservoirs, the origin of hospital infections, transmission routes, and changes in the geographical distributions of currently isolated group of bacteria and also enables the elimination of multidrug resistant infections (Drulis-Kawa et al., 2005). Its methodology is standardized and it has been widely used by many research teams. The standard phagotyping procedure is simple: the different phage lysates in the routine test dilution (RTD) or 100 s RTD are dripped onto dry Petri dishes covered with the bacterial strain. The results are read after 6 to 18 hours of incubation at 37oC (Adams, 1959; Ackermann and Dubow, 1987; Aucken and Westwell, 2002). However, more complex commercial tests utilizing bacteriophage amplification technology are now available, e.g. FASTPlaqueTB (Biotec Laboratories Ltd., Ipswich, UK), which is applied for the detection of viable Mycobacterium tuberculosis in clinical samples (Albay et al., 2003). The test uses non-pathogenic, rapidly growing Mycobacterium smegmatis to amplify D29 mycophages which replicate in the target bacteria. Typing phages are virulent, temperate, or adapted. They may all be isolated from the environment as well as from bacterial strains. Many typing sets for Gram-positive and Gramnegative bacteria have been developed since Craigie and Yen created the Vi phage-typing scheme for Salmonella typhi strains in 1938 (Craigie and Yen, 1938). The most important sets have been used for Shigella flexneri and Shigella sonnei, Salmonella enterica, Escherichia coli, Klebsiella bacilli, Pseudomonas aeruginosa, and Staphylococcus aureus (Mulczyk and Kucharewicz, 1957; Ślopek et al., 1972; Ślopek et al., 1973; Ward et al., 1987; PrzondoHessek, 1966; Ślopek et al., 1967; Sechter et al., 2000; Bergan, 1972; Dąbrowski et al., 1988; Blair and Williams, 1961).

Bacteriophages in Medicine

The extension of the technique of phage typing to paratyphoid B and Salmonella typhimurium by Felix and Callow opened a new era in the laboratory detection of salmonellas (Hardy, 2004). It enabled studying the complexity of the environmental associations of salmonella poisoning and revealing links between isolated cases, and it still does. It was shown, for example, that chicken products are the main source of human infections in the Netherlands, and that the same phage type (PT4) was the dominant strain in humans and chickens (Duijkeren et al., 2002). Evident changes in the phagotypes of the Salmonella enterica serovar Enteritidis strains isolated from different European countries has been observed lately by the Enter-net International Salmonella Database (Fisher, 2004). On the basis of phage typing 178 983 Salmonella strains, a significant increase in non-phagotype-4 strains with a parallel decreasing number of phagotype-4 strains was demonstrated. The typing scheme for Shigella flexneri and Shigella sonnei was described by Ślopek’s group (Milch et al., 1968; Ślopek et al., 1972; Ślopek et al., 1973). It was shown that the reproducibility of phage-typing results depends on the homogeneity of the phage preparations as well as the antigenic structure of the tested strains. Phage-typing allowed investigators to distinguish among strains of the same serotype isolated from different sources in several European countries using the new international phage collection which was developed as a result of typing 1474 Shigella flexneri strains with phages from Ślopek’s and Istrati’s collections (Milch et al., 1968). The proposed set of phages permitted the typing of almost 100% of the Shigella sonnei strains isolated from various sources (Ślopek et al., 1973). Phagotyping is often complemented by serotyping and antibiotyping (Riuz et al., 2003; Murugkar et al., 2005). It is also combined with molecular techniques such as pulsed-field gel electrophoresis (PFGE) and plasmid typing (Demczuk et al., 2003; Pang et al., 2005). Shiga toxin-producing Escherichia coli (STEC) O157:H7 strains are the major causative agents of human gastroenteritis complicated by hemorrhagic colitis, hemolytic-uremic syndrome, and childhood acute renal failure in Europe, Japan, and North America (Mora et al., 2004). As a result of parallel, internationally standardized phage typing and PFGE typing, it is now possible to establish the origin, animal reservoir, and the routes of transmission of human infections with the STEC O157:H7 strains. Mora et al. found that the most common phage type was isolated from humans and bovine strains and that the same phagotype-PFGE pattern belonged to two phage types, PT2 and PT8 (Mora et al., 2004). Phagotyping in conjuction with fluorescent amplified fragment length polymorphism (FAFLP) genotyping and serotyping has been used lately for characterizing Campylobacter jejuni and Campylobacter coli strains (Hopkins et al., 2004). On the basis of phage typing and FAFLP analysis it was shown that the Campylobacter strains isolated from poultry and pigs are genetically distinct and that FAFLP genotyping combined with phage typing presents the epidemiological characteristics of these strains well. Results concerning the characterization of multidrug-resistant strains of Staphylococcus aureus and coagulase-negative Staphylococcus strains for epidemiological purposes have shown the usefulness of genotyping analysis combined with phage typing, especially with international typing phages (Richardson et al., 1999; Piechowicz et al., 2005; Drulis-Kawa et al., 2005).

|

147

148

|

Górski et al.

Further perspectives We have no doubt that phage therapy may cure many cases of bacterial infection which pose difficult, or even dramatic, clinical dilemmas. However, it is clear that formal clinical trials done in accordance with current requirements are urgently required. Our experience has taught us that this is not an easy task. Present requirements on clinical trials and the need to implement the new EU regulations (EU directive 2001/20/EC) introduce regulatory requirements for all phases of study in human subjects, drawing no distinction between commercially funded drug trials and academic research. In addition, exhaustive pharmacovigilance and protection of trial subjects may be seen as a deterrent to clinical research in Europe (Grienenberger, 2004). Given the high costs of the procedures involved and the administrative burden, cooperation with industry is needed to make decisive progress in clinical trials on phage effectiveness unless public funds are allocated for this purpose (which is rather unlikely). Aside from the administrative and financial aspects of further progress in the clinical application of phage therapy, there are certain points that need more work and study, for example: better characterization of therapeutic phages (including their screening for genes potentially encoding toxins), progress in phage purification, and the introduction of techniques enabling their processing in large volumes, and methods of phage storage and their effects of phage activity. Paradoxically, it cannot be excluded that unpurified phage preparations (lysates) may continue to be used in therapy, especially their oral application. First, they may offer the additional advantage of providing immunologic stimulation, frequently needed in patients with immunodeficiency. Second, lysates may be more stable than purified phages, which are deprived of their natural milieu and nutrients. Third, our and other centers’ experience indicates that practically no side effects are seen following sterile lysate administration by the oral route, which may be much less immunogenic, and even tolerogenic (oral tolerance). Finally, the progress made in the experimental use of phage lytic enzymes does not exclude the future use of intact phages. Such purified phages may have immunosuppressive effects which could be useful in certain clinical situations; moreover, phages can exert additional beneficial effects, e.g. downregulation of excessive reactive oxygen species production. Therefore, we believe that there will be a place in the future for various forms of phage preparations for clinical use. Accordingly, in August 2006 the FDA published a regulation permitting the use of a Listeria specific bacteriophage preparation on ready-to-eat meat and poultry products. The preparation contains six phages effective against 170 different strains of L. monocytogenes (Lang, 2006). The growing menace of antibiotic-resistant infections and the exciting data presented here and elsewhere, suggests potential novel applications of phages and a more optimistic prospect for their wider applications in medicine. Acknowledgments Our studies reported in this chapter were supported by the Ministry of Science grant No. PBZ-MIN-007/P04/2003 and The Medical University of Warsaw intramural grant No. 1MG/W1/2005.

Bacteriophages in Medicine

References

Ackermann, H-W., and Dubow, M. ed. (1987). General properties of bacteriophages. In: Viruses of Prokaryotes. (CRC Press, Inc. Boca Raton, Florida) pp. 143–172. Adams, M. ed. (1959). Bacteriophages. (Interscience Publishers, Inc. New York) pp. 395–420. Ahn, K.S., and Aggarwal, B.B. (2005). Transcription factor NF-(kappa) B: a sensor for smoke and stress signals. Ann. NY Acad. Sci. 1056, 218–233. Albay, A., Kisa, O., Baylan, O., and Doganci, L. (2003). The evaluation of FASTPlaqueTB test for the rapid diagnosis of tuberculosis. Diagn. Microbiol. Infect. Dis. 46, 211–215. Andrews, R.G., Winkler, A., Potter, J., Bryant, E., Knitter, G.H., Bernstein, J.D., and Ochs, H.D. (1997). Normal immunologic response to a neoantigen, bacteriophage &X-174, in baboons with long-term lymphohematopoietic reconstitution from highly purified CD 34+ Lin- allogeneic marrow cells. Blood 90, 1701–1708. Aucken, H., and Westwell, K. (2002). Reaction difference rule for phage typing of Staphylococcus aureus at 100 times the routine test dilution. J. Clin. Microbiol. 40, 292–293. Barfoot, R., Denham, L.A., Gyure, J.G., Hall, J.G., Hobbs, S.M., Jackson, L.E.and Roberston, D. (1989). Some properties of dendritic macrophages from peripheral lymph. Immunology 68, 233–239. Barlow, J.L., van Vunakis, H., and Levine, L. (1958). Studies of the inactivation of phage by the properdin system. J. Immunol. 80, 349–355. Barrett, J.F. (2005). MRSA—what is it, and how do we deal with the problem? Expert Opin. Ther. Targets 9, 253–265. Barrow, P., Lovell, M., and Berchieri, A., Jr. (1998). Use of lytic bacteriophage for control of experimental Escherichia coli septicemia and meningitis in chickens and calves. Clin. Diagn. Lab. Immunol. 5, 294–298. Barrow, P.A., and Soothill, J.S. (1997). Bacteriophage therapy and prophylaxis: rediscovery and renewed assessment of potential. Trends Microbiol. 5, 268–271. Bastien, N., Trudel, M., and Simard, C. (1997). Protective immune responses induced by the immunization of mice with a recombinant bacteriophage displaying an epitope of the human respiratory syncytial virus. Virology 234, 118–122. Bateman, A., Eddy, S.R., and Mesyanzhinov, V.V. (1997). A member of the immunoglobulin superfamily in bacteriophage T4. Virus Genes 14, 163–165. Bearden, C.M., Agarwal, A., Book, B.K., Vieira, C.A., Sidner, R.A., Ochs, H.D., Young, M., and Pescovitz, M.D. (2005). Rituximab inhibits the in vivo primary and secondary antibody response to a neoantigen, bacteriophage phiX174. Am. J. Transpl. 5, 50–57. Bergan, T. (1972). A new bacteriophage typing set for Pseudomonas aeruginosa. 1. Selection procedure. Acta Pathol. Microbiol. Scand. Sect. B. 80, 177–188. Bernhardt, T.G., Struck, D.K., and Young, Ry. (2001a). The lysis protein E of &X174 is a specific inhibitor of the MraY-catalyzed step in peptidoglycan synthesis. J. Biol. Chem. 276, 6093–6097. Bernhardt, T.G., Wang, I.-N., Struck, D.K., and Young, Ry. (2001b). A protein antibiotic in the phage QB virion: diversity in lysis targets. Science 292, 2326–2329. Bernhardt, T.G., Wang, I.-N., Struck, D.K., and Young, Ry. (2002). Breaking free: “Protein antibiotics” and phage lysis. Res. Microbiol. 153, 493–501. Bernstein, L.J., Ochs, H.D., Wedgwood, R.J., and Rubinstein, A. (1985). Defective humoral immunity in pediatric acquired immune deficiency syndrome. J. Pediatr. 107, 352–357. Biswas, B., Adhya, S., Washart, P., Paul, B., Trostel, A.N., Powell, B., Carlton, R., and Merril, C.A. (2002). Bacteriophage therapy rescues mice bacteremic from a clinical isolate of vancomycin-resistant Enterococcus faecium. Infect. Immun. 70, 204–210. Blair, J., and Williams, R. (1961). Phage typing of staphylococci. Bull. WHO 24, 771–784. Boratyńska M., Szewczyk Z., and Weber-Dąbrowska B. (1994). The clinical evaluation of bacteriophage treatment of urinary infections. Post. Med. Klin. Dosw. 3, 7–11 (in Polish). Borecky, L. (1977). Use of bacteriophage nucleic acid inducers of interferon in man. Tex. Rep. Biol. Med. 35, 535–540. Borecky, L., Buchvald, J., Adlerova, E., Stodola, I., Obrucnikova, E., Gruntova, Z., Lackovic, V., and Doskocil, J. (1978). Results of a five-year study of the curative effect of double stranded ribonucleic acid in viral dermatoses and eye diseases. Adv. Exp. Med. Biol. 110, 175–191. Borysowski, J., Weber-Dąbrowska, B., and Górski, A. (2006). Bacteriophage endolysins as a novel class of antibacterial agents. Exp. Biol. Med. 231, 366–377.

|

149

150

|

Górski et al.

Bradbury, J. (2005). My enemy’s enemy is my friend. Using phages to fight bacteria. Lancet 363, 624– 625. Brockstedt, D.G., Bahjat, K.S., Giedlin, M.A., Liu, W., Leong, M. Luckett, W., Gao, Y., Schnupf, P., Kapadia, D., Castro, G., Lim, J.Y., Sampson-Johannes, A., Herskovits, A.A., Stassinopoulos, A., Bouwer, H.G., Hearst, J.E., Portnoy, D.A., Cook, D.N., and Dubensky, T.W. Jr. (2005). Killed but metabolically active microbes: a new vaccine paradigm for eliciting effector T-cell responses and protective immunity. Nature Med. 11, 853–860. Brown, A., and Cohen, S.N. (1973). Lack of specific inhibition of Vaccinia plaque formation by bacteriophage lambda products. Infect. Immun. 7, 862–864. Brussow, H. (2005). Phage therapy: the Escherichia coli experience. Microbiology 151, 2133–2140. Bruttin, A., and Brussow, H. (2005). Human volunteers receiving Escherichia coli phage T4 orally: a safety test of phage therapy. Antimicrob. Agents Chemother. 49, 2874–2878. Caldwell, J.A. (1928). Bacteriologic and bacteriophagic study of infected urines. J. Infect. Dis. 39, 343–362. Caroli, G., Armani, G., Levre, F. And Jefferson, T.O. (1980). Finding of “E. coli” phage in urinary tract infections. Ann. Sclavo 22, 857–860. Carrera, M.R.A., Kaufmann, G.F., Mee, J.M., Meijler, M.M., Koob, G.F., and Janda, K.D. (2004). Treating cocaine addition with viruses. Proc. Nat. Acad. Sci. USA 101, 10416–10421. Centifanto, Y.M. (1965). A therapeutic antiviral from an extract of L infected E. coli (phagicin). Proc. Soc. Exp. Biol. Med. 120, 607–611. Centifanto, Y.M. (1968). Antiviral agent from lambda-infected Escherichia coli K-12. I. Isolation. Appl. Microbiol. 16, 827–834. Cheng, Q., Nelson, D., Zhu, S., and Vischetti, V.A. (2005). Removal of group B streptococci colonizing the vagina and oropharynx of mice with a bacteriophage lytic enzyme. Antimicrob. Agents Chemother. 49, 111–117. Chibani-Chennoufi, S., Sidoti, J., Bruttin, A., Kutter, E., Sarker, S., and Brüssow, H. (2004). In vitro and in vivo bacteriolytic activities of Escherichia coli phages: implications for phage therapy. Antimicrob. Agents Chemother. 48, 2558–2569. Clark, J.R., and March, J.B. (2004). Bacterial viruses as human vaccines? Expert Rev. Vaccines 3, 463– 476. Clark, J.R., and March, J.B. (2006). Bacteriophages and biotechnology: vaccines, gene therapy and antibacterials. Trends Biotechnol. 24, 212–218. Clarke, T. (2003). Drug companies snub antibiotics as pipeline threatens to run dry. Nature 425, 225. Clewley, J.P. (2003). The day of the phage. Commun. Dis. Public Health 6, 260–262. Cowan, K.M. (1962). Studies on the coliphage neutralizing activity of normal human serum. J. Immunol. 88, 476–481. Craigie, J., and C.H. Yen. (1938). The demonstration of types of B typhosus by means of preparations of type II Vi phage. II The stability and epidemiological significance of V form types of B typhosus. Can. Public Health J. 29, 484–496. Dąbrowska, K., Opolski, A., Wietrzyk, J., Świtała-Jeleń, K., Boratyński, J., Nasulewicz, A., Lipińska, L., Chybicka, A., Kujawa, M., Zabel, M., Dolińska-Krajewska, B., Piasecki, E., Weber-Dąbrowska, B., Rybka, J., Salwa, J., Wojdat, E., Nowaczyk, M., and Górski, A. (2004a). Antitumour activity of bacteriophages in murine experimental cancer models caused possibly by inhibition of B3 integrin signaling pathway, Acta Virol. 48, 241–248. Dąbrowska, K., Opolski, A., Wietrzyk, J., Świtała-Jeleń, K., Godlewska, J., Boratyński, J., Syper, D., Weber-Dąbrowska, B., and Górski, A. (2004b). Anticancer activity of bacteriophage T4 and its mutant HAP1 in mouse experimental tumour models. Anticancer Res. 24, 3991–3995. Dąbrowska, K., Świtała-Jeleń, K., Opolski, A., and Górski, A. (2005a). Possible association between phages, Hoc protein, and the immune system. Arch. Virol. 151, 209–215. Dąbrowska, K., Świtała-Jeleń, K., Opolski, A., Weber-Dąbrowska, B., and Górski, A. (2005b). Bacteriophage penetration in vertebrates. J. Appl. Microbiol. 98, 7–13. Dąbrowska, M. (1988). Postwar Diary 1945–65. (Warszawa: Czytelnik), p. 343. Dąbrowska, K., Zembala, M., Boratyński, J., Świtała-Jeleń, K., Wietrzyk, J., Opolski, A., Szczaurska, K., Kujawa, M., Godlewska, J., and Górski, A. (2007). Hoc protein regulates the biological effects of T4 phage in mammals. Arch. Microbiol. (in press).

Bacteriophages in Medicine

Dąbrowski, M., Weber-Dąbrowska, B., and Doroszkiewicz W. (1988). The estimation of usefulness of Pseudomonas bacteriophages in epidemiological investigation of Pseudomonas aeruginosa clinical strains. Acta Microbiol. Pol. 37, 295–307. De Berardinis, P., Sartorius, R., Caivano, A., Mascolo, D., Domingo, G.J., Del Pozzo, G., Gaubin, M., Perham, R.N., Piatier-Tonneau, D., and Guardiola, J. (2003). Use of fusion proteins and procaryotic display systems for delivery of HIV-1 antigens: development of novel vaccines for HIV-1 infection. Curr. HIV Res. 1, 441–446. Delmastro P., Meola A., Monaci P., Cortese R., and Galfre G. (1997). Immunogenicity of filamentous phage displaying peptide mimotopes after oral administration. Vaccine 15, 1276–1285. Demczuk, W., Soule, G., Clark, C., Ackermann, H-W., Easy, R., Khakhria, R., Rodgers, F., and Ahmed, R. (2003). Phage-based typing scheme for Salmonella enterica serovar Heidelberg, a causative agent of food poisonings in Canada. J. Clin. Microbiol. 41, 4279–4284. Di Marzo Veronese, F., Willis, A.E., Boyer-Thompson, C., Appella, E., and Perham, R.N. (1994). Structural mimicry and enhanced immunogenicity of peptide epitopes displayed on filamentous bacteriophage. The V3 loop of HIV-1 gp120. J. Mol. Biol. 243, 167–172. Drulis-Kawa, Z., Weber-Dąbrowska, B., Lusiak-Szelachowska, M., and Doroszkiewicz, W. (2005). Potential possibilities of using phage typing in elimination of multidrug resistant staphylococci. Pol. J. Microbiol. 54, 63–67. Dubos, R.J., Straus, J.H., and Pierce, C. (1943). The multiplication of bacteriophage in vivo and its protective effects against experimental infection with Shigella dysenteria. J. Exp. Med. 20, 161–168. Duckworth, D.H., and Gulig, P.A. (2002). Bacteriophages. Potential treatment for bacterial infections. Biodrugs 16, 57–62. Duijkeren, E., Wannet, W., Houwers, D., and Pelt, W. (2002). Serotype and phage type distribution of Salmonella strains isolated from humans, cattle, pigs and chickens in the Netherlands from 1984 to 2001. J. Clin. Microbiol. 40, 3980–3985. Editorial (2004). Methicilin-resistant Staphylococcus aureus infection in children. Lancet 363, 1047. Entenza, J.M., Loeffler, J.M., Grandgirard, D., Fischetti, V.A., and Moreillon, P. (2005). Therapeutic effects of bacteriophage Cpl-1 lysin against Streptococcus pneumoniae endocarditis in rats. Antimicrob. Agents Chemother. 49, 4789–4792. Esber, H.J., Ganfield, D., and Rosenkrantz, H. (1985). Staphage lysate: an immunomodulator of the primary immune response in mice. Immunopharmacology 10, 77–82. Faruque, S.M., Islam, M.J., Ahmad, Q.S., Faruque, A.S.G., Sack, D.A., Nair, G.B., and Mekalanos, J.J. (2005). Self-limiting nature of seasonal cholera-epidemics: role of host-mediated amplification of phage. Proc. Nat. Acad. Sci. USA 102, 6119–6124. Feldmane, G., Loza, V., Duks, A., Yesyacova, I., Vinkele, R., Buike, A., Meldrajs, J., and Erele, I. (1977). Nucleic acid as interferon inducer. Its antiviral and antitumor activity. Arch. Immunol. Ther. Exp. 25, 693–700. Ferrini, U., Mileo, A.M., Nista, A., Mattei, E., and Orofino, A. (1989). Polymorphonuclear leukocyte stimulation measured by phage inactivation. Int. Arch. Allergy Appl. Immunol. 90, 207–212. Fischetti, V.A. (2001). Phage antibacterials make a comeback. Nat. Biot. 19, 734–735. Fischetti, V.A. (2003). Novel method to control pathogenic bacteria on human mucous membranes. Ann. NY Acad. Sci. 987, 207–214. Fischetti, V.A. (2005a). The use of phage lytic enzymes to control bacterial infections. In: Bacteriophages. Biology and Applications., E. Kutter, A. Sulakvelidze, ed. (Boca Raton, USA: CRC Press), pp. 321–334. Fischetti, V.A. (2005b). Bacteriophage lytic enzymes: novel anti-infectives. Trends Microbiol. 13, 491–496. Fisher, J.S.T. (2004). Dramatic shift in the epidemiology of Salmonella enterica serotype enteritidis phage types in western Europe, 1998–2003—results from the Enter-net international Salmonella database. Euro. Surveill. 9, 43–45. Fogelman, I., Davey, V., Ochs, H.D., Elaskoff, M., Feinberg, M.B., Mican, J., Siegel J.P., Sneller, M., and Lane, H.C. (2000). Evaluation of CD4+ T cell function in vivo in HIV-infected patients as measured by bacteriophage phiX174 immunization. J. Inf. Dis. 182, 435–441. Gaubin, M., Fanutti.C., Mishal, Z., Durrbach, A., De Berardinis, P., Sartorius, R., Del Pozzo, G., Guardiola, J., Perham, R.N., and Piatier-Tonneau, D. (2003). Processing of filamentous bacteriophage

|

151

152

|

Górski et al.

virions in antigen-presenting cells targets both HLA class I and class II peptide loading compartments. DNA Cell. Biol. 22, 11–8. Geier, M.R., Trigg, M.E., and Merril, C.R. (1973). Fate of bacteriophage lambda in non-immune germfree mice. Nature 246, 221–222. Górski, A., Dąbrowska, K., Świtała-Jeleń, K., Nowaczyk, M., Weber-Dąbrowska, B., Boratyński, J., Wietrzyk, J., and Opolski, A. (2003). New insights into the possible role of bacteriophages in host defense and disease. Med. Immunol. 2, 2–6. Górski, A., and Weber-Dąbrowska, B. (2005). The potential role of endogenous bacteriophages in controlling invading pathogens. Cell. Mol. Life Sci. 62, 511–519. Górski, A., Kniotek, M., Perkowska-Ptasinska, A., Mróz, A., Przerwa, A., and Gorczyca, W. (2006a). Bacteriophages and transplantation tolerance. Transplantation Proc. 38, 331–333. Górski, A., Ważna, E., Weber-Dąbrowska, B., Dąbrowska, K., Świtała-Jeleń, K., Międzybrodzki, R. (2006b). Bacteriophage translocation. FEMS Immunol. Med. Microbiol. 46, 313–319. Grienenberger, A. (2004). Establishing pan-European clinical trials: regulatory compliance and other practical considerations. J. Biolaw. Bus. 7, 58–63. Hagens, S., Habel, A., von Ahsen, U., von Gabain, A., Blasi, U. (2004). Therapy of experimental pseudomonas infections with a nonreplicating genetically modified phage. Antimicrob. Agents Chemother. 48, 3817–3822. Hanson, L.A., and Tan, E.M. (1965). Characterization of antibodies in human urine. J. Clin. Invest. 44, 703–715. Hardy, A. (2004). Salmonella: a continuing problem. Postgrad. Med. J. 80, 541–545. Hartley O., Dorgham K., Perez-Bercoff D., Cerini F., Heimann A., Gaertner H., Offord R.E., Pancino G., Debre P., and Gorochov G. (2003). Human immunodeficiency virus type 1 entry inhibitors selected on living cells from a library of phage chemokines. J. Virol. 77, 6637–6644. Hausler, T. (2006). Bug killers. Nature Medicine 12, 600–601. Hedstrom, S.A., and Kamme, C. (1973). Antibodies against staphylococcal bacteriophages in human sera. II. Assay of antibodies in exacerbation and regression of chronic staphylococcal osteomyelitis. Acta Path. Microbiol. Scand. Section B. 81, 749–752. Hermoso, J.A., Monterroso, B., Albert, A., Galan, B., Ahrazem, O., Garcia, P., Martinez-Ripoll, M., Garcia, J.L., and Menendez, M. (2003). Structural basis for selective recognition of pneumococcal cell wall by modular endolysin from phage Cp-1. Structure 11, 1239–1249. Hirszfeld, L. (1948a). The battle of the invisible with the imperceptible. Proc. of the Wroclaw Scientific Society, State Publishing Institute, 3, 3–18. (in Polish). Hirszfeld, L., Galis-Mikolajczyk A., Sembrat-Niewiadomska, Z., and Zwierz, C. (1948b). Bacteriophages and their role in the diagnosis of typhoid fever. Pol. Tyg. Lek. 14, 417–422 (in Polish). Hopkins, K., Desai, M., Frost, J., Stanley, J., and Logan, J. (2004). Fluorescent amplified fragment length polymorphism genotyping of Campylobacter jejuni and Campylobacter coli strains and its relationship with host specificity, serotyping and phage typing. J. Clin. Microbiol. 42, 229–235. Iizuka, A., Watanabe., T., Kubo, T., Yamamoto, M., Ogawa, K., Ohkuma, T., and Kaji, A. (1994). M13 bacteriophage DNA inhibits duck hepatitis B virus during acute infection. Hepatology 19, 1079–1087. Inal, J.M. (2003). Phage therapy: a reappraisal of bacteriophages as antibiotics. Arch. Immun. Ther. Exp. 51, 237–244. Jado, I., Lopez, R., Garcia, E., Fenoll, A., Casal, J., and Garcia, P. (2003). Phage lytic enzymes as therapy for antibiotic-resistant Streptococcus pneumoniae infection in a murine sepsis model. J. Antimicrob. Chemother. 52, 967–973. Jasieński, J. (1927). Attempts to apply of bacteriophages in surgery. Polska Gazeta Lekarska 4, 67–73. (in Polish). Jepson, C.D., and March, J.B. (2004). Bacteriophage lambda is a highly stable DNA vaccine delivery vehicle. Vaccine 22, 2413–2419. Jerne, K. (1956). The presence in normal serum of specific antibody against bacteriophage T4 and its increase during the earliest stages of immunization. J. Immunol. 76, 209–216. Kamme, C. (1973). Antibodies against staphylococcal bacteriophages in human sera.I. Assay of antibodies in healthy individuals and in patients with staphylococcal infections. Acta Pathol. Microbiol. Scand. [B] Microbiol. Immunol. 81, 741–748.

Bacteriophages in Medicine

Keller, R., and Zatzman, M.L. (1959). Studies on the factors concerned in the disappearance of bacteriophage particles from the animal body. J. Immunol. 83, 167- 171. Keller, R., and Engley, F.B. (1958). Fate of bacteriophage particles introduced into mice by various routes. Proc. Soc. Exp. Biol. Med. 98, 577–579. Keller, R. (1958). Passage of bacteriophage particles through intact skin of mice. Science 128, 718–719. Keyhani M. (1969). Protection against experimental Newcastle disease in chickens given Brucella bacteriophage. Vet. Rec. 84, 657–658. Kleinschmidt, W.J., Douthart, R.J., and Murphy, E.B. (1970). Interferon production by T4 coliphage. Nature 228, 27–30. Kniotek, M., Weber-Dąbrowska, B., Dąbrowska, K., Świtała-Jeleń, K., Boratyński, J., Wiszniewski, W., Glinkowski, I., Babiak, I., Górecki, M., and Nowaczyk, M. (2004a). Phages as immunomodulators of antibody production. In: Genomic Issues, Immune System Activation and Allergy (Immunology 2004), Medimond International Proceedings (Bologna: Monduzzi Editors), pp.33–36. Kniotek, M., Ahmed, A.M.A., Dąbrowska, K., Świtała-Jeleń, K., Opolski, A., and Górski, A. (2004b) Bacteriophage interactions with T cells and platelets. In: Cytokine Network, Regulatory Cells, Signaling and Apoptosis. (Immunology 2004), Medimond International Proceedings (Bologna: Monduzzi Editors), pp.189–192. Krichevsky, A., Rusnati, M., Bugatti, A., Waigmann, E., Shohat, S., and Loyter, A. (2005). The fd phage and a peptide derived from its p8 coat protein interact with the HIV-1 Tat-NLS and inhibit its biological functions. Antiviral Res. 66, 67–78. Kucharewicz-Krukowska, A., and Ślopek, S. (1987). Immunogenic effect of bacteriophage in patients subjected to phage therapy. Arch. Immunol. Ther. Exp. 35, 553–561. Lang, L.H. (2006). FDA approves use of bacteriophages to be added to meat and poultry products. Gastroenterology 131, 1370. Lee, B., Murakami, M., Mizukoshi, M., Shinomiya, N., and Yada, J. (1985). Effects of Staphylococcal Phage Lysate (SPL) on immunoglobulin production in human peripheral lymphocyte cultures. Yakugaki Zasshi 105, 574–579. Leiva, M., Moreno, E., Ruiz-Bravo, A., and Jimenez-Valera, M. (2005). Immunomodulation by non-absorbable antibiotics given by the intragastric route. Int. J. Antimicrob. Agents 25, 252–255. Lepper, P.M., Held, T.K., Schneider, E.M., Bolke, E., Gerlach, H., Trautmann, M. (2002). Clinical implications of antibiotic-induced endotoxin release in septic shock. Intensive Care Med. 28, 824–833. Leszczyński, P., Weber-Dąbrowska, B., Kohutnicka, M., Łuczak, M., Górecki, A., and Górski, A. (2006). Successful elimination of methicillin-resistant Staphylococcus aureus (MRSA) intestinal carrier status in a healthcare worker – a case report. Folia Microbiol. 51, 236–238. Levin, B.R., and Bull, J.J. (1996). Phage therapy revisited: the population biology of a bacterial infection and its treatment with bacteriophage and antibiotics. Am. Nat. 147, 881–898. Levin, B.R., and Bull, J.J. (2004). Population and evolutionary dynamics of phage therapy. Nat. Rev. Microbiol. 2, 166–173. Lityński, M. (1949). A case of diffuse peritonitis treated with bacteriophage. In: Report of the Warsaw Division of the Polish Society of Internal Medicine, Pol. Arch. Med. Wewn. 19, 304 (in Polish). Loeffler, J.M., and Fischetti, V.A. (2003a). Synergistic lethal effect of a combination of phage lytic enzymes with different activities on penicillin-sensitive and -resistant Streptococcus pneumoniae strains. Antimicrob. Agents Chemother. 47, 375–377. Loeffler, J.M., Djurkovic, S., and Fischetti, V.A. (2003b). Phage lytic enzyme Cpl-1 as a novel antimicrobial for pneumococcal bacteremia. Infect. Immun. 71, 6199–6204. Loeffler, J.M., Nelson, D., and Fischetti, V.A. (2001). Rapid killing of Streptococcus pneumoniae with a bacteriophage cell wall hydrolase. Science 294, 2170–2172. Loessner, M. (2005). Bacteriophage endolysins-current state of research and applications. Curr. Opin. Microbiol. 8, 1–8. Loessner, M., Kramer, K., Ebel, F., and Scherer, S. (2002). C-terminal domains of Listeria monocytogenes bacteriophage murein hydrolases determine specific recognition and high-affinity binding to bacterial cell wall carbohydrates. Mol. Microbiol. 44, 335–349. Low, L.Y., Yang, C., Perego, M., Osterman, A., and Liddington, R.C. (2005). Structure and lytic activity of a Bacillus anthracis prophage endolysin. J. Biol. Chem. 280, 35433–35439.

|

153

154

|

Górski et al.

Mankiewicz, E., and Liivak, M. (1967). Mycobacteriophages isolated from human sources. Nature 216, 485–6. Manoutcharian, K., Diaz-Orea, A., Gevorkian, G., Fragoso, G., Acero, G., Gonzalez, E., De Aluja, A., Villalobos, N., Gomez-Conde, E., and Sciutto, E. (2004). Recombinant bacteriophage-based multiepitope vaccine against Taenia solium pig cysticercosis. Vet. Immunol. Immunopathol. 99, 11–24. Matsuda, T., Freeman, T.A., Hilbert, D.W., Duff, M., Fuortes, M., Stapleton, P.P., Daly, J.M. (2005). Lysisdeficient bacteriophage therapy decreases endotoxin and inflammatory mediator release and improves survival in a murine peritonitis model. Surgery. 137, 639–646. Matsuzaki, S., Rashel, M., Uchiyama, J., Sakurai, S., Ujihara, T., Kuroda, M., Ikeuchi, M., Tani, T., Fujieda, M., Wakiguchi, H., and Imai, S. (2005). Bacteriophage therapy: a revitalized therapy against bacterial infectious diseases. J. Inf. Chemother. 11, 211–219. Matsuzaki, S., Yasuda, M., Nishikawa, H., Kuroda, M., Ujihara, T., Shuin, T., Shen, Y., Jin, Z., Fujimoto, S., Nasimuzzaman, M.D., Wakiguchi, H., Sugihara, S., Sugiura, T., Koda, S., Muraoka, A., and Junai, S. (2003). Experimental protection of mice against lethal Staphylococcus aureus infection by novel bacteriophage phi MR11. J. Infect. Dis. 187, 613–624. McNerney, R., and Traore, H. (2005). Mycobacteriophage and their application to disease control. J. Appl. Microbiol. 99, 223–233. Meek, E.S., and Takahashi, M. (1968). Differential inhibition by phagicin of DNA synthesis in cells infected with vaccinia. Nature 220, 822. Merril, C.R. (1977). Interactions of bacterial viruses and bacterial genes with animal systems. In: Molecular Genetic Modification of Eukaryotes, I. Rubenstein, ed. (New York: Academic Press). pp. 83–100. Merril, C.R., Scholl, D., and Adhya, S.L. (2003). The prospect for bacteriophage therapy in Western medicine. Nat. Rev. Drug Discov. 2, 489–497. Międzybrodzki, R., Fortuna, W., Weber-Dąbrowska, B., Górski, A. (2005). Bacterial viruses against viruses pathogenic for man? Virus Res. 110, 1–8. Milch, H., Laszlo, G., Ślopek, S., and Mulczyk, M. (1968). Lysotypes of Shigella flexneri. Arch. Immunol. Ther. Exp. 16, 265–279. Mora, A., Blanco, M., Blanco, J., Alonso, M., Dhabi, G., Thomson-Carter, F., Usera, M., Bartolome, R., Prats, G., and Blanco, J. (2004). Phage types and genotypes of Shiga toxin-producing Escherichia coli O157:H7 isolates from humans and animals in Spain: identification and characterization of two predominating phage types (PT2 and PT8). J. Clin. Microbiol. 42, 4007–4015. Mori, K., Kubo, T., Kibayashu, Y., Ohkuma, T., and Kaji, A. (1995). Anti-vaccinia virus effect of M13 bacteriophage DNA. Antiviral Res. 31, 79–86. Mulczyk, M., and Kucharewicz, A. (1957). Studies on phages for Shigella bacilli. Arch. Immunol. Ther. Exp. 5, 109–127. Mulczyk, M., Ślopek, S., and Marcinkowska, H. (1967). The role of specific bacteriophage in the control of the carrier state of Shigella Flexneri 6 bacilli. Epidemiol. Rev. 21, 97–100. Murugkar, H.V., Rahman, H., Kumar, A., and Bhattacharyya, D. (2005). Isolation, phage typing and antibiogram of Salmonella from man and animals in northeastern India. Indian J. Med. Res. 122, 2372–42. Nau, R., and Eiffert, H. (2002). Modulation of release of proinflammatory bacterial compounds by antibacterials: potential impact on course of inflammation and outcome in sepsis and meningitis. Clin. Microbiol. Rev. 15, 95–110. Nelson, D., Loomis, D., and Fischetti V.A. (2001). Prevention and elimination of upper respiratory colonization of mice by group A streptococci by using a bacteriophage lytic enzyme. Proc. Natl. Acad. Sci. USA 98, 4107–4112. Nelson, R. (2003). Antibiotic development pipeline runs dry—new drugs to fight organisms are not being developed. Lancet 362, 1726–1727. Nonoyama, S., Etzioni, A., Toru, H., Ruggerie, D.P, Lewis, D., Pollack, S., Aruffo, A., Yata, J.I., and Ochs, H.D. (1998). Diminished expression of CD40 ligand may contribute to the defective humoral immunity in patients with MHC class II deficiency. Eur. J. Immunol. 28, 589–598. Norrby, S.R., Nord, C.E., and Finch, R. (2005). Lack of development of new antimicrobial drugs: a potential serious threat to public health. Lancet Infect. Dis. 5, 115–119. O’Flaherty, S., Coffey, A., Meaney, W., Fitzgerald, G.F., and Ross, R.P. (2005a). The recombinant phage lysin LysK has a broad spectrum of lytic activity against clinically relevant staphylococci, including methicillin-resistant Staphylococcus aureus. J. Bacteriol. 187, 7161–7164.

Bacteriophages in Medicine

O’Flaherty, S., Ross, R.P., Meaney, W., Fitzgerald, G.F., Elbreki, M.F., and Coffey, A. (2005b). Potential of the polyvalent anti-Staphylococcus bacteriophage K for control of antibiotics-resistant staphylococci from hospitals. Appl. Environ. Microbiol. 71, 1836–1842. Ochs, H.D., Buckley, R.H., Kobayashi, R.H., Kobayashi, A.L., Sorensen, R.U., Douglas, S.D., Hamilton, B.L., and Hersfield, M.S. (1992). Antibody response to bacteriophage &X174 in patients with adenosine deaminase deficiency. Blood 80, 1163–1171. Ochs, H.D., Davis, S.D., and Wedgwood, R.J. (1971). Immunologic responses to bacteriophage &X 174 in immunodeficiency diseases. J. Clin. Invest. 50, 2559–2568. Ochs, H.D., Slichter, S.J., Harker, L.A., Von Behrens, W.E., Clark, R.A., and Wedgwood, R.J. (1980). The Wiskott-Aldrich syndrome: studies of lymphocytes, granulocytes, and platelets. Blood 55, 243–252. Olczak, M., and Strumiłło, B. (1961). Treatment of staphylococcal infections of the skin with phage. Pol. Tyg. Lek. 16, 250–252. (in Polish). Paisano, A.F., Spira, B., Cai, B. And Bombana, A.C. (2004). In vitro antimicrobial effect of bacteriophages on human dentin infected with Enterococcus faecalis ATCC 29212. Oral Microbiol. Immunol. 19, 327–330. Pang, J.C., Chiu, T.H., Chiou, C.S., Schroeter, A., Guerra, B., Helmuth, R., and Tsen, H.Y. (2005). Pulsed-field gel electrophoresis, plasmid profiles and phage types for the human isolates of Salmonella enterica serovar Enteritidis obtained over 13 years in Taiwan. J. Appl. Microbiol. 99, 1472–1483. Parent, K., and Wilson, I.D. (1973). Mycobacteriophage in Crohn’s disease. Gut 12, 1019–20. Parfitt, T. (2005). Georgia: an unlikely stronghold for bacteriophage therapy. Lancet 365, 2166–2167. Piechowicz, L., Galiński, J., Dajnowska-Stanczewa, A., and Garbacz, K. (2005). Phage types of Staphylococcus aureus isolated in Poland in 1999–2004. Med. Dosw. Mikrobiol. 57, 105–113. Price, T.H., Ochs, H.D., Gershoni-Baruch, R., Harlan, J.M., and Etzioni, A. (1994). In vivo neutrophil and lymphocyte function studies in a patient with leukocyte adhesion deficiency type II. Blood 84, 1635–1639. Pritchard, D.G., Dong, S., Baker, J.R., and Engler, J.A. (2004). The bifunctional lysin of Streptococcus agalactiae bacteriophage B30. Microbiology 150, 2079–2087. Przerwa, A., Zimecki, M., Świtała-Jeleń, K., Dąbrowska, K., Krawczyk, E., Łuczak, M., Weber-Dąbrowska, B., Międzybrodzki, R., and Górski, A. (2006). Effects of bacteriophages on free radical production and phagocytic functions. Med. Microbiol. Immunol. 195, 143–150. Przondo-Hessek, A. (1966). Bacteriophages of Klebsiella bacilli. Isolation, properties and use typing. Arch. Immunol. Ther. Exp. 14, 413–435. Pyle, N.J. (1926). The bacteriophage in relation to Salmonella Pullora infection in the domestic fowl. J. Bacteriol. 12, 245–261. Pyun, K.H., Ochs, H.D., Wedgwood, R.J., Yang, X.Q., Heller, S.R., and Reimer, C.B. (1989). Human antibody responses to bacteriophage phi X 174: sequential induction of IgM and IgG subclass antibody. Clin. Immunol. Immunopathol. 51, 252–263. Remuzzi, G., Schiepatti, A., Boissel, J.P., Garattini, S., and Horton, R. (2004). Independent clinical research in Europe. Lancet 364, 1723–1726. Reynaud, A., Cloastre, L., Bernard, J., Laveran, H., Ackermann, H.W., Licois, D., and Joly, B. (1992). Characteristics and diffusion in the rabbit of a phage for Escherichia coli 0103. Attempts to use this phage for therapy. Vet. Microbiol. 30, 203–212. Richardson, J.F., Rosdahl, V.T., van Leeuwen, W.J., Vickery, A.M., Vindel, A., and Witte, W. (1999). Phages for methicillin-resistant Staphylococcus aureus: an international trial. Edpidemiol. Infect. 122, 227–233. Rogers, C.J., Mee, J.M., Kaufmann, G.F., Dickerson, T.J., and Janda, K.D. (2005). Toward cocaine esterase therapeutics. J. Am. Chem. Soc. 127, 10016–10017. Ruiz, M., Rodriguez, J.C., Sirvent, E., Escribano, I., Cebrian, L., and Royo, G. (2003). Usefulness of different techniques in the study of the epidemiology of salmonellosis. APMIS. 111, 848–856. Russel, M. (1995). Moving through membranes with filamentous phages. Trends Microbiol. 3, 223–228. Rybniker, J., Kramme, S., and Small, P. (2006). Host range of 14 mycobacteriophages in Mycobacterium ulcerans and seven other mycobacteria including Mycobacterium tuberculosis-application for identification and susceptibility testing. J. Med. Microbiol. 55, 37–42. Schneider-Schaulies, S., and ter Meulen, V. (2002). Measles virus and immunomodulation: molecular bases and perspectives. Expert Rev. Mol. Med. 30, 1–18.

|

155

156

|

Górski et al.

Schubbert, R., Renz, D., Schmitz, B., and Doerfler, W. (1997). Foreign (M13) DNA ingested by mice reaches peripheral leukocytes, spleen, and liver via the intestinal wall mucosa and can be covalently linked to mouse DNA. Proc. Natl. Acad. Sci. USA 94, 961–966. Schuch, R., Nelson, D., and Fischetti V.A. (2002). A bacteriolytic agent that detects and kills Bacillus anthracis. Nature 418, 884–889. Schultz, I. and Neva, F.A. (1965). Relationship between blood clearance and viruria after intravenous injection of mice and rats with bacteriophage and polioviruses. J. Immunol. 94, 833–841. Sechter, J., Mestre, F., and Homsen, D.S. (2000). Twenty-three years of Klebsiella phage typing: a review of phage typing of 12 clusters of nosocomial infections and a comparison of phage typing with serotyping. Clin. Microbiol. Infect. 6, 233–238. Seyama, K., Nonoyama, S., Gangsaas, I., Hollenbaugh, D., Pabst, H.F., Aruffo, A., and Ochs, H.D. (1998). Mutations of the CD40 ligand gene and its effect on CD40 ligand expression in patients with X-linked hyper IgM syndrome. Blood 92, 2421–2434. Skurnik, M. and Strauch, E. Phage therapy: facts and fiction. Int. J. Med. Microbiol. 296,5–14. Ślopek, S., Durlakowa, I. Weber-Dąbrowska, B., Dąbrowski, M., and Kucharewicz-Krukowska, A. (1984). Results of bacteriophage treatment of suppurative bacterial infections. Detailed evaluation of the results obtained in further 150 cases. Arch. Immunol. Ther. Exp. 32, 317–335. Ślopek, S., Durlakowa, I., Kucharewicz-Krukowska, A., Krzywy, T., Ślopek, A., and Weber, B. (1972). Phage typing of Shigella flexneri. Arch. Immunol. Ther. Exp. 20, 1–60. Ślopek, S., Durlakowa, I., Kucharewicz-Krukowska, A., Krzywy, T., Ślopek, A., and Weber, B. (1973). Phage typing of Shigella sonnei. Arch. Immunol. Ther. Exp. 21, 1–161. Ślopek, S., Durlakowa, I., Weber-Dąbrowska, B., Kucharewicz-Krukowska, A., Dąbrowski, M., and Bisikiewicz, R. (1983a). Results of bacteriophage treatment of suppurative bacterial infections: general evaluation of the results. Arch. Immunol. Ther. Exp. 31, 267–291. Ślopek, S., Durlakowa, I., Weber-Dąbrowska, B., Kucharewicz-Krukowska, A., Dąbrowski, M., and Bisikiewicz, R. (1983b). Results of bacteriophage treatment of suppurative bacterial infections. Detailed evaluation of results. Arch. Immunol. Ther. Exp. 31, 293–327. Ślopek, S., Kucharewicz-Krukowska, A., Weber-Dąbrowska, B., and Dąbrowski, M. (1985a). Results of bacteriophage treatment of suppurative bacterial infections. Evaluation of the results obtained in 370 cases. Arch. Immunol. Ther. Exp. 33, 219–240. Ślopek, S., Kucharewicz-Krukowska, A., Weber-Dąbrowska, B., and Dąbrowski, M. (1985b). Results of bacteriophage treatment of suppurative bacterial infections. Analysis of treatment of suppurative staphylococcal infections. Arch. Immunol. Ther. Exp. 33, 261–263. Ślopek, S., Kucharewicz-Krukowska, A., Weber-Dąbrowska, B., and Dąbrowski, M. (1985c). Results of bacteriophage treatment of suppurative bacterial infections. Evaluation of the results obtained in children. Arch. Immunol. Ther. Exp. 33, 241–259. Ślopek, S., Przondo-Hessek, A., Milch, H., and Deak, S. (1967). A working scheme for bacteriophage typing of Klebsiella bacilli. Arch. Immunol. Ther. Exp. 15, 589–599. Ślopek, S., Weber-Dąbrowska, B., Dąbrowski, M., and Kucharewicz-Krukowska, A. (1987). Results of bacteriophage treatment of suppurative bacterial infections in the years 1981–1986. Arch. Immunol. Ther. Exp. 35, 569–583. Smith H.W., and Huggins, M.B. (1982). Successful treatment of experimental Escherichia coli infections in mice using phage: its general superiority over antibiotics. J. Gen. Microbiol. 128, 307–318. Smith, D.W. and Nagler-Anderson, C. (2005). Preventing intolerance: the induction of non-responsiveness to dietary and microbial antigens in the intestinal mucosa. J. Immunol. 174, 3851–3857. Sola, C., Cortes, P., Saka, H.A., Vindel, A.and Bocco, J.L. (2006). Evolution and molecular characterization of methicillin-resistant Staphylococcus aureus epidemic and sporadic clones in Cordoba, Argentina. J. Clin. Microbiol. 44, 192–200. Soothill, J.S. (1992). Treatment of experimental infections of mice with bacteriophages. J. Med. Microbiol. 37, 258–261. Srivastava, A.S., Kaido T., and Carrier, E. (2004). Immunological factors that affect the in vivo fate of T7 phage in the mouse. J. Virol. Meth. 115, 99–104. Stone, R. (2002). Stalin’s forgotten cure. Science 298, 728–731. Sulakvelidze, A., Alavidze, Z., and Morris, J.G. (2001). Bacteriophage therapy. Antimicrob. Agents Chemother. 45, 649–659.

Bacteriophages in Medicine

Sulakvelidze, A. (2005). Phage therapy: an attractive option for dealing with antibiotic-resistant infections. Drug Discover. Today 10, 807–809. Sulakvelidze, A., and Barrow, P. (2005). Phage therapy in animals and agribusiness. In: Bacteriophages. Biology and Applications., E. Kutter, A. Sulakvelidze, ed. (Boca Raton, USA: CRC Press), pp. 335–436. Switala- Jelen, K. (2005). Bacteriophage T4 interactions with platelets and leukocytes. Doctoral thesis. Institute of Immunology and Experimental Therapy, Polish Academy of Sciences, Wroclaw. Szarecki, B. (1942). Wound treatment with bacteriophage. Lekarz Wojskowy 34, 263. (in Polish). Tanji, Y., Shimada, T., Fukudomi, H., Miyanaga, K., Nakai, Y., and Unno, H. (2005). Therapeutic use of phage cocktail for controlling Escherichia coli O157:H7 in gastrointestinal tract of mice. J. Biosci. Bioeng. 100, 280–287. Thacker, P.D. (2003). Set a microbe to kill a microbe. Drug resistance renews interest in phage therapy. JAMA 290, 3183–3185. Thiel, K. (2004). Old dogma, new tricks—21st century phage therapy. Nature Biotechnol. 22, 31–36. Todar’s Online Textbook of Bacteriology. Available from: www.textbookofbacteriology.net. Uhr, J.W., and Weissmann, G. (1965). Intracellular distribution and degradation of bacteriophage in mammalian tissues. J. Immunol. 94, 544–550. Vales, N., Barron, B.L., Padilla, J.A., Sandoval, H., and Sotelo, J. (1991). Effect of dsRNA from J6 bacteriophage on herpetic infection in cell culture and an animal model. J. Interferon Res. 11, 271–274. Van der Meer, J.W. (2003). Immunomodulation by antimicrobial drugs. Neth. J. Med. 61, 242–248. Velicer, C.M., Heckbert, S.R., Lampe, J.W., Potter, J.D., Robertson, C.A., and Taplin, S.H. (2004). Antibiotic use in relation to the risk of breast cancer. JAMA 291, 827–835. Victor, V.M., Rocha, M., and de la Fuente, M. (2004). Immune cells: free radicals and antioxidants in sepsis. Int. Immunopharmacol. 4, 327–347. Victor, V.M., Rocha, M., Esplugues, J.V., and de la Fuente, M. (2005). Role of free radicals in sepsis: antioxidant therapy. Curr. Pharm. Dis. 11, 3141–3158. Walsh, S., Shah, A., and Mond, J. (2003). Improved pharmacokinetics and reduced antibody reactivity of lysostaphin conjugated to polyethylene glycol. Antimicrob. Agents Chemother. 47, 554–558. Wanebo, H.J., Riley, T., Katz, D., Pace, R.C., Johns, M.E., and Cantrell, R.W. (1988). Indomethacin sensitive suppressor—cell activity in head and neck cancer patients. The role of adherent mononuclear cell. Cancer 61, 462–474. Wang, J., Hu, B., Xu, M., Yan, O., Liu, S., Zhu, X., Sun, Z., Tao, D., Ding, L., Reed, E., Gong, J., Li, Q.Q., and Hu J. (2006a). Therapeutic effectiveness of bacteriophages in the rescue of mice with extendedspectrum beta-lactamase-producing Escherichia coli bacteremia. Int. J. Mol. Med. 17, 347–355. Wang, J., Hu, B., Xu, M., Yan, O., Liu, S., Zhu, X., Sun, Z., Reed, E., Ding, L., Gong, J., Li, Q.Q., and Hu, J. (2006b). Use of bacteriophage in the treatment of experimental bacteremia from imipenem-resistant Pseudomonas aeruginosa. Int. J. Mol. Med. 17, 309–317. Ward, L., De Sa, J., and Rowe, B. (1987). A phage typing scheme for Salmonella enteritidis. Epidemiol. Infect. 99, 291–294. Wildemauwe, C., Godard, C., Verschraegen, G., Claeys, G., Duyck, M.C., De Beenhouwer, H., and Vanhoof, R. (2004). Ten years phage-typing of Belgian clinical methicillin-resistant Staphylococcus aureus isolates (1992–2001). J. Hosp. Infect. 56, 16–21. Weber-Dąbrowska, B., Dąbrowski, M., and Ślopek, S. (1987). Studies on bacteriophage penetration in patients subjected to phage therapy. Arch. Immunol. Ther. Exp. 35, 563–567. Weber-Dąbrowska, B., Mulczyk, M., and Górski, A. (2000a). Bacteriophage therapy of bacterial infections: an update of our Institute’s experience. Arch. Immun. Ther. Exp. 48, 547–551. Weber-Dąbrowska, B., Zimecki, M., and Mulczyk, M. (2000b). Effective phage therapy is associated with normalization of cytokine production by blood cell cultures. Arch. Immunol. Ther. Exp. 48, 31–37. Weber-Dąbrowska, B., Mulczyk, M., and Górski, A. (2001). Therapy of cancer infections in cancer patients with bacteriophages. Clin. Appl. Immun. Rev. 1, 131–134. Weber-Dąbrowska, B., Mulczyk, M., and Górski, A. (2003). Bacteriophages as an efficient therapy for antibiotic—resistant septicemia in man. Transplantation Proc. 35, 1385–1386. Wedgwood, R.J., Ochs, H.D., and Davis, S.D. (1975). The recognition and classification of immunodeficiency diseases with bacteriophage &X 174. Birth Defects Orig. Artic. Ser. 11, 331–338. Wenzel, R.P. (2004). The antibiotic pipeline—challenges, costs, and values. N. Engl. J. Med. 351, 523–526.

|

157

158

|

Górski et al.

Westwater, C., Kasman, L.M., Schofield, D.A., Werner, P.A., Dolan, J.W., Schmidt, M.G., and Norris, J.S. (2003). Use of genetically engineered phage to deliver antimicrobial agents to bacteria: an alternative therapy for treatment of bacterial infections. Antimicrob. Agents Chemother. 47, 1301–1307. Witherspoon, R.P., Kopecky, K., Storb, R.F., Flournoy, N., Sullivan, K.M., Sosa, R., Deeg, H.J., Ochs, H.D., Cheever, M.A., Fefer, A., Thomas, E.D. (1982). Immunological recovery in 48 patients following syngeneic marrow transplantation or hematological malignancy. Transplantation 33, 143–149. Woeste, G., Zapletal, G., Wullstein G,, Golling, M., and Bechstein, W.O. (2005). Influence of MRSA carrier status in liver transplant recipients. Transplant Proc. 37, 1710–1712. Xiang, J., George, S.L., Wunschmann, S., Chang, Q., Klinzman, D., and Stapleton, J.T. (2004). Inhibition of HIV-1 replication by GB virus C infection through increases in RANTES, MIP-1alpha, MIP1beta, and SDF-1. Lancet 363, 2040–2046. Xie, H., Zhuang, X., Kong, J., Ma, G., and Zhang, H. (2005). Bacteriophage Exc-A is an efficient therapy for Escherichia coli 3-1 caused diarrhea in chickens. J. Gen. Appl. Microbiol. 51, 159–163. Yoong, P., Schuch, R., Nelson, D., and Fischetti, V.A. (2004). Identification of a broadly active phage lytic enzyme with lethal activity against antibiotic-resistant Enterococcus faecalis and Enterococcus faecium. J. Bacteriol. 186, 4808–4812. Young, R., Wang, I.N., and Roof, W.D. (2000). Phages will out: strategies of host cell lysis. Trends Microbiol. 8, 120–128. Zuercher, A.W., Miescher, S.M., Vogel, M., Rudolf, M.P., Stadler, M.B., and Stadler, B.M. (2000). Oral anti-IgE immunization with epitope-displaying phage. Eur. J. Immunol. 30, 128–135.

Phage Therapy: the Western Perspective

6

Harald Brüssow

Abstract Phage therapy has a long and colorful history. Phages have been explored as means to eliminate pathogens like Campylobacter in raw food and Listeria in fresh food or to reduce food spoilage bacteria. In agricultural practice phages were used to fight pathogens like Campylobacter, Escherichia and Salmonella in farm animals, Lactococcus and Vibrio pathogens in fish from aquaculture and Erwinia and Xanthomonas in plants of agricultural importance. The oldest use was, however, in human medicine. Phages were used against diarrheal diseases caused by E. coli, Shigella or Vibrio and against wound infections caused by facultative pathogens of the skin like staphylococci and streptococci. Phage therapy therefore looks like a platform technology. This impression is reinforced by recent extension of the phage therapy approach to systemic and even intracellular infections and the addition of non-replicating phage and isolated phage enzymes like lysins to the antimicrobial arsenal. However, despite some hope and hype in recent editorials on phage therapy, the current review documents that definitive proof for the efficiency of these phage approaches in the field or the hospital is only provided in a few cases. The impact of history and society on phage research The antibiotic crisis has created a substantial interest for alternative biological treatment options of bacterial infections. Some of the currently discussed ideas are not new. In fact, the concept of phage therapy goes back to Felix d’Hérelle, the co-discover of bacterial viruses, and his pioneering trials in the first decades of the 20th century. A renaissance of an old idea is not so rare in medicine. Phages share this fate with probiotic bacteria, where early 20th century concepts elaborated by the Nobel Prize winner E. Metchnikoff represent another biological approach to the treatment of infectious diseases. The proposal to use phages against bacterial pathogens has a long and twisted history (Alisky et al., 1998; Häusler, 2003; Summers, 1999; Sulakvelidze and Kutter, 2005; Häusler, 2003), it lives currently through a kind of renaissance (Merril et al., 2003; Sulakvelidze et al., 2001; Summers, 2001), but a final evaluation cannot yet be formulated. However, we dispose of a sufficient number of scientific publications that at least a status report can be given.

160

|

Brüssow

A major dilemma for a reviewer is the fact that phages against a number of infectious diseases affecting the intestinal tract (E. coli, Shigella, Salmonella, Proteus) or skin and wounds (Staphylococcus, Streptococcus) can be bought in Russian pharmacies, reflecting decades of phage research in the former Soviet Union. However, in the West the value of phage therapy is still a subject that awaits a definitive assessment. The reason for this paradox is that even the most ambitious clinical trials conducted in the Soviet era with this type of medicine are poorly documented in the published literature. Even translations of Russian reports were only of minor help for the present review. For example, one of the greatest and extremely well designed clinical trials of phage therapy conducted in the 1960s in Tbilisi/Georgia, which enrolled 30 000 children was only documented in a 75-line publication containing a single table (Babalova et al., 1968). Taken at face value, this trial is a convincing evidence for phage therapy. However, in view of this sparse scientific documentation, phage therapy continues to split scientists into phage-believers and phage-skeptics. Eastern scientists insist that phages demonstrated in their hands their medical value, while Western scientists claim that any evidence that is not documented with scientific publications of Western standard is suggestive at best. In fact, the dividing political line of the Cold War period persists until our days in this scientific discussion- and historians of science might argue that a healthy distrust is in place. History matters, especially in biology- the rise of the Soviet agronomist Lyssenko to power under Stalin created a special blend of Soviet biology in the political service of the communist ideology. Lyssenko delivered a death sentence to Soviet biology from which it did not recover until recent times, despite the continued scientific excellence in other less ideology-subjugated branches of knowledge in the Soviet Union (e.g. nuclear physics, space science). This deep-rooted distrust of Soviet biology impedes an objective assessment of the Soviet research in phage therapy. The current review takes a neutral “agnostic” attitude and does not try to judge this research. It concentrates instead on phage therapy approaches outside of Eastern Europe (this latter area is covered by the review of A. Gorski in this book). Before getting to this subject, let’s summarize some recent Western trends in the phage therapy field over the last years. These trends cannot be understood without some socioeconomic considerations. The author is not aware that the WHO or any other public health institutions are currently conducting trials with phages for the treatment and prevention of bacterial infections. In the West phage therapy became the activity of a few curious scientists and more recently several biotech companies. Under the pressure of markets and regulatory authorities, the latter had to formulate a research strategy that accounted for these constraints. The rise of antibiotic resistance in pathogenic bacteria coincides with the decreasing interest of the pharmaceutical industry to develop new generations of antibiotics, creating a window of opportunity for phages. Since the administrative hurdles to an introduction of phages into Western medicine are likely high and costly, many academic and industrial groups concentrated on the use of phage applications in other sectors like the food industry, in horticulture and agriculture or in animal rearing. Phages are even discussed as remedies in wastewater treatment processes (Withey et al., 2005). For example, phages were explored in “fermentation accidents” in sewage stations leading to inappropriate foaming of the tanks. Many experts therefore expect the first applications of phages in these non-medical areas where industries have only to address the safety standards of the US Environmental Protection Agency or comparable agencies and not those of the FDA.

Phage Therapy: the Western Perspective

From the perspective of the food industry, it will also be critical whether the regulatory authorities will treat phage addition to food as a “direct food additive” (like lactoferrin) having a potentially long lasting effect on the product or only as a “secondary direct food additive” that lack the long term effect (e.g. disinfection with ozone application). Furthermore, it is unknown whether the consumer will accept the idea to add a virus (even if it is only a virus infecting bacteria) to a ready-to-eat food product in order to reduce the risk of food borne illness (Kutter and Sulakvelidze, 2005, p.335–380). However, a major psychological hurdle has been removed with the approval of a Listeria phage cocktail for food sanitation by the US food and Drug Administration. Phage applications in the food industry Food microbiologists have considered four different application areas where phages could be of potential benefit. (i) Feeding to animals. Phages could be fed to the living animal to reduce the intestinal carriage of bacteria that are pathogenic to the human consumer of the animal products. This strategy could be applied during animal rearing on the farm to reduce the pathogen load already during animal rearing (e.g. Salmonella or Campylobacter reduction in poultry farming). Alternatively, phages can be fed to animals just before slaughtering (e.g. E. coli O157 reduction in cattle). In the first approach one could reduce the prevalence of the pathogen already at the farm level. This might be the preferable approach if the targeted bacterium compromises the health of the farm animal. In view of the high prevalence rate (broiler chicken are up to 76% Campylobacter positive), the value of such interventions for the human consumer might be limited since the animals might experience later cross-contamination at the slaughterhouse level. There is currently also a concern that large scale application of phages in broiler houses would quickly lead to the selection of phage-resistant bacteria (e.g. Campylobacter). Such resistance development would not be an issue when phage is only fed to animals at the slaughterhouse since this is epidemiologically a one-way situation where phage is not recycled back to the farm. (ii) Raw food. Phages can also be applied directly on raw food. This approach is currently explored with the application of Salmonella or Campylobacter phages on chicken skin or O157 phages on cattle meat. Raw food is usually cooked during food preparation and the pathogens are destroyed during this process. Therefore only undercooked food will become a problem to the consumer. (iii) Fresh-prepared food. In contrast to raw food, fresh-prepared food is consumed without any further heat treatment (“ready to eat”). Any bacterial contamination occurring during food processing will thus directly translate into an increased risk of food borne disease for the consumer. Salmonella and Listeria phages were tested in this context. (iv) Food spoilage. Without having a disease-inducing capacity, some bacteria can simply spoil the food item by giving rise to the production of offensive aromas and off-flavors when they succeed to grow in the time period between food processing in the factory and the food preparation in the kitchen. Spoilage organisms can thus substantially reduce the shelf-life of the food product and phages might help to extend it.

|

161

162

|

Brüssow

In the following, I will review recent progress in the field along these four lines. An overview on the older literature is given by Sulakvelidze and Barrow (2005 Kutter book) and a more detailed bibliography is provided by Greer (Greer, 2005). Phage treatment of raw food Poultry Ian Connerton and colleagues published recently a series of papers where they explored the possibility to reduce with phages the level of Campylobacter contamination on the skin of chicken. Campylobacter jejuni is a major cause of food-borne gastroenteritis in humans with more than 50 000 reported annual cases in the UK alone (the true incidence might be 10-times higher). Undercooked chicken is the main source of infection. Campylobacter is well adapted to the avian gut; horizontal transfer in flocks is rapid: after introduction of infected animals, nearly 100% of birds in a Campylobacter-free flock are colonized within just 2 days. In the first step, the researchers isolated Campylobacter phages from the skin of retail chicken maintained under commercial storage conditions. Ten percent of the samples were phage-positive with higher rates of phage recovery obtained from free-range chicken than from animals coming from standard and economy rearing conditions. The phages were Myoviridae with genome sizes in excess of 100 kb (some showed a curious two segment genome with 110 and 190 kb size). The phages differed dramatically for their recovery rate from chicken skin over a 10 day test period at 4°C. Recovery could be as high as 90% or as low as 2% of the initial inoculum (Atterbury et al., 2003b). In the next step the researchers inoculated chicken skin with C. jejuni, followed 30 min later with an inoculation of phage, then storage at 4°C. On control skin without phage, the added Campylobacter did not grow, but was maintained on the skin over 10 days and the cell titer showed only a slight decrease. When inoculated with a multiplicity of infection of 10 (e.g. 107 pfu of phage for 106 cfu bacteria) a 10-fold lower bacterial titer was measured on the phage-treated as compared to the untreated skin samples (Atterbury et al., 2003a). This reduction might seem modest, but when considering that phages replicate only on dividing bacteria, one could not have expected higher rates of reductions with this phage treatment. It is not unreasonable to expect that Campylobacter associated with phage might show a decreased outgrowth under food preparation conditions in the kitchen, but this effect was not investigated. The observations of the Nottingham group were confirmed by Paul Barrow from the Compton (UK) Animal Health group and colleagues (Goode et al., 2003). When Salmonella enterica serovar Enteritidis was applied to chicken skin, 60% of the bacterial inoculum could be re-isolated after 2 days of storage at 4°C. The recovery rate fell to 11% when a specific Salmonella phage was concomitantly applied. Only a low MOI of 1 was necessary to achieve this reduction. In the absence of the target bacterium, the phage titer dropped by about one log over the 2-day observation period. In the presence of the specific Salmonella strain, the phage titer increased threefold even at this low incubation temperature. When higher MOI was used, the Salmonella contamination could be further reduced. The efficacy of Salmonella phages on poultry carcasses was recently confirmed in a study by Higgins (Higgins et al.,

Phage Therapy: the Western Perspective

2005). Intentionally contaminated broiler carcasses treated with 109 pfu phage showed significantly lower Salmonella titer than untreated control carcasses. In the next step, they treated commercially processed turkeys with phages isolated on Salmonella strains, which were prevalent in the investigated turkey flock. Also under this condition, the researchers observed significantly reduced Salmonella levels with phage treatment when compared to control turkeys. Beef In order to achieve a control of E. coli O157:H7 on beef meat surfaces, O157-specific phages had first to be isolated. US researchers around Carolyne Hovde (Kudva et al., 1999) had to screen fifty feces from cattle and sheep to find three suitable phages. These phages were not only O157 serotype-specific, but needed the appropriate expression of the lipopolysaccharide and a high MOI (103 pfu/cfu) to achieve a significant decrease of bacterial titers. Only one of the phage isolates worked well. At 37°C the bacterial titers recovered over the next few days and the surviving cultures were resistant to all three phages. A mixture of the three phages was needed to sterilize the culture kept at 37°C. Notably, the sterilization was also achieved with phage-infected cultures held at 4°C. Paul Ross and colleagues from Cork (O’Flynn et al., 2004) extended these experiments with three further O157-specific phages. At 37°C, the siphophage had only moderate lytic potential, while two myophages reduced the pathogen below the detection level. However, over the next hours all E. coli cultures resumed growth. At 30°C outgrowth was only seen with the siphophage-infected culture, while at 12°C none of the phages had any lytic effect. The authors inoculated steak meat with 103 cfu/ml of E. coli O157 and superinfected the culture at 37°C with the three-phage cocktail at a massive MOI of 106. All nine control steaks showed E. coli titers of 105 cfu/ml, while seven phage-treated steaks were free of E. coli O157 and two showed less than 10 cfu/ml. As bacteriophage-resistant O157 strains invariably emerged from the phage-exposed culture, the authors investigated these cells. They found smaller and coccoid E. coli cells instead of the larger rod-shaped parental cells. Yasunori Tanjii and colleagues from the Tokyo Institute of Technology had observed this phenomenon before in continuous culture of E. coli O157 co-existing with myophage PP01 (Mizoguchi et al., 2003). Immediately after phage introduction, deep oscillations of bacterial and phage titers by a factor of 10 000 were observed. Thereafter the oscillations decayed and bacterium and phage moved to equilibrium at high concentrations for both (mutant MuL cells and PP01 phages at 108 cfu/ml and pfu/ml, respectively). Detailed analysis of the culture demonstrated the appearance of a series of phage-resistant E. coli isolates over time. Under chemostat growth conditions, mutant phages appeared that showed different host ranges than the parental PP01 phage, suggestive of an arms race between phage and bacterium before equilibrium was reached. In the initial phase the bacteria escaped from the phage predation by arrest of the synthesis of the outer membrane protein OmpC and alteration of the lipopolysaccharide composition. In the equilibrium phase, E. coli clone MuL split into two expression types: 85% were phage sensitive, while 15% resisted phage attack and showed reduced flagella expression and increased exopolysacchride production preventing the adsorption of the phage (Fischer et al., 2004).

|

163

164

|

Brüssow

Phage treatment of fresh food Over the last years a few reports addressed the potential of phages to reduce the contamination level of human food pathogens on food items that are consumed without further processing at home. Britta Leverentz from the US Department of Agriculture and colleagues contaminated fresh-cut melon and apple slices with Salmonella followed by a phage treatment. Due to the lower pH value on apple compared to melon slices (4.4 vs. 5.8), Salmonella titer increases were greater on melons than on apples. Phages were rapidly inactivated on the apple and a partial control of the Salmonella outgrowth could only be achieved with phages applied on the melon preparation. This attenuating effect was seen over a wide range of storage temperatures (5, 10, 20°C) (Leverentz et al., 2001). The researchers then explored a Listeria phage cocktail prepared by the phage biotech company Intralytix on L. monocytogenes contamination using the same fruits (Leverentz et al., 2003). On melons the authors achieved a better control of Listeria with a cocktail containing 20 lytic phages than with nisin, a bacteriocin permitted as a food preservative. On apple slices a combination of phage with 200 IU nisin showed a better reduction than 200 IU nisin alone. No potentiating effect of phage was seen when 400 IU nisin and phage was used. Phage had a somewhat greater inhibitory effect when applied as a spray than when applied with a pipette, which would facilitate a later industrial application. However, Alexander Sulakvelidze, one of the authors of this study, mentioned that this study is only the beginning: it will be important to determine the right phage (combination) applied in the right factory place at the right phage concentration under real food production conditions to achieve practical results. US researcher investigated the use of the Salmonella phage Felix on chicken frankfurters (Whichard et al., 2003). Phage Felix is a Myovirus with an unusual broad host range on smooth strains of Salmonella including those S. typhi strains causing typhoid fever and S. typhimurium strains causing many cases of food contamination in poultry products. High ratios of pfu/cfu of 10 000 were needed to inhibit Salmonella growth by about two logs on frankfurters. Canadian food scientists used a bioluminiscent Salmonella strain to trace the survival of this contaminant during Cheddar cheese production (Modi et al., 2001). The marked Salmonella strain was added at 104 cfu/ml. It increased over the next 6 hours to 106 cfu/ml and suffered only a two log decrease during 100 days of storage at 8°C. When phage was added at a concentration of 107 pfu/ml to the pasteurized milk, a significant decrease and eventually a loss of the Salmonella strain was observed over the storage period, which was accompanied by a phage titer increase. The example of the Cheddar cheese system is more of theoretical than practical interest since only two large Salmonella outbreaks were traced to cheese over the last 20 years in Canada. However, Salmonella control is obligatory for the cheese maker, thus phage application might be of value reducing product recall. Phages against spoilage bacteria The food industry has developed a range of methods to avoid food deterioration by food spoilage organisms. One of the oldest is food fermentation by lactic acid bacteria, which prevents outgrowth of the spoilage bacteria by lowering the pH in the product. Alternatives are pasteurization, application of hyperbaric pressure, gamma-irradiation and bacteriocins. Greer and Dilts tested the use of phage against food spoilage. Initial assays to extend the shelf life of beef products were unsuccessful (Greer and Dilts, 1990). In later experiments

Phage Therapy: the Western Perspective

the authors used a phage against Brochotrix thermosphacta, which generates the unpleasant “cheesy” off-odors associated with spoiled pork products. Remarkably, Gordon Greer had isolated a phage multiplying at 2°C on this psychrophilic (cold-loving) spoilage bacterium (Greer, 1983). When pork tissue was inoculated with 103 pfu/cm–2 of phage, the phage reached titers as high as 108 pfu/ml within three days. On phage-treated pork the titers of the spoilage bacterium remained for six days two to four logs lower than on phage-free pork when both were maintained under retail conditions. Phage-resistant bacteria then grew out and reached the cell count of the control pork, but no off-odor developed over this test period (Greer and Dilts, 2002). Phage treatment of live animals An attractive idea is to use phage to reduce the contamination level of livestock with food borne pathogens. Many human food pathogens do not cause disease in the animal carrier (e.g. E. coli O157 in cattle). However, in some cases the elimination of the food pathogen by a phage can also profit the animal as demonstrated by the increased hatch-rate of chicken eggs injected with Salmonella phages (quoted from (Kutter and Sulakvelidze, 2005). It was calculated that a two-log reduction in Campylobacter numbers on retail chicken could reduce the frequency of human Campylobacter infections by 30-fold. Motivated by this prospect, Ian Connerton and colleagues screened fifty Campylobacter phages isolated from retail poultry and broiler chicken on a collection of 130 Campylobacter strains to obtain broad host range phages (Connerton et al., 2004). Two such polyvalent phages were given with an antacid to 25-day-old broiler chicken. The animals were colonized with the C. jejunum test strain resulting in a titer of up to 108 cfu/g in the upper intestine, cecum and lower intestine. When the phages were given with a dose of 107 pfu, phage CP8 had only a minimal effect on the intestinal bacterial carriage, while phage CP34 caused a moderate to substantial reduction in intestinal Campylobacter counts. In vitro infection with these two phages on the challenge cell did not predict this differential in vivo lytic effect. When another Campylobacter strain was used for in vivo colonization, phage CP8 addition induced a dramatic clearance of Campylobacter over the first three days of phage inoculation. Over the next two days an outgrowth of bacteria was seen in the phage-treated chicken. However, phage-resistant Campylobacter was only recovered from these animals with a frequency of < 4%. These mutants were compromised in their ability to colonize chicken and reverted in vivo rapidly to phage sensitivity (Loc et al., 2005). In an epidemiological study the researchers followed over time phages and their Campylobacter hosts in broiler chicken from three houses. They were interested in the development of phage resistance. Some phage-resistant C. jejuni strains were isolated, but these strains did not dominate nor did they outgrow the sensitive strains. Notably, both cell types coexisted in the chicken gut in the presence of phage (Connerton et al., 2004). At the time of the writing, mostly congress reports documented the effect of oral feeding of coliphages on the carriage rate of E. coli O157:H7 in animals. Tanji’s group from Tokio observed only modest in vivo effects in a model system, the laboratory mouse (Tanji et al., 2005). Part of the failure was probably due to the limited in vitro lytic effects of the selected phages. Cocktails of phages had to be used to prevent outgrowth of resistant cells and this was only achieved aerobically. In the chemostat, no phage effect was seen

|

165

166

|

Brüssow

at a MOI of 1 and a MOI of 1000 was needed to achieve a sizable reduction of the E. coli count, which was, however, followed by an inevitable bacterial outgrowth. Repeated phage additions were needed to keep the cell titers oscillating. Daily massive doses of 1010 phages had to be given orally to maintain the fecal phage titer at 105 pfu/ml. After a single dose of 108 pfu, phage was lost from the feces of mice after two days. The fecal O157 titer of 5 × 102 cfu/g from phage-treated animals did not differ from that of control mice that had not received phage. This titer was reduced to 50 cfu/g in mice receiving daily 1010 pfu phage. The mice were sacrificed and phage was detected throughout the small intestine, cecum and colon, depending on the oral dose. The gut E. coli O157 titers were surprisingly higher in the phage-treated than in control mice challenged with the pathogen, which showed titers < 100 cfu/g. This apparent paradox was not discussed by the authors. Only extremely low target cell titer were achieved in the gut of control mice, which might not any longer permit a productive infection due to threshold phenomena described in vitro for E. coli at cell concentrations < 104 cfu/ml (Wiggins and Alexander, 1985). More promising results were obtained by members of the T4 phage lab from Elizabeth Kutter at Olympia collaborating with veterinarians from USDA when working with sheep, a natural host of O157 (Raya et al., 2006). Initially, they were surprised how quickly high challenge doses of 109 cfu of O157 dropped below the detection limit in the feces of their sheep; then they found the likely cause when detecting phage CEV1 in the feces of their sheep. CVE1 was characterized as a T4-like phage and occurred in the feces from 50% of their sheep flock. For the next challenge experiments they only used CEV1-negative sheep. When these sheep were orally challenged with 1010 cfu of O157, they observed 5 days post-challenge 106 cfu/ml of this bacterial strain in cecum and rectum. When these sheep were treated 3 days after the challenge with 1011 pfu phage CEV1, a greater than 100-fold decrease in cecum and rectum O157 titer was observed. Interestingly, the O157 titer in the rumen was only 103 cfu/ml and in accordance with the threshold hypothesis, no effect of the oral CEV1 phage was observed. Researchers from the biotech company Gangagen identified cattle transported into the slaughterhouse as ideal targets for phage therapy. Just before slaughter the cattle were treated orally with O157-specific phage and in addition sprayed with the phage. They expected to decrease the O157 cell number in the gut and on the skin, which should translate into a lower contamination level of ground beef with E. coli O157:H7. To their surprise they met great difficulties when trying to eradicate E. coli O157:H7 from naturally infected cattle by applying their phage preparation a few days before slaughtering in the abattoir (seminar 105th general meeting of ASM). As these data have not yet been published, it is difficult to identify the cause for this problem. Since cattle is naturally only contaminated with relatively low E. coli O157 cell titers, the lytic effect of phage might be less pronounced in cattle than in the sheep exposed to high O157 cell numbers. This conclusion sounds at first glance like a paradox (less killing with higher MOI), but might again reflect a target cell threshold effect for phage infection. Coexistence at low titer levels have been described for a number of phages and their target cells (e.g. cyanobacteria and their phages in the ocean (Waterbury and Valois, 1993) and some microbial ecologists believe that this threshold effect might assure the persistence of both ecological antagonists, locking them in a permanent evolutionary arms race (Levin et al., 1977; Schrag and Mittler, 1996). However,

Phage Therapy: the Western Perspective

it should also be mentioned that other biologists have contested the reality of threshold effects in phage ecology. Phages therapy in agriculture Phages as biocontrol agents of plant pathogens Somewhat characteristic for the entire phage therapy discussion is the following observation. On one side, a reviewer asks whether phages in agronomy are fact or fiction (Goodridge, 2004). At the same time, L.E. Jackson is distributing with great success his AgriPhage preparation to American farmers growing tomatoes and his small company Omnilytics has the official US approval to do so. It looks like a remake of the East-West dialog where phage preparations are sold with state approval in Russian pharmacies, while researchers in the West ask the question “myth or reality” for the medical phage application. This déjà vu situation is somewhat surprising since the agronomical phage therapy research has largely been conducted in the West and was documented in a number of publications in mainstream agronomy research journals. Part of the skepticism originates from ecological doubts. The phyllosphere, the aerial plant structures, are seen by ecologists as less hospitable to phages than the rhizosphere, represented by the plant roots and the surrounding soil. In the phyllosphere, phages should readily decay after exposure to UV and intense visible light. In addition, phages should suffer from desiccation. Phages in the soil are generally more numerous reflecting their protection from light inactivation. Furthermore, the partial hydration of soil allows some diffusion of phages. In fact, phages from Erwinia amylovora, the causal agent of fire blight, a serious disease of pome fruit (apple, pear, peach), were ten-times more frequently isolated from the soil surrounding infected trees than from aerial parts of the same trees (Gill et al., 2003). The observation that phages can play a role as biological control agents in orchards goes back to a paper from J. Erskine in 1973 (Erskine, 1973). It described an avirulent yellow saprophyte bacterium that is commonly found in association with pathogenic Erwinia strains. The saprophyte was identified as E. herbicola (now a distinct genus Patoea). It is lysogenic with a phage that can modify the severity of the fire-blight disease by serving as a lytic phage reservoir for E. amylovora. Subsequently, it was shown that the symptoms of fire blight could be attenuated in apple seedlings inoculated with Erwinia phage Ea1 when given together with its host cell, E. amylovora or E. herbicola (Ritchie and Klos, 1979). Phage Ea1 had a broad host range and infected nearly all E. amylovora isolates from apple orchards and raspberry fields in Michigan (Schnabel and Jones, 2001). Later work showed that an efficient control strategy based on phage would require a mixture of three phages since single phage preparations were unable to control Erwinia in liquid culture. Furthermore, blossoms of the fruit trees had to be treated with phage. However, phage populations decayed rapidly on blossoms due to UV light. It will thus be necessary to colonize first blossoms with an avirulent strain of Erwinia, which can then serve as an innocuous amplification strain for the phage such that the viral numbers on the blossoms can be maintained. On Jonathan apple trees, a phage mixture applied a day before or on the same day with the E. amylovora challenge reduced the number of infected blossoms from 49% in the control to 36 and 31%, respectively, in the phage-treated trees (Schnabel et al., 1999). Erwinia phages have also

|

167

168

|

Brüssow

been isolated from corn flee beetles, but it is unknown whether arthropods serve as vectors of phages linking the phyllosphere with the rhizosphere and contribute thus to the control of the pathogenic strain in nature. Also the use of Xanthomonas phage has a long history going back to work of plant pathologists from USDA starting in the 1960s. X. pruni phage protected foliage of peach seedlings from infection with Xanthomonas when the phage was applied as a spray before inoculation with the bacterium. Even when the bacterial inoculation was delayed for 24 hours, a marked symptom reduction was observed, suggesting that the phage remained active on the leaf surface. However, phage applied as short as 1 hour after the bacterial inoculation had no effect on the development of symptoms (Civerolo and Keil, 1969). Subsequent work showed that preinoculation treatment of foliage with phage lysates resulted in a higher symptom score in bacterial spot disease of plum trees than a pre-inoculation treatment with phage freshly mixed with the pathogenic host bacterium (Civerolo, 1973). A possible explanation is that the second system still leads to a phage amplification on the sprayed leaves. Xanthomonas campestris is another versatile plant pathogen. The pathovar vesicatoria causes bacterial spot disease on tomatoes and pepper and can be devastating for tomato growers in warm and damp climates. The disease was controlled with copper-containing pesticides. However, the emergence of pesticide-resistant bacteria has stimulated interest in biological control with h-mutant phages. H-mutant stands for extended host range phages that are created by selecting phage-resistant bacteria, which survived infection with the parental phage. In a second step, a mutant phage is searched in the parental phage stock able to plaque on the resistant bacterium. The phage mutant that can infect the mutant cell frequently shows an extended host range on strains non-infected by the original phage, while still infecting the original host. These h-mutant Xanthomonas phages reduced the incidence of bacterial spot disease on greenhouse-grown seedlings as well as on field-grown tomatoes over a two-growing observation period. The protective effect of phage was greater than that achieved with the copper pesticide. The phage treatment also resulted in more extra-large fruits at harvest than the pesticide treatment (Flaherty et al., 2000). Another pathovar of X. campestris, namely pv. pelargoni is a devastating disease for growers of seedling and cutting geraniums. Phage (5 s 108 pfu/ml) applied daily on infected potted geranium reduced the disease incidence from 53 to 19% and likewise the severity of disease symptoms when compared to untreated controls or plants treated with a copper pesticide (Flaherty et al., 2001). Reducing the phage application to twice a week abolished the protective effect. Identical observations were made when geranium seedlings were treated with the same protocol. Phages were also tested against a number of other plant pathogens. For example, seed potatoes infected with Streptomyces scabies were bathed in a high titered phage suspension (109 pfu/ml) of a specific phage. This treatment reduced the surface of lesions from 23% of the tubers in the control seed to a mere 1% in the phage-bathed tubers, while the weight of the harvested tubers was unaffected (McKenna et al., 2001). For more literature references see a recent review on the phage use in plants (www.apsnet.org/online/feature/phages/). This review deals with phages in the rhizosphere under a more generalized view than that of phage therapy and mentions for example also phage which had a negative impact on plant growth. The argument is somewhat indirect. Pseudomonas phage GP100 had a nega-

Phage Therapy: the Western Perspective

tive impact on the population size of P. fluorescence in the soil. Since this bacterium protects cucumber against a root disease caused by the pathogenic oomycete Phytium ultimum, phage application killed P. fluorescence and thus abolished their protective effect (Keel et al., 2002). Phage use in aquaculture The Japanese fish pathologist Toshihiru Nakai demonstrated in a series of papers the use of phages in aquaculture. His first test group was yellowtails (Seriola quinqueradiata) infected with Lactococcus garvieae. For microbiologists coming from the fermentation field a pathogenic Lactococcus sounds perhaps somewhat exotic. However, this ubiquitous opportunistic pathogen affects fish suffering from poor water quality, overcrowding and inadequate nutrition. When given by intraperitoneal injection, the phage showed a lower persistence in the body of fish when given alone as compared to co-application with live Lactococcus pathogen (1 vs. 5 days persistence in the spleen). Similar observations were made after oral feeding. Survival rates of phage-injected yellowtails were 100, 80, and 50% for groups treated 0, 1 hour and 24 hours after Lactococcus challenge, while control fish showed only 10% survival. Also feeding fish with phage reduced the mortality from 65% in controls to 10% in the phage-fed fish. A practical problem was the instability of the phage in seawater wherein it lost infectivity after 3 days, while only moderate losses were observed in sterilized seawater (Nakai et al., 1999). The next target of the Nakai lab was ayu, Plecoglossus altivelis, a popular sport fish in Japan, recently plagued by a relative of the bacterium Pseudomonas putida. The deadly disease occurs shortly after introduction of the fish into culture ponds. Since no chemotherapy is licensed for its treatment, the researchers isolated phage from diseased fish. In vitro, the phages had only modest effects on cell growth. Pathogen-inoculated fish showed a 65% mortality, while those that were fed with phage died later and with a lower frequency, namely 23%. Protection was maintained even when the phage treatment was delayed for 24 hours. The phage showed only a transient peak of replication to about 105 pfu/g, but this prevented outgrowth of the Pseudomonas inoculum (Park et al., 2000). A mixture of two phages reduced the mortality more than the application of each single phage. When fish experienced a water-borne pathogen challenge, phage applied with the feed reduced mortality in fish, kept in a water tank, from 90% to 20%. Even under commercial culture conditions, phage developed a protective effect. A facility suffered 18 kg fish loss per day before phage addition to the pond. This loss decreased to 6 kg per day about 2 weeks after the phage intervention (Park and Nakai, 2003). Recently, Nakai’s group reported success of phage therapy against bacterial infection of shellfish, namely Vibrio infection of oyster larvae, thus extending the possible use of phages in commercial aquaculture (Nakai and Park, 2002). Phage use in farm animals Chicken In veterinary practice E. coli cause severe respiratory infections in broiler chicken. Researchers from the USDA explored the possibility of phage therapy for the poultry farm. In one

|

169

170

|

Brüssow

study, phages were applied by aerosol spraying, followed by injection of 104 cfu E. coli directly into the thoracic air sac (Huff et al., 2002). Aerosol containing 107 pfu of two phages halved the mortality when the challenge was done on the day of phage spraying. When the dose of the phage was increased to 108 pfu, significant protection against infection was still observed up to 3 days after phage spraying. In a follow-up study these researchers demonstrated that multiple intra-muscular injection of phage was better than a single injection of phage with respect to mortality and recovery from experimental E. coli infection. However, for both schedules efficacy diminished when the phage treatment was delayed (Huff et al., 2003). In another study they compared bacteriophage with antibiotic treatment in chicken challenged with an E. coli pathogen, inoculated into the air sac. Infected control animals showed 68% mortality, this figure was reduced to 15% by phage treatment and to 3% by the antibiotic enrofloxacin. Interestingly, no mortality was seen in chicken treated with both antibiotic and phage, suggesting synergy between both treatment modes (Huff et al., 2004). The efficiency of phage in chicken was also documented by other labs. A British lab demonstrated the efficacy of phage applied intramuscularly against lethal E. coli infections (Barrow et al., 1998). When given at equal numbers, no morbidity was observed at all in chicken, but also 100-fold lower phage titers conferred significant protection demonstrating the in vivo multiplication of the phage. Intramuscular phage protected also against intracranial E. coli infection. Phage therapy was even effective when given at the onset of clinical symptoms. A Chinese group applied a single coliphage isolate orally to newly hatched chicken at a dose of 105 pfu/day. Over the first two weeks of life phage treatment significantly reduced the diarrhea rate and mortality when compared to untreated controls. Notably, phage treatment was also superior to antibiotic treatment (chloromycetin) (Xie et al., 2005). Already 15 years ago, Paul Barrow’s group had also explored the effect of nine Salmonella phages on the mortality rate of one-day old chicken infected with three S. typhimurium strains (Berchieri, Jr. et al., 1991). Only one strain, F2.2, induced a significantly reduced mortality in the chicken. This phage reduced the Salmonella titer by more than one log in the crop and small intestine, but only high titered phage (> 1010 pfu/ml) achieved a protection. Another phage, FAB, which coexisted with the Salmonella strain in the cecum, did not decrease the Salmonella-induced mortality. Interestingly, this phage spread also to uninoculated “in contact” chicken from the same pen. A cocktail of three Salmonella phages improved the weight gain in Salmonella-challenged chicken and reduced the pathogen titer in the gut (Toro et al., 2005). In a prevention trial, a phage delayed and reduced Campylobacter jejuni colonization in chicks. In a treatment mode the phage achieved first a 3-log reduction in the colonization level with C. jejuni. However, after the end of the phage application, only a 1-log reduction of the pathogen was maintained (Wagenaar et al., 2005). Calves In what should become classical experiments of Western phage therapy approaches, the British veterinarian Smith and colleagues infected calves with a natural bovine enteropathogenic E. coli strains causing high lethality. Convincing evidence for the efficacy of phage therapy was obtained in an extremely carefully documented series of experiments (Smith and Huggins, 1982; Smith and Huggins, 1983; Smith et al., 1987a; Smith et al., 1987b).

Phage Therapy: the Western Perspective

Diarrhea could be prevented by phage given 1 to 8 hours after infection. When phage application was delayed until the onset of symptoms, phage had no effect on diarrhea, but still largely prevented death (Smith and Huggins, 1983). Phage titers increased in the feces over time with a concomitant decrease in the enteropathogen counts. In sacrificed animals this observation was confirmed at all anatomical levels of the gastrointestinal tract. Phage counts were 10-fold lower in mucosal scrapings than in the luminal content. Phage was not recovered from blood or spleen. Phage-resistant cells were observed in most of the calves, but their titers generally remained low. Upon re-inoculation into new calves, the mutant cells were less competitive than the parental strain. In a second series of experiments, a low dose of phage (105 pfu) was given to calves at the onset of diarrhea and animals were sacrificed in a time series. Phage counts as high as 1010 were observed during the first 12 hours in the posterior parts of the small intestine, followed by a decline at 24 hours and a disappearance of phage at 40 hours when the pathogenic bacterium could not any longer be detected. Phage-resistant E. coli appeared during the experiment, but they had lost the K1 antigen, their major virulence factor. Control of the diarrhea by a low dose of phage (105 pfu) given 6 hours or 10 min before infection of the calves with E. coli was only achieved with phages showing a high in vitro bacterial lytic activity (Smith et al., 1987a). With an extremely low dose of phage (102 pfu) prophylaxis was only possible when given 10 min, but not 6 hours before bacterial challenge. Within 5 hours the phage had multiplied up to 1011 pfu in vivo demonstrating an impressive phage multiplication in vivo. Very small doses of phage (down to 20 pfu), given after the onset of diarrhea, still resulted in an amelioration of the disease. Calves held in a room previously occupied by phage-exposed calves could not any longer be infected with the enteropathogen, coming close to d’Hérelles’s initial idea of “infectious protection” by phages. Also spraying the litter of the calves in the room with a high or low dose of phage (1010 or 106 pfu) prevented an infection of the calves with the pathogenic E. coli strain, applied either before or after transfer to the phage-inoculated room. When substantial pathogen counts were measured in the feces, phage appeared with titers 10 to 100-fold higher than the bacterial counts. Phage survived in the room up to a year and at least 100 days longer than the pathogenic bacteria and was also more resistant to phenolic disinfectants than the enteropathogen (Smith et al., 1987b). Phage was sensitive to a low pH in the abomasum of the calves, but this problem could be solved in applying phage with the milk feed or shortly after feeding. Barrow et al. (Barrow et al., 1998) confirmed that intramuscular phage injection in calves orally challenged with a K1+ E. coli delayed the appearance of the bacterium in the feces and the blood and lengthened the life span of the animals. Smith et al. (Smith and Huggins, 1983) extended their phage therapy experiments to piglets and lambs and confirmed their results also in these animals. Phage use in medicine Escherichia coli diarrhea Five million children die each year as a consequence of acute diarrhea (Snyder and Merson, 1982). E. coli is the cause of a third of childhood diarrhea in developing and threshold

|

171

172

|

Brüssow

countries (Albert et al., 1995) and also the most prominent cause of diarrhea in travelers to developing countries (Black, 1990). E. coli is also prominently associated with diarrhea in pet and farm animals. Shigella, the cause of dysentery, is taxonomically a subspecies of E. coli. The mainstay of diarrhea treatment is oral rehydration solution (Bhan et al., 1994). This inexpensive measure has saved countless lives, but it has no intrinsic anti-bacterial activity. Antibiotics use is of doubtful value against E. coli and vaccines are still in the early development phase (Savarino et al., 2002). Already Félix d’Hérelle, the co-discoverer of phages, has advocated the use of phage for the treatment of bacterial diarrhea (Summers, 1999). American pharmaceutical companies sold phage-based therapy in the 1930s (Duckworth, 1999). During World War II the German and Soviet armies used phages against dysentery and the US army conducted classified research on it (Häusler, 2003). Where to get the tools? E. coli phages are commonly isolated from sewage, hospital waste water, polluted rivers and fecal samples of humans or animals. A surprising morphological diversity of coliphages is isolated from such samples (Ackermann and Nguyen, 1983). Stool from healthy subjects yielded mainly lambda-like Siphoviridae (Dhillon et al., 1976; Furuse et al., 1983), while stools of diarrhea patients gave predominantly T4-like Myoviridae (phages with contractile tails) (Furuse, 1987). However, different phages were also isolated from the same stool samples when using different indicator cells (Chibani-Chennoufi et al., 2004d). The mammalian gut seems to be the natural habitat of T4-like coliphages. Large T4 collections have been compiled (Ackermann and Krisch, 1997) and were investigated for genome evolution (Repoila et al., 1994). Important insights into the mechanism of host cell recognition were gained from the genetic analysis of the tail fibers that recognize outer membrane proteins such as OmpC as well as inner parts of the lipopolysaccharide as cellular receptors. The comparison of the phage adhesins (gp37 in T4, gp38 in T2 phage) revealed a patchwork of shared and unique protein segments. Recombination between short regions of homology led to chimeric fibers and the acquisition of new host range determinants (Tétart et al., 1998). Genetic engineering might thus offer the possibility to extend the host range of a single master T4-like phage. In fact, some coliphages use this strategy naturally: phage Mu inverses the orientation of the receptor interacting gene by a phage-encoded recombinase and another coliphage possessed two different tail fiber proteins and the combined host range of phages containing one or the other protein (Scholl et al., 2001). About 30 different O serotypes must be lysed to cover the majority of EPEC and ETEC strains worldwide (Goodridge et al., 2003; Robins-Browne, 1987). Cocktails of several phages will be necessary to get sufficient breadth and to reduce the probability of resistance developing. Phage virulence factors A number of important human bacterial pathogens owe their virulence factors to prophages integrated into the bacterial genome (for a recent review see (Brussow et al., 2004; Wagner et al., 2002). This is also true for coliphages: in the sequenced E. coli O157 strains prophages encode the major virulence factor, the Shiga-like toxin. Even the “academic” phage lambda

Phage Therapy: the Western Perspective

carries bor, which confers serum resistance to the lysogen (Barondess and Beckwith, 1990). A survey of phage and prophage genomes has revealed that mainly phages of the lambdalike genus of Siphoviridae harbor proven or potential virulence factors (Boyd and Brüssow, 2002; Canchaya et al., 2003). Many other genera of coliphages can establish lysogeny, but only few were actually identified to contain established virulence genes. Nevertheless, to be on the safe side, temperate phages should not be selected for phage therapy. As a priori, candidates for phage therapy should come from the group of “professional virulent” phages. The term “virulent” is in this context misleading since it means for phage biologists obligate lytic phages as opposed to temperate phages. T4-like phages belong to this class and no sequenced member carried known virulence genes (http://phage.bioc.tulane.edu/). These phages do not have integrases; they degrade and recycle the bacterial host genome for its own DNA synthesis and thus lack the molecular basis for a coexistence with the host (Karam, 1994). This property reduces also the possibility of in situ DNA transformation resulting from phage lysis. Sequencing of the phages can exclude undesired genes or known virulence factors. However, even in the case of T4, many genes lack known functions and database matches (Miller et al., 2003). This genetic “terra incognita” is even greater in T4like coliphages (http://phage.bioc.tulane.edu/). Safety issues Several safety issues were raised over the years with respect to the therapeutic use of phages. Extensive recombination clearly goes on among T4-like phages, as one can see by examining relationships at various genes that have been sequenced (Repoila et al., 1994). However, one does not see the same sort of reshuffling of modules (i.e. groups of genes) as described for lambdoid coliphages or dairy phages (Chibani-Chennoufi et al., 2004b). Since no toxin genes have been found in the T4-like phages, the issue of recombinants among T4-like phages is not particularly relevant for phage therapy. Another issue concerns phage gene activity in mammalian cells. For example, a galactose transferase gene incorporated into the phage lambda genome could be expressed as mRNA and translated as protein in human fibroblast cells exposed to the viable phage or phage DNA (Geier and Merril, 1972; Merril et al., 1971). However, there is no indication that the complex series of processes involved in reproducing phage T4 could be carried out in a eukaryotic cell, where the whole machinery and regulatory mechanisms are very different from those in bacteria. There are in general even substantial limitations as to the range of bacteria in which infection with a given phage can occur. Furthermore, geneticists proposed that minute amounts of orally fed phage M13 DNA were taken up by the gut and could even be integrated into the mouse chromosome (Schubbert et al., 1997). This observation was not specific for phage DNA since plasmid DNA had the same fate (Schubbert et al., 1998). It is difficult to comment on these intriguing data since they were not repeated by other groups. With all the phage and bacteria constantly present in our gut, one would think it would have been reported more often if this were a common occurrence. Phage lambda or M13 is normally found in 1 to 4% of stool samples from humans (Schluederberg et al., 1980). About every third stool sample from diarrhea patients yielded a coliphage even though only two indicator strains were being used in this particular study (Chibani-Chennoufi et al., 2004d). This percentage was

|

173

174

|

Brüssow

70% with diarrhea patients and 34% with healthy controls when 10 indicator strains were used (Furuse, 1987). Without knowing we constantly consume phages with our fermented food in yogurt (Brüssow et al., 1994), sauerkraut (Lu et al., 2003b), pickles (Lu et al., 2003a) or salami (Chibani-Chennoufi et al., 2004c). From a clinical standpoint, phages appear to be innocuous. Oral coliphages were given in controlled clinical tests to many thousand children and adults in the former Soviet Union- apparently without major adverse effects. We conducted a small safety trial with adult volunteers who received a total of 108 pfu T4 phage orally. The volunteers did not report adverse events and the levels of two enzymes diagnostic for liver cell damage remained in the normal range (Bruttin and Brüssow, 2005). Industrial phage production Problems of large scale production of dysentery phages were already addressed in a series of papers during the 1940s (Schade and Caroline, 1943; Schade and Caroline, 1944b; Schade and Caroline, 1944a). Lyophilized phages were shown to be superior to liquid preparations. The advantages included greater stability, decrease in bulk size and simpler means of oral administration as pills. Repeated cycles of freezing and thawing were not linked to activity loss, while acidity below pH of 3.5 decreased the phage activity substantially. Out of a large number of substances only egg yolk had some protective properties on the phage preparation. Under dry conditions the phage preparation resisted temperatures at least up to 55°C; at room temperature the lyophilized phages were stable over at least 14 months. Even early crude preparations containing cellular debris of the lysed E. coli did not cause much adverse reactions when applied orally, reflecting the relatively low sensitivity of the human gut to endotoxin. T4-like phages grown on a sequenced strain of K-12 devoid of inducible prophages, virulence genes and the O side chain of LPS (Chibani-Chennoufi et al., 2004d; Chibani-Chennoufi et al., 2004e) maintained their broad host range against pathogenic E. coli, so therapeutic phage production could be carried out without biohazard containment conditions and in cheap media, providing an affordable technology even where the burden of E. coli infection is the highest. Pharmacokinetics of oral phage: can oral phage reach its target cell? When given orally to adult mice and without an antacid, doses as low as 103 pfu T4/ml drinking water led to fecal phage detection. Increasing the oral phage concentration resulted in dose-dependent increases of fecal phages. Phage appeared and disappeared from the feces with a time lag of 1 and 2 days, respectively (Chibani-Chennoufi et al., 2004e). A series of elegant mouse experiments conducted in the 1940s revealed other remarkable pharmacological aspects of Shigella phages (Dubos et al., 1943). These experiments showed that phage can be carried to wherever they are needed-even across the blood-brain barrier, and multiply there, at levels that are far higher than those in blood. When 105 phage was applied intraperitoneally about 102 phage arrived in the brain of control mice. When the experiments were conducted with mice that were intra-cerebrally inoculated with Shigella, a massive increase of phage was observed in the brain to 109 phage after 8 hours indicating amplification of phage in vivo in a tissue that is protected by a tight barrier. When 109 phage was injected intraperitoneally, phage appeared with titers up to 107

Phage Therapy: the Western Perspective

in the blood, but began to fall hours after injection. Decades later it was shown that phages circulating in the blood were sequestered in the spleen, but mutant phages with prolonged circulation times could be easily obtained by serial passages in mice (Merril et al., 1996). If the mice received a Shigella strain that was not susceptible to the phage in vitro, the in vivo phage amplification was not observed (Morton and Perez-Otero, 1945). In further experiments, Morton and Engley (Morton and Engley, 1945) demonstrated that the protective effective of the phage could be diluted, limiting efficiency was reached at a 10:1 bacterium: phage injection dose. The treatment could be delayed for 3 hours after bacterial challenge and phage treatment could precede the bacterial challenge by 4 days and was still death preventing. Heat-inactivated phage had no effect on mice survival under these conditions. The authors described broad reactivity of the phages against Shigella strains, their general observation was a close correspondence between in vitro and in vivo lytic activity. Phages represent a quickly diluted medicine in case of absence of the target bacterium and an amplifiable medicine in the presence of the target pathogen. Phage therapy is thus a unique medicine, which challenges current pharmacokinetic concepts. Two types of phage therapy were distinguished: passive (where the initial phage dose removes the pathogen) and active (where the effect is due to the in vivo replication of the phage on the pathogen). In the latter case phage behaves as a self-amplifying drug, which leads to unfamiliar kinetic phenomena like treatment failure when combined with antibiotic therapy or when given too early and at a too low phage dose (Payne and Jansen, 2003). Some controversy concerns threshold phenomena. In vitro experiments using T4 phage suggested that phage amplification did not occur below a critical threshold of 104 susceptible host cells per ml (Wiggins and Alexander, 1985). The existence of such a threshold was recently disputed (Kasman et al., 2002). The field is currently dominated by the mathematical modeling of phage infections in test tubes, which do not seem to reflect the in vivo situation of phage transit in mice (Chibani-Chennoufi et al., 2004a; Chibani-Chennoufi et al., 2004e) and rats (Weld et al., 2004). Human volunteers showed a very similar fecal phage excretion pattern as mice (Bruttin and Brüssow, 2005). More than 10 per cent of the orally applied phage was recovered from the feces. When put back on phage-free drinking water, fecal phage titers quickly dropped under the threshold of detection when no infective E. coli strain was present in the gut. Host specificity and collateral damage to the commensal biota? Antibiotics kill bacteria rather broadly and can therefore lead to numerous side effects. In contrast, species specificity is the rule for phages and is commonly quoted as one of the major assets of phage therapy. A polyvalent phage refers to a virus that infects many strains within a bacterial species. Many coliphages were reported to also infect other enterobacteria than E. coli. This has been frequently seen in the phages used therapeutically in Eastern Europe and reported in many laboratories through the years. In practical terms, the host-specificity of coliphages is a major limitation for phage therapy necessitating the use of phage cocktails potentially raising problems for the commensal E. coli gut biota, which might suffer from oral T4 phage exposure. However, mice exposed to an oral four-phage cocktail did not experience a decline of their commensal E. coli biota despite the fact that the majority of the intestinal E. coli strains were susceptible

|

175

176

|

Brüssow

to the phage cocktail in vitro (Chibani-Chennoufi et al., 2004e). Likewise human volunteers orally exposed to phage T4 maintained their commensal E. coli population (Bruttin and Brüssow, 2005). Physiological aspects: starving E. coli are not a target Most studies of T4 development were conducted under typical laboratory conditions (Abedon et al., 2003; Hadas et al., 1997). The ribosome number was the most determining factor for the growth of the coliphage, followed by limitations of the transcription and then the DNA synthesis apparatus (You et al., 2002). However, it was noted already 10 years ago that the laboratory growth conditions do not correspond to those prevailing in the mammalian gut, the major habitat of T4 (Kutter et al., 1994). In the colon, E. coli has to grow under anaerobic conditions and it lacks the polysaccharide-degrading enzymes to ferment the polysaccharides available there (Chang et al., 2004). E. coli is nutritionally a bystander of anaerobes like Bacteroides that dominate the human intestinal biota. The relatively low concentration of commensal E. coli in the intestine indicates that the competition for nutrients is high. In fact, E. coli from the intestinal lumen is starving and not dividing (Poulsen et al., 1995). An actively growing commensal E. coli population with coccoid morphology was found in mucus-associated microcolonies overlaying the epithelial cells of the large intestine (Krogfelt et al., 1993). It is not clear how these factors affect the ability of T4 to infect E. coli in vivo. Notably, T4 grows similarly under aerobic and anaerobic laboratory conditions (Kutter et al., 1994). Resistance development: a practical or an academic issue? A classical paper by the founders of phage biology started with the sentences: “When a pure bacterial culture is attacked by a bacterial virus, the culture will clear after a few hours. However, after further incubation for a few hours, or sometimes days, growth of a bacterial variant (is observed) which is resistant to the action of the virus” (Luria and Delbrück, 1943). Was this already the death of phage therapy in the eggshell? In fact, the interaction of bacteria and their bacteriophages became a corner stone in community ecology and evolution. It was extensively investigated with E. coli strain B and the phages of the T series (for a recent review, see Bohannan and Lenski, 2000). In the T4-E. coli B system oscillations were observed in the chemostat community that were induced by the competition between phage-sensitive and phage-resistant bacteria, while evolution of T4 was not observed. In contrast, phage T7 starts a co-evolutionary arms race. However, the race is asymmetrical and favors the bacteria. Also phage T2 can respond with extended host range mutants. Resistance to phage T5 apparently comes without a metabolic cost to E. coli B; consequently, T5 quickly becomes extinct, raising the question how T5 could survive in nature. From a population dynamic perspective, the interaction between phage and bacteria are analogous to those of a predator and a prey. Quite detailed mathematical models were developed for this interaction and the population and evolutionary dynamics relevant for phage therapy was recently reviewed (Levin and Bull, 2004). Serial passages of Pseudomonas fluorescence and its phage in the laboratory showed that phages are not fundamentally constrained in their ability to co-evolve with bacteria. Long-term antagonistic co-evolution characterized by multiple cycles of defense (development of phage-resistant bacteria) and

Phage Therapy: the Western Perspective

counter-defense (phage populations from two transfers in the future showed consistently greater infectivity to bacteria than contemporary phage) were observed (Buckling and Rainey, 2001). Comparable data were now also obtained for E. coli and its phages. Over a 200 hour experiment of T2-like phage PP01 (Morita et al., 2002) with E. coli O157 in continuous culture (Mizoguchi et al., 2003), a series of bacterial mutants was sequentially observed. They differed in colony morphology, the nature of phage receptors OmpC and LPS and phage susceptibility. Phage PP01 evolved by broadening its host range. The system reached a coexistence of phage and bacteria, both at high titer levels, and continued to evolve. The dynamics in these interacting systems were largely determined by the trade-offs between resistance to phage (which is normally metabolically costly) and competitiveness with the parental strain for limiting resources. Some researchers arbitrarily targeted a virulence factor of E. coli like the K1 antigen with a K1-specific phage in vivo. A low number of phage-resistant E. coli strains were isolated from the calf intestine, but due to the loss of the K1 antigen the strain had lost concomitantly its pathological potential in mice (Smith and Huggins, 1982). Indeed, it might well turn out that the in vitro experiments are more of academic than of practical interest since they do not account for the complexity of phage–host interaction in the natural environment. In the defined laboratory system, the only competitor for the phage-resistant cell is its phage-susceptible ancestor cell, which is counter-selected in the presence of a phage. In the wild, the phage resistance clone must also compete with many other strains that do not feel this phage pressure. Already a microcosmos consisting of two strains (E. coli and Salmonella) largely prevented the development of the phage resistance phenotype (Harcombe and Bull, 2005). Mice Smith and Huggins started their famous series of phage therapy experiments in the 1980s, which resumed the tradition of the mouse experiments from the early 1940s, with a K1 E. coli meningitis mouse model (Smith and Huggins, 1982). Low doses of phage, given intramuscularly, protected mice against a massive dose of pathogen applied in the opposite muscle at the same time. The anti-K1 phage was in vivo more efficient than a large number of antibiotics. Multiplication of the phage occurred in the animal and phage was disseminated from the site of inoculation into the blood and the spleen where it was sequestered. However, phage treatment could not be delayed for more than 5 hours after the pathogen challenge without loss of activity. Intramuscular phage also protected against intracerebral pathogen challenge. Only phages recognizing the K1-antigen were protective. Phages with high in vitro lytic activity were also the most effective in conferring protection in vivo. Their results were reproduced recently (Bull et al., 2002; Wang et al., 2006). Efficacy trial in humans In 1963 a total of 30 769 children (6 months to 7 years old) were enrolled in Tbilisi, Georgia, in an oral phage prophylaxis trial against bacterial dysentery (Babalova et al., 1968). About half of the children living on one side of the streets received once every week Shigella phages orally in a pressed tablet form, while children on the other side of the streets received a placebo. Children were followed for 109 days. Phage application was associated with a 3.8-fold decrease in dysentery incidence (1.8 versus 6.7 episodes per 1000 children from

|

177

178

|

Brüssow

treatment and placebo group, respectively). The culture-confirmed incidence was decreased 2.6-fold by phage application (0.7 versus 1.8 episodes, respectively). Phage exposure also decreased the incidence of any form of diarrhea (15 vs. 45 episodes per 1000 children 6 to 12 month-old in treatment and placebo group, respectively). This latter observation suggests a protective effect of the anti-Shigella phage preparation against other serotypes of E. coli. Protective effects were most pronounced in children younger than three years. The most detailed reports on phage therapy published in English are from Poland (Ślopek et al., 1987) dealing with cases of septicemia (only part of them were caused by E. coli). The rate of success was greater than 90%, including patients for whom antibiotic therapy was ineffective. Quite extensive double-blinded phage prophylaxis and treatment trials were conducted in soldiers of the Red Army in four different geographical areas of the Soviet Union during 1982–1983. The authors reported that the incidence of dysentery was 10-fold less in the phage treatment as compared to the control group. These studies were only insufficiently documented in the published Russian literature that a rigorous evaluation of these trials cannot be done (for a review of the Soviet literature see (Alisky et al., 1998; Sulakvelidze et al., 2001; Sulakvelidze and Kutter, 2005). Outlook Until quite recently academic phage researchers remained rather skeptical about the medical or agricultural use of phages. The present literature survey does not paint a negative picture for the prospect of phage therapy against E. coli infections. However, one should refrain from over-interpreting the available clinical evidence and hail phages already as a panacea against bacterial infections in general. Too many clinical data with phages do not correspond to current standards of clinical and microbiological research and need to be repeated. Undue skepticism and unfounded optimism are both misplaced towards the rediscovery of phage therapy. The technology holds at least the promise for practical solutions to at least some urgent public health problems beyond E. coli. The development costs of phage therapy are much lower than for a new antibiotic. In view of the emergence of new infectious diseases at an unpredicted pace and the escape of well-known bacterial diseases from antibiotic control, we are probably well advised to develop in time the necessary phage technology. The chances to develop a successful phage approach to E. coli diarrhea control are reasonably good since it can be based on decades of research with E. coli and its phages. Research on E. coli and its phages were part of the molecular biology revolution. Why should E. coli not lead us also into the future? We could take the best of the reductionist approach (working with the most simple systems) and transfer it deliberately into the complexity of the gut of humans or animals living in their actual ecological context. This would ideally fit into the current trend towards systems biology. Much of the future of phage therapy will be determined by the attitudes of the health authorities, which have to license the use of phages. The secrets of the apparent success in the East with phage therapy were with phage cocktails. Even individualized treatments for each patient were used in surgical settings based on large phage collections and laboratory tests of phage sensitivity for the patient’s specific pathogen. However, “Intestiphage” was a fixed mix of a group of phages against E. coli and other enterobacteria. It was made in large quantity: 80% of the 2 tons of phage-therapy products made in the 1980s were shipped off

Phage Therapy: the Western Perspective

to the Soviet army, where most was used without any particular pre-testing in the individual patients; much of it was prophylactic. Injected phages against antibiotic-resistant Gram-positive bacteria Colonization of the gastrointestinal tract with vancomycin-resistant Enterococcus faecium (VRE) became prevalent in many hospitals and nursing homes. When these organisms are carried via medical devices into the circulation of intensive care patients, antibiotic treatment is difficult, sometimes even impossible. The development of injectable phage became thus an attractive treatment possibility. A collaboration between pioneers of phage therapy from the US National Institutes of Health, Carl Merril and Sankar Adhya, and scientists from the biotech company Exponential Biotherapies explored injected phages in a mouse infection model (Biswas et al., 2002). First, they isolated a Myovirus with a broad host spectrum on VRE clinical isolates showing 60% coverage on E. faecium strains. All control mice, which received VRE (but no phage) were dead after 4 days. Phage, when injected intraperitoneally, protected mice in a dose-dependent way: mice receiving from 109 to 3 s 108 pfu of phage remained 100% healthy; partial survival was seen with phage concentrations down to 3 s 104 pfu. The health status of the mice was strictly correlated with the reduction of the bacterial titer in the blood achieved by the phage replication. When treatment was delayed for 5 hours, mice showed only minimal signs of disease. Even with delays of phage application up to 24 hours after bacterial injection—a time point were mice were already moribund—50% of the mice could still be rescued with phage injection. Survival was only seen with phage preparations that lysed VRE in vitro. In fact, application of a phage that did not lyse the pathogen in vitro even increased mortality in a dose-dependent way. The authors suspected endo- and exo-toxins that contaminated the phage preparation as precipitating factors. CsCl purification of the phage eliminated this disease-enhancing effect of the phage preparation. Heat-inactivation of phage infectivity suppressed the protective effect. After the second injection, mice mounted a serum IgG response against the phage, but no anaphylactic reactions were observed after repeated phage injections. Methicillin-resistant Staphylococcus aureus (MRSA) represent in some countries half of all clinical S. aureus isolates. Researchers from Kochi/Japan induced prophages from clinical MRSA isolates and obtained Siphoviridae that lysed their target cells even in the stationary phase. They observed dose-dependent lifesaving effects of injected phages against experimental staphylococcal infection. Partial protection was observed at an MOI of 0.01 and full protection was achieved with an MOI of 1. Phage treatment could be delayed up to 1 hour after inoculation of the mice with staphylococci, a time point where mice started to show disease symptoms, without compromising protection. The selected Siphovirus FMR11 showed lytic in vivo effects against a range of clinical MRSA isolates (Matsuzaki et al., 2003). Antibiotic treatment leads in some patients to an outgrowth of Clostridium difficile in the intestine, which is difficult to treat. Ramesh et al. (Ramesh et al., 1999) explored the effect of phage directed against C. difficile, which induced ileocecitis in hamsters. The phage was in vitro acid sensitive and no phage was detected in the cecum of the animals when not given orally in a bicarbonate buffer. When applied with a buffer, 108 pfu/ml oral phage achieved 12 hours later a phage titer of 105 pfu/ml in the cecum. This titer fell 10-fold over

|

179

180

|

Brüssow

the next 12 hours to disappear the next day. While all ten C. difficile-challenged hamsters died of diarrhea and showed a hemorrhagic cecum, nine of 10 phage-treated hamsters survived the challenge. The single non-survivor had developed a phage-resistant C. difficile infection during the experiment. Vibriophages Cholera is an interesting case that links the early history of phage therapy with the recent re-discovery of phages as possible ecological drivers of seasonal cholera epidemics. F. d’Herelle conducted in the 1920s the “Bacteriophage Inquiry,” which later became known as the “Cholera Study.” Hospital studies in Calcutta showed dramatic decreases in cholera-associated mortality to 0% in phage-treated patients compared to 27% mortality in untreated patients. Asheshov, who continued the work of d’Herelle, showed tenfold lower cholera incidences in prevention studies. In these studies pilgrims served with water from phage-treated wells were compared to pilgrims served from untreated wells. These positive results encouraged J. Morison, a British military physician, to conduct a field study in Assam. During the pre-intervention period, two villages showed the characteristic seasonal cycling of cholera cases over a three-year observation period. Distribution of a limited amount of anti-cholera phages in the Naogaon village in 1928 had no significant effect on cholera incidence, while the subsequent large scale distribution of phages over a 90 miles zone led to a disappearance of cholera deaths in the treated region, which was sustained over a 5-year period. In the phage-untreated Habiganji region, cholera death tolls cycled during this time period in the usual manner known from the pre-intervention period. These impressive results motivated the government to introduce phages also in Habiganji. This in turn led to dramatic decrease in cholera mortality in Habiganji, while a nearby untreated district continued to show a substantial mortality due to cholera. In 1936 phages became a standard treatment method of cholera in India. However, cholera incidence increased again in 1937 and India experienced a big epidemic in 1944, which discredited the phage approach (for details, see Sulakvelidze and Kutter, 2005; Summers, 1999). It is unclear what led from initial success to final failure of the phage approach in India. In the late 1960s two studies were conducted with anti-cholera phages in Bangladesh. In the first study, 8 cholera patients were treated with massive doses of phages exceeding 1015 pfu per patient (Monsur et al., 1970). In four patients V. cholerae was quickly eliminated from the stools and the total stool output and the duration of diarrhea after treatment was reduced compared to a control group. In four other patients the massive phage dose had no effect on the presence of the pathogen in the stool and clinical effects of phage were not observed. Even in the first four patients, the phage effect was inferior to the impact of tetracycline treatment. The second study was a Russian-Bangladeshi collaboration using phages produced in the Soviet Union. Fifty adult patients were randomized into four groups: oral phage; injected phage plus oral phage; placebo; tetracycline treatment. With respect to stool volume and diarrhea duration after treatment and Vibrio excretion no significant difference was seen between phage treatments and placebo. Only the tetracycline group showed significantly shortened values compared to the controls. In a smaller pediatric group of 10 patients, phage was also less effective than tetracycline treatment. The authors considered the possibility of an incompatibility of the infecting strain (El Tor, classical) with the therapeutic phage. However, no difference was seen between patients treated with the “correct”

Phage Therapy: the Western Perspective

and “incorrect” phages. The authors concluded that phage has no place in the treatment of cholera. Thirty years later an epidemiological study swings the pendulum back by providing ecological data for a model formulated 1931 in an Indian medical journal stating “that the rise and fall of phage in Nature brings the cholera epidemic to a close” and to a new start, respectively. The ecological data were elaborated by a collaboration of the International Diarrheal Diseases Research Center (ICDDR) in Dhaka/Bangladesh and the Harvard University, led by B. Nair and J. Mekalanos, respectively. The researchers studied the characteristic cycling of Vibrio cholerae epidemics in Dhaka between 2001 and 2004. The basic observation was simple. When the researchers studied 221 water samples from the city area they found 114 samples that contained either a vibriophage or an epidemic V. cholerae strain, but only 15 samples that contained both—10 of the vibrios were actually resistant to the phages contained in the same sample (Faruque et al., 2005b). As the ICDDR identifies routinely the causative agent in 2% of the hospitalized patients, the researchers could overlay the phage isolation data on data from the concurrent O1- and O139-cholera epidemics in Dhaka. The peaks of phage isolation from environmental water samples coincided with the troughs in the cholera incidences. The isolated virulent environmental phages plated on O1, O139 or on non-O1, non-O139 environmental strains. One phage was temperate and related prophages were found in epidemic and environmental V. cholerae strains. The time curves in a follow-up study (Faruque et al., 2005a) showed a characteristic cycling as if the accumulation of cholera phages in the environmental water ended the epidemic. During the early phase of the epidemic, phage titers were low in the water. As the epidemic progressed an increasing number of patients excreted phage with their stools. At the end of the epidemic, nearly all stools of patients showed phages, sometimes in high titers. In the following weeks, the phage peak decreased considerably, reflecting the relative instability of the phage in environmental waters. This loss of phage opens the possibility for a renewed build-up of epidemic vibrios in the environment and the start of the next epidemic cycle. Vibrio cholerae is not the only representative of its genus that thrives in tropical and temperate estuarine environments. V. vulnificus is found in filter-feeding shellfish like oysters. Consumption of contaminated oysters can lead in susceptible people to septicemia and fasciitis. In a mouse model, US researchers explored the use of intravenously injected phage (Cerveny et al., 2002). Phage application reduced the build-up of V. vulnificus in the skin lesion and the liver and prevented the drop of the body temperature observed in the untreated animals challenged with the pathogen. The protective effect was not seen with a phage that needed high level of salts found in seawater for in vitro lysis of the pathogen. Phages that did not lyse the pathogen in vitro also lacked in vivo efficacy in the mouse model. Delaying the treatment for 6 hours compromised the therapeutic effect and phage concentrations below 108 pfu failed to afford protection. The phage was relatively stable in blood and showed a half-life of 2 hours. Phage against skin and wound infections Eastern Europe has a long tradition in using phages against wound infection that goes back to the large-scale application of phages on injured soldiers in the Red Army during World War II. A stronghold of this tradition is still today Tbilisi in Georgia as demonstrated

|

181

182

|

Brüssow

by a recent publication in a Western dermatology journal. The Georgian authors, one of them an expert in phage therapy working at the University of Maryland, demonstrated striking effects with a biodegradable film soaked with phages on ulcers refractory to standard treatment (Markoishvili et al., 2002). Seventy per cent of 96 patients showed ulcer healing. Microbiological assessment was done in 22 patients: healing was associated with the elimination of the pathogen from the wound. Impressive as these results are, they still suffer from formal shortcomings that make a rigorous scientific evaluation impossible. For example, no placebo group was used. Furthermore, the film contained not only 106 pfu/cm2 “PyoPhage” (a mixture of Pseudomonas aeruginosa, E. coli, Staphylococcus aureus, Streptococcus and Proteus phages), but also the antibiotic ciprofloxacin. Many patients were in addition treated with a wound-healing drug containing hespiridin as active ingredient. It is thus difficult to differentiate phage effects from that of the other active ingredients. In the West, James Soothill took up phage use against skin infections. This British scientist used skin grafts in a guinea pig model. While untreated grafts took, split grafts inoculated with 106 cfu P. aeruginosa consistently failed. However, when these inoculated grafts were treated with 107 pfu of specific phage, 6 out of 7 grafts took (Soothill, 1994). He extended the experiments to mice injected with three bacterial pathogens that infect the wounds of burned patients. A phage directed against Acinetobacter showed a remarkable lytic activity both in vitro and in vivo: as few as 100 pfu of phage protected mice against 5 lethal doses of Acinetobacter (corresponding to 108 cfu) (Soothill, 1992). Acinetobacter phage showed an impressive 100 000-fold in vivo amplification in mice, which were inoculated with the pathogen. Phage was found in the tissues and the GI tract. Also the Pseudomonas phage showed protective activity, but it needed high phage doses (> 107 pfu) to assure the survival of the mice. In contrast, Staphylococcus phage F-131, which was only weakly lytic in vitro, failed to protect mice against S. aureus challenge. The latter result was surprising since phage F-131 was the major phage used in apparently successful, but uncontrolled Polish phage therapy studies conducted by S. Ślopek in the 1980s. The reason for the use of phage F-131 in the Polish trials was its broad host range on S. aureus strains. Recently, Soothill and colleagues demonstrated efficacy with a different phage isolate in a rabbit model of S. aureus-induced abscesses (Wills et al., 2005). The sewage-derived phage reduced the abscess area and the count of S. aureus in the abscess in a phage dose-dependent way. Over a 100-fold titer difference of inoculated phage, only minimal differences were observed in the wounds with respect to phage titers suggesting that phage had amplified in vivo. However, phage application could not be delayed without loss of protection, even a delay of phage application for 6 hours after pathogen inoculation led to abscess areas and pathogen counts in the wound that were identical to those from S. aureus-infected control animals. A recent publication reported that a Pseudomonas phage rescued mice from drug-resistant P. aeruginosa infection when the phage was applied intraperitoneally at low dose, but within 3 hours after bacterial challenge (Wang et al., 2006). The survival rate of the mice fell from 100 to 50 and 20% when the phage application was delayed for 60, 180 and 360 minutes. Heat-inactivated phage or phage that lacked in vitro lytic activity against the challenge strain lacked protective activity. In the surviving animals, phage appeared in the circulation 2 hours after i.p. injection and remained at elevated titers up to 2 days until bacteria were eradicated.

Phage Therapy: the Western Perspective

Phages against intracellular bacteria Mycobacterium avium, associated with AIDS infections, and M. tuberculosis, the causative agent of tuberculosis, are not obvious targets for phage therapy. Their residence within macrophages and their dormant nature protect them against phage attack. Yet, the need of alternative treatment options for these infections is great since M. avium is resistant to most of the anti-tuberculosis drugs. Thanks to the efforts of Graham Hatfull in Pittsburgh the biology of mycobacteriophages is well developed. One of these phages is TM4, a virulent mycobacteriophage that kills extracellular M. tuberculosis efficiently. However, intracellular M. avium is left unharmed by extracellular TM4 phage. Here a group of US researchers around Luiz Bermudez used a nice trick: they infected a non-pathogenic mycobacterium, M. smegmatis, with TM4 phage and used the infected mycobacterium as a vehicle to reach the intracellular M. avium in the macrophage (Broxmeyer et al., 2002). Here the phage did what the researchers hoped it would do, phage was released from the Trojan horse and could now reach the adjacent pathogenic mycobacterium. Indeed, after 4 days of incubation the intracellular pathogenic bacteria experienced a 100-fold titer decrease, which is quite remarkable since most antimicrobial agents (phages included) need their microbial target in active replication to be protective. G. Hatfull speculated that several mycobacteriophages elaborate a “wake-up” protein that stimulates dormant mycobacteria into replication only to kill them (Pedulla et al., 2003). Killing with a non-replicating phage Even if phages as such do not present a health risk to a healthy subject, the phage might develop indirect toxic side effects when confronting target bacteria in the body of infected patients, especially during intravenous application. Phage replication will eventually lead to the lysis of the bacterial target cell, which then liberates cell wall components. Endotoxin (LPS), released from gram-negative bacterial cells, can elicit a circulatory shock in patients. To address this problem Udo Bläsi’s lab from Vienna modified the filamentous Pseudomonas phage Pf3 for its non-lytic, but lethal job (Hagens et al., 2004). Filamentous phages naturally do not lyse their target cell during progeny phage release. However, this advantage is at the same time a major disadvantage for phage therapy: filamentous phage infections slows the growth of the infected bacterium, but does not kill it. To achieve that goal, the researchers replaced an export protein gene in the phage genome by a restriction endonuclease gene. They could propagate the recombinant phage on a production strain harboring the corresponding methylase gene as a modification enzyme. The infected Pseudomonas cell suspension showed a > 99% decline in cfu, but no decrease in OD suggesting that non-viable cells remained structurally intact. Endotoxin release was reduced 10-fold in comparison with an infection by a virulent Pseudomonas RNA phage. The recombinant, but not the parental phage rescued mice from lethal Pseudomonas infection. The recombinant phages achieved that protection by inducing a significantly lower inflammatory response than the likewise protective RNA phage. Phage lysins The next logical step after a non-replicating phage is to reduce the killing principle to a single phage protein, the phage lysin. These enzymes were known for a long time. Phages

|

183

184

|

Brüssow

need them to get out of the infected cell. Some phages also possess a lytic activity associated with the tail structure, when they drill a small hole into the cell wall during the adsorption phase of phage infection. Actually, virologists knew that the target cell could be killed by an overdose of extracellular phage, a phenomenon known as “lysis from without.” The structure of the phage lysins, consisting of a enzymatic cell wall-lysing domain and a cell-wall recognizing domain conferring bacterial species-specificity to the protein, had fascinated biologists and biochemists alike. Biologists studied the cooperation of the lysin with an export protein (holin) and the precise timing of the expression of the lysis genes during the infection process. However, the obvious step to exploit this phage enzyme for controlling infectious diseases was only done quite recently. This idea came to Vincent Fischetti and his group at the Rockefeller University in New York. Their first target was Streptococcus pyogenes probably because V. Fischetti had isolated the lysin from the streptococcal phage C1 more than 30 years ago (Fischetti et al., 1971). A pure enzyme was prepared by a combination of three column chromatography steps. Ten nanograms of this enzyme sterilized 107 cfu of S. pyogenes within 5 seconds. Morphologically, the cell wall of this pathogen weakened under the enzyme action leading to the extrusion of the cytoplasm and subsequent death of the cell. The enzyme was remarkably specific for group A streptococci, with low activity against group C and D streptococci, but no activity against oral streptococcal commensals. Pretreatment of the oral cavity of mice with this enzyme conferred protection against subsequent colonization with the pathogen. In addition, enzyme given as a treatment to already colonized mice reduced the mortality to 25% compared to 100% in untreated controls (Nelson et al., 2001). The Rockefeller University wrote history of science with the discovery of Avery that DNA is the chemical basis of the genetic material. Avery worked his entire life with pneumococci (Streptococcus pneumoniae), it is thus understandable that these bacteria became the next target of phage lysins. Since pneumococcal phages grow only poorly in broth, the experiments were done with Pal, a phage lysin cloned by the group of Rubens Lopez in Madrid. The enzyme was specific for pneumococci, including highly antibiotic-resistant strains, but did not attack other streptococcal species, except S. oralis and to a lesser extent S. mitis. The phage lysin could eliminate S. pneumoniae from the oropharynx of colonized mice in a dose-dependent way. Notably, no lysin-resistant colonies developed under the treatment protocol (Loeffler et al., 2001). The cell wall of Gram-positive bacteria is a chemically very complex peptidoglycan structure and offers a number of scissable bonds to different lytic enzymes. Pal is an amidase, which cuts the bond to the interpeptide bridge linking two glycan chains (for a recent review see (Loessner, 2005). V. Fischetti’s group subsequently cloned the phage lysin Cpl-1, a muraminidase, which cuts within the glycan chain. Despite the different attack site, Cpl-1 showed the same target specificity as Pal. As pneumococci also cause invasive infection, the Rockefeller researchers explored now intravenous application of the enzyme. A single bolus of 2 mg purified enzyme reduced the pneumococcal blood titer by 3-logs. When applied 1 hour after intravenous challenge with the pathogen, a single lysin injection saved all mice from a lethal infection. When treatment was delayed to 5 hours after challenge, two lysin injections could substantially delay, but not prevent death. After multiple injections, mice developed antibodies against the phage lysin, but they did not neutralize the lytic activity of the enzyme in vivo (Loeffler et al., 2003). In a subsequent study the researchers demonstrated synergistic lethal effects

Phage Therapy: the Western Perspective

when the Pal and Cpl-1 lysins were combined and tested against several serotypes of S. pneumoniae (Loeffler and Fischetti, 2003). Very similar results were reported by R. Lopez’ group with a pneumococcal sepsis model in mice ( Jado et al., 2003). As the mainstay of the treatment of invasive diseases are antibiotics, the Rockefeller scientists teamed up with a clinician from a nearby hospital and tested combinations of the phage lysin with standard antibiotics. In vitro they observed synergistic action between Cpl-1 and gentamycin when pneumococci showed a high susceptibility to penicillin. In contrast, synergistic action between lysin and penicillin was only observed in highly penicillin-resistant strains (Djurkovic et al., 2005). In a rat endocarditis model, a low-concentration continuous infusion of Cpl-1 had only a moderate effect on pneumococcal blood titers and none on pneumococci on the vegetations of the heart valves. A high-concentration lysin infusion eliminated the pathogen from the blood and also reduced the bacteria from the vegetations. However, the latter preparation induced also the release of cytokines into the circulation. Vancomycin induced less cytokine production probably due to the slower release of the pathogen’s cell wall fragments during the treatment (Entenza et al., 2005). In the meanwhile, phage lysin applications were extended to other important pathogens. Recently, a lysin was cloned from a group B streptococcal (GBS) phage that showed a dual catalytic domain combining an endopeptidase and a muraminidase domain. It showed a broad activity against GBS, but also against a few oral streptococcal commensals. However, it showed no lytic activity against lactobacilli and could thus be of value for intrapartum prophylaxis to reduce vaginal GBS colonization in pregnant women. The phage lysin reduced indeed the GBS colonization level when applied to the vagina of mice. The enzyme was less active against oropharynx GBS colonization, which is probably explained by its low pH optimum (Cheng et al., 2005). When the Rockefeller scientists studied a lysin from an enterococcal phage, they were surprised to find that this enzyme did not only kill vancomycin-resistant Enterococcus faecalis in 15 minutes, but also E. faecium, and in addition a broad range of pathogenic streptococci and even some staphylococci, while maintaining a low lytic activity against commensal bacteria (Yoong et al., 2004). Paul Ross’ group from Cork/Ireland isolated lysin LysK from staphylococcal myovirus K that showed a broad inhibitory activity against different species of staphylococci involved in human and veterinary infections, including bovine mastitis. As this enzyme could be expressed in the dairy strain Lactococcus lactis, it might be a cheap source of phage lysin for veterinary application (O’Flaherty et al., 2005). The wide application of phage lysin is not yet exhausted with these examples. V. Fischetti’s group made headlines with a Nature paper wherein they reported that a lysin cloned from a Bacillus anthracis phage specifically kills the target bacterium in vitro (Schuch et al., 2002). When injected intraperitoneally, this enzyme impressively reduced mortality in B. cereus-infected mice. In contrast to antibiotics, even mutagenized B. cereus could not develop resistance to this lysin. As B. anthracis is a dreaded bioterrorist agent, the experiments were conducted for safety reasons with a proxy strain, namely B. cereus, which is genetically closely related to B. anthracis. Remarkably, both vegetative cells and germinating spores were susceptible to the phage lysin. This lysin found also diagnostic use. Spores were germinated on a filter, the lysin was added leading to the release of cytoplasmic ATP, which via the luciferin/luciferase system led to light emission. In this way the Rockefeller group did not

|

185

186

|

Brüssow

only provide a highly appreciated “biological disinfectant” against a top bioterrorism agent, they also showed how fundamental phage research can lead to numerous applications addressing urgent questions confronting our societies like the antibiotic crisis. However, like in the case of replication-competent phages, phage lysins still have to demonstrate their use in carefully controlled clinical trials to become a real therapeutic adjunct to antibiotics in clinical practice. Acknowledgment I thank Anne Bruttin for help with the literature references. References

Abedon, S.T., Hyman, P., and Thomas, C. (2003). Experimental examination of bacteriophage latentperiod evolution as a response to bacterial availability. Appl. Environ. Microbiol. 69, 7499–7506. Ackermann, H.W., and Krisch, H.M. (1997). A catalogue of T4-type bacteriophages. Arch. Virol. 142, 2329–2345. Ackermann, H.W., and nguyen, T.M. (1983). Sewage coliphages studied by electron microscopy. Appl. Environ. Microbiol. 45, 1049–1059. Albert, M.J., Faruque, S.M., Faruque, A.S., Neogi, P.K., Ansaruzzaman, M., Bhuiyan, N.A., Alam, K., and Akbar, M.S. (1995). Controlled study of Escherichia coli diarrheal infections in Bangladeshi children. J. Clin. Microbiol. 33, 973–977. Alisky, J., Iczkowski, K., Rapoport, A., and Troitsky, N. (1998). Bacteriophages show promise as antimicrobial agents. J. Infect. 36, 5–15. Atterbury, R.J., Connerton, P.L., Dodd, C.E., Rees, C.E., and Connerton, I.F. (2003a). Application of hostspecific bacteriophages to the surface of chicken skin leads to a reduction in recovery of Campylobacter jejuni. Appl. Environ. Microbiol. 69, 6302–6306. Atterbury, R.J., Connerton, P.L., Dodd, C.E., Rees, C.E., and Connerton, I.F. (2003b). Isolation and characterization of Campylobacter bacteriophages from retail poultry. Appl. Environ. Microbiol. 69, 4511–4518. Babalova, E.G. Katsitadze, K.T., Sakvarelidze, L.A., Imnaishvili, N.S., Sharashidze, T.G., Badashvili, V.A., Kiknadze, G.P., Meipariani, A.N., Gendzekhadze, N.D., Machavariani, E.V., Gogoberidze, K.L., Gozalov, E.I., and Dekanosidze, N.G. (1968). [Preventive value of dried dysentery bacteriophage]. Zh. Mikrobiol. Epidemiol. Immunobiol. 45, 143–145. Barondess, J.J., and Beckwith, J. (1990). A bacterial virulence determinant encoded by lysogenic coliphage lambda. Nature 346, 871–874. Barrow, P., Lovell, M., and Berchieri, A., Jr. (1998). Use of lytic bacteriophage for control of experimental Escherichia coli septicemia and meningitis in chickens and calves. Clin. Diagn. Lab. Immunol. 5, 294–298. Berchieri, A., Jr., Lovell, M.A., and Barrow, P.A. (1991). The activity in the chicken alimentary tract of bacteriophages lytic for Salmonella typhimurium. Res. Microbiol. 142, 541–549. Bhan, M.K., Mahalanabis, D., Fontaine, O., and Pierce, N.F. (1994). Clinical trials of improved oral rehydration salt formulations: a review. Bull. World Health Organ. 72, 945–955. Biswas, B., Adhya, S., Washart, P., Paul, B., Trostel, A.N., Powell, B., Carlton, R., and Merril, C.R. (2002). Bacteriophage therapy rescues mice bacteremic from a clinical isolate of vancomycin-resistant Enterococcus faecium. Infect. Immun. 70, 204–210. Black, R.E. (1990). Epidemiology of travelers’ diarrhea and relative importance of various pathogens. Rev. Infect. Dis. 12 Suppl 1:S73–9., S73-S79. Bohannan, B.J.M., and Lenski, R.E. (2000). Linking genetic change to community evolution: insights from studies of bacteria and bacteriophages involved. Ecol. Lett. 3, 362–377. Boyd, E.F., and Brüssow, H. (2002). Common themes among bacteriophage-encoded virulence factors and diversity among the bacteriophages involved. Trends Microbiol. 10, 521–529. Broxmeyer, L., Sosnowska, D., Miltner, E., Chacon, O., Wagner, D., McGarvey, J., Barletta, R.G., and Bermudez, L.E. (2002). Killing of Mycobacterium avium and Mycobacterium tuberculosis by a my-

Phage Therapy: the Western Perspective

cobacteriophage delivered by a nonvirulent mycobacterium: a model for phage therapy of intracellular bacterial pathogens. J. Infect. Dis. 186, 1155–1160. Brüssow, H., Canchaya, C., and Hardt, W.D. (2004). Phages and the evolution of bacterial pathogens: from genomic rearrangements to lysogenic conversion. Microbiol. Mol. Biol. Rev. 68, 560–602, table. Brüssow, H., Fremont, M., Bruttin, A., Sidoti, J., Constable, A., and Fryder, V. (1994). Detection and classification of Streptococcus thermophilus bacteriophages isolated from industrial milk fermentation. Appl. Environ. Microbiol. 60, 4537–4543. Bruttin, A., and Brüssow, H. (2005). Human volunteers receiving Escherichia coli phage T4 orally: a safety test of phage therapy. Antimicrob. Agents Chemother. 49, 2874–2878. Buckling, A., and Rainey, P.B. (2001). Antagonistic coevolution between a bacterium and a bacteriophage. Proc. R. Soc. Lond. B. 269, 931–936. Bull, J.J., Levin, B.R., DeRouin, T., Walker, N., and Bloch, C.A. (2002). Dynamics of success and failure in phage and antibiotic therapy in experimental infections. BMC. Microbiol. 2, 35. Canchaya, C., Proux, C., Fournous, G., Bruttin, A., and Brüssow, H. (2003). Prophage genomics. Microbiol. Mol. Biol. Rev. 67, 238–76. Cerveny, K.E., DePaola, A., Duckworth, D.H., and Gulig, P.A. (2002). Phage therapy of local and systemic disease caused by Vibrio vulnificus in iron-dextran-treated mice. Infect. Immun. 70, 6251–6262. Chang, D.E., Smalley, D.J., Tucker, D.L., Leatham, M.P., Norris, W.E., Stevenson, S.J., Anderson, A.B., Grissom, J.E., Laux, D.C., Cohen, P.S., and Conway, T. (2004). Carbon nutrition of Escherichia coli in the mouse intestine. Proc. Natl. Acad. Sci. USA 101, 7427–7432. Cheng, Q., Nelson, D., Zhu, S., and Fischetti, V.A. (2005). Removal of group B streptococci colonizing the vagina and oropharynx of mice with a bacteriophage lytic enzyme. Antimicrob. Agents Chemother. 49, 111–117. Chibani-Chennoufi, S., Bruttin, A., Dillmann, M.L., and Brüssow, H. (2004a). Phage–host interaction: an ecological perspective. J. Bacteriol. 186, 3677–3686. Chibani-Chennoufi, S., Canchaya, C., Bruttin, A., and Brüssow, H. (2004b). Comparative genomics of the T4-Like Escherichia coli phage JS98: implications for the evolution of T4 phages. J. Bacteriol. 186, 8276–8286. Chibani-Chennoufi, S., Dillmann, M.L., Marvin-Guy, L., Rami-Shojaei, S., and Brüssow, H. (2004c). Lactobacillus plantarum bacteriophage LP65: a new member of the SPO1-like genus of the family Myoviridae. J. Bacteriol. 186, 7069–7083. Chibani-Chennoufi, S., Sidoti, J., Bruttin, A., Dillmann, M.-L., Kutter, E., Qadri, F., Sarker, S.A., and Brüssow, H. (2004d). Isolation of T4-like Bateriophages from the Stool of Pediatric Patients in Bangladesh. J. Bacteriol. J. Bacteriol. 186, 8287–8294. Chibani-Chennoufi, S., Sidoti, J., Bruttin, A., Kutter, E., Sarker, S., and Brüssow, H. (2004e). In vitro and in vivo bacteriolytic activities of Escherichia coli phages: implications for phage therapy. Antimicrob. Agents Chemother. 48, 2558–2569. Civerolo, E.L. (1973). Relationships of Xanthomonas pruni Bacteriophages to bacterial Spot Disease in Prunus. Phytopathology 63, 1279–1284. Civerolo, E.L., and Keil, H.L. (1969). Inhibition of Bacterial Spot of Peach Foliage by Xanthomonas pruni Bacteriophage. Phytopathological Notes 1966–1967. Connerton, P.L., Loc Carrillo, C.M., Swift, C., Dillon, E., Scott, A., Rees, C.E., Dodd, C.E., Frost, J., and Connerton, I.F. (2004). Longitudinal study of Campylobacter jejuni bacteriophages and their hosts from broiler chickens. Appl. Environ. Microbiol. 70, 3877–3883. Dhillon, T.S., Dhillon, E.K., Chau, H.C., Li, W.K., and Tsang, A.H. (1976). Studies on bacteriophage distribution: virulent and temperate bacteriophage content of mammalian feces. Appl. Environ. Microbiol. 32, 68–74. Djurkovic, S., Loeffler, J.M., and Fischetti, V.A. (2005). Synergistic killing of Streptococcus pneumoniae with the bacteriophage lytic enzyme Cpl-1 and penicillin or gentamicin depends on the level of penicillin resistance. Antimicrob. Agents Chemother. 49, 1225–1228. Dubos, R.J., Hookey Straus, J., and Pierce, C. (1943). The multiplication of bacteriophage in vivo and its protective effect against an experimental indfection with Shigella Dysenteriae. J. Exp. Med. 78, 161–168. Duckworth, D.H. (1999). History of virology: Bacteriophages. In: Encyclopedia of Virology, A. Granoff and R.G. Webster, eds. (Memphis, USA: Academic Press), pp. 725–730.

|

187

188

|

Brüssow

Entenza, J.M., Loeffler, J.M., Grandgirard, D., Fischetti, V.A., and Moreillon, P. (2005). Therapeutic effects of bacteriophage Cpl-1 lysin against Streptococcus pneumoniae endocarditis in rats. Antimicrob. Agents Chemother. 49, 4789–4792. Erskine, J.M. (1973). Characteristics of Erwinia amylovora bacteriophage and its possible role in the epidemology of fire blight. Can. J. Microbiol. 19, 837–845. Faruque, S.M., Islam, M.J., Ahmad, Q.S., Faruque, A.S., Sack, D.A., Nair, G.B., and Mekalanos, J.J. (2005a). Self-limiting nature of seasonal cholera epidemics: Role of host-mediated amplification of phage. Proc. Natl. Acad. Sci. USA 102, 6119–6124. Faruque, S.M., Naser, I.B., Islam, M.J., Faruque, A.S., Ghosh, A.N., Nair, G.B., Sack, D.A., and Mekalanos, J.J. (2005b). Seasonal epidemics of cholera inversely correlate with the prevalence of environmental cholera phages. Proc. Natl. Acad. Sci. USA 102, 1702–1707. Fischer, C.R., Yoichi, M., Unno, H., and Tanji, Y. (2004). The coexistence of Escherichia coli serotype O157:H7 and its specific bacteriophage in continuous culture. FEMS Microbiol. Lett. 241, 171–177. Fischetti, V.A., Gotschlich, E.C., and Bernheimer, A.W. (1971). Purification and physical properties of group C streptococcal phage-associated lysin. J. Exp. Med. 133, 1105–1117. Flaherty, J.E., Harbaugh, B.K., Jones, J.B., and Somodi, G.C. (2001). H-mutant bacteriophages as a potential biocontrol of bacterial blight of geranium. Hort. Science 36, 98–100. Flaherty, J.E., Jones, J.B., Harbaugh, B.K., Somodi, G.C., and Jackson, L.E. (2000). Control of bacterial spot on tomato in the greenhouse and field with H-mutant bacteriophages. Hort. Science 35, 882–884. Furuse, K. (1987). Distribution of coliphages in the general environment: general considerations. In: Phage Ecology, Goyal S.M., Gerba C.P., and Bitton G., eds. John Wiley and sons), pp. 87–124. Furuse, K., Osawa, S., Kawashiro, J., Tanaka, R., Ozawa, A., Sawamura, S., Yanagawa, Y., Nagao, T., and Watanabe, I. (1983). Bacteriophage distribution in human faeces: continuous survey of healthy subjects and patients with internal and leukaemic diseases. J. Gen. Virol. 64 (Pt 9), 2039–2043. Geier, M.R., and Merril, C.R. (1972). Lambda phage transcription in human fibroblasts. Virology 47, 638–643. Gill, J.J., Svircev, A.M., Smith, R., and Castle, A.J. (2003). Bacteriophages of Erwinia amylovora. Appl. Environ. Microbiol. 69, 2133–2138. Goode, D., Allen, V.M., and Barrow, P.A. (2003). Reduction of experimental Salmonella and Campylobacter contamination of chicken skin by application of lytic bacteriophages. Appl. Environ. Microbiol. 69, 5032–5036. Goodridge, L., Gallaccio, A., and Griffiths, M.W. (2003). Morphological, host range, and genetic characterization of two coliphages. Appl. Environ. Microbiol. 69, 5364–5371. Goodridge, L.D. (2004). Bacteriophage biocontrol of plant pathogens: fact or fiction? Trends Biotechnol. 22, 384–385. Greer, G.G. (1983). Psychrotrophic Brocothrix thermosphacta bacteriophages isolated from beef. Appl. Environ. Microbiol. 46, 245–251. Greer, G.G. (2005). Bacteriophage control of foodborne bacteriat. J. Food Prot. 68, 1102–1111. Greer, G.G., and Dilts, B.D. (1990). Inability of a bacteriophage pool to control beef spoilage. Int. J. Food Microbiol. 10, 331–342. Greer, G.G., and Dilts, B.D. (2002). Control of Brochothrix thermosphacta spoilage of pork adipose tissue using bacteriophages. J. Food Prot. 65, 861–863. Hadas, H., Einav, M., Fishov, I., and Zaritsky, A. (1997). Bacteriophage T4 development depends on the physiology of its host Escherichia coli. Microbiology 143 (Pt 1), 179–185. Hagens, S., Habel, A., von Ahsen, U., von Gabain, A., and Blasi, U. (2004). Therapy of experimental pseudomonas infections with a nonreplicating genetically modified phage. Antimicrob. Agents Chemother. 48, 3817–3822. Harcombe, W.R. and Bull, J.J. (2005). Impact of phages on two-species bacterial communities. Appl. Environ. Microbiol. 71, 5254–5259. Häusler, T. (2003). Gesund durch Viren. (Piper Verlag: München). Higgins, J.P., Higgins, S.E., Guenther, K.L., Huff, W., Donoghue, A.M., Donoghue, D.J., and Hargis, B.M. (2005). Use of a specific bacteriophage treatment to reduce Salmonella in poultry products. Poult. Sci. 84, 1141–1145. Huff, W.E., Huff, G.R., Rath, N.C., Balog, J.M., and Donoghue, A.M. (2002). Prevention of Escherichia coli infection in broiler chickens with a bacteriophage aerosol spray. Poult. Sci. 81, 1486–1491.

Phage Therapy: the Western Perspective

Huff, W.E., Huff, G.R., Rath, N.C., Balog, J.M., and Donoghue, A.M. (2003). Bacteriophage treatment of a severe Escherichia coli respiratory infection in broiler chickens. Avian Dis. 47, 1399–1405. Huff, W.E., Huff, G.R., Rath, N.C., Balog, J.M., and Donoghue, A.M. (2004). Therapeutic efficacy of bacteriophage and Baytril (enrofloxacin) individually and in combination to treat colibacillosis in broilers. Poult. Sci. 83, 1944–1947. Jado, I., Lopez, R., Garcia, E., Fenoll, A., Casal, J., and Garcia, P. (2003). Phage lytic enzymes as therapy for antibiotic-resistant Streptococcus pneumoniae infection in a murine sepsis model. J. Antimicrob. Chemother. 52, 967–973. Karam, J.D. (1994). Molecular Biology of Bacteriophage T4. (Washington DC: ASM Press). Kasman, L.M., Kasman, A., Westwater, C., Dolan, J., Schmidt, M.G., and Norris, J.S. (2002). Overcoming the phage replication threshold: a mathematical model with implications for phage therapy. J. Virol. 76, 5557–5564. Keel, C., Ucurum, Z., Michaux, P., Adrian, M., and Haas, D. (2002). Deleterious impact of a virulent bacteriophage on survival and biocontrol activity of Pseudomonas fluorescens strain CHAO in natural soil. Mol. Plant Microbe Interact. 15, 567–576. Krogfelt, K.A., Poulsen, L.K., and Molin, S. (1993). Identification of coccoid Escherichia coli BJ4 cells in the large intestine of streptomycin-treated mice. Infect. Immun. 61, 5029–5034. Kudva, I.T., Jelacic, S., Tarr, P.I., Youderian, P., and Hovde, C.J. (1999). Biocontrol of Escherichia coli O157 with O157-specific bacteriophages. Appl. Environ. Microbiol. 65, 3767–3773. Kutter, E., Kellenberger, E., Carlson, K., Eddy, S., Neitzel, J., Messinger, L., North, J., and Guttman, B. (1994). Effects of bacterial growth conditions and physiology on T4 infection. In: Molecular Biology of Bacteriophage T4, ed. J.D. Karam, ed. (Washington, DC: ASM Press), pp. 406–418. Kutter, E., and Sulakvelidze, A. (2005). Bacteriophages, Biology and Applications., E. Kutter and A. Sulakvelidze, eds. (Boca Raton: CRC Press). Leverentz, B., Conway, W.S., Alavidze, Z., Janisiewicz, W.J., Fuchs, Y., Camp, M.J., Chighladze, E., and Sulakvelidze, A. (2001). Examination of bacteriophage as a biocontrol method for salmonella on freshcut fruit: a model study. J. Food Prot. 64, 1116–1121. Leverentz, B., Conway, W.S., Camp, M.J., Janisiewicz, W.J., Abuladze, T., Yang, M., Saftner, R., and Sulakvelidze, A. (2003). Biocontrol of Listeria monocytogenes on fresh-cut produce by treatment with lytic bacteriophages and a bacteriocin. Appl. Environ. Microbiol. 69, 4519–4526. Levin, B.R., and Bull, J.J. (2004). Population and evolutionary dynamics of phage therapy. Nat. Rev. Microbiol. 2, 166–173. Levin, B.R., Stewart, F., and Chao, L. (1977). Resource-limited growth, competition, and predation: a model and experimental studies with bacteria and bacteriophage. Am. Nat. 111, 3–24. Loc, C.C., Atterbury, R.J., el Shibiny, A., Connerton, P.L., Dillon, E., Scott, A., and Connerton, I.F. (2005). Bacteriophage therapy to reduce Campylobacter jejuni colonization of broiler chickens. Appl. Environ. Microbiol. 71, 6554–6563. Loeffler, J.M., Djurkovic, S., and Fischetti, V.A. (2003). Phage lytic enzyme Cpl-1 as a novel antimicrobial for pneumococcal bacteremia. Infect. Immun. 71, 6199–6204. Loeffler, J.M., and Fischetti, V.A. (2003). Synergistic lethal effect of a combination of phage lytic enzymes with different activities on penicillin-sensitive and -resistant Streptococcus pneumoniae strains. Antimicrob. Agents Chemother. 47, 375–377. Loeffler, J.M., Nelson, D., and Fischetti, V.A. (2001). Rapid killing of Streptococcus pneumoniae with a bacteriophage cell wall hydrolase. Science 294, 2170–2172. Loessner, M.J. (2005). Bacteriophage endolysins—current state of research and applications. Curr. Opin. Microbiol. 8, 480–487. Lu, Z., Breidt, F., Jr., Fleming, H.P., Altermann, E., and Klaenhammer, T.R. (2003a). Isolation and characterization of a Lactobacillus plantarum bacteriophage, phiJL-1, from a cucumber fermentation. Int. J. Food Microbiol. 84, 225–235. Lu, Z., Breidt, F., Plengvidhya, V., and Fleming, H.P. (2003b). Bacteriophage ecology in commercial sauerkraut fermentations. Appl. Environ. Microbiol. 69, 3192–3202. Luria, S.E., and Delbrück, M. (1943). Mutations of bacteria from virus sensitivity to virus resistance. Genetics 28, 491–511. Markoishvili, K., Tsitlanadze, G., Katsarava, R., Morris, J.G., Jr., and Sulakvelidze, A. (2002). A novel sustained-release matrix based on biodegradable poly(ester amide)s and impregnated with bacteriophages

|

189

190

|

Brüssow

and an antibiotic shows promise in management of infected venous stasis ulcers and other poorly healing wounds. Int. J. Dermatol. 41, 453–458. Matsuzaki, S., Yasuda, M., Nishikawa, H., Kuroda, M., Ujihara, T., Shuin, T., Shen, Y., Jin, Z., Fujimoto, S., Nasimuzzaman, M.D., Wakiguchi, H., Sugihara, S., Sugiura, T., Koda, S., Muraoka, A., and Imai, S. (2003). Experimental protection of mice against lethal Staphylococcus aureus infection by novel bacteriophage phi MR11. J. Infect. Dis. 187, 613–624. McKenna, F., El-Tarabily, K.A., Hardy, G.E.ST.J., and Dell, B. (2001). Novel in vivo use of polyvalent Streptomyces phage to disinfest Streptomyces scabies-infected seed potatoes. Plant Pathol. 50, 666–675. Merril, C.R., Biswas, B., Carlton, R., Jensen, N.C., Creed, G.J., Zullo, S., and Adhya, S. (1996). Longcirculating bacteriophage as antibacterial agents. Proc. Natl. Acad. Sci. USA 93, 3188–3192. Merril, C.R., Geier, M.R., and Petricciani, J.C. (1971). Bacterial virus gene expression in human cells. Nature 233, 398–400. Merril, C.R., Scholl, D., and Adhya, S.L. (2003). The prospect for bacteriophage therapy in Western medicine. Nat. Rev. Drug Discov. 2, 489–497. Miller, E.S., Kutter, E., Mosig, G., Arisaka, F., Kunisawa, T., and Ruger, W. (2003). Bacteriophage T4 genome. Microbiol. Mol. Biol. Rev. 67, 86–156, table. Mizoguchi, K., Morita, M., Fischer, C.R., Yoichi, M., Tanji, Y., and Unno, H. (2003). Coevolution of bacteriophage PP01 and Escherichia coli O157:H7 in continuous culture. Appl. Environ. Microbiol. 69, 170–176. Modi, R., Hirvi, Y., Hill, A., and Griffiths, M.W. (2001). Effect of phage on survival of Salmonella enteritidis during manufacture and storage of cheddar cheese made from raw and pasteurized milk. J. Food Prot. 64, 927–933. Monsur, K.A., Rahman, M.A., Huq, F., Islam, M.N., Northrup, R.S., and Hirschhorn, N. (1970). Effect of massive doses of bacteriophage on excretion of vibrios, duration of diarrhoea and output of stools in acute cases of cholera. Bull. World Health Organ. 42, 723–732. Morita, M., Tanji, Y., Mizoguchi, K., Akitsu, T., Kijima, N., and Unno, H. (2002). Characterization of a virulent bacteriophage specific for Escherichia coli O157:H7 and analysis of its cellular receptor and two tail fiber genes. FEMS Microbiol. Lett. 211, 77–83. Morton, H.E., and Engley, F.B. J. (1945). The protective action of dysentery bacteriophage in experimental infections in mice. J. Bacteriol. 49, 245–255. Morton, H.E., and Perez-Otero, J.E. (1945). The increase of bacteriophage in vivo during experimental infections with Shigella paradysenteriae, flexner, in mice. J. Bacteriol. 49, 237–244. Nakai, T., and Park, S.C. (2002). Bacteriophage therapy of infectious diseases in aquaculture. Res. Microbiol. 153, 13–18. Nakai, T., Sugimoto, R., Park, K.H., Matsuoka, S., Mori, K., Nishioka, T., and Maruyama, K. (1999). Protective effects of bacteriophage on experimental Lactococcus garvieae infection in yellowtail. Dis. Aquat. Organ 37, 33–41. Nelson, D., Loomis, L., and Fischetti, V.A. (2001). Prevention and elimination of upper respiratory colonization of mice by group A streptococci by using a bacteriophage lytic enzyme. Proc. Natl. Acad. Sci. USA 98, 4107–4112. O’Flaherty, S., Coffey, A., Meaney, W., Fitzgerald, G.F., and Ross, R.P. (2005). The recombinant phage lysin LysK has a broad spectrum of lytic activity against clinically relevant staphylococci, including methicillin-resistant Staphylococcus aureus. J. Bacteriol. 187, 7161–7164. O’Flynn, G., Ross, R.P., Fitzgerald, G.F., and Coffey, A. (2004). Evaluation of a cocktail of three bacteriophages for biocontrol of Escherichia coli O157:H7. Appl. Environ. Microbiol. 70, 3417–3424. Park, S.C., and Nakai, T. (2003). Bacteriophage control of Pseudomonas plecoglossicida infection in ayu Plecoglossus altivelis. Dis. Aquat. Organ 53, 33–39. Park, S.C., Shimamura, I., Fukunaga, M., Mori, K.I., and Nakai, T. (2000). Isolation of bacteriophages specific to a fish pathogen, Pseudomonas plecoglossicida, as a candidate for disease control. Appl. Environ. Microbiol. 66, 1416–1422. Payne, R.J., and Jansen, V.A. (2003). Pharmacokinetic principles of bacteriophage therapy. Clin. Pharmacokinet. 42, 315–325. Pedulla, M.L., Ford, M.E., Houtz, J.M., Karthikeyan, T., Wadsworth, C., Lewis, J.A., Jacobs-Sera, D., Falbo, J., Gross, J., Pannunzio, N.R., Brucker, W., Kumar, V., Kandasamy, J., Keenan, L., Bardarov, S., Kriakov, J., Lawrence, J.G., Jacobs, W.R., Jr., Hendrix, R.W., and Hatfull, G.F. (2003). Origins of highly mosaic mycobacteriophage genomes. Cell 113, 171–182.

Phage Therapy: the Western Perspective

Poulsen, L.K., Licht, T.R., Rang, C., Krogfelt, K.A., and Molin, S. (1995). Physiological state of Escherichia coli BJ4 growing in the large intestines of streptomycin-treated mice. J. Bacteriol. 177, 5840–5845. Ramesh, V., Fralick, J.A., and Rolfe, R.D. (1999). Prevention of Clostridium difficile-induced ileocecitis with bacteriophage. Anaerobe 5, 69–78. Raya, R.R., Varey, P., Oot, R.A., Dyen, M.R., Callaway, T.R., Edrington, T.S., Kutter, E.M., and Brabban, A.D. (2006). Isolation and characterization of a new T-even bacteriophage, CEV1, and determination of its potential to reduce Escherichia coli O157:H7 levels in sheep. Appl. Environ. Microbiol. 72, 6405–6410. Repoila, F., Tétart, F., Bouet, J.Y., and Krisch, H.M. (1994). Genomic polymorphism in the T-even bacteriophages. EMBO J. 13, 4181–4192. Ritchie, D.F., and Klos, E.J. (1979). Some properties of Erwinia amylovora bacteriophages. Phytopathology 69, 1078–1083. Robins-Browne, R.M. (1987). Traditional enteropathogenic Escherichia coli of infantile diarrhea. Rev. Infect. Dis. 9, 28–53. Savarino, S.J., Hall, E.R., Bassily, S., Wierzba, T.F., Youssef, F.G., Peruski, L.F., Jr., Abu-Elyazeed, R., Rao, M., Francis, W.M., El Mohamady, H., Safwat, M., Naficy, A.B., Svennerholm, A.M., Jertborn, M., Lee, Y.J., and Clemens, J.D. (2002). Introductory evaluation of an oral, killed whole cell enterotoxigenic Escherichia coli plus cholera toxin B subunit vaccine in Egyptian infants. Pediatr. Infect. Dis. J. 21, 322–330. Schade, A.L., and Caroline, L. (1943). The preparation of a polyvalent dysentery bacteriophage in a dry and stable form. I. Preliminary investigations and general procedures. J. Bacteriol. 46, 463–473. Schade, A.L., and Caroline, L. (1944a). The preparation of a polyvalent dysentery bacteriophage in a dry and stable form. II. Factors affecting the stabilization of dysentery bacteriophage during lyophilization. J. Bacteriol. 48, 179–190. Schade, A.L., and Caroline, L. (1944b). The preparation of a polyvalent dysentery bacteriophage in a dry and stable form. III. Stability of the dried bacteriophage towards heat, humidity, age and acidity. J. Bacteriol. 48, 243–251. Schluederberg, S.A., Marshall, B., Tachibana, C., and Levy, S.B. (1980). Recovery frequency of phages lambda and M13 from human and animal faeces. Nature 283, 792–794. Schnabel, E.L., Fernando, W.G. D., Meyer, M.P., Jones, A.L., and Jackson, L.E. Bacteriophage of Erwinia amylovora and their Potential for Biocontrol. [Eds. M.T.Momol, H. Saygili], 649–653. 1999. ISHS. Ref Type: Conference Proceeding Schnabel, E.L., and Jones, A.L. (2001). Isolation and characterization of five Erwinia amylovora bacteriophages and assessment of phage resistance in strains of Erwinia amylovora. Appl. Environ. Microbiol. 67, 59–64. Scholl, D., Rogers, S., Adhya, S., and Merril, C.R. (2001). Bacteriophage K1–5 encodes two different tail fiber proteins, allowing it to infect and replicate on both K1 and K5 strains of Escherichia coli. J. Virol. 75, 2509–2515. Schrag, S.J., and Mittler, J.E. (1996). Host-Parasite Coexistence: The role of spatial refuges in stabilizing bacteria-phage interactions. Am. Natur. 148, 348–377. Schubbert, R., Hohlweg, U., Renz, D., and Doerfler, W. (1998). On the fate of orally ingested foreign DNA in mice: chromosomal association and placental transmission to the fetus. Mol. Gen. Genet. 259, 569–576. Schubbert, R., Renz, D., Schmitz, B., and Doerfler, W. (1997). Foreign (M13) DNA ingested by mice reaches peripheral leukocytes, spleen, and liver via the intestinal wall mucosa and can be covalently linked to mouse DNA. Proc. Natl. Acad. Sci. USA 94, 961–966. Schuch, R., Nelson, D., and Fischetti, V.A. (2002). A bacteriolytic agent that detects and kills Bacillus anthracis. Nature 418, 884–889. Ślopek, S., Weber-Dąbrowska, B., Dąbrowski, M., and Kucharewicz-Krukowska, A. (1987). Results of bacteriophage treatment of suppurative bacterial infections in the years 1981–1986. Arch. Immunol. Ther. Exp. (Warsz.) 35, 569–583. Smith, H.W., and Huggins, M.B. (1982). Successful treatment of experimental Escherichia coli infections in mice using phage: its general superiority over antibiotics. J. Gen. Microbiol. 128, 307–318. Smith, H.W., and Huggins, M.B. (1983). Effectiveness of phages in treating experimental Escherichia coli diarrhoea in calves, piglets and lambs. J. Gen. Microbiol. 129, 2659–2675.

|

191

192

|

Brüssow

Smith, H.W., Huggins, M.B., and Shaw, K.M. (1987a). Factors influencing the survival and multiplication of bacteriophages in calves and in their environment. J. Gen. Microbiol. 133, 1127–1135. Smith, H.W., Huggins, M.B., and Shaw, K.M. (1987b). The control of experimental Escherichia coli diarrhoea in calves by means of bacteriophages. J. Gen. Microbiol. 133, 1111–1126. Snyder, J.D., and Merson, M.H. (1982). The magnitude of the global problem of acute diarrhoeal disease: a review of active surveillance data. Bull. World Health Organ. 60, 605–613. Soothill, J.S. (1992). Treatment of experimental infections of mice with bacteriophages. J. Med. Microbiol. 37, 258–261. Soothill, J.S. (1994). Bacteriophage prevents destruction of skin grafts by Pseudomonas aeruginosa. Burns 20, 209–211. Sulakvelidze, A., Alavidze, Z., and Morris, J.G., Jr. (2001). Bacteriophage therapy. Antimicrob. Agents Chemother. 45, 649–659. Sulakvelidze, A., and Kutter, E. (2005). Bacteriophage therapy in humans. In: Bacteriophages: Biology and Applications, e.E. Kutter and A. Sulakvelidze, eds (Boca Raton: CRC Press), pp. 381–436. Summers, W.C. (1999). Félix d’Hérelle and the origins of molecular biology. (Yale University Press). Summers, W.C. (2001). Bacteriophage therapy. Annu. Rev. Microbiol. 55, 437–451. Tanji, Y., Shimada, T., Fukudomi, H., Miyanaga, K., Nakai, Y., and Unno, H. (2005). Therapeutic use of phage cocktail for controlling Escherichia coli O157:H7 in gastrointestinal tract of mice. J. Biosci. Bioeng. 100, 280–287. Tétart, F., Desplats, C., and Krisch, H.M. (1998). Genome plasticity in the distal tail fiber locus of the T-even bacteriophage: recombination between conserved motifs swaps adhesin specificity. J. Mol. Biol. 282, 543–556. Toro, H., Price, S.B., McKee, A.S., Hoerr, F.J., Krehling, J., Perdue, M., and Bauermeister, L. (2005). Use of bacteriophages in combination with competitive exclusion to reduce Salmonella from infected chickens. Avian Dis. 49, 118–124. Wagenaar, J.A., Van Bergen, M.A., Mueller, M.A., Wassenaar, T.M., and Carlton, R.M. (2005). Phage therapy reduces Campylobacter jejuni colonization in broilers. Vet. Microbiol. 109, 275–283. Wagner, P.L., Livny, J., Neely, M.N., Acheson, D.W., Friedman, D.I., and Waldor, M.K. (2002). Bacteriophage control of Shiga toxin 1 production and release by Escherichia coli. Mol. Microbiol. 44, 957–970. Wang, J., Hu, B., Xu, M., Yan, Q., Liu, S., Zhu, X., Sun, Z., Tao, D., Ding, L., Reed, E., Gong, J., Li, Q.Q., and Hu, J. (2006). Therapeutic effectiveness of bacteriophages in the rescue of mice with extended spectrum beta-lactamase-producing Escherichia coli bacteremia. Int. J. Mol. Med. 17, 347–355. Waterbury, J.B., and Valois, F.W. (1993). Resistance to co-occurring phages enables marine synechococcus communities to coexist with cyanophages abundant in seawater. Appl. Environ. Microbiol. 59, 3393–3399. Weld, R.J., Butts, C., and Heinemann, J.A. (2004). Models of phage growth and their applicability to phage therapy. J. Theor. Biol. 227, 1–11. Whichard, J.M., Sriranganathan, N., and Pierson, F.W. (2003). Suppression of Salmonella growth by wildtype and large-plaque variants of bacteriophage Felix O1 in liquid culture and on chicken frankfurters. J. Food Prot. 66, 220–225. Wiggins, B.A., and Alexander, M. (1985). Minimum bacterial density for bacteriophage replication: implications for significance of bacteriophages in natural ecosystems. Appl. Environ. Microbiol. 49, 19–23. Wills, Q.F., Kerrigan, C., and Soothill, J.S. (2005). Experimental bacteriophage protection against Staphylococcus aureus abscesses in a rabbit model. Antimicrob. Agents Chemother. 49, 1220–1221. Withey, S., Cartmell, E., Avery, L.M., and Stephenson, T. (2005). Bacteriophages—potential for application in wastewater treatment processes. Sci. Total Environ. 339, 1–18. Xie, H., Zhuang, X., Kong, J., Ma, G., and Zhang, H. (2005). Bacteriophage Esc-A is an efficient therapy for Escherichia coli 3–1 caused diarrhea in chickens. J. Gen. Appl. Microbiol. 51, 159–163. Yoong, P., Schuch, R., Nelson, D., and Fischetti, V.A. (2004). Identification of a broadly active phage lytic enzyme with lethal activity against antibiotic-resistant Enterococcus faecalis and Enterococcus faecium. J. Bacteriol. 186, 4808–4812. You, L., Suthers, P.F., and Yin, J. (2002). Effects of Escherichia coli physiology on growth of phage T7 in vivo and in silico. J. Bacteriol. 184, 1888–1894.

Bacteriophage–Host Interaction in Lactic Acid Bacteria

7

Christina Skovgaard Vegge, John Gerard Kenny, Lone Brøndsted, Stephen Mc Grath, and Douwe van Sinderen

Abstract The first contact between an infecting phage and its bacterial host is the attachment of the phage to the host cell. This attachment is mediated by the phage’s receptor binding protein (RBP), which recognizes and binds to a receptor on the bacterial surface. RBPs are also referred to as: host specificity protein, host determinant, and anti-receptor. For simplicity, the RBP term will be used here. A variety of molecules have been suggested to act as host receptors for bacteriophages infecting lactic acid bacteria (LAB); among those are polysaccharides (lipo)teichoic acids as well as a single membrane protein. A number of RBPs of LAB phages have been identified by the generation of hybrid phages with altered host range. These studies, however, also found additional phage proteins to be important for successful phage infection. Analysis of the crystal structure of several RBPs indicates that these proteins share a common tertiary folding as well as supporting previous indications of the saccharide nature of the host receptor. The Gram-positive LAB have a thick peptidoglycan layer, which must be traversed in order to inject the phage genome into the bacterial cytoplasm. Peptidoglycan-degrading enzymes are expected to facilitate this penetration and such enzymes have been found as structural elements of a number of LAB phages. Lactic acid bacteria and bacteriophages Lactic acid bacteria (LAB) are a heterogeneous group of Gram-positive bacteria with low genomic mol percentage of guanine and cytosine, which are non-sporing, non-respiring cocci or rods producing lactic acid as the major end-product of the fermentation of carbohydrates. The metabolic capabilities of species of Lactobacillus, Lactococcus, Leuconostoc, as well as Streptococcus thermophilus are widely used in the dairy industry for the production of fermented food products such as cheese and yogurt. Even though phages infecting Lactococcus were identified as a cause of dairy fermentation failure more than 70 years ago, they remain the largest single cause of fermentation disruption in the dairy industry despite major technological advances in the intervening period. The economic burden of phage infections in the dairy industry has increased the attention on the molecular processes underlying the interactions between the LAB hosts and their phages. Consequently the dairy phages represent one of the best characterized groups with respect to diversity and genome data (Brüssow and Hendrix, 2002). The identified phages of LAB have genomes of double

194

|

Vegge et al.

stranded DNA and belong to the order of Caudovirales, i.e. phages with a tail (Ackermann et al., 1984; Jarvis et al., 1991). Tailed phages are classified into three families according to the morphology of their tails; Myoviridae with long contractile tails, Siphoviridae with long non-contractile tails, and Podoviridae with short tails. Each family is furthermore subdivided into different morphotypes according to head shapes (isometric, prolate and large prolate heads), and the individual phages are finally grouped into species based on their relative DNA homology (Murphy et al., 1995; Ackermann, 1998). The predominant morphology among all phages is that of the isometric headed Siphoviridae phages, and the phage ecology of the dairy industry almost exclusively contains phages of this family. In recent years, increasing attention has been dedicated to the interactions between LAB phages and their bacterial hosts. The main focus areas have been the host receptors, phage-associated cell wall degrading enzymes, and the receptor recognition process by receptor binding proteins. The current knowledge on these perspectives of host interactions of LAB phages will be the topic of the present chapter. Host receptors of LAB phages A prerequisite for a phage infection is the attachment of the phage to the host. For tailed phages this attachment is mediated by the phage tail, which recognizes and binds to specific receptors on the surface of the bacteria. Following the binding of a receptor, the phage genome is injected into the host, and the empty phage virion is left idle on the cell surface. The bacterial receptor is a natural component of the cell envelope and may be polysaccharides, lipopolysaccharides (LPS), teichoic and lipoteichoic acids, or proteins present in the outer membrane. Phage receptors have primarily been studied in Gram-negative bacteria; particularly the receptors recognized by a few well characterized phages such as the Escherichia coli phages T4, L and T5, and these receptors will be briefly described. Myoviridae phage T4 binds initially and reversibly to B-type LPS or to the outer membrane porin OmpC via its long tail fibers (Montag et al., 1990; Heller, 1992). Following this interaction the short tail fibers bind irreversibly to the LPS core region (Makhov et al., 1993; Leiman et al., 2000). Phage L of the Siphoviridae family interacts first reversibly and then irreversibly with the outer membrane protein LamB (Randall-Hazelbauer and Schwartz, 1973; Wang et al., 2000). While Siphoviridae phage T5 undergoes the initial binding to polymannose O antigens in LPS by the long tail fibers, after which the receptor binding protein binds irreversibly to the ferrichrome transporter FhuA (Heller and Braun, 1979; Heller and Braun, 1982; Böhm et al., 2001). The structure of the cell envelope of the Gram-positive lactic acid bacteria (LAB) is very different from the Gram-negative bacteria and consists generally of a two layered structure composed of an inner plasma membrane and an outer cell wall. The cell wall is composed of a thick multilayered peptidoglycan shell, in which polysaccharides, teichoic and lipoteichoic acids, as well as proteins are embedded (Delcour et al., 1999). There are indications that some LAB phages first adsorb reversibly to cell wall carbohydrates and secondly bind irreversibly to a cell membrane receptor (Archibald, 1980; Monteville et al., 1994), however, it can not be concluded that a two step mechanism is employed by all LAB phages. Electron microscopy analysis of the adsorption of Lactococcus lactis phages from the

Bacteriophage–Host Interaction in Lactic Acid Bacteria

936, c2 and P034 species showed that a few of the examined phages adsorbed uniformly over the whole cell surface, while the majority of the nine investigated phages only adsorbed to distinct spots (Budde-Niekiel and Teuber, 1987). This indicates that the phages either adsorbed to receptors present on the entire cell surface, e.g. a cell wall carbohydrate, or specific receptors with limited cell numbers, e.g. a membrane protein. A chromosomally encoded Phage Infection Protein (PIP) has been shown to be required for the adsorption and DNA injection of lactococcal phages of the c2 species (Valyasevi et al., 1991; Geller et al., 1993; Monteville et al., 1994). PIP is an integral membrane protein with a multiple-membrane spanning region but no covalent attachment to the cell wall

(Mooney et al., 2006). Most lactococcal phages of the 936, P335 and 949 species do not use PIP as receptor as they are not affected in their ability to infect host cells that carry a mutated version of the pip gene (Kraus and Geller, 1998). However, the 936 species phage kh is an exception to this rule as it does require the PIP protein for infection (Monteville et al., 1994; Babu et al., 1995). A number of pip gene homologs have been identified amongst LAB, and insertional inactivation of a PIP analog in S. thermophilus Sfi11 conferred complete resistance to the strain against infection by all tested phages (Lucchini et al., 2000). Targets for cell binding by other lactococcal phages of various species have been shown to include different carbohydrates on the cell surface (Valyasevi et al., 1990; Schäfer et al., 1991; Sijtsma et al., 1991; Monteville et al., 1994; Valyasevi et al., 1994). Binding of phages to glucosamine and galactosamine moieties has also been demonstrated (Keogh and Pettingill, 1983). Random insertional mutagenesis of two L. lactis strains has identified a group of genes involved in the process of adsorption by two lactococcal phages of the 936 species (Dupont et al., 2004a). The genes, which when mutated, resulted in a reduced ability of the phage to adsorb, encode putative membrane spanning proteins and glycosyltransferases and were shown to be located on the same operon. These proteins are predicted to be involved in the biosynthesis and transport of cell wall polysaccharides indicating that these moieties are targets for phage binding. This is in agreement with other studies showing that lactococcal phages of the 936 species adsorb efficiently to cell wall components, while no adsorption was observed to either cell membrane fractions or exopolysaccharides (Deveau et al., 2002; Geller et al., 2005). These observations indicate that 936-type lactococcal phages do not require membrane proteins for infection and that the receptors are located exclusively in the bacterial cell wall. This hypothesis is supported and further refined by the finding that glycerol, which is a component of (lipo)teichoic acid, binds tightly to the crystallized receptor-binding proteins of two lactococcal phages of the 936 and P335 species, respectively (Spinelli et al., 2006; Tremblay et al., 2006). Although it could no be excluded that other polysaccharides may also be part of the receptors for the investigated phages (lipo)teichoic acids were argued the best receptor candidates, as these polymers point out of the cell wall and furthermore can ensure strain specificity by molecular variation (Spinelli et al., 2006; Tremblay et al., 2006).

|

195

196

|

Vegge et al.

A specific host receptor of streptococcal phages has so far not been identified, but adsorption studies of six phages isolated from dairy productions showed that these phages bound to carbohydrate components of the host cell envelopes, and that the phage adsorption could be reduced by the addition of glucosamine, N-acetylglucosamine, and rhamnose (Quiberoni et al., 2000; Binetti et al., 2002). Equivalently, studies with lactobacilli phages indicated that the phage receptors of these bacteria are components of the peptidoglycan or are linked to it (Yokokura, 1977; Quiberoni et al., 2004). In support of this, two Lactobacillus delbrueckii phages (LL-H and JCL1032) were shown to be inactivated by poly (glycerophosphate) lipoteichoic acids isolated from their common bacterial host, thus indicating that lipoteichoic acids are components of the host receptor for these phages (Räisänen et al., 2004). Taken together there have been a variety of reported molecules as putative host receptors for LAB phages, and these apparent disparities between different phages highlights the complexity of different mechanisms employed by the plethora of phages infecting LAB. Phage-associated lytic enzymes The peptidoglycan (PG) layer of the cell wall plays a vital role in providing structural support, shape and protection from the often perilous environment encountered by the bacterial cell. Phages employ muralytic enzymes which degrade this PG, thereby allowing the release of the progeny phage at the end of the lytic infection cycle through host cell lysis. While the PG functions as a hurdle to the eventual release of these phage it also serves as a barrier through which the phage genome must be passed in order to inject the phage genome into the host cell. This particular step in the phage life cycle remains poorly understood and has only been investigated for the small number of phages, which have been shown to possess phage-associated lytic enzymes. These enzymes are believed to function at the initiation of the phage infection process by boring holes in the PG layer through which the DNA injection machinery can pass to contact the inner membrane of the host cell. In all cases the particular location of these lytic enzymes within the phage structures is attuned to the different methods of host cell entry employed by the various phages. The attachment of the lytic enzymes to the phage is believed to allow for tight control of the enzymes. Excessive activity of the PG-hydrolases would be counterintuitive for the phage, since it would be expected to affect the overall viability of the host cell at the beginning of infection. Equivalent to the knowledge on phage host receptors, the knowledge on phage-associated lytic enzymes of LAB phages is limited. One therefore has to turn to phages infecting Gram-negative bacteria in order to illustrate how phages employ such lytic enzymes in the host infection process. The most studied phages in this regard are the phages T4 and T7, which infect Escherichia coli and phage PRD1, which can infect various bacterial species of the Enterobacteriaceae. The structural proteins of these phages have been found to include a lysozyme and two transglycosylases, which are located at the tail, within the phage head, and in the nucleocapsid, respectively (Moak and Molineux, 2000; Rydman and Bamford, 2002; Kanamaru et al., 2005). In the case of the Myoviridae phage T4, the lysozyme gp5 is expressed as a precursor from which a 20 kDa C-terminal domain is cleaved off. However, the C-terminal domain remains associated with the protein and assembles into the phage virion in a needle-like complex (Kanamaru et al., 1999). During T4 infection of the bacterial cell, major structural rearrangements in the phage tail forces the needle-shaped C-ter-

Bacteriophage–Host Interaction in Lactic Acid Bacteria

minal part of gp5, to penetrate the outer cell membrane. Upon contact with the periplasmic peptidoglycan layer, the needle dissociates from the tail and activates the lysozyme domains of gp5. The latter digest the peptidoglycan in the cell wall, which facilitate the subsequent injection of the phage DNA into the host cell (Kanamaru et al., 2002; Leiman et al., 2003; Rossmann et al., 2004; Leiman et al., 2004; Kanamaru et al., 2005). This system differs significantly from that employed by the Podoviridae phage T7 where four copies of the transglycosylase gp16 are incorporated into the capsid. At the initiation of the T7 infection process, gp16 as well as four other proteins are ejected out via the very short phage tail (Kemp et al., 2005), and gp16 has been shown to both control the ejection of the phage genome as well as to be required for infection under conditions believed to cause increased cross linking of the PG layer (Moak and Molineux, 2000; Struthers-Schlinke et al., 2000). Phage PRD1 belongs to the Tectiviridae family, i.e. the virion consists of an icosahedral capsid surrounding an internal membrane, which encloses the double-stranded DNA genome ( Jaatinen et al., 2004). The phage does not have a tail and circumvents the PG barricade via an alternative mechanism to those cited above. In this instance the PRD1 particle binds to host receptors using spike structures located at the vertices of the capsid. This allows the formation of a portal in the capsid through which the membrane passes in a tube-like conformation. It is at this point that the phage-associated lytic enzyme P7 is believed to degrade the PG layer allowing access of the injection machinery to the inner membrane. Interestingly, a second lytic enzyme, the endolysin P15, is also present within the PRD1 virion and may also play a role in penetrating host cell walls (Rydman and Bamford, 2000; Rydman and Bamford, 2002). The thickness of the PG layer in Gram-positive bacteria is much more than that of their Gram-negative counterparts, with estimated values ranging between approximately 20–50 nm and 2.5–7.5 nm, respectively (Beveridge and Graham, 1991; Labischinski et al., 1991). Therefore, one would expect phages infecting Gram-positive bacteria to be accordingly equipped to passage their DNA across this obstacle, since this requirement would be more stringent for phages where the hurdle is greater. A number of phages infecting LAB have been shown to contain PG-degrading enzymes as elements of their phage structure (Kenny et al., 2004; Moak and Molineux, 2004; Kenny, 2005). Among these, the proteins with the lytic activity have only been identified in the Lactococcus lactis phage Tuc2009 and the Streptococcus thermophilus phages ø01205 and ø7201 (Kenny et al., 2004; Kenny, 2005). Moreover, the Tail-Associated Lysin (Tal2009) from Tuc2009 has been shown to form a central tail fiber at the very distal part of the phage tail and antibodies directed against Tal2009 inhibited the ability of Tuc2009 to infect host cells (Kenny et al., 2004; Vegge et al., 2005; Mc Grath et al., 2006). These results would appear to indicate an optimal location of the lytic proteins within the phage structure to contact the PG layer of the host cell. Receptor recognition of LAB phages The first step of a phage lifecycle is the binding of the phage virion to the host cell. For tailed phages this initial interaction is mediated by a receptor-binding protein (RBP) located in the distal part of the tail. The RBP recognizes and binds to a receptor on the surface of the host cell—a process which is highly specific and the initial determinant of host specificity for any phage. Thus the RBP and host receptor interaction is a prerequisite for infection

|

197

198

|

Vegge et al.

even though other factors such as cytoplasmic phage resistance systems may subsequently prevent a successful phage infection. RBPs of tailed phages have primarily been studied for phages infecting the Gramnegative species Escherichia coli, and most of the knowledge on phage–host interactions therefore derives from studies of these phages. However, experimental investigations of host interactions and RBPs of several LAB phages have been reported recently, and these studies are in the progress of unraveling intriguing molecular interactions for phages infecting Gram-positive bacteria. Streptococcus phages Streptococcus thermophilus phages described to date all belong to the Siphoviridae B1 morphotype, which is characterized by a long non-contractile tail and isometric head containing the double stranded DNA genome (Ackermann, 1998). The phages are classified into two general groups according to the DNA packing mechanism and the number of major structural proteins (Le et al., 1997). Phages infecting S. thermophilus bacteria have been shown to adsorb efficiently and irreversibly to their host cells, essentially regardless of temperature, pH value, and host cell viability (Quiberoni et al., 2000; Binetti et al., 2002; Duplessis et al., 2005). The first RBPs of LAB phages were identified from the streptococcal phages DT1 and MD4 by Duplessis and Moineau (2001). In this study, a method, previously developed for E. coli Teven phages (Tétart et al., 1996; Tétart et al., 1998), was used in order to generate chimeric DT1 phages by homologous recombination between orf18DT1 (Tremblay and Moineau, 1999) and the corresponding orf18MD4 gene from MD4. This gene exchange resulted in an altered host range, as the chimeric phages had gained the ability to infect the MD4 host strain while lost the ability to infect the host strains of the parental DT1 phage. It was therefore concluded that the respective ORF18 of DT1 and MD4 were the RBPs of these phages (Duplessis and Moineau, 2001). However, the chimeric phages were still able to adsorb to their original host strain, indicating that other proteins of DT1 were involved in the adsorption process. This was subsequently confirmed, as mutations in two putative tail genes (orf15DT1 and orf17 DT1) were found to affect the host range of phage DT1 (Duplessis et al., 2006). Sequence analysis of the chimeric phages revealed that the host specificity of DT1 and MD4 was determined by a 145 amino acids variable domain (VR2) in the carboxy-terminal end of the RBPs (Duplessis and Moineau, 2001). This finding has subsequently been extended, as a degree of correlation has been noted between VR2 domain homology and the host range of several S. thermophilus phages (Binetti et al., 2005). Alignments of ORF18-like proteins of S. thermophilus phages have further revealed these proteins to contain between one and three variable domains, which are surrounded by more conserved regions and collagen-like repeats (Lucchini et al., 1999; Duplessis and Moineau, 2001; Levesque et al., 2005). This modular construction resembles the mosaic nature of the receptor binding tail fiber proteins of the T-even phages (Tétart et al., 1998) and indicates the RBPs of S. thermophilus phages have evolved by significant exchange of gene domains. Lactococcus phages Lactococcus lactis phages are grouped into 12 different species, but mainly phages from the three Siphoviridae species: 936, c2, and P335 have been experimentally investigated due

Bacteriophage–Host Interaction in Lactic Acid Bacteria

to their prevalence in industrial dairy phage ecology. Members of the 936 and c2 species are virulent phages with isometric- and moderately elongated (prolate) heads, respectively; while the phages of the P335 species are a heterogeneous group of both virulent and temperate isometric-headed phages ( Jarvis et al., 1991; Moineau et al., 1992; Moineau et al., 1996; Brüssow, 2001). Lactococcus 936 species phages The RBPs of phages sk1 and bIL170 of the 936 species (Chandry et al., 1997; Crutz-Le Coq et al., 2002) were identified by Dupont et al. (2004b), who in a manner similar to Duplessis and Moineau (2001) generated chimeric phages with an altered host range. These bIl170-derived phages had orf20bIL170 exchanged with orf18sk1 from phage sk1 (Chandry et al., 1997) and were consequently found to generate plaques on the sk1 host strain L. lactis MG1614. However, the plaques formed were pinpoint and turbid and the phages were also deficient in further lytic propagation on the new host strain. The authors therefore argued these findings to indicate that ORF20bIL170 and ORF18sk1 are RBPs of bIL170 and sk1, respectively, but that incompatibility between bIL170 and internal host factors of L. lactis MG1614 most likely prevents efficient propagation of the chimeric phages on the sk1 host strain (Dupont et al., 2004b). In addition, ORF18sk1 was identified at the tip of the sk1 tail by transmission electron microscopy and immunogold labeled antibodies raised against ORF18sk1 (Dupont et al., 2004b), which supports the ORF18sk1-RBP conclusion, as RBPs theoretically must be situated in the distal tail structure in order to come in contact with the host receptor in the bacterial envelope. The observations for sk1 are in agreement with subsequent findings for another 936-type phage, p2. In a study by De Haard et al. (2005) a phage display library was generated with single-domain antibody fragments raised in a llama against whole p2 virions. Antibodies of this library were tested for their ability to inhibit p2 infection of the bacterial host, and a single antibody fragment was accordingly found to efficiently prevent this infection (De Haard et al., 2005). This antibody fragment was subsequently shown to bind specifically to the p2 protein ORF18p2, which previously had been recognized as the putative RBP of this phage (Dupont et al., 2005; Spinelli et al., 2005). ORF18p2 of phage p2 shows 97% amino acid sequence identity with the RBP (ORF18sk1) of sk1 identified by Dupont et al. (2004b), and infection by sk1 was equally prevented by the antibody fragment identified in the p2 inhibition assay (De Haard et al., 2005). Moreover, electron microscopy with immunogold labeled antibody fragments was used to localize ORF18p2 at the tip of the p2 tail equivalent to ORF18sk1 of sk1 (De Haard et al., 2005). Fascinatingly, ORF18p2 of phage p2 was the first RBP from a LAB phage, for which the crystal structure was solved (Spinelli et al., 2005). This structure revealed that ORF18p2 is a homotrimeric protein composed of three domains: shoulders, a neck and the receptor-recognizing head (Spinelli et al., 2005). The shoulder domain is formed from a B-sandwich, which attaches the protein to the distal phage tail, while the neck is an interlaced B-prism that connects to the head domain. The head is a B-barrel made from seven anti-parallel B-strands in which the receptor-recognizing site is located (Spinelli et al., 2005). Glycerol and different saccharides were shown to bind with high affinity to four specific amino acids of the RBP, which strongly indicate a saccharide nature of the bacterial receptor recognized by phage p2 (Tremblay et al., 2006).

|

199

200

|

Vegge et al.

Lactococcus P335 species phages Phages of the P335 species are not the most numerous or the most prevalent in milk fermentation (Madera et al., 2004), but this species contain several temperate phages, which have been studied in great detail. TP901-1 and Tuc2009 are among the most significant of these phages; their genomes are fully sequenced and many aspects of the TP901-1 and Tuc2009 lifecycles have been examined in detail, such as prophage integration and excision, replication, transcription, and regulation of the lytic/lysogenic switch (Brøndsted et al., 2001; Seegers et al., 2004; Brøndsted and Hammer, 2006). Additionally, and very important in the context of host receptor recognition, are the tail structures of TP901-1 and Tuc2009, currently the best characterized for phages infecting LAB. Both phages contain a baseplate structure and a central tail fiber at the distal end of the tail, and the protein content of these structures have been examined in several studies ( Johnsen et al., 1995; Pedersen et al., 2000; Kenny et al., 2004; Vegge et al., 2005; Mc Grath et al., 2006). Furthermore, the RBPs from TP901-1 and Tuc2009 have been identified by the generation of a chimeric phage (Vegge et al., 2006) equivalent to the above mentioned studies with the lactococcal 936 species phages and S. thermophilus phages. The TP901-1 derived chimeric phage (TP9011C) had the lower baseplate protein bppLTP901-1 exchanged with the corresponding protein bppL2009 from phage Tuc2009, and this protein exchange was shown to confer a completely different host range. The novel chimera TP901-1C showed efficient adsorption to and infection of the Tuc2009 host strain L. lactis UC509.9, while very limited adsorption and complete lack of infection was observed with the TP901-1 host strain L. lactis 3107, thus identifying bppLTP901-1 and bppL2009 as the RBPs of TP901-1 and Tuc2009, respectively (Vegge et al., 2006). In contrast to the previous RBP studies using chimeric LAB phages, TP901-1C was isolated as a prophage integrant of the Tuc2009 host strain, and the chimeric phage was subsequently propagated and purified in order to analyze the protein content and tail morphology. This analysis showed the exchange of the lower baseplate protein not only resulted in altered host range but also gave a changed morphology of the baseplate structure (Vegge et al., 2006). A genomic analysis of the TP901-1C genome revealed that the chimeric phage had an altered sequence in the region of the late promoter and the Pac-site. This new sequence was most likely obtained by homologous recombination with another prophage on the bacterial chromosome, and it was speculated that this sequence exchange was necessary to overcome potential inhibition mechanisms acting subsequent to the host receptor binding to the new host (Vegge et al., 2006). Analysis of the crystal structure of the TP901-1 RBP (BppLTP901-1) revealed a trimeric structure highly similar to the RBPs of phages p2 and bIL170. Furthermore, bppLTP901-1 was found to bind glycerol and saccharides equivalently to the p2 and bIL170 RBPs (Spinelli et al., 2005; Spinelli et al., 2006; Ricagno et al., 2006). Together, these observations support the theory that host cell wall components act as receptors for lactococcal phages. Additionally, a comparison of the distal TP901-1 tail on electron micrographs with the crystal structure of bppLTP901-1 revealed the lower baseplate to be composed of six trimeric RBP complexes responsible for the host receptor recognition (Spinelli et al., 2006). The identification of the TP901-1 and Tuc2009 RBPs as baseplate structures is interesting from the viewpoint that most identified RBPs from E. coli phages are located in tail fibers, e.g. the straight tail fiber protein gpJ of Siphoviridae phage L and the adhesion

Bacteriophage–Host Interaction in Lactic Acid Bacteria

proteins in the long tail fibers from T-even Myoviridae phages (Heller, 1992; Wang et al., 2000). The tail fiber of TP901-1 and Tuc2009 is protruding below the RBP carrying baseplate, and the tail fiber proteins from these phages have been identified as TalTP901-1 and Tal2009, respectively (Vegge et al., 2005; Mc Grath et al., 2006). As stated previously, Tal2009 has experimentally been shown to contain a lytic activity, which degrades the cell-walls of L. lactis (Kenny et al., 2004). An equivalent activity is expected for TalTP901-1 because of high sequence identity and consistency in processing between the two Tal proteins (Vegge et al., 2005). Apparently the tail fibers of TP901-1 and Tuc2009 have no host-specific function in the infection process, but a comparison with the distal tail structure and infection mechanism of E. coli phage T5 can give an indication of a possible function. The T5 phage do not contain a baseplate, but in a manner similar to TP901-1 and Tuc2009 the RBP is located just above a straight tail fiber, and upon binding to the host receptor the T5 tail fiber is considered to traverse the cell envelope, thus allowing safe transfer of the phage genome into the host bacteria (Letellier et al., 1999; Böhm et al., 2001). The tail fibers of TP901-1 and Tuc2009 are only one-third of the length of the T5 fiber and it is unlikely that these fibers can traverse the thick peptidoglycan layer of the Gram-positive cell. Nevertheless, it has been speculated that the Tal proteins participate in the infection by degradation of the cell wall peptidoglycan in order to promote access to the host receptor or by assisting the injection of the phage DNA subsequent to binding of the RBP to the host receptor (Kenny et al., 2004; Vegge et al., 2005). Structural characterizations of TP901-1 and Tuc2009 have revealed that these phages carry highly identical tail proteins and that their tail morphologies are very alike (Vegge et al., 2006; Mc Grath et al., 2006). However, Tuc2009 contains an extra structural protein (ORF522009), which is encoded between the upper and lower baseplate proteins (Seegers et al., 2004). The function of ORF522009 is yet unknown, but due to the encoding genomic position has it been speculated that this protein is likely to be situated in the baseplate and somehow to participate in the infection process (Seegers et al., 2004; Vegge et al., 2006; Mc Grath et al., 2006). The tail proteins of TP901-1 and Tuc2009 show similarity to proteins of other lactococcal phages. A tail protein module almost identical to that of Tuc2009 is found in phage ul36, which is a virulent phage of the P335 species (Labrie and Moineau, 2002). Proteins with moderate to minor similarity to the RBPs bppLTP901-1 and bppL2009 are found in more distantly related phages, such as FLC3 and r1t of the P335 species, BK5T, and several phages of the 936 species (van Sinderen et al., 1996; Chandry et al., 1997; Mahanivong et al., 2001; Blatny et al., 2004; Dupont et al., 2004a; Dupont et al., 2005). The bppL-homologous proteins of 936 species phages include putative RBPs (Dupont et al., 2004a; Dupont et al., 2005), as well as the experimentally determined RBPs (ORF18sk1 and ORF18p2) of phages sk1 and p2 (Dupont et al., 2004b; Spinelli et al., 2005), which could indicate a common ancestor for RBPs of lactococcal phages. Lactococcus c2 species phages Lactococcal phages of the c2 species are characterized by having prolate head structures, and these phages are among the most prevalent species observed in both raw milk and in cheese plants (Moineau et al., 1992; Madera et al., 2004). Investigations of receptor recognition by c2 species phages have been carried out mainly with emphasis on the bacterial

|

201

202

|

Vegge et al.

host receptor, and as described previously the membrane protein, PIP, was found to be essential for infection by these phages (Geller et al., 1993; Babu et al., 1995). Competitive inhibition of phage infections with cell wall monosaccharides have revealed that rhamnose in particular, and to a lesser extent glucose and galactose, can inhibit the infection of several c2 species phages (Monteville et al., 1994), which indicate that cell wall components also play a role in the receptor recognition by these phages. In support of this, adsorption assays using envelope fractions showed that the adsorption to the cell wall was reversible, while the binding to the cell membrane was irreversible after 15 min (Monteville et al., 1994). This suggests that receptor recognition by phages of the c2 species is a two step process: first a reversible adsorption to a cell wall component and subsequently an irreversible binding to the cell membrane protein PIP. As is the case for 936 and P335 species phage, RBPs of phages from the c2 species have been identified by the generation of chimeric phages with altered host range. The RBPs from phages bIL69 and CHL92 were identified as ORF35bIL69 and ORF2CHL92, respectively (Stuer-Lauridsen et al., 2003), but the chimeric bIL69/CHL92 phages were very inefficiently propagated on the new host strain, which is comparable to the observations by Dupont et al. for chimeric 936 phages (Dupont et al., 2004b). This indicates that other factors, in addition to RBPs, may be crucial for successful phage propagation (StuerLauridsen et al., 2003). Rakonjac et al. (2005) confirmed this suspicion by identifying phage sequences which were fundamental for host infection but not involved in host receptor recognition. This study focused on the phages c2 and 923 together with their host strains L. lactis MG1363 and 112, respectively. Both c2 and 923 can adsorb to MG1363 and 112, but they can only form plaques on their respective host strain at 30oC, which indicates that infection of the non-permissive strain is inhibited subsequent to binding of the host receptor (Rakonjac et al., 2005). In order to investigate the infection inhibition of the nonpermissive phage/host pairs, a new protocol was developed for obtaining and analyzing phage recombinants with altered host rage (Rakonjac et al., 2005). Mixed infections were carried out with both phages (c2 and 923) alternately and successively on both host strains (112 and MG1363). This resulted in an enrichment of phage-phage recombinants, which were able to infect both 112 and MG1363. Subsequent analysis revealed that infection of strain 112 by phage c2 is inhibited at the level of cos-end ligation and possibly by binding of a host protein to the cos-site of the c2 genome subsequent to injection of the DNA. While phage 923 infection of strain MG1363 is inhibited at the level of DNA injection (Rakonjac et al., 2005). A correlation was observed between the c2 gene gL10 and the ability to infect MG1363, which suggests that the c2 protein pL10 is involved in DNA injection into the host cells (Rakonjac et al., 2005). This finding is supported by a previous identification of pL10 at the tip of the c2 tail (Lubbers et al., 1995). These results by Rakonjac et al. (2005) provide experimental evidence for previous indications of host range being determined by more factors than a perfect match of host receptor and phage RBP (Stuer-Lauridsen et al., 2003; Dupont et al., 2004b; Vegge et al., 2006). Lactobacillus phages Investigations of Lactobacillus phages are very limited compared to the much better studied streptococcal and lactococcal phages; however there are a few studies investigating the host

Bacteriophage–Host Interaction in Lactic Acid Bacteria

infection of these phages. Ravin et al. (2002) examined spontaneous phage-resistant strains of Lactobacillus delbrueckii together with phage mutants, which could infect these resistant bacteria. These analyses identified the RBPs of the isometric-headed LL-H phage and the prolate-headed JCL1032 phage of the Siphoviridae family (Alatossava and Pyhtila, 1980; Forsman, 1993; Ravin et al., 2002). These RBPs were found to be highly similar in the Cterminal region thus indicating a similar structural location in the phage virions. However, the two phages were found to use different host receptors of the same strain (Ravin et al., 2002). Interestingly, it was also observed that single nucleotide mutations in the RBP gene could alter the host receptor specificity of phage LL-H such that mutant phages could infect a phage-resistant host derivative (Ravin et al., 2002). Another L. plantarum phage, LP65, also capable of infecting strains of Carnobacterium, was characterized by Chibani-Chennoufi et al. (2004). LP65 belongs to the Myoviridae family of phages, which are defined by having a contractile tail. From electron microscopy it was illustrated that LP65 changes baseplate formation upon tail sheath contraction and that the extruded tail tube tunnels through the cell wall equivalent to that of the E. coli phage T4 (Chibani-Chennoufi et al., 2004; Leiman et al., 2004). Interestingly, it was observed that LP65 had a tendency to adsorb next to the septum of dividing cells, and that membrane convolutions were found inside the infected cells (Chibani-Chennoufi et al., 2004). However, the authors stated that these convolutions resembled structural alterations, which had been observed in L. plantarum that had been exposed to bacteriocins, and therefore suggested it to be indicative of a stress reaction (Chibani-Chennoufi et al., 2004). This section has described studies on receptor recognition by phages infecting strains of Streptococcus, Lactococcus and Lactobacillus. These studies have primarily focused on identification of phage-encoded RBPs, which recognize and bind to a receptor in the host cell envelope, and most studies have identified these proteins by generation of chimeric phages with altered host range. A few studies have related the RBP to a specific tail structure, but for the majority of phages the structural locations of the RBPs remain to be elucidated. Concluding remarks Phages are generally considered to interact with their hosts in a two step process. The first step involves a reversible interaction with a cell surface component, followed by a secondary irreversible interaction with a membrane associated component. However, except for lactococcal phages of the c2 species, this two-step binding mechanism has not been convincingly shown for any other LAB phages. This discrepancy may be related to the fact that the two different kinds of binding primarily have been reported in relation to E. coli phages with long tail fibers, which promotes the first reversible interaction (Heller, 1992). When the present knowledge on host interactions of LAB phage is compared with the corresponding information on phages infecting E. coli, it appears that there are no reports of LAB phages using long tail fibers for this interaction. At present, these fibers seem to be missing in the LAB phages, but it is possible their existence is yet to be discovered. The term “host specificity” is often used in relation to host interactions. But, what does the term actually mean in relation to the interactions between phages and their bacterial hosts? If host specificity denotes all factors influencing whether a phage can successfully

|

203

204

|

Vegge et al.

infect a bacterial cell, then all the different steps of the phage infection process have to be taken into consideration. For instance, LAB have been found to contain many different systems of phage resistance (Coffey and Ross, 2002), but even though these systems greatly affect the infection ability of phages they are generally not considered in relation to host specificity. On the contrary, host specificity is generally used to describe the initial steps of the host infection process, which take place on the outside of the bacterial membrane. However, several studies reviewed in this chapter have shown that a successful infection requires more than a perfect match between the bacterial receptor and the phage’s RBP. The last decades have expanded the knowledge on host interactions of LAB phages considerably and provided insight to some of the molecular processes underlying these interactions. This knowledge provides a solid basis for unraveling a more comprehensive picture of the infection mechanism of LAB phages. In order to improve the understanding, focus needs to be directed to some of the more unexplored steps of the infection process, i.e. obtain a more explicit idea on how the phages actually overcome the PG barrier and how the phage genome is transferred from the phage head and into the bacterial cytoplasm. The recent reports outlining the crystal structure of several RBPs from lactococcal phage have provided very interesting insights into the process of host receptor recognition. Further investigations of the other tail proteins that compose the Siphoviridae phage tail will undoubtedly enable us to answer some of the above questions. References

Ackermann, H.W. (1998). Tailed bacteriophages: the order caudovirales. Adv. Virus Res. 51, 135–201. Ackermann, H.W., Cantor, E.D., Jarvis, A.W., Lembke, J., and Mayo, J.A. (1984). New species definitions in phages of gram-positive cocci. Intervirology 22, 181–190. Alatossava, T., and Pyhtila, M.J. (1980). Characterization of a new Lactobacillus lactis bacteriophage. IRCS Med. Sci. 8, 297–298. Archibald, A.R. (1980). Phage receptors in Gram-positive bacteria. In: Virus Receptors Part 1 (receptors and recognition, series B, vol 7), L.L. Randall and L. Philipson, ed. (London and new York: Chapman & Hall), pp. 6–26. Babu, K.S., Spence, W.S., Monteville, M.R., and Geller, B.L. (1995). Characterization of a cloned gene (pip) from lactococcus lactis required for phage infection. Dev. Biol. Stand. 85, 569–575. Beveridge, T.J., and Graham, L.L. (1991). Surface layers of bacteria. Microbiol. Rev. 55, 684–705. Binetti, A.G., Del, R.B., Martin, M.C., and Alvarez, M.A. (2005). Detection and characterization of Streptococcus thermophilus bacteriophages by use of the antireceptor gene sequence. Appl. Environ. Microbiol. 71, 6096–6103. Binetti, A.G., Quiberoni, A., and Reinheimer, J.A. (2002). Phage adsorption to Streptococcus thermophilus. Influence of enviromental factors and characterization of cell-receptors. Food Res. Int. 35, 73–83. Blatny, J.M., Godager, L., Lunde, M., and Nes, I.F. (2004). Complete genome sequence of the Lactococcus lactis temperate phage phiLC3: comparative analysis of phiLC3 and its relatives in lactococci and streptococci. Virology 318, 231–244. Böhm, J., Lambert, O., Frangakis, A.S., Letellier, L., Baumeister, W., and Rigaud, J.L. (2001). FhuAmediated phage genome transfer into liposomes: a cryo-electron tomography study. Curr. Biol. 11, 1168–1175. Brøndsted, L., and Hammer, K. (2006). Phages of Lactococcus lactis. In: The Bacteriophages, R. Calendar, ed. (New York: Oxford University Press), pp. 572–592. Brøndsted, L., Østergaard, S., Pedersen, M., Hammer, K., and Vogensen, F.K. (2001). Analysis of the complete DNA sequence of the temperate bacteriophage TP901-1: evolution, structure, and genome organization of lactococcal bacteriophages. Virology 283, 93–109. Brüssow, H. (2001). Phages of dairy bacteria. Annu. Rev. Microbiol. 55, 283–303. Brüssow, H., and Hendrix, R.W. (2002). Phage genomics: small is beautiful. Cell 108, 13–16.

Bacteriophage–Host Interaction in Lactic Acid Bacteria

Budde-Niekiel, A., and Teuber, M. (1987). Electron microscopy of the adsorption of the bacteriophages to lactic acid streptococci. Milchwissenschaft 42, 551–554. Chandry, P.S., Moore, S.C., Boyce, J.D., Davidson, B.E., and Hillier, A.J. (1997). Analysis of the DNA sequence, gene expression, origin of replication and modular structure of the Lactococcus lactis lytic bacteriophage sk1. Mol. Microbiol. 26, 49–64. Chibani-Chennoufi, S., Dillmann, M.L., Marvin-Guy, L., Rami-Shojaei, S., and Brussow, H. (2004). Lactobacillus plantarum bacteriophage LP65: a new member of the SPO1-like genus of the family Myoviridae. J. Bacteriol. 186, 7069–7083. Coffey, A., and Ross, R.P. (2002). Bacteriophage-resistance systems in dairy starter strains: molecular analysis to application. Antonie Van Leeuwenhoek 82, 303–321. Crutz-Le Coq, A.-M., Cesselin, B., Commissaire, J., and Anba, J. (2002). Sequence analysis of the lactococcal bacteriophage bIL170: insights into structural proteins and HNH endonucleases in dairy phages. Microbiology 148, 985–1001. De Haard, H.J., Bezemer, S., Ledeboer, A.M., Muller, W.H., Boender, P.J., Moineau, S., Coppelmans, M.C., Verkleij, A.J., Frenken, L.G., and Verrips, C.T. (2005). Llama antibodies against a lactococcal protein located at the tip of the phage tail prevent phage infection. J. Bacteriol. 187, 4531–4541. Delcour, J., Ferain, T., Deghorain, M., Palumbo, E., and Hols, P. (1999). The biosynthesis and functionality of the cell-wall of lactic acid bacteria. Antonie Van Leeuwenhoek 76, 159–184. Deveau, H., Van Calsteren, M.R., and Moineau, S. (2002). Effect of exopolysaccharides on phage–host interactions in Lactococcus lactis. Appl. Environ. Microbiol. 68, 4364–4369. Duplessis, M., Levesque, C.M., and Moineau, S. (2006). Characterization of Streptococcus thermophilus host range phage mutants. Appl. Environ. Microbiol. 72, 3036–3041. Duplessis, M., and Moineau, S. (2001). Identification of a genetic determinant responsible for host specificity in Streptococcus thermophilus bacteriophages. Mol. Microbiol. 41, 325–336. Duplessis, M., Russell, W.M., Romero, D.A., and Moineau, S. (2005). Global gene expression analysis of two Streptococcus thermophilus bacteriophages using DNA microarray. Virology 340, 192–208. Dupont, K., Janzen, T., Vogensen, F.K., Josephsen, J., and Stuer-Lauridsen, B. (2004a). Identification of Lactococcus lactis genes required for bacteriophage adsorption. Appl. Environ. Microbiol. 70, 5825–5832. Dupont, K., Vogensen, F.K., and Josephsen, J. (2005). Detection of lactococcal 936-species bacteriophages in whey by magnetic capture hybridization PCR targeting a variable region of receptor-binding protein genes. J. Appl. Microbiol. 98, 1001–1009. Dupont, K., Vogensen, F.K., Neve, H., Bresciani, J., and Josephsen, J. (2004b). Identification of the receptorbinding protein in 936-species lactococcal bacteriophages. Appl. Environ. Microbiol. 70, 5818–5824. Forsman, P. (1993). Characterization of a prolate-headed bacteriophage of Lactobacillus delbrueckii subsp. lactis, and its DNA homology with isometric-headed phages. Arch. Virol. 132, 321–330. Geller, B.L., Ivey, R.G., Trempy, J.E., and Hettinger-Smith, B. (1993). Cloning of a chromosomal gene required for phage infection of Lactococcus lactis subsp. lactis C2. J. Bacteriol. 175, 5510–5519. Geller, B.L., ngo, H.T., Mooney, D.T., Su, P., and Dunn, N. (2005). Lactococcal 936-species phage attachment to surface of Lactococcus lactis. J. Dairy Sci. 88, 900–907. Heller, K., and Braun, V. (1979). Accelerated adsorption of bacteriophage T5 to Escherichia coli F, resulting from reversible tail fiber-lipopolysaccharide binding. J. Bacteriol. 139, 32–38. Heller, K., and Braun, V. (1982). Polymannose O-antigens of Escherichia coli, the binding sites for the reversible adsorption of bacteriophage T5+ via the L-shaped tail fibers. J. Virol. 41, 222–227. Heller, K.J. (1992). Molecular interaction between bacteriophage and the Gram-negative cell envelope. Arch. Microbiol. 158, 235–248. Jaatinen, S.T., Viitanen, S.J., Bamford, D.H., and Bamford, J.K. (2004). Integral membrane protein P16 of bacteriophage PRD1 stabilizes the adsorption vertex structure. J. Virol. 78, 9790–9797. Jarvis, A.W., Fitzgerald, G.F., Mata, M., Mercenier, A., Neve, H., Powell, I.B., Ronda, C., Saxelin, M., and Teuber, M. (1991). Species and type phages of lactococcal bacteriophages. Intervirology 32, 2–9. Johnsen, M.G., Neve, H., Vogensen, F.K., and Hammer, K. (1995). Virion positions and relationships of lactococcal temperate bacteriophage TP901-1 proteins. Virology 212, 595–606. Kanamaru, S., Gassner, N.C., Ye, N., Takeda, S., and Arisaka, F. (1999). The C-terminal fragment of the precursor tail lysozyme of bacteriophage T4 stays as a structural component of the baseplate after cleavage. J. Bacteriol. 181, 2739–2744.

|

205

206

|

Vegge et al.

Kanamaru, S., Ishiwata, Y., Suzuki, T., Rossmann, M.G., and Arisaka, F. (2005). Control of bacteriophage T4 tail lysozyme activity during the infection process. J. Mol. Biol. 346, 1013–1020. Kanamaru, S., Leiman, P.G., Kostyuchenko, V.A., Chipman, P.R., Mesyanzhinov, V.V., Arisaka, F., and Rossmann, M.G. (2002). Structure of the cell-puncturing device of bacteriophage T4. Nature 415, 553–557. Kemp, P., Garcia, L.R., and Molineux, I.J. (2005). Changes in bacteriophage T7 virion structure at the initiation of infection. Virology 340, 307–317. Kenny, J.G. Molecular characterisation of the lactococcal bacteriophage Tuc2009. 2005. National University of Ireland, Cork. Ph.D. thesis. Ref Type: Thesis/Dissertation Kenny, J.G., Mc Grath, S., Fitzgerald, G.F., and van Sinderen, D. (2004). Bacteriophage Tuc2009 encodes a tail-associated cell wall-degrading activity. J. Bacteriol. 186, 3480–3491. Keogh, B.P., and Pettingill, G. (1983). Adsoption of bacteriophage eb7 on Streptococcus cremoris EB7. Appl. Environ. Microbiol. 45, 1946–1948. Kraus, J., and Geller, B. (1998). Membrane receptor for prolate phages is not required for infection of Lactococcus lactis small or large isometric phages. Int. Dairy J. 81, 2329–2335. Labischinski, H., Goodell, E.W., Goodell, A., and Hochberg, M.L. (1991). Direct proof of a “more-thansingle-layered” peptidoglycan architecture of Escherichia coli W7: a neutron small-angle scattering study. J. Bacteriol. 173, 751–756. Labrie, S., and Moineau, S. (2002). Complete genomic sequence of bacteriophage ul36: demonstration of phage heterogeneity within the P335 quasi-species of lactococcal phages. Virology 296, 308–320. Le, M.C., van, S.D., Walsh, L., Stanley, E., Vlegels, E., Moineau, S., Heinze, P., Fitzgerald, G., and Fayard, B. (1997). Two groups of bacteriophages infecting Streptococcus thermophilus can be distinguished on the basis of mode of packaging and genetic determinants for major structural proteins. Appl. Environ. Microbiol. 63, 3246–3253. Leiman, P.G., Chipman, P.R., Kostyuchenko, V.A., Mesyanzhinov, V.V., and Rossmann, M.G. (2004). Three-dimensional rearrangement of proteins in the tail of bacteriophage T4 on infection of its host. Cell 118, 419–429. Leiman, P.G., Kanamaru, S., Mesyanzhinov, V.V., Arisaka, F., and Rossmann, M.G. (2003). Structure and morphogenesis of bacteriophage T4. Cell Mol. Life Sci. 60, 2356–2370. Leiman, P.G., Kostyuchenko, V.A., Shneider, M.M., Kurochkina, L.P., Mesyanzhinov, V.V., and Rossmann, M.G. (2000). Structure of bacteriophage T4 gene product 11, the interface between the baseplate and short tail fibers. J. Mol. Biol. 301, 975–985. Letellier, L., Plancon, L., Bonhivers, M., and Boulanger, P. (1999). Phage DNA transport across membranes. Res. Microbiol. 150, 499–505. Levesque, C., Duplessis, M., Labonte, J., Labrie, S., Fremaux, C., Tremblay, D., and Moineau, S. (2005). Genomic organization and molecular analysis of virulent bacteriophage 2972 infecting an exopolysaccharide-producing Streptococcus thermophilus strain. Appl. Environ. Microbiol. 71, 4057–4068. Lubbers, M.W., Waterfield, N.R., Beresford, T.P., Le Page, R.W., and Jarvis, A.W. (1995). Sequencing and analysis of the prolate-headed lactococcal bacteriophage c2 genome and identification of the structural genes. Appl. Environ. Microbiol. 61, 4348–4356. Lucchini, S., Desiere, F., and Brüssow, H. (1999). Comparative genomics of Streptococcus thermophilus phage species supports a modular evolution theory. J. Virol. 73, 8647–8656. Lucchini, S., Sidoti, J., and Brussow, H. (2000). Broad-range bacteriophage resistance in Streptococcus thermophilus by insertional mutagenesis. Virology 275, 267–277. Madera, C., Monjardin, C., and Suarez, J.E. (2004). Milk contamination and resistance to processing conditions determine the fate of Lactococcus lactis bacteriophages in dairies. Appl. Environ. Microbiol. 70, 7365–7371. Mahanivong, C., Boyce, J.D., Davidson, B.E., and Hillier, A.J. (2001). Sequence analysis and molecular characterization of the Lactococcus lactis temperate bacteriophage BK5-T. Appl. Environ. Microbiol. 67, 3564–3576. Makhov, A.M., Trus, B.L., Conway, J.F., Simon, M.N., Zurabishvili, T.G., Mesyanzhinov, V.V., and Steven, A.C. (1993). The short tail-fiber of bacteriophage T4: molecular structure and a mechanism for its conformational transition. Virology 194, 117–127. Mc Grath, S., Neve, H., Seegers, J.F., Eijlander, R., Vegge, C.S., Brøndsted, L., Heller, K.J., Fitzgerald, G.F., Vogensen, F.K., and van Sinderen, D. (2006). Anatomy of a lactococcal phage tail. J. Bacteriol. 188, 3972–3982.

Bacteriophage–Host Interaction in Lactic Acid Bacteria

Moak, M., and Molineux, I.J. (2000). Role of the Gp16 lytic transglycosylase motif in bacteriophage T7 virions at the initiation of infection. Mol. Microbiol. 37, 345–355. Moak, M., and Molineux, I.J. (2004). Peptidoglycan hydrolytic activities associated with bacteriophage virions. Mol. Microbiol. 51, 1169–1183. Moineau, S., Ackermann, H.W., and Pandian, S. (1992). Characterization of lactococcal bacteriophages from Quebec cheese plants. Can. J. Microbiol. 38, 875–82. Moineau, S., Borkaev, M., Holler, B.J., Walker, S.A., Kondo, J.K., Vedamuthu, E.R., and Vandenbergh, P.A. (1996). Isolation and characterization of lactococcal bacteriophages from cultured buttermilk plants in the United States. J. Dairy Sci. 79, 2104–2100. Montag, D., Hashemolhosseini, S., and Henning, U. (1990). Receptor-recognizing proteins of T-even type bacteriophages. The receptor-recognizing area of proteins 37 of phages T4 TuIa and TuIb. J. Mol. Biol. 216, 327–334. Monteville, M.R., Ardestani, B., and Geller, B. (1994). Lactococcal bacteriophages require a host cell wall carbohydrate and a plasma membrane protein for adsorption and ejection of DNA. Appl. Environ. Microbiol. 60, 3204–3211. Mooney, D.T., Jann, M., and Geller, B.L. (2006). Subcellular location of phage infection protein (Pip) in Lactococcus lactis. Can. J. Microbiol. 52, 664–672. Murphy, F.A., Fauquet, C.M., Bishop, D.H.L., Ghabrial, S.A., Jarvis, A.W., Martelli, G.P., Mayo, M.A., and Summers (eds.), M.D. (1995). Virus taxonomy. Classification and nomenclature of viruses. Sixth report of the international committee on taxonomy of viruses. Arch. Virol. Suppl. 10. Pedersen, M., Østergaard, S., Bresciani, J., and Vogensen, F.K. (2000). Mutational analysis of two structural genes of the temperate lactococcal bacteriophage TP901-1 involved in tail length determination and baseplate assembly. Virology 276, 315–328. Quiberoni, A., Guglielmotti, D., Binetti, A., and Reinheimer, J. (2004). Characterization of three Lactobacillus delbrueckii subsp. bulgaricus phages and the physicochemical analysis of phage adsorption. J. Appl. Microbiol. 96, 340–351. Quiberoni, A., Stiefel, J.I., and Reinheimer, J.A. (2000). Characterization of phage receptors in Streptococcus thermophilus using purified cell walls obtained by a simple protocol. J. Appl. Microbiol. 89, 1059–1065. Räisänen, L., Schubert, K., Jaakonsaari, T., and Alatossava, T. (2004). Characterization of lipoteichoic acids as Lactobacillus delbrueckii phage receptor components. J. Bacteriol. 186, 5529–5532. Rakonjac, J., O’Toole, P.W., and Lubbers, M. (2005). Isolation of lactococcal prolate phage-phage recombinants by an enrichment strategy reveals two novel host range determinants. J. Bacteriol. 187, 3110–3121. Randall-Hazelbauer, L., and Schwartz, M. (1973). Isolation of the bacteriophage lambda receptor from Escherichia coli. J. Bacteriol. 116, 1436–1446. Ravin, V., Raisanen, L., and Alatossava, T. (2002). A conserved C-terminal region in gp71 of the small isometric-head phage LL-H and ORF474 of the prolate-head phage JCL1032 is implicated in specificity of adsorption of phage to its host, Lactobacillus delbrueckii. J. Bacteriol. 184, 2455–2459. Ricagno, S., Campanacci, V., Blangy, S., Spinelli, S., Tremblay, D., Moineau, S., Tegoni, M., and Cambillau, C. (2006). Crystal structure of the receptor-binding protein head domain from Lactococcus lactis phage bIL170. J. Virol. 80, 9331–9335. Rossmann, M.G., Mesyanzhinov, V.V., Arisaka, F., and Leiman, P.G. (2004). The bacteriophage T4 DNA injection machine. Curr. Opin. Struct. Biol. 14, 171–180. Rydman, P.S., and Bamford, D.H. (2000). Bacteriophage PRD1 DNA entry uses a viral membrane-associated transglycosylase activity. Mol. Microbiol. 37, 356–363. Rydman, P.S., and Bamford, D.H. (2002). The lytic enzyme of bacteriophage PRD1 is associated with the viral membrane. J. Bacteriol. 184, 104–110. Schäfer, A., Geis, A., Neve, H., and Teuber, M. (1991). Bacteriophage receptors of Lactococcus lactis subsp. “diacetylactis” F7/2 and Lactococcus lactis subsp. cremoris Wg2–1. FEMS Microbiol. Lett. 62, 69–73. Seegers, J.F., Mc Grath, S., O’connell-Motherway, M., Arendt, E.K., van de Guchte, M., Creaven, M., Fitzgerald, G.F., and van Sinderen, D. (2004). Molecular and transcriptional analysis of the temperate lactococcal bacteriophage Tuc2009. Virology 329, 40–52. Sijtsma, L., Hellingwerf, K.J., and Wouters, J.T. (1991). Composition and phage binding capacity of cell walls isolated from Lactococcus lactis subsp. cremoris SK110 and SK112. Neth. Milk Dairy J. 45, 81–95.

|

207

208

|

Vegge et al.

Spinelli, S., Campanacci, V., Blangy, S., Moineau, S., Tegoni, M., and Cambillau, C. (2006). Modular structure of the receptor binding proteins of Lactococcus lactis phages. The RBP structure of the temperate phage TP901-1. J. Biol. Chem. 281, 14256–14262. Spinelli, S., Desmyter, A., Verrips, C.T., De Haard, H.J., Moineau, S., and Cambillau, C. (2005). Lactococcal bacteriophage p2 receptor-binding protein structure suggests a common ancestor gene with bacterial and mammalian viruses. Nat. Struct. Mol. Biol. Struthers-Schlinke, J.S., Robins, W.P., Kemp, P., and Molineux, I.J. (2000). The internal head protein Gp16 controls DNA ejection from the bacteriophage T7 virion. J. Mol. Biol. 301, 35–45. Stuer-Lauridsen, B., Janzen, T., Schnabl, J., and Johansen, E. (2003). Identification of the host determinant of two prolate-headed phages infecting Lactococcus lactis. Virology 309, 10–17. Tétart, F., Desplats, C., and Krisch, H.M. (1998). Genome plasticity in the distal tail fiber locus of the T-even bacteriophage: recombination between conserved motifs swaps adhesin specificity. J. Mol. Biol. 282, 543–556. Tétart, F., Repoila, F., Monod, C., and Krisch, H.M. (1996). Bacteriophage T4 host range is expanded by duplications of a small domain of the tail fiber adhesin. J. Mol. Biol. 258, 726–731. Tremblay, D.M., and Moineau, S. (1999). Complete genomic sequence of the lytic bacteriophage DT1 of Streptococcus thermophilus. Virology 255, 63–76. Tremblay, D.M., Tegoni, M., Spinelli, S., Campanacci, V., Blangy, S., Huyghe, C., Desmyter, A., Labrie, S., Moineau, S., and Cambillau, C. (2006). Receptor-binding protein of Lactococcus lactis phages: identification and characterization of the saccharide receptor-binding site. J. Bacteriol. 188, 2400–2410. Valyasevi, R., Sandine, W.E., and Geller, B.L. (1990). The bacteriophage kh receptor of Lactococcus lactis subsp. cremoris KH is the rhamnose of the extracellular wall polysaccharide. Appl. Environ. Microbiol. 56, 1882–1889. Valyasevi, R., Sandine, W.E., and Geller, B.L. (1991). A membrane protein is required for bacteriophage c2 infection of Lactococcus lactis subsp. lactis C2. J. Bacteriol. 173, 6095–6100. Valyasevi, R., Sandine, W.E., and Geller, B.L. (1994). Lactococcus lactis ssp. lactis C2 bacteriophage sk1 receptor involving rhamnose and glucose moieties in the cell wall. J. Dairy. Sci. 77, 1–6. van Sinderen, D., Karsens, H., Kok, J., Terpstra, P., Ruiters, M.H., Venema, G., and Nauta, A. (1996). Sequence analysis and molecular characterization of the temperate lactococcal bacteriophage r1t. Mol. Microbiol. 19, 1343–1355. Vegge, C.S., Brøndsted, L., Neve, H., Mc Grath, S., van Sinderen, D., and Vogensen, F.K. (2005). Structural characterization and assembly of the distal tail structure of the temperate lactococcal bacteriophage TP901-1. J. Bacteriol. 187, 4187–4197. Vegge, C.S., Vogensen, F.K., Mc, G.S., Neve, H., van, S.D., and Brøndsted, L. (2006). Identification of the lower baseplate protein as the antireceptor of the temperate lactococcal bacteriophages TP901-1 and Tuc2009. J. Bacteriol. 188, 55–63. Wang, J., Hofnung, M., and Charbit, A. (2000). The C-terminal portion of the tail fiber protein of bacteriophage lambda is responsible for binding to LamB, its receptor at the surface of Escherichia coli K-12. J. Bacteriol. 182, 508–512. Yokokura, T. (1977). Phage receptor material in Lactobacillus casei. J. Gen. Microbiol. 100, 139–145.

Transfer of DNA from Phage to Host

8

Lucienne Letellier, Laure Plançon, and Pascale Boulanger

Abstract Phage DNA transport is atypical among membrane transport and thus poses a fascinating problem: transport is unidirectional; it concerns a unique molecule the size of which may represent 50 times that of the bacterium. The rate of DNA transport can reach values as high as 3 to 4 thousands base pairs/sec. This raises many questions, which will be addressed in this chapter. Is there a single mechanism of transport for all types of phages? How does the phage genome overcome the hydrophobic barrier of the host envelope? Is DNA transported as a free molecule or in association with proteins? Is such transport dependent on phage and/or host cell components? What is the driving force for transport? Data will be presented for a few selected tailed phages for which DNA transport has been most extensively studied. Introduction Largely unexplored for a long time the field of phage genome transfer has given rise to a renewed interest during the last few years. Most of the former studies focused on doublestranded DNA (dsDNA) tailed phages (Benson et al., 1999; Casjens, 2005) which represent the largest group of bacteriophages presently identified (Ackermann, 2003) and were mainly restricted to phages from Gram-negative bacteria. More recently results have started to emerge from the study of bacteriophages from other taxonomic groups and infecting Gram-positive bacteria. If molecular biology and genetics have been essential tools to understand this early stage of the infection process, physical techniques have led, during the last decade, to remarkable progresses in characterizing the mechanisms of DNA ejection. The topic of phage genome transport has been covered in several reviews (Dreiseikelmann, 1994; Goldberg et al., 1994; Letellier et al., 2003; Letellier et al., 2004; Molineux, 2006; Poranen et al., 2002). This review will therefore rather focus on recent advances in this field with the purpose of having some general concepts/ideas/questions emerging. Some general features on the transport of phage DNA A common feature of dsDNA tailed phages is that only their genome is transferred to the host cytoplasm while the phage capsid and tail remain bound to the cell surface. In the case

210

|

Letellier et al.

of viruses infecting Gram-negative bacteria the DNA has to overcome two hydrophobic barriers: the outer and the inner membrane. Furthermore, it has to cross the periplasm, the space separating the two membranes that contains nucleases and the rigid polymeric sugar structure of the peptidoglycan. In the case of Gram-positive bacteria the rigid and in some cases thick peptidoglycan represents a major obstacle that the phage genome has to overcome. Phage nucleic acid transport poses a fascinating problem: transport is unidirectional; it concerns a unique polyanionic molecule the size of which may represent 50 times that of the bacterium (50 µm of contour length for the dsDNA of phage T4). The rate of DNA transport, although varying from one phage to another, can reach values as high as 3 to 4 thousand base pairs/sec, a value significantly larger than that attained during the transport of DNA in conjugation and natural transformation (100 bases/sec). This raises many questions. Is there a unique mechanism of transport for all types of phages? How does the phage genome overcome the hydrophobic barrier of the host envelope? Is DNA transported as a free molecule or in association with proteins? Is such transport dependent on phage and/or host cell components? What is the driving force for transport? To understand the constraints that are associated with DNA ejection from the phage particle one has to take into consideration the initial state of this polymer inside the viral capsid. The dsDNA is highly condensed in the capsid where it occupies the whole volume. Although the size of the DNA strands strongly varies from one phage to the other—ranging from 4 kbp to 640 kbp (a mycobacteriophage) (Brussow and Hendrix, 2002)—its concentration within the capsid remains relatively constant, about 450 mg/ml (Bloomfield, 1996). As a mater of example phage T4 DNA (172 kbp) occupies a volume of 5 s 105 nm3 when confined in the 120 s 86 nm prolate capsid. When released in aqueous solution this volume is increased by a factor of 104 and the DNA has a radius of gyration of about 1000 nm. Cryo-electron microscopy and image reconstruction of isolated capsids from phage T7 (Cerritelli et al., 1997) (Agirrezabala et al., 2005a), T4 (Olson et al., 2001), epsilon 15 ( Jiang et al., 2006) and T5 (Effantin et al., 2006) have shown that the DNA adopts a quasi-crystalline structure within the capsid with the DNA strands spaced by 25 Å. As a corollary de-condensation of the DNA that occurs upon its ejection has to be highly regulated for the DNA not be entangled and to allow transcription to occur during its transfer into the host. The determination of an increasing number of high resolution structures of phage connectors should help understanding how the DNA is traversing the head to tail connection. Nearly thirty years ago, it was hypothesized that the forces driving DNA ejection from bacteriophage into the host cell, could come from the strong repulsions that the neighboring DNA portions experience due to being locally confined in their capsids (Earnshaw and Casjens, 1980; Riemer and Bloomfield, 1978). In other words, the DNA packaging machinery might introduce DNA under sufficient pressure in the capsid during morphogenesis to allow spontaneous ejection of the genome upon infection. This attractive hypothesis was recently reconsidered both theoretically and experimentally. We shall see that although not in contradiction with experimental data obtained on some phages this hypothesis does not hold up for others.

Bacteriophage DNA Transport

The different stages of transfer of dsDNA into the host Phage–host receptor recognition Gram-negative bacteria Infection of Gram-negative bacteria by tailed phages follows almost the same general scheme: the tail fibers make first contact with the cell surface. This reversible adsorption is followed by specific and irreversible binding of one or of a cluster of phage tail proteins to an outer membrane (OM) component. Almost all surface components including flagella, pili, capsules, lipopolysaccharides and proteins serve as receptor for phages. For the purpose of simplicity we will call the OM protein the “OM receptor” and the phage protein that binds to the OM protein the “adhesin.” Many OM receptors have been identified and their interactions with adhesins demonstrated mainly by genetic approaches or investigated starting from the purified proteins. Phages have developed very different host specificities through mutation and both intra- and inter-species recombination that are highlighted by the existence of pleiotropic adhesins that can recognize different outer membrane components [reviewed in (Heller, 1992) (Letellier et al., 1999; Vinga et al., 2006)]. The examples given below cover the three families of the Caudovirales order: the Syphoviridae, Podovirida and Myoviridae. Phages belonging to these families are characterized by long non contractile tails, short tails and long contractile tails respectively (Ackermann, 2003). The Syphoviridae phages lambda and T5 infect Escherichia coli. They are the only phages for which is was demonstrated that interaction with their purified receptor is sufficient to trigger DNA release from the phage particle thus allowing a mechanistic analysis of the DNA ejection process in vitro (Boulanger et al., 1996; Novick and Baldeschwieler, 1988) (see “What can be learned from in vitro studies of DNA ejection?”). The interaction between phage lambda and its receptor, the maltoporin LamB, is mediated by the tail protein gpJ (Randall-Hazelbauer and Schwartz, 1973). Analysis of lambda mutants has shown that host range mutations occurred in the last distal (5–10%) portion of gene J (Werts et al., 1994). Only the last 249 amino acids of the carboxy-terminal part of the adhesin protein could bind to LamB (Wang et al., 2000). The interaction between the C-terminal fragment of protein J and LamB was recently investigated using lipid bilayer experiments. This electrophysiological study confirmed the critical role of this domain for binding to LamB (Berkane et al., 2006). Host recognition by phage T5 is initiated by reversible binding of the long tail fibers to the O-antigen of the lipopolysaccharide (LPS) (Heller and Braun, 1982). Then the phage irreversibly binds to the iron-siderophore receptor FhuA by means of pb5, a 67.8 kDa protein located at the distal end of the phage tail (Heller, 1984; Heller and Schwarz, 1985; Krauel and Heller, 1991). The binding site for T5 and two other phages T1 and phi80 is located on one of the hydrophilic loops of FhuA (L4) facing the external medium (Killmann et al., 1995). The receptor-binding domain of pb5 is confined to ca. 200 aa within the N-terminal half of pb5 (Mondigler et al., 1996; Mondigler et al., 1995). pb5 was recently purified and characterized (Plançon, 2002). It has a high beta sheet content (51%), a property shared with the adsorption proteins of phage P22, PRD1 (Grahn et al.,

|

211

212

|

Letellier et al.

1999; Steinbacher et al., 1997; Xu et al., 2003) and of adenovirus (van Raaij et al., 1999) suggesting that this type of secondary structure could be relevant for the function of these proteins in phage infection. The functionality of purified pb5 was attested by its capacity to impair infection of E. coli cells by phage T5 and &80 and to inhibit growth of bacteria on iron-ferrichrome. pb5 formed an equimolecular complex with FhuA in vitro that showed remarkable stability since it was not dissociated by 2% of the detergent sodium lauryl sulfate even if the temperature was raised up to 70°C. The strength of the association between pb5 and FhuA is reminiscent of the irreversible nature of the phage adsorption step in the infectious process. The Salmonella phage P22 belongs to the Podoviridae family. Its capsid is surrounded by six tailspike proteins. Each of the tailspike is formed of a unique homotrimeric protein, gp9 consisting of a N-terminal head binding domain and a C-terminal domain that recognizes and cleave the O-antigen repeating units of the LPS (Iwashita and Kanegasaki, 1976). Crystal structures of both domains were solved at 1.8 Å resolution allowing the identification of the active-site topology (Steinbacher et al., 1997). These structures were recently docked into the three-dimensional map of the tail determined by cryo-electron microscopy and image reconstruction providing a molecular view of the events taking place during host attachment (Tang et al., 2005). The Myoviridae phage T4 is a typical example of host range adaptation. Recognition of the host is initiated by binding of six long tail fibers either to the terminal glucose of E. coli B-type LPS or to the outer membrane protein OmpC, found on K but not B strains of E. coli (Goldberg et al., 1994). The six short tail fibers which are folded under the baseplate then extend and attached irreversibly to the core region of the LPS. These tail fibers consist of an elongated trimer of gp12. Crystal structure at 1.5 Å resolution of a protease-stable fragment of gp12 has revealed the structure of the receptor-binding domain and allowed to identify the LPS-binding region (Thomassen et al., 2003). Gram-positive bacteria Studies of phages from Gram-positive bacteria represent an important challenge for the dairy industry since lytic phages are often responsible for failures of fermentation during the process of fermented milk products. Receptors from these phages differ from those of Gram-negative bacteria ranging from peptidoglycan elements to teichoic acids and lipoteichoic acid and associated proteins. Recently, several receptors from phages infecting Streptococcus thermophilus and Lactococcus lactis have been characterized (see Chapter 7, S. Mc Grath). Duplessis and Moineau have carried out the first bioinformatics analysis of the lytic S. thermophilus phage DT1 genome that allowed identifying the genetic determinant (adhesin) involved in the recognition of S. thermophilus hosts (Duplessis and Moineau, 2001). This analysis was extended to seven additional S. thermophilus phages. The deduced ORF18 was shown to be divided into three domains. The N-terminal part of the protein was conserved in all phages. A second domain contained collagen-like repeats. The third domain consisted of the C-terminal section of the protein and a variable region (VR2) carrying the host specificity. This three-domain organization was recently confirmed by the first description of the crystal structure of the adhesin of the Siphoviridae phage p2, a phage which infects L. lactis (Spinelli et al., 2006a; Spinelli et al., 2006b). The adhesin/re-

Bacteriophage DNA Transport

ceptor-binding protein (RBP) is located at the tip of the phage tail. It is homotrimeric and composed of three domains: the shoulder, a beta-sandwich attached to the phage; the neck, an interlaced beta-prism; and the receptor-recognition head, a seven-stranded beta-barrel. Structural similarity between the recognition-head domain of phage p2 adhesin and those of adenoviruses and reoviruses suggests that these viruses, despite evolutionary distant targets, lack of sequence similarity and the different chemical nature of their genomes (DNA versus RNA), might have a common ancestral gene (Spinelli et al., 2006a; Spinelli et al., 2006b). The receptor from the Bacillus subtilis phage SPP1 was recently identified and purified. A search of mutations in B. subtilis that blocked SPP1 infection identified the pha-2 locus and yueB as the gene coding for the SPP1 receptor. YueB is a 120 kDa membrane protein (Sao-José et al., 2004). Interestingly, YueB is an ortholog of Pip, a membrane protein previously shown to be required for infection of L. lactis by c2-species phages (Geller et al., 1993); (Monteville et al., 1994). Both receptors are composed of a large extracellular domain (ectodomain) and five predicted transmembrane segments localized at the carboxyl terminus region. A database survey for YueB/Pip-like proteins showed that they are widely distributed in Gram-positive bacteria. YueB and Pip are the sole membrane protein receptors currently known to be directly implicated in phage adsorption to Gram-positive hosts suggesting that YueB-like proteins represent a major family of receptors for virus infection of this group of bacteria. The YueB ectodomain was recently purified (Sao-Jose et al., 2006). Its properties are consistent with a dimeric YueB molecule forming an elongated fiber bound to the cytoplasmic membrane via ten transmembrane segments. The length of the fiber (36.5 nm) would be sufficient to span the ~30 nm thick peptidoglycan cell wall of the Gram-positive bacterium to expose a receptor domain to the medium. The purified protein was functional in vitro since it could bind to the tail tip of SPP1 and triggered the release of phage DNA. These novel structural features likely apply to the large family of proteins homologous to YueB that are found in Gram-positive bacteria and that share the same predicted membrane topology and secondary structure. Some general features of tail spike proteins Many of the phage tailspike adhesins studied so far (T4, p22) and attachment fibers of adenovirus and reovirus although having highly divergent sequences appear to share common structural characteristics. They are homotrimeric and rich in beta-sheet structure. They also share unusual topologies having an elongated shape, regions of extensive wrapping about a long, 3-fold symmetry axis and inter-subunit hydrophobic interactions. All these structural features may contribute to their thermostability, detergent-resistance and protease-resistance, conditions that have contributed to their survival in a wide range of environments (Mitraki et al., 2002; Weigele et al., 2005). Crossing the host envelope Once bound to its host the phage tail undergoes major conformational changes that are transmitted to the head-tail connector allowing the DNA to be ejected from the capsid and to be transferred to the host cytoplasm within a few seconds following binding. Available evidence indicates that the DNA of tailed phage crosses the host membranes linearly

|

213

214

|

Letellier et al.

nucleotide pair by nucleotide pair and in the direction opposite to that in which it was packaged (Gonzalez-Huici et al., 2004; Molineux, 2001; Saigo, 1975). The phage genome has to overcome several obstacles. In the case of Gram-negative bacteria it has to penetrate the outer membrane, avoid the endonucleases that are present in the periplasm, finds its way through the rigid structure of the peptidoglycan and then cross the hydrophobic barrier of the cytoplasmic membrane. Crossing the cell wall Crossing the cell wall requires the integrity of the peptidoglycan (or murein) to be locally ruptured. The cut-off value for the passage of globular proteins through the peptidoglycan is about 50 kDa (Demchick and Koch, 1996). The peptidoglycan is also transiently interrupted during its recycling. Although not excluded the passage of the DNA through these gaps appears not to be its main route. Indeed, increasing evidence exists that phages carry their own cell-wall degrading activities. Search in protein databases has revealed sequence similarities between catalytic domains of transglycosylases/murein hydrolases of bacteria and bacteriophages (Lehnherr et al., 1998). Muralytic activity was found in several phages from Gram-positive and Gram-negative bacteria (Moak and Molineux 2004). In phage T7 gp16 is part of the internal core of the capsid that is ejected into the cell at the initiation of infection (Molineux, 2001). In phage P22 gp4 is part of the neck of the particle and is essential for infectivity. The activity associated with Bacillus subtilis phage ø29 and its relatives lies in the terminal protein gp3 (Moak and Molineux 2004). In phage T4 a tail lysosyme encoded by gene 5 is responsible for digestion of the peptidoglycan. Its crystal structure was solved (Kanamaru et al., 2002). This protein is well conserved protein among the T4-type phages (Rossmann et al., 2004). It is part of the T4 puncturing device that allows the DNA to cross the envelop (Kanamaru et al., 2002) (see “The Myoviridae phage T4”). In Gram-positive bacteria the tail fiber of the L. lactis phage Tuc2009 was shown to possess a hydrolase activity (Kenny et al., 2004). Phages tail proteins as DNA “injectisome” In a number of cases it is now established that phage tail proteins protect the DNA throughout the envelope traversal. The participation of phage proteins in forming a DNA channel was suggested by fractionation of the envelope of phage T5-infected E. coli cells: pb2, the unique protein forming the T5 straight fiber was recovered in a membrane fraction containing proteins of both the inner and the outer membrane (Guihard et al., 1992). Furthermore pb2 was only present in the envelope under conditions where the DNA was transferred. Since isolated pb2 had a porin-like activity in planar lipid bilayer (Feucht et al., 1990) we concluded that pb2 would be able to cross both membranes and form the DNA channel. A corollary of this observation was that this channel would protect the DNA from periplasmic endonucleases. This proposal was further supported by in vitro reconstitution of the transport of T5 DNA in liposomes in which FhuA, the T5 OM receptor was reconstituted (Letellier et al., 1997; Plançon et al., 1997). Cryo-electron microscopy enabled to unequivocal demonstrated the delivery of the phage genome into the liposome (Lambert et al., 2000; Lambert et al., 1998). Furthermore, cryo-electron tomography associated with images reconstructions allowed visualizing pb2 crossing the

Bacteriophage DNA Transport

Figure 8.1 Cryo-electron micrograph illustrating the transfer of phage T5 DNA into a double lipid layer liposome reconstituted with the outer membrane receptor FhuA and loaded with 50 mM spermine. The DNA of several phages has crossed the two lipid layers and is condensed into a unique toroid occupying a volume smaller than the liposome The tip of the tails can apparently cross two juxtaposed membranes separated by not more than 5–6 nm. (Bar = 100 nm). Taken from (Böhm et al., 2001).

liposome membrane (Böhm et al., 2001) (see Figure 8.1). Interestingly, the tail straight fiber appeared shorter and larger when inserted into the liposome than on the isolated phage suggesting that its passage through the membrane was accompanied by large conformational changes. Yet even in its shorter conformation the straight fiber might be long enough to cross the whole E. coli envelope! pb2 (Mr = 123 kDa) is present in 5 to 6 copies in the phage structure (McCorquodale and Warner, 1988). Secondary structure predictions deduced from the sequence [Genbank Accession Number AY303686 (Letellier et al., 2004) and Jacquot et al., to be published] showed that pb2 contains many heptads repeats and is mainly organized as a coiled-coil structure except for a short stretch of hydrophobic amino acids (30 residues) in the C-terminal part. The amino acid composition of the C-terminal hydrophobic stretch resembles that of viral fusion peptides (Callebaut et al., 1997). We expressed in E. coli a truncated form of pb2 missing 80% of the N-terminal sequence but including this hydrophobic stretch. The protein, which is oligomeric in solution, showed multiple interesting features. It induced instantaneous aggregation of bacteria. It inserted spontaneously into liposomes, and permeabilized them to small solutes. Light scattering and fluorescence microscopy experiments showed an important increase in the liposomes

|

215

216

|

Letellier et al.

size upon addition of pb2 suggesting that the peptide triggered the fusion of the vesicles ( Jacquot et al., to be published). What would be the in vivo function of this fusogenic sequence? pb2 is known to change conformation upon binding of the phage to FhuA (Feucht et al., 1990). Furthermore its morphology is modified (Böhm et al., 2001). It is thus tempting to propose that this change of conformation renders the fusogenic peptide accessible to the host envelope and allows its hydrophobic domain to insert in the outer membrane and to fuse the outer and cytoplasmic membrane. The fusogenic domain of pb2 would have two functions: ensuring the contact between the two membranes so that to protect the DNA from periplasmic nucleases and by spanning both membranes it would form the DNA channel (Letellier et al., 2004). pb2 shares striking common characteristics with gpH (Mr = 92 kDa), the tape measure protein that determines the length of phage lambda tail (Heller, 1992; Hendrix, 1988; Katsura, 1990). Both proteins are present at the same number of copies per phage particle and lose about a hundred amino acids by proteolytic cleavage at their C-terminal end during phage assembly. Secondary structure predictions suggest gpH to be mainly organized as a coiled-coil structure except for a short stretch of hydrophobic amino acids in the Cterminal part that also resembles that of viral fusion peptides (Callebaut et al., 1997). gpH is not only involved in determining the length of the phage tail but also in DNA transfer into the host cell (Roessner and Ihler, 1984; Scandella and Arber, 1976). gpH like pb2 is resistant to detergent treatment in phage particles but not in phage ghosts or purified tails. On the basis of the analogy between pb2 and gpH, we speculated that the pb2 oligomer forms a five or six-stranded, partially coiled coil structure that spans the tail of phage T5, its N-terminal domain contacting the head-tail connector and its C-terminal domain forming the straight fiber (Letellier et al., 2004) ( Jacquot et al., to be published). These tape-measure proteins would therefore be good candidates for triggering the ejection signal to the phage head, allowing its opening and the release of the DNA. Recent data obtained on the Podoviridae phage P22 strongly support this proposal. The three dimensional structure of the P22 tail recently determined by cryo-electron microscopy has revealed the localization of the different components of the tail among which gp26 (24.7 kDa) which forms a needle at the distal end of the tail This needle has a length of 21.4 nm and a diameter of about 3.8 nm (large enough for the DNA to pass through) (Tang et al., 2005). Sequence analysis and secondary structure prediction indicate the presence of long helices and multiple heptad repeats (Andrews et al., 2005) (Tang et al., 2005). In agreement with this analysis, Andrews et al. showed that the recombinant gp26 protein exists in solution as a very stable elongated trimeric coiled-coil ~21 nm in length. By analogy with the coiled-coil structures that have been observed in viral proteins the authors speculate the hydrophobic regions of gp26 may represent a prototype of phage fusion peptide. Tang et al. further discussed the insertion of the gp26 needle into the host envelope underline that the needle would be long enough to span the envelope. Tape-measure proteins are found in almost all phages with flexible non-contractile tails, including mycobacteriophages (Abuladze et al., 1994) and lactococcal bacteriophages (Pedersen et al., 2000), and a linear relationship between the number of residues and the length of the tail is always observed. In a very recent review (Cornelis et al., 2006) presented a comparative analysis of the length control of bacteriophage tails by tape measure proteins,

Bacteriophage DNA Transport

of the hook of the flagellum and of the needle of the type III secretion injectisome. The bacterial flagellum allows motility, whereas the injectisome allows bacteria docked on the surface of a eukaryotic cell membrane to inject effector proteins across the bacterial membranes and the eukaryotic cell membrane. The injectisome needle is functionally related to the tail of bacteriophages, in that they both represent hollow tubes for the injection of macromolecules—DNA or proteins—into cells. The authors conclude that the length of these 3 structures has to be controlled: for the hook of the flagellum, it is controlled for mechanical reasons; for the injectisome, the needle length evolved to match specific structures at the bacterial and host cell surfaces; for control of the length of the tail of bacteriophages, the reason is presumably the same as for the injectisome; since the phage tail protein fulfils a function similar to that of the needle of the injectisome: its length has probably also evolved to match the lipopolysaccharides and proteins of the host envelope. It remains to be now clearly established whether phage tape measure proteins are indeed injectisomes. Energetics of DNA transport The inward movement of the long DNA molecule cannot occur without being driven by an input of energy. Two general hypothesis of DNA transport have been formulated. The first one proposed that the packaging machinery might simply introduce DNA under sufficient pressure in the capsid during morphogenesis to allow spontaneous ejection of the genome upon infection (Earnshaw and Harrison, 1977; Riemer and Bloomfield, 1978). The second hypothesis proposed that the electrochemical gradient of protons across the host cytoplasmic membrane would be the driving force for DNA transport (Grinius, 1980). The chemiosmotic theory of DNA transport On the basis of the chemiosmotic theory of Peter Mitchell (Mitchell and Moyle, 1967), (Grinius, 1980) postulated a “universal” mechanism of DNA transport in which the electrochemical gradient of protons generated across the host cytoplasmic membrane by the electron transfer chain would be the driving force for DNA transport. The polyanionic DNA molecule would be co-transported with protons down their gradient. This attractive hypothesis was supported by experiments showing that phage T4 DNA transport took place only in energized membranes and above a threshold of membrane potential of –90 mV (Labedan and Goldberg, 1979). However, further experiments revealed that the membrane potential was not required for transport of the DNA but rather for opening of a voltage-gated “DNA channel” (Boulanger and Letellier, 1988). This hypothesis also failed to explain T5 DNA transport since it was shown that the DNA could cross the cytoplasmic membrane of de-energized cells (Filali Maltouf and Labedan, 1983). Obviously, the chemiosmotic theory did not provide a general framework to understand phage DNA transport. Pressure inside the capsid and DNA ejection Nearly twenty years ago, it was hypothesized that forces driving DNA ejection from bacteriophage into the host cell could come from the strong repulsions that the neighboring DNA strands experience as a consequence of their confinement in the capsid (Earnshaw and Casjens, 1980; Riemer and Bloomfield, 1978). Several models have recently been

|

217

218

|

Letellier et al.

elaborated (Evilevitch et al., 2003; Kindt et al., 2001; Purohit et al., 2004; Purohit et al., 2003). They propose that DNA ejection is directly governed by the difference between the pressure inside and outside the capsid. The outside pressure is related, in vivo, to the high concentrations of proteins inside the bacterial cytoplasm and in vitro, to the osmotic pressure of the solution surrounding the phage. The possibility to package phage &29 DNA in vitro has allowed the group of Bustamante to measure the forces generated by the motor of encapsidation throughout the packaging of a single DNA molecule into an isolated capsid (Smith et al., 2001). At maximum the motor pulled with a force of about 60 picoNewtons roughly corresponding to a capsid pressure of 6 MPa (60 Atmospheres). In vitro experiments were reported that support the pressure dependent model of DNA ejection. Evilevitch et al. (Evilevitch et al., 2005; Evilevitch et al., 2003) measured the in vitro ejection of phage lambda DNA triggered by binding of the phage to its purified receptor LamB. Ejection appeared complete. The interpretation was that capsids are permeable to water and salt ions, so that there is no difference in hydrostatic pressure between the inside and outside of the capsid, and osmotic equilibrium is also maintained. The pressure difference between the capsid and the solution is therefore only associated with the confinement of the DNA. When the capsid is opened, ejection proceeds until this pressure difference falls to zero. The corollary is that an increase of external osmotic pressure should decrease and even suppress DNA ejection. This is indeed what they observed upon increasing the external osmotic pressure by addition of an osmotic stress agent (polyethylene glycol 8000): the fraction of lambda DNA ejected drops from 1 to 0 on increase of the external osmotic pressure from 0 to 20 Atmospheres. In view of these data it is tempting to conclude that phage DNA ejection is a passive process, being only driven by the pressure gradient. However, this hypothesis felt to explain the in vitro complex behavior of phage T5 DNA whose ejection occurs stepwise (see below). Moreover, it seems difficult to extrapolate the in vitro observations made on phage lambda to in vivo situations since, as shown below, phages have developed very different strategies for injecting their DNA into the host. An overview of the different strategies adopted by phages to inject their DNA into the host The Myoviridae phage T4 The coliphage T4 is one of the most complex tailed phages. Its prolate capsid contains a dsDNA of 172 kbp. T4 shares with other members of the Myoviridae family one of the most elaborated tail structure that consists of more than 20 proteins each present in multiple copies. A combination of X-ray crystallography data fitted into cryo-electron microscopy maps has allowed deciphering the structure of the components of the phage baseplate and tail structure providing the first molecular view of the contraction mechanism [(Kanamaru et al., 2002; Kostyuchenko et al., 2005; Rossmann et al., 2004) and reference herein]. Infection is initiated by interaction of the long tail fibers with the LPS or the porin OmpC. This is followed by the irreversible attachment of the short fibers and the conformational changes of the baseplate from the hexagonal to star conformation. This initiates the contraction of the tail sheath which drives the inner tail tube through the envelope close to the cytoplasmic membrane. The DNA then crosses the membranes in about 30 sec at 37°C.

Bacteriophage DNA Transport

This represents the highest rate (approximately 4000 bp/sec) observed for DNA transport. As a matter of comparison, the rate of the ATP-dependent DNA translocation through the connector during packaging is of the order of 140 bp/sec (Guasch et al., 2002). The crystal structure of the centre of the baseplate called by Rossmann and collaborators “the cell-puncturing device” was solved (Kanamaru et al., 2002). It consists of two proteins, gp5 and gp27, that form a trimeric complex. The structure of the complex resembles a torch with a length of 19 nm. The gp27 trimer forms the head, a hollow cylinder with an internal diameter of 3 nm. The gp5 trimer, which forms the handle, consists of three domains: the carboxy-terminal domain is a triple-stranded long beta-helix, the middle is a lysozyme domain which serves to digest the peptidoglycan and the amino-terminal, an antiparallel beta-barrel domain that inserts into the cylinder formed by gp27. The diameter of gp27 is of a size that can accommodate a double-stranded DNA helix. The authors proposed that upon attachment of the base plate to the cell surface, the tail sheath contraction exerts a force onto the tail tube towards the cell membrane. This force would be transmitted through the gp27 cylinder and the N-terminal domain of gp5 to the beta-helix, allowing the later to act as a membrane-puncturing device. As the contraction of the tail sheath progresses, the beta helix would span the outer membrane allowing the lyzozyme domains to digest the peptydoglycan and the penetration of the tail tube close to the inner membrane for injection of the DNA into the host cytoplasm. How then is the DNA crossing the cytoplasmic membrane is not discussed by the authors. Could it be that gp5 also crosses the inner membrane? Its N-terminal triple-stranded beta-helix domain—11 nm—would be long enough. We previously concluded that T4 DNA transport takes place through a phage protein forming a channel that opened above a threshold of membrane potential and remained opened only during DNA transport (Boulanger and Letellier, 1988).We speculate that the membrane potential dictates the conformation of gp5 and its insertion in the membrane. What is then the “driving force” for T4 DNA transport? We are having no satisfying answer to that question. A theoretical model was proposed that described the dynamics of DNA release as a reptation process through the phage tail, the driving force being the decrease of the condensed DNA free energy (Gabashvili and Grosberg, 1992). T4 DNA transport takes place within 30 sec, a too short delay to permit transcription of the phage genome so that it is unlikely that the DNA is mechanically pulled by its own polymerase. Alternatively, the DNA might be pulled and condensed while entering the cytoplasm by attachment of host histones like HU (Rouviere-Yaniv et al., 1979). The Podoviridae phage T7 T7 consists of an icosahedral capsid containing a morphologically distinct structure, the internal core which is formed of multimers of three proteins: gp16, gp15, gp14. The dsDNA (40 kbp) is spooled around the long axis of the core structure in about six co-axial shells (Agirrezabala et al., 2005b; Cerritelli et al., 1997) Apart from six fibers each formed by gp17, the tail is composed by two proteins gp11 and gp12. T7 is the only one phage for which a detailed description of the mechanism of entry of its genome has been possible (Molineux, 2001). In brief, the assay used to follow the kinetics of T7 DNA entry in the cytoplasm consists of an E. coli strain overproducing Dam, an enzyme that methylates the GATC sites of the incoming T7 DNA. Methylation renders the T7 genome sensitive to the enzyme

|

219

220

|

Letellier et al.

DpnI, which specifically cuts methylated GATC sequences. Therefore, methylation defines the time at which the different GATC sequences reach the host cytoplasm. The T7 genome enters the cell in three distinct phases. Following binding of the short tail fibers to the host surface and to a yet unknown OM receptor, the internal core proteins gp14, gp15 and gp16 are ejected and recovered in the host envelope. It has been proposed that they would form the DNA conduit across the envelope with the transglycolytic activity of gp16 contributing in crossing the peptidoglycan. About 850 bp of the left end of the genome are transferred into the cytoplasm by a transcription-independent mechanism allowing the internalization of three promoters for the E. coli RNA polymerase (RNAP). DNA translocation is then shut down. The internalized promoters allow the host RNAP to initiate transcription and to pull about 20% (7 kbp) of the genome into the cell. T7 RNA polymerase is then synthesized and the remaining DNA is transcripted by the phage enzyme [reviewed in (Molineux, 2001)]. The internalization of the genome takes about 10 min at 30°C which is slow compared to phage T4 (30 sec). This slow rate suggests that the limiting step in transport could be the rate of transcription. In vitro studies using laser tweezers (Wang et al., 1998) have elegantly demonstrated that a transcribing RNA polymerase is a strong and efficient molecular motor. The characteristics of this motor would permit the phage genome to be mechanically pulled into the cytoplasm by the polymerase. However, this proposal failed to explain how T7 mutants that did not require transcription internalize their genome. Interestingly mutations that uncoupled DNA transport from transcription were all found in a 130 residue segment of the 1318 amino acid sequence of gp16. Many questions remain unsolved: How do gp16 and the other internalized core proteins contribute to pull/clamp the DNA? How does internalization of the leading sequence take place and by which mechanism is DNA translocation shut down after transfer of this sequence? Whatever the answer been we have to admit that the above observations are not consistent with the proposal that the gradient of pressure between the inside and the outside of the capsid governs the transfer of T7 DNA (Kindt et al., 2001). The Syphoviridae phage Ø29 The Bacillus subtilis phage Ø29 has a prolate capsid 41.5 nm long and 31.5 nm width containing a 19 kbp dsDNA and a short tail 32.5 nm long. The mechanism of encapsidation of its genome was analyzed extensively and the development of an efficient in vitro model of encapsidation has allowed determining the forces involved in packaging of the DNA into single capsids (see above). In contrast, it is only recently that the transfer of its DNA into the host has started to be studied. Ø29 DNA entry into the cell was followed by monitoring the binding of the viral protein p6 to the entering DNA. Ø29 DNA enters the cell with a right to left polarity. It is a two-step process. The transport of the first 60% of the genome appears to occur independently of an external energy source suggesting that it is driven by the packaging pressure built inside the capsid. This region of the genome codes for several proteins including p17 that was proposed to be a component of the molecular machinery that pulls the remaining DNA into the cytoplasm (Gonzalez-Huici et al., 2004). The use of several energy poisons and inhibitors suggest that the second step is transcription-independent, requiring negatively supercoiled DNA and an electrochemical gradient of protons (Gonzalez-Huici et al., 2006). More data are required to precisely elucidate this complex injection mechanism.

Bacteriophage DNA Transport

The Syphoviridae phage T5 T5 targets E. coli. It is characterized by a 90 nm large icosahedral capsid, containing a 121 750 bp dsDNA and by a flexible non-contractile tail 250 nm in length. Reconstruction images from cryo-electron micrographs have highlighted the unusual trimeric symmetry of the tail (Effantin et al., 2006). Several features make T5 an unusual and remarkable phage [reviewed in (McCorquodale and Warner, 1988)]. Its genome is the largest of the T-odd viruses. It carries single-stranded interruptions at genetically defined positions on one of the DNA strands as well as large terminal redundancies in the form of 10 160 direct repeats. T5 also contains some of the strongest known prokaryotic promoters (McCorquodale and Warner, 1988). Remarkably, most of these features including the two-steps mechanism of DNA entry (see below) have been known since the 1960s and 1970s, but we still have a poor understanding of the molecular events occurring during the infection process. Its genome has been recently sequenced (NCBI Entrez IDs AY587007, AY692264 and AY543070) (Wang et al., 2005) thus offering new perspectives in the comprehension of its infectious process. A unique feature of phage T5 is that its genome is transferred in two steps (Lanni, 1968); 8% of the DNA (First Step Transfer or FST DNA) enters first the cytoplasm; then, there is a pause of about 4 minutes (at 37°C) during which proteins encoded by this fragment are synthesized. Two of these proteins (A1 and A2) are required for the transfer of the remaining DNA (92%) or second step transfer (SST) DNA. A1 (Mr = 64 kDa) is implicated in at least three functions: the degradation of host DNA, the shutoff of preearly transcription and together with the product of gene A2 in completion of phage DNA transfer. A2 (Mr = 15 kDa) is a homodimeric DNA-binding polypeptide that appears to form an oligomeric structure with A1 (McCorquodale et al., 1977; Snyder and Benzinger, 1981). If one prevents synthesis of A1 and A2, then the SST DNA still connected to the phage head and attached to the injected FST fragment, crosses the outer and inner membranes without being degraded suggesting that the DNA is protected from periplasmic nucleases during its transport (Labedan et al., 1973). The reason why phage T5 transfers its DNA in two steps remains unclear. The arrest of transfer at the FST stage is not due to newly synthesized phage proteins encoded by the FST DNA since it takes place in the presence of inhibitors of protein synthesis. Cloning and sequencing of the terminal region of the FST DNA fragment revealed that it contained several large inverted repeats forming potential stem-and-loop structures that could jam the DNA during its transfer (Heusterspreute et al., 1987). Data from Davison and Brunel (Davison and Brunel, 1979a; Davison and Brunel, 1979b) suggest that T5 specifies a restriction protection function, encoded by the FST DNA that may protect the viral DNA against several bacterial restriction systems (EcoRI, EcoKI, EcoPI). Whether host factors are involved in the transfer in two-steps in vivo remains to be elucidated since phage T5 DNA ejection also proceeds in steps when triggered in vitro by interaction with its receptor (see below). (Filali Maltouf and Labedan, 1983) previously demonstrated that both the FST and SST DNA could be transferred in cells deprived of metabolic energy sources (electrochemical gradient of protons and ATP). This again raises the question of the force driving DNA transport. Relief of the capsid internal pressure is likely to explain the passive transfer of the FST DNA but not that of the SST DNA. Indeed, the capsid and tail of phage T5 can be sheared off the bacterial cell surface once the FST DNA has been transported into the cytoplasm.

|

221

222

|

Letellier et al.

Under these conditions the SST DNA (approximately 27 micrometer in length) is floating in the surrounding medium, but is still attached to the FST DNA and crossing the envelope (Labedan et al., 1973). If one provides the energy for the synthesis of A1 and A2, then the free DNA is transported through the membranes. It is therefore obvious that the DNA has to be pulled inside the cytoplasm. One reasonable hypothesis is that bacterial histones and/or the DNA binding protein A2 encoded by the FST DNA, contribute to this transport. The possibility to mimic T5 DNA transport into liposomes should help to validate this hypothesis. T5 DNA transport is accompanied by a transient permeabilization of the inner membrane to cytoplasmic potassium that strictly follows the timing of DNA penetration. The characteristics of this efflux led us to propose that is was due to the opening in the cytoplasmic membrane of a channel, the function of which was to transfer the DNA (Boulanger and Letellier, 1992). The participation of the tail fiber phage protein pb2 in forming the DNA channel was proposed on the basis of electrophysiological studies (Feucht et al., 1990) and further supported by fractionation of the envelope of T5-infected E. coli cells (Guihard et al., 1992) (see also Phages tail proteins as DNA "injectisome"). How this channel-forming protein inserts in the cytoplasmic membrane, opens upon FST DNA transfer then closes during the synthesis of the FST-encoded proteins and reopens transiently during the transfer of the SST DNA remains speculative. We observed that calcium regulated the membrane permeability changes occurring upon the transfer of the DNA. Decreasing calcium below 0.1 mM depleted the bacteria of K+, caused a complete membrane depolarization and a decrease of cytoplasmic ATP. To account for these observations we propose that calcium controls the conformation of the protein forming the DNA channel and that below 0.1 mM the channel remained opened. Alternatively calcium could be required for the proper interaction of A1 and A2 to incoming DNA This would create an energetic state of the host unfavorable to the synthesis of phage components and lead to abortion of the infectious process (Bonhivers and Letellier, 1995). What can be learned from in vitro studies of DNA ejection? An attractive feature of T5 is that DNA release from the phage can be triggered in vitro by mere interaction of the virus with its purified OM receptor FhuA solubilized in detergent (Boulanger et al., 1996). DNA release was demonstrated spectroscopically using a fluorescent DNA intercalant YO-PRO-1 (Carlsson et al., 1994) the fluorescence of which is increased in proportion to the amount of DNA freed from the capsid. It was concluded that each T5 molecule ejected virtually all its DNA in the surrounding medium in less than a few seconds but for all phages to eject their DNA more than 20 min were necessary. We have now a better understanding of these results. We investigated by fluorescence microscopy and in real time the dynamics of FhuA-induced DNA ejection of individual T5 phage particles adsorbed onto the chamber of a microfluidic cell (Mangenot et al., 2005). The length of the fluorescently stained DNA was measured at different stages of its ejection after being stretched in a hydrodynamic flow. DNA ejection was more complex than expected. The rate of DNA ejection was high reaching 75000 bp/sec. For some phages DNA release occurred in one step in a few hundred msec. For others, ejection proceeded in a stepwise fashion and was arrested for variable times at discrete positions on the DNA

Bacteriophage DNA Transport

near the genetically defined single-stranded interruptions on the genome. We concluded that the peculiar arrangement of the DNA at the nicks may represent an energetic barrier that must be overcome for ejection to take place. The time elapsed between the different stages of ejection would be related to the pressure inside the capsid and consequently to the amount of non-ejected DNA. These data were confronted to those obtained on an ensemble of phage particles using light scattering which allowed determining the energetic parameters associated with DNA ejection (de Frutos et al., 2005). An activation energy of the order of 70 kT must be overcome to allow the complete DNA ejection. The complex shape of the kinetics of ejection was analyzed using a phenomenological model based on the multistep process described above. It was concluded that passing from one stage to another requires thermal activation of the pressurized DNA. Both thermal and pressure effects contribute to shorten or to lengthen the pause time between the different stages of ejection explaining why the T5 DNA ejection is so slow compared to other types of phages. How this stepwise ejection observed in vitro is relevant to the complex physiological process remains to be established. The physiological role of the nicks is also still not elucidated. The stop sites may be involved in regulating DNA encapsidation/transfer in vivo. Nickless mutants of T5 have been described that should permit testing the different hypothesis (Rogers et al., 1979a; Rogers et al., 1979b). Concluding remarks We are far from understanding the mechanism by which a viral genome can be delivered into a bacterial cell. In vitro studies have been particularly stimulating allowing to approach, at a molecular level, the functioning of receptors as well as the mechanism of DNA transport and condensation. Yet multiple approaches are necessary. First, as shown above, studies on an ensemble of phage particles may obscure the behavior of individual viral particles. Second, this reductionist approach does not always allow taking in account the eventual contribution of other viral or host partners. In vivo studies on phage DNA delivery are only slowly progressing. This is in contrast to the increasing knowledge we have on the encapsidation process. Genetics of phage has indeed contributed to identify and characterize phage receptors but similar approaches are lacking to analyze DNA transport since only few mutations affecting this specific step have until now been characterized. With the increasing interest that scientists for all fields show to phages we expect some of them to shade some light on the mechanism of DNA delivery! Acknowledgments Much of the work described has beneficiated from the support of the CNRS mostly through the programs “Physique et Chimie du Vivant” and “Dynamique et Réactivité des Assemblages Biologiques.” We thank W. Baumeister, M. Bonhivers, A. Ghazi, A. Gillings, P. Jacquot, C. Janmot, O. Lambert, M. le Maire, S. Mangenot, G. Moeck, M. Santamaria, J-L Rigaud, E. Raspaud and M. de Frutos for their contributions at different phases of the work on phages and receptors. References

Abuladze, N.K., Gingery, M., Tsai, J., and Eiserling, F.A. (1994). Tail length determination in bacteriophage T4. Virology 199, 301–310.

|

223

224

|

Letellier et al.

Ackermann, H.W. (2003). Bacteriophage observations and evolution. Res. Microbiol. 154, 245–251. Agirrezabala, X., Martin-Benito, J., Caston, J.R., Miranda, R., Valpuesta, J.M., and Carrascosa, J.L. (2005a). Maturation of phage T7 involves structural modification of both shell and inner core components. Embo J. 24, 3820–3829. Agirrezabala, X., Martin-Benito, J., Valle, M., Gonzalez, J.M., Valencia, A., Valpuesta, J.M., and Carrascosa, J.L. (2005b). Structure of the connector of bacteriophage T7 at 8A resolution: structural homologies of a basic component of a DNA translocating machinery. J. Mol. Biol. 347, 895–902. Andrews, D., Butler, J.S., Al-Bassam, J., Joss, L., Winn-Stapley, D.A., Casjens, S., and Cingolani, G. (2005). Bacteriophage P22 tail accessory factor GP26 is a long triple-stranded coiled-coil. J. Biol. Chem. 280, 5929–5933. Benson, S.D., Bamford, J.K., Bamford, D.H., and Burnett, R.M. (1999). Viral evolution revealed by bacteriophage PRD1 and human adenovirus coat protein structures. Cell 98, 825–833. Berkane, E., Orlik, F., Stegmeier, J.F., Charbit, A., Winterhalter, M., and Benz, R. (2006). Interaction of bacteriophage lambda with its cell surface receptor: an in vitro study of binding of the viral tail protein gpJ to LamB (Maltoporin). Biochemistry 45, 2708–2720. Bloomfield, V. (1996). DNA condensation. Curr. Opin. Struct. Biol. 6, 331–341. Böhm, J., Lambert, O., A.S., F., Letellier, L., W., B., and Rigaud, J.L. (2001). FhuA-mediated phage genome transfert into liposomes: A cryo-electron tomography study. Curr. Biol. 11, 1168–1175. Bonhivers, M., and Letellier, L. (1995). Calcium controls phage T5 infection at the level of the Escherichia coli cytoplasmic membrane. FEBS Lett. 374, 169–173. Boulanger, P., le Maire, M., Bonhivers, M., Dubois, S., Desmadril, M., and Letellier, L. (1996). Purification and structural and functional characterization of FhuA, a transporter of the Escherichia coli outer membrane. Biochemistry 35, 14216–14224. Boulanger, P., and Letellier, L. (1988). Characterization of ion channels involved in the penetration of phage T4 DNA into E. coli cells. J. Biol. Chem. 263, 9767–9775. Boulanger, P., and Letellier, L. (1992). Ion channels are likely to be involved in the two steps of phage T5 DNA penetration into Escherichia coli cells. J. Biol. Chem. 267, 3168–3172. Brussow, H., and Hendrix, R.W. (2002). Phage genomics: small is beautiful. Cell 108, 13–16. Callebaut, I., Labesse, G., Durand, P., Poupon, A., Canard, L., Chomilier, J., Henrissat, B., and Mornon, J.P. (1997). Deciphering protein sequence information through hydrophobic cluster analysis (HCA): current status and perspectives. Cell Mol. Life Sci. 53, 621–645. Carlsson, C., Larsson, A., Jonsson, M., Albinsson, B., and Norden, B. (1994). Optical and photophysical properties of the oxazole yellow DNA probes YO and YOYO. J. Phys Chem. 98, 10313–10321. Casjens, S.R. (2005). Comparative genomics and evolution of the tailed-bacteriophages. Curr. Opin. Microbiol. 8, 451–458. Cerritelli, M.E., Cheng, N., Rosenberg, A., McPherson, C., Booy, F., and Steven, A. (1997). Encapsidated conformation of bacteriophage T7 DNA. Cell 91, 271–280. Cornelis, G.R., Agrain, C., and Sorg, I. (2006). Length control of extended protein structures in bacteria and bacteriophages. Curr. Opin. Microbiol. 9, 201–206. Davison, J., and Brunel, F. (1979a). Restriction insensitivity in bacteriophage T5. I. Genetic characterization of mutants sensitive to EcoRI restriction. J. Virol. 29, 11–16. Davison, J., and Brunel, F. (1979b). Restriction insensitivity in bacteriophage T5. II. Lack of EcoRI modification in T5+ and T5ris mutants. J. Virol. 29, 17–20. de Frutos, M., Letellier, L., and Raspaud, E. (2005). DNA ejection from bacteriophage T5: analysis of the kinetics and energetics. Biophys. J. 88, 1364–1370. Demchick, P., and Koch, A.L. (1996). The permeability of the wall fabric of Escherichia coli and Bacillus subtilis. J. Bacteriol. 178, 768–773. Dreiseikelmann, B. (1994). Translocation of DNA across bacterial membranes. Microbiol. Rev. 58, 293–316. Duplessis, M., and Moineau, S. (2001). Identification of a genetic determinant responsible for host specificity in Streptococcus thermophilus bacteriophages. Mol. Microbiol. 41, 325–336. Earnshaw, W.C., and Casjens, S.R. (1980). DNA packaging by the double-stranded DNA bacteriophages. Cell 21, 319–331. Earnshaw, W.C., and Harrison, S.C. (1977). DNA arrangement in isometric phage heads. Nature 268, 598–602.

Bacteriophage DNA Transport

Effantin, G., Boulanger, P., Neumann, E., Letellier, L., and Conway, J. (2006). Bacteriophage T5 structure reveals similarities with HK97 and T4 suggesting evolutionary relationships. J. Mol. Biol. 361, 993–1002. Evilevitch, A., Gober, J., Philips, M., Knobler, C.M., and Gelbart, W.M. (2005). Measurements of DNAlengths remaining in a viral capsid after osmotically suppressed partial ejection. Biophys. J. 88, 751–756. Evilevitch, A., Lavelle, L., Knobler, C.M., Raspaud, E., and Gelbart, W.M. (2003). Osmotic pressure inhibition of DNA ejection from phage. Proc. Natl. Acad. Sci. USA 100, 9292–9295. Feucht, A., Schmid, A., Benz, R., Schwarz, H., and Heller, K.J. (1990). Pore formation associated with the tail-tip protein pb2 of bacteriophage T5. J. Biol. Chem. 265, 18561–18567. Filali Maltouf, A., and Labedan, B. (1983). Host cell metabolic energy is not required for injection of bacteriophage T5 DNA. J. Bacteriol. 153, 124–133. Gabashvili, I.S., and Grosberg, A. (1992). Dynamics of double-stranded DNA reptation from bacteriophage. J. Biomol. Struct. Dynamics 9, 911–920. Geller, B.L., Ivey, R.G., Trempy, J.E., and Hettinger-Smith, B. (1993). Cloning of a chromosomal gene required for phage infection of Lactococcus lactis subsp. lactis C2. J. Bacteriol. 175, 5510–5519. Goldberg, E., Grinius, L., and Letellier, L. (1994). Recognition, Attachement, and Injection. In: Molecular Biology of Bacteriophage T4, K.J.D. ed. (Washington, American Society of Microbiology). Gonzalez-Huici, V., Salas, M., and Hermoso, J. (2004). The push-pull mechanism of bacteriophage phi29 DNA injection. Mol. Microbiol. 52, 529–540. Gonzalez-Huici, V., Salas, M., and Hermoso, J.M. (2006). Requirements for Bacillus subtilis bacteriophage Phi29 DNA ejection. Gene. 374, 19–25. Grahn, A.M., Caldentey, J., Bamford, J.K., and Bamford, D.H. (1999). Stable packaging of phage PRD1 DNA requires adsorption protein P2, which binds to the IncP plasmid-encoded conjugative transfer complex. J. Bacteriol. 181, 6689–6696. Grinius, L. (1980). Nucleic acid transport driven by ion gradient across cell membrane. FEBS Lett. 113, 1–10. Guasch, A., Pous, J., Ibarra, B., Gomis-Ruth, F.X., Valpuesta, J.M., Sousa, N., Carrascosa, J.L., and Coll, M. (2002). Detailed architecture of a DNA translocating machine: the high-resolution structure of the bacteriophage phi29 connector particle. J. Mol. Biol. 315, 663–676. Guihard, G., Boulanger, P., and Letellier, L. (1992). Involvement of phage T5 tail proteins and contact sites between the outer and inner membrane of Escherichia coli in phage T5 DNA injection. J. Biol. Chem. 267, 3173–3178. Heller, K.J. (1984). Identification of the phage gene for host receptor specificity by analyzing hybrid phages of T5 and BF23. Virology 139, 11–21. Heller, K.J. (1992). Molecular interaction between bacteriophage and the gram-negative cell envelope. Arch. Microbiol. 158, 235–248. Heller, K.J., and Braun, V. (1982). Polymannose O-antigens of Escherichia coli, the binding sites for the reversible adsorption of bacteriophage T5+ via the L-shaped tail fibers. J. Virol. 41, 222–227. Heller, K.J., and Schwarz, H. (1985). Irreversible binding to the receptor of bacteriophages T5 and BF23 does not occur with the tip of the tail. J. Bacteriol. 162, 621–625. Hendrix, R.W. (1988). Tail length determination in double-stranded DNA bacteriophages. Curr. Top. Microbiol. Immunol. 136, 21–29. Heusterspreute, M., Ha-Thi, V., Tournis-Gamble, S., and Davison, J. (1987). The first-step transfer-DNA injection-stop signal of bacteriophage T5. Gene 52, 155–164. Iwashita, S., and Kanegasaki, S. (1976). Enzymic and molecular properties of base-plate parts of bacteriophage P22. Eur. J. Biochem. 65, 87–94. Jiang, W., Chang, J., Jakana, J., Weigele, P., King, J., and Chiu, W. (2006). Structure of epsilon15 bacteriophage reveals genome organization and DNA packaging/injection apparatus. Nature 439, 612–616. Kanamaru, S., Leiman, P.G., Kostyuchenko, V.A., Chipman, P.R., Mesyanzhinov, V.V., Arisaka, F., and Rossmann, M.G. (2002). Structure of the cell-puncturing device of bacteriophage T4. Nature 415, 553–557. Katsura, I. (1990). Mechanism of length determination in bacteriophage lambda tails. Adv. Biophys. 26, 1–18. Kenny, J.G., Mc Grath, S., Fitzgerald, G.F., and van Sinderen, D. (2004). Bacteriophage Tuc2009 encodes a tail-associated cell wall-degrading activity. J. Bacteriol. 186, 3480–3491.

|

225

226

|

Letellier et al.

Killmann, H., Videnov, G., Jung, G., Schwarz, H., and Braun, V. (1995). Identification of receptor binding sites by competitive peptide mapping: phages T1, T5 and &80 and colicin M bind to the gating loop of FhuA. J. Bacteriol. 177, 694–698. Kindt, J., Tzlil, S., Ben-Shaul, A., and Gelbart, W.M. (2001). DNA packaging and ejection forces in bacteriophage. Proc. Natl. Acad. Sci. USA 98, 13671–13674. Kostyuchenko, V.A., Chipman, P.R., Leiman, P.G., Arisaka, F., Mesyanzhinov, V.V., and Rossmann, M.G. (2005). The tail structure of bacteriophage T4 and its mechanism of contraction. Nat. Struct. Biol. 12, 810–813. Krauel, V., and Heller, K.J. (1991). Cloning, sequencing, and recombinational analysis with bacteriophage BF23 of the bacteriophage T5 oad gene encoding the receptor-binding protein. J. Bacteriol. 173, 1287–1297. Labedan, B., Crochet, M., Legault-Demare, J., and Stevens, B.J. (1973). Location of the first step transfer fragment and single-strand interruptions in T5stO bacteriophage DNA. J. Mol. Biol. 75, 213–234. Labedan, B., and Goldberg, E.B. (1979). Requirement for membrane potential in injection of phage T4 DNA. Proc. Natl. Acad. Sci. USA 76, 4669–4673. Lambert, O., Letellier, L., Gelbart, W.M., and Rigaud, J.L. (2000). DNA delivery by phage as a strategy for encapsulating toroidal condensates of arbitrary size into liposomes. Proc. Natl. Acad. Sci. USA 97, 7248–7253. Lambert, O., Plancon, L., Rigaud, J.L., and Letellier, L. (1998). Protein-mediated DNA transfer into liposomes. Mol. Microbiol. 30, 761–765. Lanni, Y.T. (1968). First-step-transfer deoxyribonucleic acid of bacteriophage T5. Bacteriol. Rev. 32, 227–242. Lehnherr, H., Hansen, A.M., and Ilyina, T. (1998). Penetration of the bacterial cell wall: a family of lytic transglycosylases in bacteriophages and conjugative plasmids. Mol. Microbiol. 30, 454–457. Letellier, L., Boulanger, P., De Frutos, M., and Jacquot, J.P. (2003). Channeling phage DNA through membranes: from in vivo to in vitro. Res. Microbiol. 154, 283–287. Letellier, L., Boulanger, P., Plancon, L., Jacquot, P., and Santamaria, M. (2004). Main features on tailed phage, host recognition and DNA uptake. Front. Biosci. 9, 1228–1339. Letellier, L., Locher, K.P., Plancon, L., and Rosenbusch, J.P. (1997). Modeling ligand-gated receptor activity. FhuA-mediated ferrichrome efflux from lipid vesicles triggered by phage T5. J. Biol. Chem. 272, 1448–1451. Letellier, L., Plancon, L., Bonhivers, M., and Boulanger, P. (1999). Phage DNA transport across membranes. Res. Microbiol. 150, 499–505. Mangenot, S., Hochrein, M., Radler, J., and Letellier, L. (2005). Real time imaging of DNA ejection from single phage particles. Curr. Biol. 15, 1–20. McCorquodale, D.J., Shaw, A.R., Shaw, P.K., and Chinnadurai, G. (1977). Pre-early polypeptides of bacteriophages T5 and BF23. J. Virol. 22, 480–488. McCorquodale, J.D., and Warner, H.R. (1988). Bacteriophages T5 and related phages. In: The Viruses, R. Calendar, ed. (New York, Plenum Press), pp. 439–476. Mitchell, P., and Moyle, J. (1967). Chemiosmotic hypothesis of oxidative phosphorylation. Nature 213, 137–139. Mitraki, A., Miller, S., and van Raaij, M.J. (2002). Review: conformation and folding of novel beta-structural elements in viral fiber proteins: the triple beta-spiral and triple beta-helix. J. Struct. Biol. 137, 236–247. Moak, M., and Molineux, I. (2004). Peptidoglycan hydrolytic activities associated with bacteriophage virions. Mol. Microbiol. 51, 1169–1183. Molineux, I.J. (2001). No syringes please, ejection of phage T7 DNA from the virion is enzyme driven. Mol. Microbiol. 40, 1–8. Molineux, I.J. (2006). Fifty-three years since Hershey and Chase; much ado about pressure but which pressure is it? Virology 344, 221–229. Mondigler, M., Holz, T., and Heller, K.J. (1996). Identification of the receptor-binding regions of pb5 proteins of bacteriophages T5 and BF23. Virology 219, 19–28. Mondigler, M., Vogele, R.T., and Heller, K.J. (1995). Overproduced and purified receptor binding protein pb5 of bacteriophage T5 binds to the T5 receptor protein FhuA. FEMS Microbiol. Lett. 130, 293–300.

Bacteriophage DNA Transport

Monteville, M.R., Ardestani, B., and Geller, B.L. (1994). Lactococcal Bacteriophages Require a Host Cell Wall Carbohydrate and a Plasma Membrane Protein for Adsorption and Ejection of DNA. Appl. Environ. Microbiol. 60, 3204–3211. Novick, S.L., and Baldeschwieler, J.D. (1988). Fluorescence measurement of the kinetics of DNA injection by bacteriophage lambda into liposomes. Biochemistry 27, 7919–7924. Olson, N.H., Gingery, M., Eiserling, F.A., and Baker, T.S. (2001). The structure of isometric capsids of bacteriophage T4. Virology 279, 385–391. Pedersen, M., Ostergaard, S., Bresciani, J., and Vogensen, F.K. (2000). Mutational analysis of two structural genes of the temperate lactococcal bacteriophage TP901-1 involved in tail length determination and baseplate assembly. Virology 276, 315–328. Plançon, L., Chami, M., and Letellier, L. (1997). Reconstitution of FhuA, an Escherichia coli outer membrane protein, into liposomes—Binding of phage T5 to FhuA triggers the transfer of DNA into the proteoliposomes. J. Biol. Chem. 272, 16868–16872. Plançon, L., Janmot, C., le Maire, M., Desmadril, M., Bonhivers, M., Letellier, L., Boulanger, P. (2002). characterization of a high affinity complex between the bacterial outer membrane protein FhuA and the phage T5 protein pb5. J. Mol. Biol. 318, 557–569. Poranen, M.M., Daugelavicius, R., and Bamford, D.H. (2002). Common principles in viral entry. Annu. Rev. Microbiol. 56, 521–538. Purohit, P., Inamdar, M., Grayson, P., Squires, T., Konder, J., and Phillips, M. (2004). Forces during bacteriophage DNA packaging and ejection. Biophys. J. Purohit, P., Kondev, J., and Philips, R. (2003). Mechanics of DNA packaging in viruses. Proc. Natl. Acad. Sci. USA 100, 3173–3178. Randall-Hazelbauer, L., and Schwartz, M. (1973). Isolation of the bacteriophage lambda receptor from Escherichia coli. J. Bacteriol. 116, 1436–1446. Riemer, S.C., and Bloomfield, V.A. (1978). Packaging of DNA in bacteriophage heads: some considerations on energetics. Biopolymers 17, 785–794. Roessner, C.A., and Ihler, G.M. (1984). Proteinase sensitivity of bacteriophage lambda tail proteins gpJ and pH in complexes with the lambda receptor. J. Bacteriol. 157, 165–170. Rogers, S.G., Godwin, E.A., Shinosky, E.S., and Rhoades, M. (1979a). Interruption-deficient mutants of bacteriophage T5: isolation and general properties. J. Virol. 29, 716–725. Rogers, S.G., Hamlett, N.V., and Rhoades, M. (1979b). Interruption-deficient mutants of bacteriophage T5. II. Properties of a mutant lacking specific interruption. J. Virol. 29, 726–734. Rossmann, M.G., Mesyanzhinov, V.V., Arisaka, F., and Leiman, P.G. (2004). The bacteriophage T4 DNA injection machine. Curr. Opin. Struct. Biol. 14, 171–180. Rouviere-Yaniv, J., Yaniv, M., and Germond, J.E. (1979). E. coli DNA binding protein HU forms nucleosomelike structure with circular double-stranded DNA. Cell 17, 265–274. Saigo, K. (1975). Tail-DNA connection and chromosome structure in bacteriophage T5. Virology 68, 154–165. Sao-José, C., Baptista, C., and Santos, M.A. (2004). B.Subtilis operon encoding a membrane receptor for bacteriophage SPP1. J. Bacteriol. 186, 8337–8346. Sao-Jose, C., Lhuillier, S., Lurz, R., Melki, R., Lepault, J., Santos, M.A., and Tavares, P. (2006). The ectodomain of the viral receptor YueB forms a fiber that triggers ejection of bacteriophage SPP1 DNA. J. Biol. Chem. 281, 11464–11470. Scandella, D., and Arber, W. (1976). Phage lambda DNA injection into Escherichia coli pel- mutants is restored by mutations in phage genes V or H. Virology 69, 206–215. Smith, D.E., Tans, S.J., Smith, S.B., Grimes, S., Anderson, D.L., and Bustamante, C. (2001). The bacteriophage straight phi29 portal motor can package DNA against a large internal force. Nature 413, 748–752. Snyder, C.E., Jr., and Benzinger, R.H. (1981). Second-step transfer of bacteriophage T5 DNA: purification and characterization of the T5 gene A2 protein. J. Virol. 40, 248–257. Spinelli, S., Campanacci, V., Blangy, S., Moineau, S., Tegoni, M., and Cambillau, C. (2006a). Modular structure of the receptor binding proteins of lactococcus lactis phages: The RBP structure of the temperate phage TP901-1. J. Biol. Chem. 281, 14256–14262. Spinelli, S., Desmyter, A., Verrips, C.T., de Haard, H.J., Moineau, S., and Cambillau, C. (2006b). Lactococcal bacteriophage p2 receptor-binding protein structure suggests a common ancestor gene with bacterial and mammalian viruses. Nat. Struct. Mol. Biol. 13, 85–89.

|

227

228

|

Letellier et al.

Steinbacher, S., Miller, S., Baxa, U., Budisa, N., Weintraub, A., Seckler, R., and Huber, R. (1997). Phage P22 tailspike protein: crystal structure of the head-binding domain at 2.3 A, fully refined structure of the endorhamnosidase at 1.56 A resolution, and the molecular basis of O-antigen recognition and cleavage. J. Mol. Biol. 267, 865–880. Tang, L., Marion, W.R., Cingolani, G., Prevelige, P.E., and Johnson, J.E. (2005). Three-dimensional structure of the bacteriophage P22 tail machine. Embo J. 24, 2087–2095. Thomassen, E., Gielen, G., Schutz, M., Schoehn, G., Abrahams, J.P., Miller, S., and van Raaij, M.J. (2003). The structure of the receptor-binding domain of the bacteriophage T4 short tail fibre reveals a knitted trimeric metal-binding fold. J. Mol. Biol. 331, 361–373. van Raaij, M.J., Mitraki, A., Lavigne, G., and Cusack, S. (1999). A triple beta-spiral in the adenovirus fibre shaft reveals a new structural motif for a fibrous protein. Nature 401, 935–938. Vinga, I., Sao-José, C., Tavares, P., and Santos, M.A. (2006). Bacteriophage entry in the host cell. In: Modern Bacteriophage Biology and Biotechnology, G. Wegrzyn, ed., pp. 163–196. Wang, J., Hofnung, M., and Charbit, A. (2000). The C-terminal portion of the tail fiber protein of bacteriophage lambda is responsible for binding to LamB, its receptor at the surface of Escherichia coli K-12. J. Bacteriol. 182, 508–512. Wang, J., Jiang, Y., Vincent, M., Sun, Y., Yu, H., Bao, Q., Kong, H., and Hu, S. (2005). Complete genome sequence of bacteriophage T5. Virology 332, 45–65. Wang, M.D., Schnitzer, M.J., Yin, H., Landick, R., Gelles, J., and Block, S.M. (1998). Force and velocity measured for single molecules of RNA polymerase. Science 282, 902–907. Weigele, P.R., Haase-Pettingell, C., Campbell, P.G., Gossard, D.C., and King, J. (2005). Stalled folding mutants in the triple beta-helix domain of the phage P22 tailspike adhesin. J. Mol. Biol. 354, 1103–1117. Werts, C., Michel, V., Hofnung, M., and Charbit, A. (1994). Adsorption of bacteriophage lambda on the LamB protein of Escherichia coli K-12: point mutations in gene J of lambda responsible for extended host range. J. Bacteriol. 176, 941–947. Xu, L., Benson, S.D., Butcher, S.J., Bamford, D.H., and Burnett, R.M. (2003). The receptor binding protein P2 of PRD1, a virus targeting antibiotic-resistant bacteria, has a novel fold suggesting multiple functions. Structure 11, 309–322.

Prophages and their Contribution to Host Cell Phenotype

9

W. Michael McShan and Joseph J. Ferretti

Abstract In many bacterial species, prophages figure prominently in the biology of these cells, often conferring key phenotypes that can convert a non-pathogenic strain into a pathogen. The source of these phenotypic changes can be through prophage-encoded toxins, bacterial cell surface alterations, or resistance to the human immune system. Further, prophage integration into the host genome can inactivate or alter the expression of host genes. In addition to these direct genetic alterations associated with the addition or inactivation of genes, prophages can also alter the phenotype of bacteria at the population level by facilitating the spread of favorable genes through transduction. Phages as vectors for genes that modify host phenotypes The prophage lifestyle and the host phenotype Soon after the discovery of bacteriophages by Twort and d’Herelle in the second decade of the twentieth century, “filterable agents” were identified that could alter the phenotype of their bacterial hosts through toxigenic conversion. The earliest examples of these filterable agents came from the mid-1920s, demonstrating that sterile culture filtrates from “scarlatina” (erythrogenic toxin A producing) strains of Streptococcus pyogenes (group A streptococcus, GAS) could cause the toxigenic conversion of non-toxin producing strains (Cantacuzene and Boncieu, 1926; Frobisher and Brown, 1927) to toxin producing strains. These observations were the first in a series of discoveries where the transmissibility of streptococcal erythrogenic A toxin predated the actual discovery of its phage association (Zabriskie, 1964) and ultimate molecular characterization ( Johnson and Schlievert, 1984; Weeks and Ferretti, 1984). Similarly, the Corynebacterium diphtheriae diphtheria toxin and the Clostridium botulinum beta-toxin were identified as being transmissible and phage associated years before their molecular characterizations (Eklund et al., 1971; Freeman, 1951; Freeman and Morse, 1952; Frobisher et al., 1947; Fujii et al., 1988; Holmes and Barksdale, 1969; Inoue and Iida, 1971; Sakaguchi et al., 2005; Uchida et al., 1971). In the last several decades, numerous examples of phages capable of altering their host’s phenotype have been discovered and characterized, and the list is growing every year. This chapter presents an

230

|

McShan and Feretti

overview of what has been learned in this field with emphasis upon the most recent findings. Morons The regulation of lysogeny in phage lambda, the paradigm of temperate phages, basically predicts that once the lysogenic state has been achieved that only one gene, the repressor, will be expressed with the remainder of the phage’s genes silent until induction into the lytic cycle (Ptashne, 2004). And yet, it was discovered that other genes encoding proteins that altered the bacterial host surface were expressed during lysogeny, both being known to be dispensable for lambda growth and replication (Barondess and Beckwith, 1990). The term “moron” has been proposed to describe genetic additions to phage genomes that may not be involved with “normal” phage functions like integration or the assembly of capsids but possibly act to increase the evolutionary fitness of the phage (often indirectly by increasing the fitness of the host) ( Juhala et al., 2000). Transcriptionally, morons are predicted to be independent of the phage genetic programs associated with lysogeny or lysis. However, recent evidence shows that the expression of these morons may be influenced by bacterial signals, including ones resulting from the interaction of a bacterium with its human host. The origins of the toxin genes and other morons associated with phages remains unresolved. Supposedly, such genes, which typically occupy the end of an integrated prophage, were picked up by some flawed excision event that captured a portion of the adjoining bacterial chromosome. The separate evolutionary origin of these morons perhaps may be glimpsed by the fact that the moron often has a different %G+C DNA composition (usually lower) from the remainder of the phage genome (Cerdeno-Tarraga et al., 2003; Ferretti et al., 2001). However, the non-phage source for such morons is almost never identified and suggests that the origins of these elements must be from some very distant time. Indeed, the divergent %G+C content may be merely coincidental and reflect optimal codon usage or structural constraints since morons, unlike the remainder of the phage genome except for the repressor, are often expressed during lysogeny (McShan, 2005). A recent analysis of multiple sequenced genomes has further suggested that these genes with anomalous sequence composition are characteristic of phages and other mobile elements and probably were not captured bacterial genes (Daubin et al., 2003). Transmissibility, modular exchange, and horizontal transfer The linking of a virulence factor to a mobile genetic element is a common evolutionary strategy in the bacterial world whether the vector is a phage, plasmid or transposon. A temperate phage is in many ways an ideal vector for spreading a favorable gene throughout a bacterial population providing mechanisms for DNA replication, packaging in an environmentally safe delivery system, and incorporation into the host genome. However, the system has its restrictions; phages have a limited host range and often cannot infect all strains of their hosts. Indeed, these restrictions in host range are the basis for phage typing used to differentiate individual strains within a species. This restriction of potential host places a barrier upon the widespread dissemination of a favorable phage-encoded gene throughout a species. However, recent genetic and epidemiological evidence shows that at least regions or functional modules of phage genomes can become widely distributed across

Prophages and their Contribution to Host Cell Phenotype

strain barriers. For example, over the last two decades a new variant of a phage-associated toxin associated with severe streptococcal disease has disseminated widely worldwide (Cleary et al., 1992). The theory that bacteriophage genomes could be considered as functional modules that could be exchanged with equivalent modules from another phage was first proposed by Botstein to explain the results of heteroduplex analysis of phage genomes (Botstein, 1980). The era of rapid DNA sequencing has allowed the acquisition of multiple temperate phage genomes from a species; this wealth of data argues in favor of modular exchange. For example, comparison of the E. coli lambdoid phages HK97 and HK022 genomes showed that these phages were genetic mosaics, having regions of homology sharply divided from the surrounding non-homologous DNA ( Juhala et al., 2000). Another series of modular DNA exchanges that crossed phylogenetic boundaries in the Enterobacteriaceae is observed in phages e14, SfV, and P22. Prophage e14, an element of E. coli K12, shares a large part of its sequence with the Shigella flexneri phage SfV, including the repressor and Cro proteins (Mehta et al., 2004). Phage SfV is a host serotype converting phage that has an attP-int-xis region that is identical to those genes in phage P22, a generalized transducing phage of Salmonella typhimurium (Huan et al., 1997a). Genome sequencing has revealed that prophages are prominent genetic elements in S. pyogenes, providing many examples of modular shuffling. For example, a strikingly conFigure 1 served region found in S. pyogenes prophages is shown in Figure 9.1. This region contains

Phage

S. pyogenes attB

Virulence Gene(s)

Ref.

Phi Sda

HAD-like hydrolase

sda

(Smoot et al., 2002)

MGAS10394.6

Intergenic region

sda

(Banks et al., 2004)

Phi MF2

recX

speC MF2

http://www.sanger.ac.u k/

MGAS315.4

yesN

speK

(Beres et al., 2002)

SPsP3

Iron dependent repressor

speL

(Nakagawa et al., 2003)

SF370.1

Dipeptidase

speC MF2

(Ferretti et al., 2001)

Figure 9.1 The S. pyogenes prophages sharing a conserved region encompassing the late genes were compared at the DNA level for homology. The heavily shaded and boxed block marks the region containing the genes for DNA packaging, structural proteins for phage heads and tails, and lysis; the general organization of the genomes is indicated above. The conserved region has been recombined with a variety of other genetic modules as evidenced by the diversity in attB sites and associated virulence genes (below). Sequence analysis was done and the graphics generated using the software package Base-by-Base (Brodie et al., 2004).

|

231

232

|

McShan and Feretti

many of the predicted genes for DNA packaging, head and tail assembly, and lysis, and appears to have been recombined with several different site-specific recombination modules and virulence genes. Positionally, each of the six phages is integrated into different sites in the bacterial chromosome even though they may share the same virulence genes (e.g. SF370.1 and Manfredo phi MF2 both have toxin genes speC and MF2, but integrate into different attB sites). The preservation of this module may reflect a particularly evolutionary successful module that has been disseminated widely. However, it should be noted that each individual phage has some level of variation within this region, showing insertions, deletions, or replacements relative to the other members. Such alterations may reflect selection for optimal function within a specific genetic context or perhaps just indicate genetic drift over time, as individual populations remain isolated. The timing and control of phage-encoded virulence gene expression also may play a role in transmission of prophages. Broudy and Fischetti found that a strain of S. pyogenes that was phenotypically a toxin non-expresser when grown on artificial media was stimulated to secrete pyrogenic exotoxin SpeC and a phage-encoded DNase (Spd1 or MF2) after being co-cultured with human pharyngeal cells, and this event was mediated by a pharyngeal cell soluble factor (Broudy et al., 2001; Broudy et al., 2002). Further, contact with the eukaryotic cells mediated induction of the toxin-producing phages, facilitating the conversion of non-toxigenic S. pyogenes to a toxigenic strain (Broudy and Fischetti, 2003). Similarly, the co-culture of the M3 streptococcal strain mgAS315 with human epithelial pharyngeal cells induced the prophage encoded proteins speA and phospholipase A(2), although simultaneous induction of the associated prophages was seen in one but not both prophages (Banks et al., 2003). Finally, in a study of clinical M1 strains, it was shown that erythrogenic toxin SpeA production was very low in about half the strains, but culture of these strains in Teflon chambers implanted into mice for 5 days induced the expression of SpeA. The induction of SpeA expression occurred with a simultaneous inhibition of the chromosomally encoded cysteine protease SpeB, possibly preventing the degradation of the phage-encoded protein (Aziz et al., 2004; Chatellier et al., 2000; Kazmi et al., 2001). The mechanism(s) responsible for prophage induction in these experiments remains to be determined, however it is well known that virtually all other phage inductions occur via damage to DNA and the increased synthesis of RecA, which stimulates autocleavage of the repressor. All these studies point out that the regulation of these toxin genes and phage induction in the presence of eukaryotic cells is still an area that needs intense investigation to determine the relative roles played by the human host, the bacterium and the phage. While direct transfer of virulence factors through linkage to temperate phage genomes is an important mechanism of horizontal gene transfer, the indirect transfer of numerous host genes, including virulence factors, also can occur through phage transduction. The toxic shock syndrome toxin (TSST-1) is contained on a 15.2 kb pathogenicity island containing an integrase gene and flanked by 17 nucleotide direct repeats. In the presence of phage phi13 and phi80A, the TSST-1 element is excised and circularized, and growth of phi80A causes replication of the circular form of the pathogenicity island and leads to high frequency transduction by this phage (Lindsay et al., 1998). Similar-phage mediated transfer ultimately may be found to facilitate the spread of other non-phage associated toxins as in the recent severe variant of C. difficile (McDonald et al., 2005).

Prophages and their Contribution to Host Cell Phenotype

Resistance to serum killing and phagocytosis A number of temperate phages act by altering the surface of the bacterial host to confer resistance to killing by normal human serum or to phagocytosis. In Gram-negative bacteria, the basis for this resistance is often a modification to the lipopolysaccharide. An early report of phage-mediated resistance to phagocytosis detailed the lysogenization of Pseudomonas aeruginosa by phage D3. A phage-encoded gene was found to alter a surface antigen of the bacteria that resulted in protection against superinfection by new D3 phage through loss of the ability to adsorb the phage and the simultaneous change in the opsonizing capacity of normal mouse serum against the bacterium (Holloway and Cooper, 1962). Subsequently, the basis of the change was identified as alterations in the O antigen mediated by a phage encoded fucosamine O-acetylase (Kropinski, 2000; Kuzio and Kropinski, 1983). Another phage of P. aeruginosa, FIZ15, has been identified from a clinical isolate that is also able to increase resistance to phagocytosis and killing by serum while simultaneously increasing adhesion to human buccal epithelial cells. As with phage D3, these phenotypic changes appear to be due to alterations in the O-antigen (Vaca-Pacheco et al., 1999). A similar phage-encoded system is found in serogroup C1 Salmonella choleraesuis. Here, phage 14 (P14) was found to convert the O-antigen of its host bacterium from O-6,7 to O6,7,14. The O-antigen polysaccharide from lysogens was estimated to be six times as long as those from a non-lysogen, and further, lysogens were serum-resistant whereas non-lysogens were serum-sensitive. This serum-resistance appeared to correlate with increased virulence since about 10 times more colony forming units of a lysogen were recovered from the livers and spleens of mice after intraperitoneal inoculation as compared to a non-lysogen (Nnalue et al., 1990). In a related member of the Enterobacteriaceae, S. flexneri bacteriophage SfV converts serotype Y hosts to serotype 5a through a phage-encoded glucosyl transferase gene (Huan et al., 1997b). Phage-mediated alteration of LPS can also be accomplished through an indirect route. Phase variation in the Legionella pneumophila lipopolysaccharide causes the organism to switch between a virulent, serum-resistant phase and a non-virulent, serum-sensitive one. The genes responsible for the virulent phenotype are found on a 30 kb genetic element that appears to be at least partially derived from a prophage remnant. The element is capable of excising from the host chromosome and replicating as a high-copy plasmid with the resulting change to the avirulent LPS phenotype. Re-integration into the bacterial chromosome restores the virulent phenotype, and it is possible that the avirulent phenotype results from over-expression of some gene in the plasmid phase that interferes with the LPS synthesis pathway. The mechanism that mediates the switch between the episomal and integrated forms of the element is unclear since no member of the phage integrase family was found on the prophage and the process was found to operate independently of RecA. However, homologs of RecE, RecT, and RusA are encoded by the element and may play a role in this process (Luneberg et al., 2001). In addition to encoding genes that alter the composition of the host O-antigen, phage encoded proteins can confer serum resistance by specifying new proteins that are located on the bacterial surface. The analysis of dispensable genes in the phage lambda genome and the search for “wild” examples of lambdoid phages uncovered virulence genes that existed in this most well-studied of phages. Analysis of genes (other than the repressor) expressed

|

233

234

|

McShan and Feretti

during lysogeny by Barondess and Beckwith identified two genes in the “dispensable” region of the chromosome, lom and bor, which were actively transcribed. The bor gene, closely related to the iss locus of plasmid CoIV,I-K94, encodes a lipoprotein located in the E. coli outer membrane and confers bacterial resistance to serum complement killing in vitro and virulence in animals. A survey of clinical strains of E. coli showed that bor-related genes were not uncommon, suggesting that linkage to a mobile genetic element had been a successful strategy in the dissemination of this serum-resistance phenotype (Barondess and Beckwith, 1990; Barondess and Beckwith, 1995). Prophage conversion can also alter the Gram-positive bacterial surface to increase resistance to phagocytosis. Cleary and his co-workers examined the relation between S. pyogenes lysogeny and the expression of the antiphagocytic M protein. The M protein is the major surface antigen of group A streptococci, and some strains of the organism give rise to frequent M– colonial variants. These spontaneous variations in M protein expression were related to the presence of an extrachromosomal element by the observation that high numbers of M– variants were generated by growing streptococci under conditions that typically cure bacteria of plasmids or phage (Cleary et al., 1975; Spanier and Cleary, 1983). Subsequently, a temperate bacteriophage, SP24, was isolated and when introduced into an M– emm76 strain restored emm expression to levels similar to the wild type, suggesting that the phage introduces a transcription regulator (Spanier and Cleary, 1980). While the above mechanisms of prophage-mediated immune avoidance are somewhat passive in mechanism, increasing the bacterium’s resistance to phagocytosis or Ab-complement killing, the Panton-Valentine leukocidins (PVL) of S. aureus are active aggressins against immune cells, acting as pore-forming toxins that kill polymorphonuclear leukocytes, monocytes, and macrophages, possibly through targeting the mitochondria (Genestier et al., 2005; Gladstone and Van Heyningen, 1957). These toxin genes were first shown to be prophage associated by transfer of the toxin to non-toxigenic strains through phages released by mitomycin C induction of a PVL+ strain (Kaneko et al., 1997). In contrast to many phage-associated toxins, PVL is a two-component toxin specified by the genes lukS-PV lukF-PV, two of the 63 ORFs found on the 41 401 bp phage phiPVL genome (Kaneko et al., 1998). Although strains of S. aureus harboring PVL-containing phages are uncommon in general (0.6%), a markedly higher incidence of PVL+ strains is found in staphylococcal arthritis and abscesses (38.9%) and in staphylococcal pneumonia strains (90%) (Foster, 2004; Gillet et al., 2002; Melles et al., 2004). Phages and host invasiveness A number of prophage-encoded systems have been identified that facilitate bacterial binding to host cells or invasiveness; several play important roles in the pathogenic potential of Salmonella. S. typhimurium has a specialized type III secretion system encoded by pathogenicity island 1 (SPI1) that translocates effector proteins into host cells, modulating host cell signal transduction. Prophage SopE&, a member of the P2 family of phages, was identified as the vector for SopE, a SPI1-dependent effector protein. Significantly, only a small subset of Salmonella strains carry this gene, and the ones that do are epidemic strains that are responsible for a large number of human and animal disease. The role of phage SopE& in

Prophages and their Contribution to Host Cell Phenotype

the horizontal transfer of sopE is demonstrated by its ability to produce infectious particles that can infect new strains of S. typhimurium (Mirold et al., 1999). The prophages Gifsy-1 and Gifsy-2 have also been shown to play roles in Salmonella virulence. Gifsy-1 carries the gene gipA that becomes transcriptionally active when the bacteria infects the small intestine of mice and is important in the survival of the bacterium in Peyer’s patches. The role of GipA appears to be specific for survival in Peyer’s patches since it is dispensable for the later, systemic phase of the S. typhimurium infection (Stanley et al., 2000). The Gifsy-2 prophage carries a gene for a periplasmic Cu/Zn superoxide dismutase (SodC) that significantly improves the ability of the bacteria to survive in mice by defending against killing in macrophages (Figueroa-Bossi and Bossi, 1999). In addition to sodC, phage Gifsy-2 carries at least one additional virulence factor, a Salmonella pathogenicity island 2 type III secreted effector protein. Identified as GtgE, this protein is required for full virulence in mice (Ho et al., 2002). The carriage of prophages with multiple genes that improve the ability of S. typhimurium to establish or maintain an infection in animals appears to be a characteristic of virulent strains. Prophage Fels-1 also carries a gene for a novel superoxide dismutase, sodCIII, as well as the nanH gene that encodes a neuraminidase (Figueroa-Bossi et al., 2001). Taken together, these various studies suggest that lysogenic conversion is a major mechanism in the dissemination of virulence traits associated with survival in the animal host in Salmonella species. Several phage encoded genes have been proposed to be “spreading factors” that allow a bacterium to invade host tissues. The capsule of group A streptococci is composed of hyaluronic acid and many of its bacteriophages encode a hyaluronidase (hyaluronate lyase (Smith et al., 2005)) to aid in penetration of the capsule. Human synthesized hyaluronic acid is an important component of ground substance, synovial fluid, and the vitreous humor of the eye, and it has been suggested that the phage-encoded hyaluronidase might also assist in the break-down of host tissues, allowing the streptococcus to spread (Hynes and Ferretti, 1989; Hynes et al., 1995). A recent study compared a wild type group A streptococcus with an isogenic hyaluronidase knock-out. The two strains were injected subcutaneously into mice with or without high-molecular-weight dextran blue. The hylP– strain produced small lesions with dye concentrated in close proximity. The hyaluronidase-producing strain produced identical lesions, but in contrast, the dye diffused subcutaneously and the bacteria were not isolated from unaffected skin stained by dye diffusion. Therefore, the HylP enzyme was able to digest tissue hyaluronic acid and facilitate the spread of large molecules, but did not appear to be sufficient to cause subcutaneous diffusion of bacteria or to affect lesion size (Rivera Starr and Engleberg, 2006). A similar role has been suggested for the phage-encoded staphylokinase of S. aureus (Coleman et al., 1989; Sako et al., 1983), although the actual effects are untested. Another means for promoting infection is through binding to specific host tissues. In infective endocarditis, the binding of platelets by bacteria is thought to be crucial to pathogenesis. Platelet binding in Streptococcus mitis is mediated in part by two surface proteins, PblA and PblB. Bensing and her co-workers found the genes for these proteins encoded in the 35 kb genome of prophage SM1, being related to phages r1t of Lactococcus lactis and SF370.3 of S. pyogenes (Bensing et al., 2001; Siboo et al., 2003). Phage SM1 is inducible

|

235

236

|

McShan and Feretti

by mitomycin C or UV light, and induction causes an increase in transcription of pblA, providing another example of how the expression of some phage virulence-associated genes appears to be linked to the induction of the lytic cycle (Banks et al., 2003; Broudy et al., 2001). Toxigenic phages Gram-positive species Prophage-mediated toxigenic conversion is associated with many of the best known and widely feared diseases caused by Gram-positive bacteria, including diphtheria, botulism, and streptococcal toxic shock syndrome. Although the transmissibility of a toxigenic phenotype was observed early in phage studies, the first linkage of toxin production to a bacteriophage was made by Freeman for diphtheria toxin, demonstrating that a bacteriophage could convert a non-toxigenic, avirulent strain of C. diphtheriae into a toxigenic one (Freeman, 1951; Freeman and Morse, 1952). Subsequent genome sequencing showed that the toxin gene is contained on the 36.6 kb genome of the temperate prophage beta of C. diphtheriae (Cerdeno-Tarraga et al., 2003); the attB site for the phage is the gene for an Arg tRNA (Ratti et al., 1997). The diphtheria toxin is synthesized in response to iron-limiting conditions and is regulated by the bacterially encoded DtxR, a dimeric DNA binding protein that functions as an iron-dependent, negative global regulator. Under high-iron conditions, DtxR represses the synthesis of diphtheria toxin as well as the corynebacterial siderophore and other components of the high-affinity iron uptake system (Lee et al., 1997). The link between virulence and prophages is well-established in S. pyogenes and has existed since Zabriskie demonstrated that phage T12 was responsible for the toxigenic conversion of S. pyogenes to produce scarlet fever toxin, also variously known as erythrogenic toxin A and streptococcal pyrogenic exotoxin A, SpeA (Zabriskie, 1964). One of the most prominent characteristics of the multiple S. pyogenes genomes that have been sequenced is the number of phage-associated virulence genes, and indeed, a majority of these genes were unknown prior to genome sequencing. A considerable range of genes are carried by GAS prophages and prophage-like elements, including superantigens (speA, speC, speG, speH, speI, speJ, speK, speL, SSA, and variants), DNases (MF2, MF3, and MF4), phospholipase A2 (sla), and macrolide resistance (mefA). It is quite common to find that one prophage carries more than one virulence gene such as phage SF370.1 that contains both speC and MF2. The known and suspected GAS phage-associated virulence genes are all positioned in the phage genomes in the region between the lysis module and the right attachment site with the bacterial chromosome. The importance of phage encoded genes to the virulence of group A streptococci probably cannot be overestimated since no other species studied has the number and variety of such genes as does S. pyogenes. A number of toxin and virulence genes have been associated with prophages of S. aureus, including the exfoliative toxin, staphylokinase, and enterotoxin A (Table 9.1). As in the case of S. pyogenes prophages, staphylococcal prophages can carry multiple virulence genes such as staphylokinase and enterotoxin A (Coleman et al., 1989; Iandolo et al., 2002). Interestingly, staphylococcal prophages often integrate into bacterial ORFs, resulting in simultaneous gene conversion by addition and inactivation (discussed below). The

Prophages and their Contribution to Host Cell Phenotype

Table 9.1 Toxin genes associated with prophages Species

Toxin (references)

Gram negative E. coli

Shiga toxin (Huang et al., 1987; Newland et al., 1985)

P. aeruginosa

Cytotoxin (Hayashi et al., 1990; Nakayama et al., 1999)

S. enterica s. Typhimurium

ArtAB ADP-ribosyltransferase toxin (Saitoh et al., 2005)

Shigella dysenteriae

Shiga toxin

V. cholerae

Cholera toxin (Waldor and Mekalanos, 1996)

Gram positive C. diphtheriae

Diphtheria toxin (Cerdeno-Tarraga et al., 2003; Holmes and Barksdale, 1969; Uchida et al., 1971)

C. botulinum

Botulinum Neurotoxins, Exoenzyme C3 (Eklund et al., 1971; Fujii et al., 1988; Inoue and Iida, 1971; Sakaguchi et al., 2005)

S. pyogenes

Erythrogenic toxins SpeA and SpeC (Goshorn and Schlievert, 1989; Johnson and Schlievert, 1984; Weeks and Ferretti, 1984) Phage-associated superantigens and other virulence factors identified by genome sequencing (Banks et al., 2004; Beres et al., 2004; Ferretti et al., 2001; Green et al., 2005; Nakagawa et al., 2003; Smoot et al., 2002; Sumby et al., 2005)

S. aureus

Enterotoxin A (Sea) (Betley and Mekalanos, 1985) Exfoliative toxin ETA (Yamaguchi et al., 2000) Staphylokinase (Coleman et al., 1989; Iandolo et al., 2002) Toxin shock syndrome toxin (phage mobilized and transduced) (Lindsay et al., 1998)

staphylococcal toxic shock syndrome toxin gene (TSST-1), while contained on a small pathogenicity island (SaPI1) and not on a prophage genome, is mobilized by staphylococcal phages F13 and 80alpha. Upon induction by these phages, SaPI1 is excised and circularized and replicates during the growth of phage 80alpha, which transduces it at very high frequency. Another related toxin-containing element, SaPI2, is mobilized by phage 80 but not by 80alpha, suggesting specific replicative pairs exist (Lindsay et al., 1998). The staphylococcal exfoliative toxins (ETs) are extracellular proteins that cause human skin to split at the epidermal layer, causing scalded skin syndrome. Two distinct exfoliative toxins have been identified in S. aureus, one (ETA) encoded by a prophage and the other (ETB) by a large plasmid. The FETA prophage has a circularly permuted linear dsDNA genome of 43 kb (Yamaguchi et al., 2000). Additional phage-encoded virulence genes may exist in staphylococci. For example, prophage F13, in addition to carrying the staphylokinase gene, also encodes fib, a gene for a soluble fibronectin-binding protein (Iandolo et al., 2002). The contribution of fib to virulence is, however, still unknown (Boden and Flock, 1994). The botulism neurotoxin produced by Clostridium botulinum, one of the most toxic molecules known, can be encoded either on the bacterial chromosome or on a prophage.

|

237

238

|

McShan and Feretti

Seven types of botulinum neurotoxins are known, and the genes for type C1 and D toxins are carried by prophages (Eklund et al., 1972; Eklund et al., 1971). Interestingly, the closely related tetanus toxins of Clostridium tetani are encoded on plasmids (Eklund et al., 1988; Hauser et al., 1995). The genome of the type-C toxin carrying phage c-st has been completely determined and found to be one of the largest temperate phage DNA molecules known at 185 682 bp. The prophage was found to exist both in the integrated form and as a free circular plasmid, accounting for the frequently observed unstable nature of these lysogens (Sakaguchi et al., 2005). Gram-negative species A number of important toxins associated with diarrheal and gastrointestinal disease such as hemorrhagic colitis are found as elements of Gram-negative prophages. The Shiga toxins, produced by E. coli O157:H7 and many Shigella species, exist in the antigenically distinct forms Stx-1 and Stx-2, both being phage encoded. These phages can have a broad range of hosts within these two closely related species of bacteria; for example, Shigella sonnei phage 7888 can infect a broad spectrum of Shigella strains of different species and serotypes, including Stx-producing Shigella dysenteriae type 1. Additionally, this phage can infect E. coli strains with a rough lipopolysaccharide phenotype (Strauch et al., 2001). The genome of E. coli Stx2 phage 933W was found to contain, in addition to the Shiga toxin gene, several other morons that either may increase bacterial survival or virulence, including serine/threonine kinase Stk, three novel tRNA genes, and the outer membrane proteins Lom and Bor, conferring the ability to survive in macrophages and serum, respectively (Plunkett et al., 1999). The Stx1 prophage bp-4795 was also found to carry a type III effector that is translocated by the enterocyte effacement-encoded type III secretion system into HeLa cells, where it localizes with the Golgi apparatus (Creuzburg et al., 2005). The type III secretion system of Shigella flexneri is capable of injecting proteins into human cells via a molecular syringe structure, thus promoting bacterial invasion and leading to an extensive inflammatory reaction in the gut (dysentery or Shigellosis). West and colleagues (2005) identified two colonization defective mutants of S. flexneri that contained mutations in the gtrABV operon of a resident bacteriophage. This operon is normally responsible for addition of glucose residues to each O-antigen repeat unit and results in a shortening of the LPS surface molecule by about one-half. The shortened LPS molecule has a conformational change from the non-glucosylated forms that makes it more compact, but still protective to the bacterial cell against host inflammatory mediators. Thus bacteriophage genes enhance Shigella virulence while at the same time evading the host innate immunity (West et al., 2005). While most toxigenic phages are lambda-like members of the Siphoviridae, the cholera toxin encoding phages (CTXF) of V. cholerae are filamentous phages. However, unlike the well-characterized filamentous phage M13, CTXF integrates into a specific attB site on the bacterial chromosome by site-specific recombination, employing a single-stranded, circular DNA molecule (Val et al., 2005; Waldor and Mekalanos, 1996). Virulence in V. cholerae involves a network of both phage and bacterial encoded genes that interact in a highly adapted system. Two membrane localized transcription factors, ToxR and TcpP, activate the promoter for ToxT, a cytoplasmic transcription factor. ToxT then activates promoters

Prophages and their Contribution to Host Cell Phenotype

required for colonization of the host and virulence, including the promoters for the cholera toxin and the toxin-coregulated pilus (TCP). The TCP, in addition to being a colonization factor, also serves as the CTXF receptor. Remarkably, many of these virulence associated genes that are found on another phage-like pathogenicity island (Vibrio pathogenicity island, VPI) that encodes ToxT, TCP, and the ToxR-regulated genes aldA and tagA. VPI is 39.5 kb in size and, like CTXF, appears to be another filamentous phage (Karaolis et al., 1998; Karaolis et al., 1999). In addition to the VPI-encoded genes, other bacterial genes play a role in pathogenesis of cholera. The release of cholera toxin is mediated by the extracellular protein secretion type II secretion system, and an outer membrane component of this system, EpsD, is required for secretion of the phage as well (Davis et al., 2000). These two mobile genetic elements, CTXF and VPI, differ in their response to induction. Phage CTXF once integrated into the bacterial chromosome, becomes a stable and permanent genetic element. Induction by DNA damaging agents does not cause excision of the phage; rather, a burst of phage replication and release of new virions ensues. In contrast, VPI is capable of excising from the chromosome and replicating extrachromosomally (Rajanna et al., 2003). Another virulence-related filamentous phage has recently been identified that plays a role in the invasive phenotype of Neisseria meningitidis. While its genetic structure is similar to the CTXF phage of Vibrio cholerae, no toxin or specific-virulence gene has been identified yet, although one ORF specifies a protein with some relation to the Zot protein of V. cholerae and several ORFs of unknown function are present in the location where the cholera toxin is found. The phage is able to excise from the bacterial chromosome, and infectious phages are secreted from the bacteria via the type IV pilin secretin. Epidemiological analysis showed a strong association between strains causing invasive disease and the presence of the phage, especially in infections of young adults (Bille et al., 2005). Conversion of host gene phenotypes by other mechanisms Targets for prophage integration Although bacteriophage lambda has been the paradigm for many aspects of prophage biology, its integration into an intergenic region of the E. coli chromosome is not representative for the choice of attachment sites (attB) for many prophages. Instead, it is quite common to find prophages integrated into an ORF, very typically the 3a translated portion, and providing through the region of sequence duplication with the host a means of maintaining the integrity of the gene and not altering its normal expression. Highly conserved sequences are often the sites for integration, probably providing stable and invariant targets for attachment by an infecting prophage. In particular, tRNA and tmRNA genes have been particularly favored, and the crossover during site-specific recombination can occur at three specific locations: the anticodon loop, the T-loop tDNA, and the 3a asymmetric end. Phylogenetically, these sites are favored by a phages infecting a wide range of prokaryotes, including Gram negatives, both high and low %G+C Gram positives, and mycobacteria (Williams, 2002). Although tRNA-like sites seem to be favored, other genes can be targeted for integration including genes involved in basic metabolic pathways such as citrate dehydrogenase and glucose 6-phosphate isomerase (Campbell et al., 1992; Shimizu-Kadota et al., 2000), DNA

|

239

240

|

McShan and Feretti

recombination and repair like recX and mutL (Canchaya et al., 2002; Ferretti et al., 2001; Nakagawa et al., 2003), and potential or known virulence factors such as the staphylococcal beta toxin and group A streptococcal dipeptidase (Canchaya et al., 2002; Mason and Allen, 1975). Every strain of the group A streptococcus examined in genomic detail has contained multiple prophage genomes, and perhaps because such a large population exists these prophages use a wide variety of attachment sites including the histone-like protein, gamma-glutamyl kinase (proB), dTDP-glucose-4,6-dehydratase, an iron dependent repressor, recombination protein recO, excinuclease ABC subunit A, and a HAD-like hydrolase (McShan, 2005). Most bacteriophage integrases appear to have a high degree of DNA sequence specificity for their associated attachment site. If the primary attachment site for phage lambda is deleted, for example, the phage integrase is only able to mediate site-specific recombination at secondary sites with a 100- to 1000-fold reduction in efficiency (Weisberg and Landy, 1983). Secondary attachment sites for lambda have been identified in populations of E. coli, but a strong preference remains for the primary lambda site (Kuhn and Campbell, 2001). This strong sequence preference appears to be typical for phage integrases, and the increasing number of genomic DNA sequences from multiple isolates of the same bacterium illustrate that a given integrase gene is strongly associated with a specific att site. However, exceptions do exist. Coliphage Mu is well known for its lack of target sequence specificity, and ten different attB sites have been identified for phage P2, its integrase tolerating up to 37% mismatches from the core sequence (Barreiro and Haggard-Ljungquist, 1992). Prophages and altered host gene expression In most prophages, the recombination between the host and phage chromosomes occurs such that the function of the host genetic material is unimpeded by the presence of the phage. This result is accomplished by two factors: (1) duplication of the host DNA sequence at the site of crossover by the phage genome and (2) integration at the 3a end of the targeted gene so that the duplication can complete the original bacterial ORF. The consequences of 5a versus 3a integration are shown in Figure 9.2; it is easy to see how integration into the 5a end or the middle of a gene could result in the disruption of normal transcription and a loss of gene function. Yet remarkably, a number of examples are now known where phage integration does occur at the 5a end or the middle of a gene, potentially altering gene expression (Table 9.2). The selective advantage to a host bacterium through the acquisition of a phage-encoded toxin or exoenzyme that improves survival against human immune defenses or allows acquisition of resources is often readily apparent. In contrast, the loss or modification of a host gene by a prophage suggests that such loss must be favorable under some unusual or alternative environmental conditions to which the organism is occasionally exposed. A number of examples of phage-mediated gene inactivation have been discovered, although whatever advantage is conferred upon a bacterium by most of the events is not well understood. Bacterial gene inactivation may result from the occasional occupation of alternate or secondary attachment sites by prophages. In strains of E. coli containing a deleted lambda attB site, several secondary sites may be used for integration by the phage, although at a

Figure 2.

Prophages and their Contribution to Host Cell Phenotype

Figure 9.2 The consequences of 3a versus 5a prophage integration. (A) The prophage undergoes site-specific recombination with the 3a end of a bacterial gene, duplicating the end of the coding region and ensuring that gene function is maintained. (B) The prophage here targets the 5a end of a gene, interrupting the ORF and creating a polar mutation upon integration.

greatly reduced frequency. About 0.025% of the lysogens in these strains have lambda integrated into guaB, the gene specifying IMP dehydrogenase. This site occurs within the coding region of guaB, and integration results in a polar mutation (Thomas and Drabble, 1986). The E. coli phages 21 and e14 both insert within the 3a end of the isocitrate dehydrogenase gene (icd) and, instead of duplicating the bacterial DNA, each introduces an alternative and distinct 3a end for that gene, producing a modified version of encoded protein (Campbell et al., 1992). Whether this alteration creates a selective advantage for the bacterial host under some conditions or simply reflects genetic drift of long-standing chromosomal elements is unknown. In addition to potential alterations to icd, e14 also encodes the functional genes lit (T4 exclusion), mcrA (a modified cytosine restriction activity) and the pin recombinase that alter the phenotype of its host.

|

241

242

|

McShan and Feretti

Table 9.2 Prophage integration sites with potential to alter the host phenotype Host

Phage

attB

Ref.

E. coli

P21

3a end of Isocitrate dehydrogenase

(Campbell et al., 1992)

Lambda

guaB (Secondary attB)

(Thomas and Drabble, 1986)

933W

Tryptophan repressor-binding protein (WrbA)

(Plunkett et al., 1999)

L54a

Lipase ORF

(Lee and Iandolo, 1986)

phi12

Unknown ORF

(Iandolo et al., 2002)

phi13

Beta-toxin (hlb) ORF

(Coleman et al., 1991; Mason and Allen, 1975)

SF370.1

5a end of dipeptidase

(Canchaya et al., 2002; Ferretti et al., 2001)

SF370.4

5a of mutL

''

MGAS315.4

Promoter of yesN

(Beres et al., 2002)

MGAS315.5

5a end of conserved hypothetical protein

''

MGAS315.6

5a end of gamma-glutamyl kinase (proB)

''

SPsP2

5a of recX

(Nakagawa et al., 2003)

SPsP3

Promoter of Iron dependent repressor SPy0450

''

MGAS10394.7

Promoter of Excinuclease ABC subunit A

(Banks et al., 2004)

Phi Sda

5a end of HAD-like hydrolase

(Smoot et al., 2002)

S. aureus

S. pyogenes

In S. aureus, two examples of bacterial virulence gene regulation through prophage integration into an attB site contained within an ORF have been described. Several phages have been identified that target the beta toxin (beta lysin) gene, causing simultaneous inactivation of the beta toxin and insertion of staphylokinase alone or in combination with enterotoxin A (Sea) (Coleman et al., 1991; Mason and Allen, 1975). A similar system has been found to control the expression of lipase by a phage that integrates into its structural gene, causing a polar mutation (Lee and Iandolo, 1986). An unusual number of S. pyogenes prophages have attachment sites positioned at the promoter or 5a end of genes that potentially could alter gene expression or create polar mutations. The predicted polycistronic message containing the essential genes for DNA mismatch repair (MMR), mutS and mutL, as well as several additional DNA repair genes are interrupted by a prophage in a number of the genome strains. For example, prophage SF370.4 is integrated into the 5a end of mutL, thus separating it from its promoter up-

Prophages and their Contribution to Host Cell Phenotype

stream of mutS. The presence of the prophage is predicted to inactivate MMR and lead to a mutator phenotype. Integration at the 5a end of a host gene is also seen in other S. pyogenes genome prophages, including a dipeptidase gene, recombination protein recX, proB, a HAD-like hydrolase, and a conserved hypothetical protein (Table 9.2). Other prophages appear to target the promoter region immediately upstream of the actual ORF, including the promoter regions of yesN, gamma-glutamyl kinase, and an iron-dependent repressor. Similarly, the prophage-like transposon mgAS10394.4 separates the comE operon proteins 2 and 3, probably creating a polar mutation that silences protein 3. In all of these cases, the insertion of a prophage at the 5a end of a gene has the potential for altering host gene expression by introducing a polar mutation or an alternative promoter, and such transcription-altering prophages appear to be an important class of genetic regulatory elements in S. pyogenes in modifying the biology of their hosts. Finally, prophages may alter the expression of host genes by prophage encoded regulatory proteins such as repressors. Chen and co-workers (Chen et al., 2005) recently described how the well-characterized lambda repressor acts to alter the expression of an essential E. coli gene. While several bacterial genes appeared to be downregulated by the presence of lambda, the most inhibited was an essential gene for gluconeogenesis, phosphoenolpyruvate carboxykinase (pckA). In the presence of ATP, PckA converts oxaloacetate to phosphoenolpyruvate, and the inhibition of this enzyme results in an inability of the bacterium to grow on succinate as a carbon source. The regulatory region of pckA is homologous to the lambda operator and provides a binding site for the cI repressor, downregulating its expression. The effect of this interaction is to lower the growth rate of E. coli in energy-poor environments, a response that may be adaptive for the bacterium in some contexts (Chen et al., 2005). Transduction of antibiotic resistance While many virulence-associated genes are carried by prophages, no examples of antibiotic resistance genes have been found as elements of phage chromosomes. However, phages can mobilize antibiotic resistance genes through generalized transduction. Indeed, the role played by transduction was observed early on for many important pathogens (Coetzee and Sacks, 1960; Malke, 1969; Pattee and Baldwin, 1961; Watanabe and Watanabe, 1959). While many transducing phages are lytic phages, a number of examples of temperate phage transduction have been observed in the staphylococci and streptococci. For example, Ubukata and co-workers observed the transfer of resistance to tetracycline, chloramphenicol, macrolides, lincomycin and clindamycin by induced temperate phages from serotype T-12 group A streptococcal strains (Ubukata et al., 1975). Antibiotic resistance was transferred either alone or in combination, suggesting that the donor strain may have been harboring a genetic element conferring multiple drug resistance such as the transposon TN916. Malke similarly observed transduction of multiple antibiotics in group A streptococci as well as in groups C and G (Malke et al., 1975). The transfer of erythromycin and streptomycin resistance by temperate phages following mitomycin C induction has been also observed (Hyder and Streitfeld, 1978). S. aureus prophage F11de is a derivative of phage F11 formed after some rare recombination events with plasmid pI258 (Bachi, 1980; Novick, 1991) and is capable of high-frequency transduction of erythromycin resistance (Lofdahl et al., 1981). Transduction of antibiotic resistance by prophages has also been observed in several other

|

243

244

|

McShan and Feretti

bacterial species, including Actinobacillus actinomycetemcomitans, Pseudomonas aeruginosa, and Lactococcus lactis (Hupkova et al., 1994; Koch et al., 1997; Willi et al., 1997). Conclusions The interaction of bacterium and bacteriophage ranges from a simple predator-prey model to a complex, almost symbiotic relationship that promotes the survival and evolutionary success of both. In the case of temperate phages, this full range of interactions can occur depending upon whether the lytic or lysogenic pathway has been triggered, and alteration to the host phenotype can result throughout this range. Infection and lysis without the induction of lysogeny can influence host populations through negative selection, identifying resistant mutants that may represent minority phenotypes with respect to nutrient utilization or variations in virulence or colonization factors in the bacterial population. Furthermore, generalized transduction depends upon the lysis of one host to transfer genetic material to a new cell. However, it is in the establishment of lysogeny that some of the most remarkable changes to bacterial host phenotype occur. The range of prophage-mediated alterations to host phenotype has continued to increase as more investigations have delved into this area. The early discoveries that many toxin genes were carried by prophages have been expanded to include prophage-encoded genes that increase bacterial resistance to serum bactericidal or phagocyte killing, promotion of bacterial binding or invasion into human host tissues, and survival in special ecological niches. Further, the choice of integration site by the prophage into the bacterial chromosome may alter the host genotype through gene inactivation or through replacement of normal promoter elements with phage-encoded ones. The similarity that prophages have to pathogenicity islands can hardly be overlooked, and the range of prophage-mediated characteristics that add to host survival or virulence can be easily predicted to increase as new investigations occur. Indeed, even within the bacteriophages that have been investigated, many genes of unknown function often exist that may act as regulators of bacterial genes, and the range of new potential host-altering genes expands quickly when the prophage genomes discovered by genome sequencing are considered since most of these remain little characterized and contain numerous, often unique genes encoding unknown functions. Continued study of the interactions between prophages and bacteria will be certain to reveal many novel and perhaps surprising relationships. References

Aziz, R.K., Pabst, M.J., Jeng, A., Kansal, R., Low, D.E., Nizet, V., and Kotb, M. (2004). Invasive M1T1 group A Streptococcus undergoes a phase-shift in vivo to prevent proteolytic degradation of multiple virulence factors by SpeB. Mol. Microbiol. 51, 123–134. Bachi, B. (1980). Physical mapping of the BglI, BglII, PstI and EcoRI restriction fragments of staphylococcal phage phi 11 DNA. Mol. Gen. Genet. 180, 391–398. Banks, D.J., Lei, B., and Musser, J.M. (2003). Prophage induction and expression of prophage-encoded virulence factors in group A Streptococcus serotype M3 strain mgAS315. Infect. Immun. 71, 7079– 7086. Banks, D.J., Porcella, S.F., Barbian, K.D., Beres, S.B., Philips, L.E., Voyich, J.M., DeLeo, F.R., Martin, J.M., Somerville, G.A., and Musser, J.M. (2004). Progress toward characterization of the group A Streptococcus metagenome: complete genome sequence of a macrolide-resistant serotype M6 strain. J. Infect. Dis. 190, 727–738.

Prophages and their Contribution to Host Cell Phenotype

Barondess, J.J., and Beckwith, J. (1990). A bacterial virulence determinant encoded by lysogenic coliphage lambda. Nature (London) 346, 871–874. Barondess, J.J., and Beckwith, J. (1995). bor gene of phage lambda, involved in serum resistance, encodes a widely conserved outer membrane lipoprotein. J. Bacteriol. 177, 1247–1253. Barreiro, V., and Haggard-Ljungquist, E. (1992). Attachment sites for bacteriophage P2 on the Escherichia coli chromosome: DNA sequences, localization on the physical map, and detection of a P2-like remnant in E. coli K-12 derivatives. J. Bacteriol. 174, 4086–4093. Bensing, B.A., Siboo, I.R., and Sullam, P.M. (2001). Proteins PblA and PblB of Streptococcus mitis, which promote binding to human platelets, are encoded within a lysogenic bacteriophage. Infect. Immun. 69, 6186–6192. Beres, S.B., Sylva, G.L., Barbian, K.D., Lei, B., Hoff, J.S., Mammarella, N.D., Liu, M.Y., Smoot, J.C., Porcella, S.F., Parkins, L.D., et al. (2002). Genome sequence of a serotype M3 strain of group A Streptococcus: phage-encoded toxins, the high-virulence phenotype, and clone emergence. Proc. Natl. Acad. Sci. USA 99. Beres, S.B., Sylva, G.L., Sturdevant, D.E., Granville, C.N., Liu, M., Ricklefs, S.M., Whitney, A.R., Parkins, L.D., Hoe, N.P., Adams, G.J., et al. (2004). Genome-wide molecular dissection of serotype M3 group A Streptococcus strains causing two epidemics of invasive infections. Proc. Natl. Acad. Sci. USA 101, 11833–11838. Betley, M.J., and Mekalanos, J.J. (1985). Staphylococcal enterotoxin A is encoded by phage. Science 229, 185–187. Bille, E., Zahar, J.R., Perrin, A., Morelle, S., Kriz, P., Jolley, K.A., Maiden, M.C., Dervin, C., Nassif, X., and Tinsley, C.R. (2005). A chromosomally integrated bacteriophage in invasive meningococci. J. Exp. Med. 201, 1905–1913. Boden, M.K., and Flock, J.I. (1994). Cloning and characterization of a gene for a 19 kDa fibrinogen-binding protein from Staphylococcus aureus. Mol. Microbiol. 12, 599–606. Botstein, D. (1980). A theory of modular evolution for bacteriophages. Ann. NY Acad. Sci. 354, 484– 491. Brodie, R., Smith, A.J., Roper, R.L., Tcherepanov, V., and Upton, C. (2004). Base-By-Base: single nucleotide-level analysis of whole viral genome alignments. BMC Bioinformatics 5, 96. Broudy, T.B., and Fischetti, V.A. (2003). In vivo lysogenic conversion of Tox(-) Streptococcus pyogenes to Tox(+) with Lysogenic Streptococci or free phage. Infect. Immun. 71, 3782–3786. Broudy, T.B., Pancholi, V., and Fischetti, V.A. (2001). Induction of lysogenic bacteriophage and phageassociated toxin from group A streptococci during coculture with human pharyngeal cells. Infect. Immun. 69, 1440–1443. Broudy, T.B., Pancholi, V., and Fischetti, V.A. (2002). The in vitro interaction of Streptococcus pyogenes with human pharyngeal cells induces a phage-encoded extracellular DNase. Infect. Immun. 70, 2805–2811. Campbell, A., Schneider, S.J., and Song, B. (1992). Lambdoid phages as elements of bacterial genomes. Genetica 86, 259–267. Canchaya, C., Desiere, F., McShan, W., Ferretti, J., Parkhill, J., and Brussow, H. (2002). Genome analysis of an inducible prophage and prophage remnants integrated in the Streptococcus pyogenes strain SF370. Virology 302, 245–258. Cantacuzene, J., and Boncieu, O. (1926). Modifications subies pare des streptococques d’origine non-scarlatineuse qu contact des produits scarlatineux filtres. CR Acad. Sci. 182, 1185. Cerdeno-Tarraga, A.M., Efstratiou, A., Dover, L.G., Holden, M.T., Pallen, M., Bentley, S.D., Besra, G.S., Churcher, C., James, K.D., De Zoysa, A., et al. (2003). The complete genome sequence and analysis of Corynebacterium diphtheriae NCTC13129. Nucleic Acids Res. 31, 6516–6523. Chatellier, S., Ihendyane, N., Kansal, R.G., Khambaty, F., Basma, H., Norrby-Teglund, A., Low, D.E., McGeer, A., and Kotb, M. (2000). Genetic relatedness and superantigen expression in group A streptococcus serotype M1 isolates from patients with severe and nonsevere invasive diseases. Infect. Immun. 68, 3523–3534. Chen, Y., Golding, I., Sawai, S., Guo, L., and Cox, E.C. (2005). Population fitness and the regulation of Escherichia coli genes by bacterial viruses. PLoS Biol. 3, e229. Cleary, P.P., Johnson, Z., and Wannamaker, L. (1975). Genetic instability of M protein and serum opacity factor of group A streptococci: evidence suggesting extrachromosomal control. Infect. Immun. 12, 109–118.

|

245

246

|

McShan and Feretti

Cleary, P.P., Kaplan, E.L., Handley, J.P., Wlazlo, A., Kim, M.H., Hauser, A.R., and Schlievert, P.M. (1992). Clonal basis for resurgence of serious Streptococcus pyogenes disease in the 1980s. Lancet 339, 518–521. Coetzee, J.N., and Sacks, T.G. (1960). Transduction of streptomycin resistance in Proteus mirabilis. J. Gen. Microbiol. 23, 445–455. Coleman, D., Knights, J., Russell, R., Shanley, D., Birkbeck, T.H., Dougan, G., and Charles, I. (1991). Insertional inactivation of the Staphylococcus aureus beta-toxin by bacteriophage f13 occurs by site- and orientation-specific integration of the f13 genome. Mol. Microbiol. 5, 933–939. Coleman, D.C., Sullivan, D.J., Russell, R.J., Arbuthnott, J.P., Carey, B.F., and Pomeray, H.M. (1989). Staphylococcus aureus bacteriophages mediating the simultaneous lysogenic conversion of beta-lysin, staphylokinase and enterotoxin A: molecular mechanism of triple conversion. J. Gen. Microbiol. 135, 1679–1697. Creuzburg, K., Recktenwald, J., Kuhle, V., Herold, S., Hensel, M., and Schmidt, H. (2005). The Shiga toxin 1-converting bacteriophage bp-4795 encodes an NleA-like type III effector protein. J. Bacteriol. 187, 8494–8498. Daubin, V., Lerat, E., and Perriere, G. (2003). The source of laterally transferred genes in bacterial genomes. Genome Biol. 4, R57. Davis, B.M., Lawson, E.H., Sandkvist, M., Ali, A., Sozhamannan, S., and Waldor, M.K. (2000). Convergence of the secretory pathways for cholera toxin and the filamentous phage, CTXphi. Science 288, 333–335. Eklund, M.W., Poysky, F.T., Mseitif, L.M., and Strom, M.S. (1988). Evidence for plasmid-mediated toxin and bacteriocin production in Clostridium botulinum type G. Appl. Environ. Microbiol. 54, 1405– 1408. Eklund, M.W., Poysky, F.T., and Reed, S.M. (1972). Bacteriophage and the toxigenicity of Clostridium botulinum type D. Nat. New Biol. 235, 16–17. Eklund, M.W., Poysky, F.T., Reed, S.M., and Smith, C.A. (1971). Bacteriophage and the toxigenicity of Clostridium botulinum type C. Science 172, 480–482. Ferretti, J.J., McShan, W.M., Ajdic, D., Savic, D.J., Savic, G., Lyon, K., Primeaux, C., Sezate, S., Suvorov, A.N., Kenton, S., et al. (2001). Complete Genome Sequence of an M1 Strain of Streptococcus pyogenes. Proc. Natl. Acad. Sci. USA 98, 4658–4663. Figueroa-Bossi, N., and Bossi, L. (1999). Inducible prophages contribute to Salmonella virulence in mice. Mol. Microbiol. 33, 167–176. Figueroa-Bossi, N., Uzzau, S., Maloriol, D., and Bossi, L. (2001). Variable assortment of prophages provides a transferable repertoire of pathogenic determinants in Salmonella. Mol. Microbiol. 39, 260–271. Foster, T.J. (2004). The Staphylococcus aureus “superbug.” J. Clin. Invest 114, 1693–1696. Freeman, V.J. (1951). Studies on the virulence of bacteriophage-infected strains of Corynebacterium diphtheriae. J. Bacteriol. 61, 675–688. Freeman, V.J., and Morse, I.U. (1952). Further observations on the change to virulence of bacteriophageinfected avirulent strains of Corynebacterium diphtheriae. J. Bacteriol. 63, 407–414. Frobisher, M., and Brown, J.H. (1927). Transmissible toxicogenicity of streptococci. Bull Johns Hopkins Hosp 41, 167–173. Frobisher, M., Parsons, E.I., and Updyke, E. (1947). The correlation of laboratory and clinical evidence of virulence of C. diphtheriae. Am. J. Pub. Health 37, 543–548. Fujii, N., Oguma, K., Yokosawa, N., Kimura, K., and Tsuzuki, K. (1988). Characterization of bacteriophage nucleic acids obtained from Clostridium botulinum types C and D. Appl. Environ. Microbiol. 54, 69–73. Genestier, A.L., Michallet, M.C., Prevost, G., Bellot, G., Chalabreysse, L., Peyrol, S., Thivolet, F., Etienne, J., Lina, G., Vallette, F.M., et al. (2005). Staphylococcus aureus Panton-Valentine leukocidin directly targets mitochondria and induces Bax-independent apoptosis of human neutrophils. J. Clin. Invest. 115, 3117–3127. Gillet, Y., Issartel, B., Vanhems, P., Fournet, J.C., Lina, G., Bes, M., Vandenesch, F., Piemont, Y., Brousse, N., Floret, D., and Etienne, J. (2002). Association between Staphylococcus aureus strains carrying gene for Panton-Valentine leukocidin and highly lethal necrotising pneumonia in young immunocompetent patients. Lancet 359, 753–759.

Prophages and their Contribution to Host Cell Phenotype

Gladstone, G.P., and Van Heyningen, W.E. (1957). Staphylococcal leucocidins. Br. J. Exp. Pathol. 38, 123–137. Goshorn, S.C., and Schlievert, P.M. (1989). Bacteriophage association of streptococcal pyrogenic exotoxin type C. J. Bacteriol. 171, 3068–3073. Green, N.M., Zhang, S., Porcella, S.F., Nagiec, M.J., Barbian, K.D., Beres, S.B., Lefebvre, R.B., and Musser, J.M. (2005). Genome sequence of a serotype M28 strain of group A Streptococcus: potential new insights into puerperal sepsis and bacterial disease specificity. J. Infect. Dis. 192, 760–770. Hauser, D., Gibert, M., Marvaud, J.C., Eklund, M.W., and Popoff, M.R. (1995). Botulinal neurotoxin C1 complex genes, clostridial neurotoxin homology and genetic transfer in Clostridium botulinum. Toxicon 33, 515–526. Hayashi, T., Baba, T., Matsumoto, H., and Terawaki, Y. (1990). Phage-conversion of cytotoxin production in Pseudomonas aeruginosa. Mol. Microbiol. 4, 1703–1709. Ho, T.D., Figueroa-Bossi, N., Wang, M., Uzzau, S., Bossi, L., and Slauch, J.M. (2002). Identification of GtgE, a novel virulence factor encoded on the Gifsy-2 bacteriophage of Salmonella enterica serovar Typhimurium. J. Bacteriol. 184, 5234–5239. Holloway, B.W., and Cooper, G.N. (1962). Lysogenic conversion in Pseudomonas aeruginosa. J. Bacteriol. 84, 1321–1324. Holmes, R.K., and Barksdale, L. (1969). Genetic analysis of tox+ and tox- bacteriophages of Corynebacterium diphtheriae. J. Virol. 3, 586–598. Huan, P.T., Bastin, D.A., Whittle, B.L., Lindberg, A.A., and Verma, N.K. (1997a). Molecular characterization of the genes involved in O-antigen modification, attachment, integration and excision in Shigella flexneri bacteriophage SfV. Gene 195, 217–227. Huan, P.T., Whittle, B.L., Bastin, D.A., Lindberg, A.A., and Verma, N.K. (1997b). Shigella flexneri typespecific antigen V: cloning, sequencing and characterization of the glucosyl transferase gene of temperate bacteriophage SfV. Gene 195, 207–216. Huang, A., Friesen, J., and Brunton, J.L. (1987). Characterization of a bacteriophage that carries the genes for production of Shiga-like toxin 1 in Escherichia coli. J. Bacteriol. 169, 4308–4312. Hupkova, M., Blahova, J., and Krcmery, V. (1994). Transfer, by conjugation and transduction, of resistance to ceftazidime and cefotaxime from the same clinical isolate of Pseudomonas aeruginosa. Zentralbl. Bakteriol. 281, 196–200. Hyder, S.L., and Streitfeld, M.M. (1978). Transfer of erythromycin resistance from clinically isolated lysogenic strains of Streptococcus pyogenes via their endogenous phage. J. Infect. Dis. 138, 281–286. Hynes, W.L., and Ferretti, J.J. (1989). Sequence analysis and expression in Escherichia coli of the hyaluronidase gene of Streptococcus pyogenes bacteriophage H4489A. Infect. Immun. 57, 533–539. Hynes, W.L., Hancock, L., and Ferretti, J.J. (1995). Analysis of a second bacteriophage hyaluronidase gene from Streptococcus pyogenes: Evidence for a third hyaluronidase involved in extracellular enzymatic activity. Infect. Immun. 63, 3015–3020. Iandolo, J.J., Worrell, V., Groicher, K.H., Qian, Y., Tian, R., Kenton, S., Dorman, A., Ji, H., Lin, S., Loh, P., et al. (2002). Comparative analysis of the genomes of the temperate bacteriophages phi 11, phi 12 and phi 13 of Staphylococcus aureus 8325. Gene 289, 109–118. Inoue, K., and Iida, H. (1971). Phage-conversion of toxigenicity in Clostridium botulinum types C and D. Jpn J. Med. Sci. Biol. 24, 53–56. Johnson, L.P., and Schlievert, P.M. (1984). Group A streptococcal phage T12 carries the structural gene for pyrogenic exotoxin type A. Mol. Gen. Genet. 194, 52–56. Juhala, R.J., Ford, M.E., Duda, R.L., Youlton, A., Hatfull, G.F., and Hendrix, R.W. (2000). Genomic sequences of bacteriophages HK97 and HK022: pervasive genetic mosaicism in the lambdoid bacteriophages. J. Mol. Biol. 299, 27–51. Kaneko, J., Kimura, T., Kawakami, Y., Tomita, T., and Kamio, Y. (1997). Panton-valentine leukocidin genes in a phage-like particle isolated from mitomycin C-treated Staphylococcus aureus V8 (ATCC 49775). Biosci. Biotechnol. Biochem. 61, 1960–1962. Kaneko, J., Kimura, T., Narita, S., Tomita, T., and Kamio, Y. (1998). Complete nucleotide sequence and molecular characterization of the temperate staphylococcal bacteriophage phiPVL carrying PantonValentine leukocidin genes. Gene 215, 57–67. Karaolis, D.K., Johnson, J.A., Bailey, C.C., Boedeker, E.C., Kaper, J.B., and Reeves, P.R. (1998). A Vibrio cholerae pathogenicity island associated with epidemic and pandemic strains. Proc. Natl. Acad. Sci. USA 95, 3134–3139.

|

247

248

|

McShan and Feretti

Karaolis, D.K., Somara, S., Maneval, D.R., Jr., Johnson, J.A., and Kaper, J.B. (1999). A bacteriophage encoding a pathogenicity island, a type-IV pilus and a phage receptor in cholera bacteria. Nature (London) 399, 375–379. Kazmi, S.U., Kansal, R., Aziz, R.K., Hooshdaran, M., Norrby-Teglund, A., Low, D.E., Halim, A.B., and Kotb, M. (2001). Reciprocal, temporal expression of SpeA and SpeB by invasive M1T1 group a streptococcal isolates in vivo. Infect. Immun. 69, 4988–4995. Koch, B., Christiansen, B., Evison, T., Vogensen, F.K., and Hammer, K. (1997). Construction of specific erythromycin resistance mutations in the temperate lactococcal bacteriophage TP901-1 and their use in studies of phage biology. Appl. Environ. Microbiol. 63, 2439–2441. Kropinski, A.M. (2000). Sequence of the genome of the temperate, serotype-converting, Pseudomonas aeruginosa bacteriophage D3. J. Bacteriol. 182, 6066–6074. Kuhn, J., and Campbell, A. (2001). The bacteriophage lambda attachment site in wild strains of Escherichia coli. J. Mol. Evol. 53, 607–614. Kuzio, J., and Kropinski, A.M. (1983). O-antigen conversion in Pseudomonas aeruginosa PAO1 by bacteriophage D3. J. Bacteriol. 155, 203–212. Lee, C.Y., and Iandolo, J.J. (1986). Lysogenic conversion of staphylococcal lipase is caused by insertion of the bacteriophage L54a genome into the lipase structural gene. J. Bacteriol. 166, 385–391. Lee, J.H., Wang, T., Ault, K., Liu, J., Schmitt, M.P., and Holmes, R.K. (1997). Identification and characterization of three new promoter/operators from Corynebacterium diphtheriae that are regulated by the diphtheria toxin repressor (DtxR) and iron. Infect. Immun. 65, 4273–4280. Lindsay, J.A., Ruzin, A., Ross, H.F., Kurepina, N., and Novick, R.P. (1998). The gene for toxic shock toxin is carried by a family of mobile pathogenicity islands in Staphylococcus aureus. Mol. Microbiol. 29, 527–543. Lofdahl, S., Sjostrom, J.E., and Philipson, L. (1981). Cloning of restriction fragments of DNA from staphylococcal bacteriophage phi 11. J. Virol. 37, 795–801. Luneberg, E., Mayer, B., Daryab, N., Kooistra, O., Zahringer, U., Rohde, M., Swanson, J., and Frosch, M. (2001). Chromosomal insertion and excision of a 30 kb unstable genetic element is responsible for phase variation of lipopolysaccharide and other virulence determinants in Legionella pneumophila. Mol. Microbiol. 39, 1259–1271. Malke, H. (1969). Transduction of Streptococcus pyogenes K 56 by temperature-sensitive mutants of the transducing phage A25. Z. Naturforsch. B 24, 1556–1561. Malke, H., Starke, R., Kohler, W., Kolesnichenko, G., and Totolian, A.A. (1975). Bacteriophage P13234mo-mediated intra- and intergroup transduction of antibiotic resistance among streptococci. Zentralblatt Fur Bakteriologie, Parasitenkunde, Infektionskrankheiten Und Hygiene—Erste Abteilung Originale—Reihe A: Medizin. Mikrobiol. Parasitol. 233, 24–34. Mason, R.E., and Allen, W.E. (1975). Characteristics of Staphylococcus aureus associated with lysogenic conversion to loss of beta-hemolysin production. Can. J. Microbiol. 21, 1113–1116. McDonald, L.C., Killgore, G.E., Thompson, A., Owens, R.C., Jr., Kazakova, S.V., Sambol, S.P., Johnson, S., and Gerding, D.N. (2005). An epidemic, toxin gene-variant strain of Clostridium difficile. N. Engl. J. Med. McShan, W.M. (2005). The bacteriophages of group A streptococci. In: Gram-Positive Pathogens, V.A. Fischetti, R.P. Novick, J.J. Ferretti, D.A. Portnoy, and J.I. Rood, eds. (Washington DC: American Society for Microbiology). Mehta, P., Casjens, S., and Krishnaswamy, S. (2004). Analysis of the lambdoid prophage element e14 in the E. coli K-12 genome. BMC Microbiol. 4, 4. Melles, D.C., Gorkink, R.F., Boelens, H.A., Snijders, S.V., Peeters, J.K., Moorhouse, M.J., van der Spek, P.J., van Leeuwen, W.B., Simons, G., Verbrugh, H.A., and van Belkum, A. (2004). Natural population dynamics and expansion of pathogenic clones of Staphylococcus aureus. J. Clin. Invest. 114, 1732–1740. Mirold, S., Rabsch, W., Rohde, M., Stender, S., Tschape, H., Russmann, H., Igwe, E., and Hardt, W.D. (1999). Isolation of a temperate bacteriophage encoding the type III effector protein SopE from an epidemic Salmonella typhimurium strain. Proc. Natl. Acad. Sci. USA 96, 9845–9850. Nakagawa, I., Kurokawa, K., Yamashita, A., Nakata, M., Tomiyasu, Y., Okahashi, N., Kawabata, S., Yamazaki, K., Shiba, T., Yasunaga, T., et al. (2003). Genome sequence of an M3 strain of Streptococcus pyogenes reveals a large-scale genomic rearrangement in invasive strains and new insights into phage evolution. Genome Res. 13, 1042–1055.

Prophages and their Contribution to Host Cell Phenotype

Nakayama, K., Kanaya, S., Ohnishi, M., Terawaki, Y., and Hayashi, T. (1999). The complete nucleotide sequence of phi CTX, a cytotoxin-converting phage of Pseudomonas aeruginosa: implications for phage evolution and horizontal gene transfer via bacteriophages. Mol. Microbiol. 31, 399–419. Newland, J.W., Strockbine, N.A., Miller, S.F., O’Brien, A.D., and Holmes, R.K. (1985). Cloning of Shigalike toxin structural genes from a toxin converting phage of Escherichia coli. Science 230, 179–181. Nnalue, N.A., Newton, S., and Stocker, B.A. (1990). Lysogenization of Salmonella choleraesuis by phage 14 increases average length of O-antigen chains, serum resistance and intraperitoneal mouse virulence. Microb. Pathog. 8, 393–402. Novick, R.P. (1991). Genetic systems in staphylococci. Methods Enzymol. 204, 587–636. Pattee, P.A., and Baldwin, J.N. (1961). Transduction of resistance to chlortetracycline and novobiocin in Staphylococcus aureus. J. Bacteriol. 82, 875–881. Plunkett, G., 3rd, Rose, D.J., Durfee, T.J., and Blattner, F.R. (1999). Sequence of Shiga toxin 2 phage 933W from Escherichia coli O157:H7: Shiga toxin as a phage late-gene product. J. Bacteriol. 181, 1767–1778. Ptashne, M. (2004). A genetic switch: phage lambda revisited, 3rd edn (Cambridge, MA: Cell Press: Blackwell Scientific Publications). Rajanna, C., Wang, J., Zhang, D., Xu, Z., Ali, A., Hou, Y.M., and Karaolis, D.K. (2003). The vibrio pathogenicity island of epidemic Vibrio cholerae forms precise extrachromosomal circular excision products. J. Bacteriol. 185, 6893–6901. Ratti, G., Covacci, A., and Rappuoli, R. (1997). A tRNA(2Arg) gene of Corynebacterium diphtheriae is the chromosomal integration site for toxinogenic bacteriophages. Mol. Microbiol. 25, 1179–1181. Rivera Starr, C., and Engleberg, N.C. (2006). Role of Hyaluronidase in Subcutaneous Spread and Growth of Group A Streptococcus. Infect. Immun. 74, 40–48. Saitoh, M., Tanaka, K., Nishimori, K., Makino, S., Kanno, T., Ishihara, R., Hatama, S., Kitano, R., Kishima, M., Sameshima, T., et al. (2005). The artAB genes encode a putative ADP-ribosyltransferase toxin homologue associated with Salmonella enterica serovar Typhimurium DT104. Microbiology 151, 3089–3096. Sakaguchi, Y., Hayashi, T., Kurokawa, K., Nakayama, K., Oshima, K., Fujinaga, Y., Ohnishi, M., Ohtsubo, E., Hattori, M., and Oguma, K. (2005). The genome sequence of Clostridium botulinum type C neurotoxin-converting phage and the molecular mechanisms of unstable lysogeny. Proc. Natl. Acad. Sci. USA 102, 17472–17477. Sako, T., Sawaki, S., Sakurai, T., Ito, S., Yoshizawa, Y., and Kondo, I. (1983). Cloning and expression of the staphylokinase gene of Staphylococcus aureus in Escherichia coli. Mol. Gen. Genet. 190, 271–277. Shimizu-Kadota, M., Kiwaki, M., Sawaki, S., Shirasawa, Y., Shibahara-Sone, H., and Sako, T. (2000). Insertion of bacteriophage phiFSW into the chromosome of Lactobacillus casei strain Shirota (S-1): characterization of the attachment sites and the integrase gene. Gene 249, 127–134. Siboo, I.R., Bensing, B.A., and Sullam, P.M. (2003). Genomic organization and molecular characterization of SM1, a temperate bacteriophage of Streptococcus mitis. J. Bacteriol. 185, 6968–6975. Smith, N.L., Taylor, E.J., Lindsay, A.M., Charnock, S.J., Turkenburg, J.P., Dodson, E.J., Davies, G.J., and Black, G.W. (2005). Structure of a group A streptococcal phage-encoded virulence factor reveals a catalytically active triple-stranded beta-helix. Proc. Natl. Acad. Sci. USA 102, 17652–17657. Smoot, J.C., Barbian, K.D., Van Gompel, J.J., Smoot, L.M., Chaussee, M.S., Sylva, G.L., Sturdevant, D.E., Ricklefs, S.M., Porcella, S.F., Parkins, L.D., et al. (2002). Genome sequence and comparative microarray analysis of serotype M18 group A Streptococcus strains associated with acute rheumatic fever outbreaks. Proc. Natl. Acad. Sci. USA 99. Spanier, J.G., and Cleary, P.P. (1980). Bacteriophage control of antiphagocytic determinants in group A streptococci. J. Exp. Med. 152, 1393–1406. Spanier, J.G., and Cleary, P.P. (1983). A restriction map and analysis of the terminal reduncacy in the Group A streptococcal bacteriophage SP 24. Virology 130, 502–513. Stanley, T.L., Ellermeier, C.D., and Slauch, J.M. (2000). Tissue-specific gene expression identifies a gene in the lysogenic phage Gifsy-1 that affects Salmonella enterica serovar Typhimurium survival in Peyer’s patches. J. Bacteriol. 182, 4406–4413. Strauch, E., Lurz, R., and Beutin, L. (2001). Characterization of a Shiga toxin-encoding temperate bacteriophage of Shigella sonnei. Infect. Immun. 69, 7588–7595. Sumby, P., Porcella, S.F., Madrigal, A.G., Barbian, K.D., Virtaneva, K., Ricklefs, S.M., Sturdevant, D.E., Graham, M.R., Vuopio-Varkila, J., Hoe, N.P., and Musser, J.M. (2005). Evolutionary origin and emer-

|

249

250

|

McShan and Feretti

gence of a highly successful clone of serotype M1 group A Streptococcus involved multiple horizontal gene transfer events. J. Infect. Dis. 192, 771–782. Thomas, M.S., and Drabble, W.T. (1986). Secondary attachment site for bacteriophage lambda in the guaB gene of Escherichia coli. J. Bacteriol. 168, 1048–1050. Ubukata, K., Konno, M., and Fujii, R. (1975). Transduction of drug resistance to tetracycline, chloramphenicol, macrolides, lincomycin and clindamycin with phages induced from Streptococcus pyogenes. J. Antibiotics 28, 681–688. Uchida, T., Gill, D.M., and Pappenheimer, A.M., Jr. (1971). Mutation in the structural gene for diphtheria toxin carried by temperate phage. Nat. New Biol. 233, 8–11. Vaca-Pacheco, S., Paniagua-Contreras, G.L., Garcia-Gonzalez, O., and de la Garza, M. (1999). The clinically isolated FIZ15 bacteriophage causes lysogenic conversion in Pseudomonas aeruginosa PAO1. Curr. Microbiol. 38, 239–243. Val, M.E., Bouvier, M., Campos, J., Sherratt, D., Cornet, F., Mazel, D., and Barre, F.X. (2005). The singlestranded genome of phage CTX is the form used for integration into the genome of Vibrio cholerae. Mol. Cell 19, 559–566. Waldor, M.K., and Mekalanos, J.J. (1996). Lysogenic conversion by a filamentous phage encoding cholera toxin. Science 272, 1910–1914. Watanabe, T., and Watanabe, M. (1959). Transduction of streptomycin resistance in Salmonella typhimurium. J. Gen. Microbiol. 21, 16–29. Weeks, C.R., and Ferretti, J.J. (1984). The gene for type A streptococcal exotoxin (erythrogenic toxin) is located in bacteriophage T12. Infect. Immun. 46, 531–536. Weisberg, R.A., and Landy, A. (1983). Site-specific recombination in phage lambda. In: Lambda II, J.W.R., Roger W. Hendrix, Franklin W. Stahl and Robert A. Weisberg, ed. (Cold Spring Harbor, NY: Cold Spring Harbor Press), pp. 211–250. West, N.P., Sansonetti, P., Mounier, J., Exley, R.M., Parsot, C., Guadagnini, S., Prevost, M.C., ProchnickaChalufour, A., Delepierre, M., Tanguy, M., and Tang, C.M. (2005). Optimization of virulence functions through glucosylation of Shigella LPS. Science 307, 1313–1317. Willi, K., Sandmeier, H., Kulik, E.M., and Meyer, J. (1997). Transduction of antibiotic resistance markers among Actinobacillus actinomycetemcomitans strains by temperate bacteriophages Aa phi 23. Cell Mol. Life Sci. 53, 904–910. Williams, K.P. (2002). Integration sites for genetic elements in prokaryotic tRNA and tmRNA genes: sublocation preference of integrase subfamilies. Nucleic Acids Res. 30, 866–875. Yamaguchi, T., Hayashi, T., Takami, H., Nakasone, K., Ohnishi, M., Nakayama, K., Yamada, S., Komatsuzawa, H., and Sugai, M. (2000). Phage conversion of exfoliative toxin A production in Staphylococcus aureus. Mol. Microbiol. 38, 694–705. Zabriskie, J.B. (1964). The role of temperate bacteriophage in the production of erythrogenic toxin by group A streptococci. J. Exp. Med. 119, 761–779.

Prophage Induction of Phage L John W. Little

10

Abstract The gene regulatory circuitry of phage L is among the best-understood circuits at the mechanistic level. This circuitry involves several interesting regulatory behaviors. An infected cell undergoes a decision between two alternative pathways, the lytic and lysogenic pathways. If the latter is followed, the lysogenic state is established and maintained. While this state is highly stable, it can switch to the lytic pathway in the process of prophage induction, which occurs when the host SOS response is triggered by DNA damage. We first review events that stabilize the lysogenic pathway and that affect the lysis–lysogeny decision. This is followed by a more extensive description of prophage induction. This process displays threshold behavior, and a model for this behavior is presented. Other aspects of systems behavior in the L circuitry are briefly described. Finally, we discuss insights into the evolution of complex regulatory circuits afforded by studies with the L circuitry. Introduction The gene regulatory circuitry of bacteriophage L is one of the best-understood complex circuits (Ptashne, 2004; Little ). L has been studied intensively for fifty years, and many general features of gene regulatory circuitry that we now take for granted were first discovered in L. Our understanding of the L circuitry is deep at the mechanistic level, and it is beginning to be utilized as a vehicle for systems approaches as well. L has a more complex life cycle than most phages (Figure 10.1), and it displays a correspondingly wide range of interesting gene regulatory behavior. Three primary aspects of its behavior have received wide attention. First, L undergoes a choice, soon after infecting a cell, between two alternative pathways, the lytic pathway and the lysogenic pathway (Herskowitz and Hagen, 1980; Echols, 1986b). Second, if the cell follows the lysogenic pathway, the viral genome is integrated into the host, and the lytic genes are repressed by a master regulatory protein, termed CI. This gene regulatory state, termed the lysogenic state, is extremely stable. Third, a lysogen can be induced to switch from the lysogenic state to the lytic state in an epigenetic switch termed prophage induction (Roberts and Devoret, 1983; Little, 1993). This switch of regulatory state involves a host gene regulatory system called the SOS system (Little and Mount, 1982), and results in execution of the lytic pathway.

Fig 1. 252

|

Little

Figure 10.1 Life cycle of phage L. An infected cell can follow either of two pathways, as depicted. The lytic pathway takes about 1 hour, and results in cell lysis. The lysogenic pathway results in a highly stable regulatory state, the lysogenic state. Although highly stable, it can switch to the lytic pathway when the host SOS system is induced by DNA damage. See text for details.

It is the goal of this chapter to outline the molecular mechanisms underlying each of these regulatory decisions, states and switches. This will serve as background to a more detailed consideration of prophage induction. We will also speculate about several less well-understood aspects of prophage induction, with the intent of generating testable hypotheses. The lytic pathway Lytic growth of L, like that of most phages, follows a temporal pathway that will not be described here in detail (Herskowitz and Hagen, 1980; Echols, 1986a; Ptashne, 2004). Upon infection, gene expression starts from two early promoters, termed PL and PR (Figure 10.2); the only essential gene expressed from PL is N, encoding an anti-termination protein (Das, 1992; Roberts, 1993). From PR are expressed two DNA replication genes, O and P, and the Q gene (not depicted), which is required for transcription of the late genes (Roberts, 1993; Kobiler et al., 2005). Accordingly, to block the lytic pathway, it is necessary only to block transcription from PL and PR. CI protein and stabilization of the lysogenic state Although the lysis–lysogeny decision is the first regulatory decision to be made upon infection by L, it is simpler to begin with the events that stabilize the lysogenic state. This will form a basis for understanding what needs to happen during the lysis-lysogeny decision. As stated, the lysogenic state is stabilized by the action of the CI protein, often termed L repressor. CI is a specific DNA binding protein with many different activities (Figure 10.3).

Prophage Induction of Phage L

Figure 10.2 Maps of L and critical regulatory regions. Maps are to scale. (A) Map of the regulatory region of L, showing most of the sites of action of regulatory proteins. To the left of the depicted portion lies the att site, at which site-specific recombination occurs with a cognate site in the host chromosome; to the right of the region shown lies the Q gene, expressed from the PR promoter and encoding a factor required for expression of the late genes. The “R” and “L” in PR and PL stand respectively for rightward and leftward. The OL region contains three binding sites, OL1, OL2 and OL3, to which both CI and Cro bind (not depicted); see Figure 3. (B) Map of the cI–cro interval. Depicted are the location of the cI and cro genes; the locations of three promoters driving expression of these genes; and the location of the OR region. (C) Detailed map of the OR region, showing the location of operators to which CI and Cro bind, and the two promoters regulated by these proteins. See text for details.

These properties contribute in various ways to the behavior of the L circuitry. We first present an overview of these activities, then describe them and explore their consequences in more detail. CI is organized into two domains (Pabo et al., 1979) (Figure 10.3A), an N-terminal DNA-binding (DBD) domain and a C-terminal domain that carries out multiple functions as described below. These domains are separated by a connector or hinge region. It has been believed that this hinge region is perhaps 40 residues in length, but recent structural evidence (Luo et al., 2001) with a homologous protein, the E. coli LexA repressor, suggests that the hinge is much shorter and that most of the region previously described as a hinge is probably organized into a folded structure that helps modulate one of the activities of

|

253

254

|

Little

3.

Figure 10.3 CI—Organization and functions. (A) Organization of CI. CI is pronounced “seeone,” not “see-eye.” Location of N-terminal and C-terminal domains is shown. The hinge region is short and its exact size is unknown. Location of the active site for cleavage, including the active site Ser and Lys residues, is indicated. CSR, cleavage site region; clv site, cleavage site. See text for details. (B) Activities of CI. See text for details. The detailed organization of the looped structures is not known, and several different arrangements are possible; one plausible arrangement is depicted. Part B modified from Michalowski and Little, 2005.

this protein. Many of the features shown in Figure 10.3 are described here or later in the section on prophage induction. CI has multiple activities (Figure 10.3). First, it forms dimers by specific contacts in its C-terminal domain (Bell et al., 2000). Dimerization is relatively strong (Kdimer ≈5–10 nM)

Prophage Induction of Phage L

(Koblan and Ackers, 1991), but some monomers are present. Second, CI dimers bind specifically to six operator sites, located in two regulatory regions termed the OL and OR regions (Ptashne, 2004). Molecular events at the latter region (Figure 10.2D) are the most crucial determinant of the regulatory behavior, as detailed below. The affinities of CI for each of these operators differs somewhat (Koblan and Ackers, 1992), contributing to the detailed regulatory behavior. Third, CI binds cooperatively to adjacent binding sites in each of the OL and OR regions (Burz and Ackers, 1994). This form of binding is called “pairwise cooperativity” ( Johnson et al., 1979). Cooperativity results from protein-protein interactions in the C-terminal domain (Bell and Lewis, 2001; Bell et al., 2000). Since CI binds most tightly to OL1 and OR1 ( Johnson et al., 1979), and more weakly to the remaining four operators, pairwise cooperative binding results in occupancy of OR1 and OR2, and of OL1 and OL2. Fourth, CI acts to stimulate its own expression from a weak promoter, termed PRM, located in the OR region ( Jain et al., 2004). This action occurs when CI is bound to the OR2 site, and is mediated by a weak protein-protein interaction between the DNA-binding domain and the S70 subunit of RNA polymerase (Li et al., 1994; Jain et al., 2004). Fifth, CI undergoes a specific proteolytic cleavage reaction during prophage induction. Cleavage separates the DNA-binding domain from the dimerization domain, and hence it inactivates the DNA-binding function of CI. The cleavage reaction is unusual, in that CI cleaves itself, rather than being subject to the action of a specific protease (Little, 1984; Little, 1993); this activity is modulated in the cell by the action of an activated form of RecA protein. Sixth, in addition to pairwise cooperativity, CI can undergo long-range cooperative interactions between a pair of dimers at OL and a pair of dimers at OR, forming a looped structure (Révet et al., 1999; Dodd et al., 2001). Finally, two more dimers can bind to OL3 and OR3, still another form of cooperativity. These activities of CI are crucial to its roles in L regulation. As stated, the crucial regulatory events occur in the OR region. This region contains three binding sites for CI, and two promoters, PR and PRM, which act to favor the lytic and lysogenic states, respectively. Because of cooperative DNA binding and the high affinity for OR1, CI generally occupies OR1 and OR2 at the same time. This binding has two different consequences. First, the PR promoter is repressed, preventing expression of lytic genes, and blocking lytic growth. Second, CI bound to OR2 acts to stimulate its own expression from PRM about ten-fold (Meyer et al., 1980; Meyer and Ptashne, 1980; Hochschild et al., 1983). This activity, termed positive autoregulation or positive control, acts to increase the level of CI in the cell, further stabilizing the lysogenic state. Looping between OL and OR is thought to play two related roles in regulatory circuitry. First, at relatively low CI levels, looping increases the occupancy of OR1 and OR2, affording greater repression of the PR promoter (Révet et al., 1999). This acts to stabilize the regulatory circuitry by preventing occasional gene expression from this promoter. Increased occupancy also stimulates PRM at somewhat lower CI levels than occurs in the absence of looping. Second, at higher CI levels, cooperative binding to OR3 acts to repress the PRM promoter (Dodd et al., 2001). This negative autoregulation likely serves the purpose of damping out fluctuations in the level of CI in a lysogen (see also Becskei and Serrano, 2000).

|

255

256

|

Little

To summarize, in the lysogenic state CI acts at the OR region to perpetuate its own expression and prevent the expression of the lytic promoter PR. The PL promoter is also repressed by CI binding to OL1 and/or OL2, and cooperative binding serves to increase repression by the same mechanisms as described above for the OR region. Accordingly, in a cell containing CI, the lysogenic state is highly stable, both because lytic genes are almost completely repressed, and because CI acts to maintain a relatively constant and high level of CI. Cro protein and the anti-immune state In addition to CI, L has a second repressor, termed Cro. It is expressed from the PR promoter. Strikingly, Cro binds to the same six operators as those to which CI binds. Cro binds to these sites with differing affinities, and its pattern of relative affinities is different from that of CI. In the OR region, Cro binds tightly to OR3, and weakly to OR1 and OR2 ( Johnson et al., 1978; Darling et al., 2000). When bound to OR3, Cro represses PRM but has no effect on PR (Meyer et al., 1980). When Cro is present at higher levels, it binds to OR1 and/or OR2, repressing PR as well; accordingly, Cro is also under negative autoregulation. Cro is required for lytic growth (Eisen et al., 1970; Echols et al., 1973; Eisen et al., 1975; Folkmanis et al., 1977), and it is believed that this requirement involves partial repression of PL and PR; the molecular basis is not completely clear. The other apparent role of Cro is to favor the lytic pathway, by binding to OR3 and repressing the weak PRM promoter. The lytic pathway does not apparently need to be highly stable, since the cells lyse after ~50–60 min. However, if lytic functions are knocked out by mutation (in N and in a DNA-replication gene (O or P)), the cells can persist in an “anti-immune” state in which Cro is made and represses cI (Eisen et al., 1970; Calef et al., 1971; Neubauer and Calef, 1970). This state is less stable than the lysogenic state, and it is not believed to play a role in normal L gene regulation. However, its behavior is parallel to that of CI: If the cell contains Cro and no CI, that pattern can be perpetuated. Conversely, in a cell with CI and no Cro that alternative pattern is maintained. Hence, the L regulatory circuitry has the property of “bistability”—it can exist in two alternative stable states. The existence and activities of Cro raise an apparent paradox. When a cell is infected with L, there is no CI present, and Cro is promptly made. How can the lysogenic state ever be achieved? How can the cell make any CI? The answer, to which we now turn, is that CI is made initially from another promoter, termed PRE (Figure 10.2C), that is regulated by a different set of regulatory controls. The lysis–lysogeny decision Soon after a cell is infected with L, it undergoes a decision (Herskowitz and Hagen, 1980; Echols, 1986a) between the lytic and lysogenic pathways (Figure 10.1). This decision is not made immediately, but is deferred for 10–15 min, by which time the L DNA has begun replicating. It is believed that this delay provides an opportunity for the infecting phage to “assess the conditions.” It is not clear exactly what this means, nor how the phage evaluates the conditions, nor what criteria are employed. However, it is known that the decision is influenced by the physiological state of the cell and by the number of phage infecting

Prophage Induction of Phage L

the cell. We first describe the mechanisms underlying the decision, then return to some possible mechanisms for regulating the likelihood of each response. The critical protein for determining whether the infected cell will follow the lysogenic pathway is another phage protein termed CII. CII is a strong transcriptional activator. If high levels of CII are present (see below), it activates three different promoters (Kobiler et al., 2005; Little, 2005); transcription from these promoters works in various ways to favor the lysogenic response. The most direct mechanism is stimulation of the PRE promoter (Figure 10.2C), which transcribes cI. Transcription of cI from PRE produces a large burst of CI, giving a level of CI 10–15 times higher than that found in a lysogen (Reichardt and Kaiser, 1971; Reichardt, 1975). This overproduction probably results both because PRE is a stronger promoter than PRM, and because the PRE mRNA contains a Shine-Dalgarno sequence, unlike the leaderless PRM mRNA (Ptashne et al., 1976), which is translated less efficiently (Shean and Gottesman, 1992). Overproduction of CI results in complete shutoff of transcription from PL and PR, and probably of PRM as well; since CI is stable in vivo, the levels decline slowly over several generations until the level approaches that found in a lysogen. As noted, the lysis-lysogeny decision is not made immediately upon infection. Recent studies with PRE:GFP fusions have examined the kinetics of CII function after infection (Kobiler et al., 2005). CII function appears only after about 15–20 minutes, even in cells that are driven strongly towards the lysogenic response by infection with several phages (see also below). This finding strongly suggests that the decision is deferred for this length of time. Although both PRM and PRE drive the synthesis of CI, the two promoters have different regulatory functions and different meanings. The function of PRE is to establish the lysogenic state (hence the “E” in its name), while that of PRM is to maintain this state (hence the “M”). The “meaning” of expression from PRE is that CII levels are high, while that of PRM is that CI levels are high. How then are CII levels regulated? CII is made from the PR promoter; hence this promoter also leads to expression of a gene favoring the lysogenic pathway. CII is highly unstable, with a half-life of ~2 minutes (Gottesman et al., 1981; Kobiler et al., 2002). CII is degraded by a host protease, variously termed FtsH and HflB (Ito and Akiyama, 2005). FtsH is an integral membrane protein, complicating analysis of mechanisms for regulating its activity. The action of FtsH appears to be modulated both by a phage protein, CIII, and by other membrane-bound proteins termed HflKC and HflD. CIII is made from the PL promoter, and acts as a competitive inhibitor, binding to FtsH but undergoing slow proteolysis (Herman et al., 1997). The actions of HflKC and HflD are poorly understood at present, and their localization to the membrane is likely to complicate biochemical analysis. It is plausible that the level of CII can be modulated by regulating the activity of FtsH in various ways, and that such regulation can occur in response to the physiology of the cell. At present, the mechanisms of such regulation are obscure. It is known that lysogenization is favored by poor growth conditions, or by starvation of the cells (Kourilsky, 1973; Herskowitz and Hagen, 1980). In addition, high multiplicity of infection favors the lysogenic response (Kourilsky, 1973), presumably by increasing the levels of CII and CIII by gene dosage effects.

|

257

258

|

Little

To some extent, one can rationalize the above responses to various conditions in terms of the following framework. It is difficult to know how to assess the overall fitness of L, since it can propagate either as a lysogen or by lytic growth, and one cannot readily assess the weight to be given each of these modes, particularly since we do not study L or E. coli in their natural environments. However, it is plausible that, if conditions do not favor lytic growth, the infected cell would be channeled towards the lysogenic pathway. Hence, if growth conditions are poor or the cells are starving, lytic growth will likewise be poor. If the ratio of phage to cells in the environment is high, it is likely that the number of susceptible cells is low, again limiting the capacity of the population to grow lytically. The SOS regulatory system and prophage induction Prophage induction is an epigenetic switch of a gene regulatory state from the lysogenic state to the lytic state (Roberts and Devoret, 1983). This process ordinarily proceeds at a low rate (see below), but is extremely efficient when a host regulatory circuit called the SOS response has been activated by conditions that damage DNA or inhibit DNA replication. We first describe this system, then turn to a description of its role in prophage induction. The SOS response (Figure 10.4) is controlled by two critical regulatory proteins, LexA and RecA (Little and Mount, 1982). LexA is homologous to L CI; its domain organization is the same as that of CI, and it has some (but not all) of the activities carried out by CI. Under normal growth conditions, LexA negatively regulates a set of perhaps 40 SOS genes. The SOS functions carry out various activities that help the cell recover from inducing treatments, most notably by repairing DNA damage. Unlike repression by CI, LexA does not

Figure 10.4 SOS regulatory circuit. Depicted schematically are the activities of the two critical regulatory proteins, LexA repressor and RecA co-protease, at various stages of the SOS regulatory cycle. See text for details.

Prophage Induction of Phage L

completely repress its target genes; instead, these are expressed at low basal levels (Peterson and Mount, 1987), and presumably act to repair low sporadic levels of DNA damage. RecA protein is the central protein in genetic recombination (Kowalczykowski et al., 1994), catalyzing a wide variety of strand transfer reactions between homologous DNA molecules. These reactions result in recombination, and play important roles in several types of DNA repair. In addition to these complex strand transfer reactions, RecA also plays a regulatory role in the SOS system. When DNA is damaged, RecA protein is converted to an activated form, probably by binding to single-stranded DNA at sites of damage and at stalled replication forks. It is known that activation of RecA requires DNA replication (Sassanfar and Roberts, 1990), and it is believed that RecA is activated when replication forks encounter sites of DNA damage. However, at present, it is unclear exactly what structures are present at sites of damage, or whether multiple structures exist that can activate RecA. Some evidence (Rupp and Howard-Flanders, 1968; Heller and Marians, 2006) suggests that replication forks can pass sites of damage, leaving daughter-strand gaps, to which RecA could bind. Conversely, a body of evidence (Cox et al., 2000) indicates that replication forks stall at sites of damage; that the structures of protein–DNA complexes at stalled forks are dynamic and can vary; and that mechanisms exist for restarting stalled forks. Whether RecA is present in activated form at all stalled forks, or only at a subset, is not yet known. Accordingly, it is difficult to be more specific about how RecA is activated when damaged DNA is replicated. Extensive in vitro studies on RecA indicate that the activated form of the protein is a filament of RecA polymerized on single-stranded DNA, with ATP bound to the filament (Kowalczykowski et al., 1994). ATP[S], a poorly hydrolysable analog, can substitute for ATP. It is likely that this polymeric form of RecA also exists in vivo at the sites just discussed. When activated in this way, RecA catalyzes the cleavage of LexA, inactivating its repressor function and leading to derepression of the SOS regulon. Activated RecA catalyzes the cleavage of LexA. However, its role in this reaction is indirect (Little, 1984; Little, 1993). The actual catalytic active site for proteolysis of LexA lies in the LexA molecule itself, not in RecA. Hence, we term RecA a “co-protease” to emphasize the indirect nature of its action (Little, 1991). Evidence that LexA undergoes self-cleavage is that this reaction can proceed spontaneously in vitro at high pH, cutting the same bond as is cleaved in the RecA-dependent reaction. Mutations blocking RecAmediated cleavage also impair self-cleavage. LexA cleavage proceeds by action of a serine hydroxyl group, Ser119 in LexA, acting as a nucleophile to attack the peptide bond in a way analogous to the active site serine in trypsin (Slilaty and Little, 1987). However, the general base for activating the serine is a lysine, Lys156 in LexA, rather than a histidine as in trypsin. In LexA (and in other proteins of this class), Ser119 and Lys156 are adjacent and lie at the end of a shallow cleft, which serves as a binding pocket for the substrate (Peat et al., 1996; Bell et al., 2000; Luo et al., 2001). Crystal structures of several mutant forms of LexA (Luo et al., 2001) reveal the mechanism by which self-cleavage occurs, and by which it is ordinarily restrained. Two different structures were observed. In the non-cleavable (NC) form, the cleavage site lies ~20 Å from the catalytic center, and lies in a small folded structure we term the cleavage site region or CSR (Figure 10.2A). In the cleavable (C) form, the cleavage site is juxtaposed with the active site serine nucleophile that attacks the peptide bond; cleavage is prevented by mutation

|

259

260

|

Little

of Lys156. We believe that the equilibrium between these two forms lies strongly toward the NC form, explaining why the reaction is very slow in vivo. It is not known at present how RecA acts to stimulate cleavage; our favored model is that RecA acts somehow to stabilize the C form, but it is unclear how it does so. Activation of RecA is a reversible process. When the DNA is repaired, the levels of activated RecA fall, as can be observed by monitoring LexA cleavage as a function of time after ultraviolet (UV) irradiation (Little, 1983). If the DNA damage is not too severe, the cells can recover; LexA becomes stable, and acts once again to repress the SOS genes. Returning now to L, RecA mediates cleavage of CI in a completely parallel reaction (Little, 1984; Bell et al., 2000). The active site Ser and Lys residues (S149 and K192 in L CI) are conserved, and cleavage occurs at a conserved Ala-Gly bond located near the center of the protein (Figure 10.2A). Accordingly, when the SOS system is activated, CI cleavage begins. One significant difference between LexA and CI is that LexA is cleaved much more rapidly than CI. Cleavage of LexA is essentially complete within 5 min of DNA damage by ultraviolet (UV) irradiation; in contrast, CI cleavage takes about 30 min to be complete (Bailone et al., 1979). Since, as stated above, activation of RecA is reversible, the cell will not switch if activated RecA does not persist long enough to make cleavage of CI complete. We return to this in detail below. These differential sensitivities likely reflect the differing regulatory niches of LexA and CI. The SOS system is designed to be freely reversible, and probably to be activated promptly upon DNA damaging treatments. In contrast, L may have evolved to avoid efficient induction by low levels of DNA damage, levels from which the cell can recover. Prophage induction becomes efficient only at doses of UV light that start to kill the host cell. In this view, L induction can be likened to rats deserting a sinking ship; it will exit the host only if cell death is likely. As stated above, however, one must take such arguments with a grain of salt, since the natural environment differs so much from laboratory conditions, and we do not know what selective pressures have operated on L. We return to this question in detail below. In addition to prophage induction induced by DNA damage, L lysogens can undergo spontaneous switching to the lytic state. This switching results in release of free phage particles into the growth medium and occurs at a rate of perhaps 10–5 per generation. This spontaneous switching requires the SOS system, as judged both by the fact that it occurs at barely detectable levels in a recA– host (Brooks and Clark, 1967) (see also below) and that it is also blocked by mutations in cI, termed ind–, that block RecA-mediated cleavage (Gimble and Sauer, 1985). Since this reaction involves the SOS system, regardless of the stimulus provoking it, we term it “spontaneous induction.” It is plausible that this process is triggered by sporadic damage to DNA in the absence of overt damage supplied by the investigator. Recent evidence indicates that replication forks stall in a significant fraction of cells, and must be restarted (Cox et al., 2000). In addition, examination of single cells carrying a LexA-controlled GFP reporter shows that a small fraction of the cells are highly induced for GFP expression (McCool et al., 2004). We speculate that a still smaller fraction of the population has suffered sporadic damage severe enough to allow CI cleavage to proceed to completion, leading to spontaneous induction. One line of evidence in favor of this comes from studies of lambda-like (or lambdoid) phages 933W and H-19B, which undergo spontaneous induction at higher rates (Livny and Friedman, 2004). The

Prophage Induction of Phage L

key observation utilized also a L derivative termed LOR323, which also has a high rate of spontaneous induction (Little et al., 1999). In cells undergoing spontaneous induction of H-19B, LOR323 usually switches as well (Livny and Friedman, 2004), consistent with the expectation that spontaneous induction results from a property of the host cell and its metabolism, as opposed to stochastic fluctuations in phage regulatory proteins; the latter model would predict that switching events would not usually be correlated. Threshold behavior of prophage induction When lysogens are exposed to graded amounts of DNA damage, in this case by irradiation with graded doses of UV light (Little et al., 1999; Michalowski et al., 2004), the switching response shows threshold behavior (Figure 10.5). At low doses of UV, switching occurs only in a small fraction of the cells. At a particular dose, termed the set point, switching rather abruptly becomes efficient. Hence, this behavior raises several questions: First, has the set point been optimized during the course of evolution, and under what sort of selective pressure? Second, what determines the location of the set point? And third, what is the mechanistic basis for threshold behavior? A definitive answer to the first question is not possible at present. Evidence described below indicates that the set point can be modulated by mutation, indicating that evolution of its value is certainly possible. We have suggested above that the set point has evolved so that L can exit from cells with severe DNA damage that may not survive. As mentioned Fig 5.

Phage/cell

150

OR323

100

Wild-type

50

set point 0

0

10

20

30

40

UV dose (J/m2) Figure 10.5 Threshold behavior of prophage induction. Exponentially growing cells are irradiated with various doses of UV light, diluted into growth medium, shaken 2 hours, and treated with chloroform. The samples are titered to measure the production of phage. The set point for wild-type is indicated; it is defined as the dose giving half the maximum yield of phage (14 J/m2 in this experiment). The set point for LOR323 is far lower, about 1.8 J/m2. See Text for details.

|

261

262

|

Little

above, other lambda-like phages (933W and H-19B) have shown that these have markedly lower set points than that of L (Livny and Friedman, 2004). These phages carry Shiga toxin, which is expressed only during lytic growth. The authors speculate that a high rate of spontaneous induction, which leads to lytic growth, confers a selective advantage to lysogens of these phages, since it leads to diarrhea and dispersal of the gut flora into the environment. A general conclusion is that the set point is likely to be optimized, and that the selective forces likely differ for different phages. The set point can be modulated by mutations in L. Various types of mutations can affect its value. It is likely that most classes of mutation work by affecting the level of CI in the cell. First, changes in the strength of the PRM promoter alter the set point, and there is a fairly good correlation between the set point and PRM strength (Michalowski et al., 2004). Second, as noted above changes in the operators in the OR region can affect it. For instance, changing OR1 to the sequence of OR3, creating LOR323, drastically lowers the set point (Little et al., 1999). In this case, this probably results both because CI levels are reduced and because CI binds less tightly, and in an altered pattern, to the OR operators. Third, mutations in OR3 or in OL3 that weaken CI-mediated looping also shift the set point (our unpublished data), presumably by raising the CI level through loss of negative autoregulation (Dodd et al., 2004). Other mutations, in cI itself, probably increase the set point by slowing the rate of CI cleavage so that it is more difficult to reduce CI levels below a critical value. We now turn to the molecular basis of threshold behavior. We have proposed (Michalowski et al., 2004; Atsumi and Little, 2006a) that this systems property results, at the mechanistic level, from the kinetics with which RecA is activated. Here we spell this model out in more detail. The model proposes that threshold behavior results from the timing with which RecA protein is activated by DNA damage. As noted above, RecA activation is reversible. According to the model, induction requires RecA to remain activated for a period of time long enough to cleave nearly all the CI, a process known to take roughly 30 min in vivo (Bailone et al., 1979). At all doses of DNA damage, the initial amount of RecA* is about the same. However, at low levels of DNA damage, RecA does not remain activated long enough to reduce CI levels sufficiently to cause switching, while at high doses it does. This leads to the threshold behavior. We next elaborate on each aspect of this model. First, RecA is activated in vitro by binding to single-stranded DNA (Craig and Roberts, 1980). As described above, DNA replication is required for activation of RecA in vivo (Sassanfar and Roberts, 1990). Hence, it is believed that RecA is activated at replication forks that have encountered DNA damage. At the molecular level, it is not completely clear what happens when forks encounter damage. A popular current model is that forks stall, and must be restarted by a variety of means. However, a recent paper (Heller and Marians, 2006), and old data (Rupp and Howard-Flanders, 1968), suggest that, to the contrary, replication continues, leaving “daughter-strand” gaps of an average size of ~1000 nt at the sites of damage. The actual situation may lie somewhere in between; this is a very active area of work and the situation is almost certain to be complex. For our present purposes, the distinction between the two extremes may be important. If the forks stall at all or most

Prophage Induction of Phage L

sites of damage, the amount of RecA* is likely to be constant, since each fork will quickly stall at the UV doses we use. If the forks keep moving, leaving daughter-strand gaps in their wake, RecA* may be present at each daughter-strand gap and the total amount of RecA* depends on the number of gaps and, hence, on the amount of damage. As an intermediate model, possibly a fork can pass many sites of damage but eventually stalls; if so, the initial amount of RecA* might be similar at high and low doses, but might vary substantially for different forks and perhaps for different cells in a population. Second, the initial amount of RecA* is about the same with different amounts of damage, as indicated by our studies on in vivo cleavage of LexA (Little, 1983), another substrate for RecA-mediated cleavage. We found that the initial kinetics of LexA degradation were about the same at 2, 5 and 10 J/m2, and that the stability of LexA at 15 min after UV treatment was about the same at a range of doses. At later times, the rate of cleavage slowed markedly; importantly, this happened earlier at lower UV doses than at higher doses (some data not shown). The simplest interpretation is that the initial amount of RecA* is about the same at various doses. Third, cleavage of CI is not instantaneous; it takes 20–30 minutes, even after a high UV dose, for the cells to switch, as judged by several lines of evidence, including direct examination of CI activity (Bailone et al., 1979). Accordingly, we suggest that if RecA does not remain active long enough to reduce CI levels to a critical value, the cell will not switch. This model accounts qualitatively for threshold behavior. It predicts that at low levels of DNA damage, most cells will not switch, because RecA is no longer active enough at 20–30 min. With more DNA damage, more cells switch, giving rise to the threshold behavior observed. The rate of RecA-mediated CI cleavage is likewise complex. Only monomers of CI are good substrates (Cohen et al., 1981; Gimble and Sauer, 1989). CI has a dimer dissociation constant (Kdimer) of 5–10nM in vitro, depending on the conditions (Koblan and Ackers, 1991); its in vivo value is unknown. At the measured in vivo concentration of roughly 250nM (Reichardt, 1975), most of the protein is dimeric. It is also likely that not all the CI that is not bound to operators is free in solution, since most specific DNA-binding proteins also bind non-specifically to any DNA (Von Hippel and Berg, 1986; Pray et al., 1998). If we assume for the sake of discussion that Kdimer is 5nM and that the concentration of free CI is 100nM, there will be ~15nM monomers. The Km for RecA-mediated CI cleavage (Gimble and Sauer, 1989) is far higher, roughly 3 µM. Accordingly, the concentration of monomers « Km in vivo, and the rate of the reaction is proportional to the concentration of monomers. Hence, cleavage should reduce CI levels rapidly at first, and more slowly at later times. We now consider in more detail the various forces affecting the level of CI that are thought to take place as CI levels fall due to RecA-mediated cleavage. At the level of CI present in a lysogen, PRM is repressed roughly 50% from its maximally activated value due to negative autoregulation conferred by looping (Dodd et al., 2001). As CI levels begin to fall, negative autoregulation diminishes, partially counteracting the loss due to cleavage. At a substantially lower value of CI, CI begins to vacate OR2 (and OR1), resulting in loss of

|

263

264

|

Little

positive autoregulation. Unlike the relief of negative feedback, however, this loss of positive feedback tends to drive the system towards switching, because the rate of synthesis rather abruptly drops. As OR2 and OR become free, PR is derepressed, leading to expression of Cro and the lytic genes. As CI levels fall to a critical value (Michalowski and Little, 2005; Atsumi and Little, 2006a), which we term the “switching threshold” (Babić and Little, 2007), switching becomes likely. This switching threshold is probably in the range of 10% the lysogen level of CI. However, another factor enters into play that complicates our intuitive understanding of this process. Lysogens contain roughly 250nM CI; taking as a round number that 1 molecule/cell is 1nM, this means that at the switching threshold there are only about 25 molecules of CI in the cell. At such low numbers of molecules, the behavior of cells becomes stochastic—that is, a certain element of chance enters in, as we shall describe below—such that different cells with the same numbers of molecules can follow different fates (McAdams and Arkin, 1997). Stochastic behavior arises for several reasons, and can best be illustrated by example. If we consider a typical promoter, it may fire once every minute. But this is an average. In some one-minute periods, no transcripts are started; in others, two or more transcripts initiate. In addition, a given transcript may be translated to make perhaps one, five, or twenty molecules of protein before the mRNA molecule is degraded. Accordingly, two different cells starting with the same number of transcripts may make widely differing numbers of molecules, and may follow different fates. These effects become progressively more severe as the numbers become smaller. The result is that the fate of a cell with say 25 molecules of CI is not rigorously determined; rather, there is a certain probability that it will switch or return to the lytic state. It seems likely that the probability of switching becomes progressively larger as the number of CI molecules is lower. Another stochastic factor, related to RecA-mediated cleavage, results from the fact that this reaction is far below the Km for the reaction; hence its rate is very slow, and stochastic effects will tend to dominate. Taking stochastic behavior into account complicates a description of a pathway or a course of events. The description given above for the forces acting to raise or lower CI levels during prophage induction is typical of a molecular geneticist’s outline for a pathway. It leaves the misleading impression that all cells will follow the same regulatory fate. However, it is almost certain that this is not the case, at least in a situation like prophage induction. Does Cro play a role in prophage induction? As noted above, Cro binds tightly to OR3. It has long been believed, without direct evidence, that this binding makes switching irreversible by repressing basal expression of PRM. Our recent evidence (Atsumi and Little, 2006a) reveals that the situation is more interesting and more complicated. We replaced Cro with a dimeric form of Lac repressor, but left intact the OR3 site, so that repression of PRM by Lac repressor was not possible. Strikingly, we found that in this phage the set point was increased substantially compared to its parent phage containing Cro. We interpret this finding to mean that Cro is not essential for switching to occur, but that it does play a role in cells that are undecided or “on the edge” and can go either way, returning to the lysogenic state or switching to lytic growth. This model of prophage induction predicts that different cells will induce at different times after an inducing treatment, due to stochastic effects that determine when expression of PL and PR begins and PRM is no longer active. It also predicts that cells with higher CI

Prophage Induction of Phage L

levels will switch at later times than those with lower CI levels, although this effect may be obscured by the asynchrony postulated in the first prediction. Systems behavior of the L regulatory circuitry With the recent rise of systems biology has come an awareness that a full understanding of a regulatory circuit requires being able to predict its behavior, and particularly to predict the behavior of mutants that make rather modest changes in the interactions among the components of the system. It is our view that L offers fertile ground for applying this approach, due to our detailed mechanistic understanding of many aspects, and to the range of regulatory behavior it exhibits. Threshold behavior is a clear example of this. This type of behavior is ubiquitous in biology. It plays important roles in signal transduction, partly by preventing low levels of input due to noise from triggering a full-scale response. Mechanisms such as positive feedback, double-negative feedback, and non-linearity are believed to be necessary for bistable circuits to exist (Ferrell, Jr., 2002). All these mechanisms operate in the L circuit. Non-linearity—a response that is more than proportional to the signal—results from weak dimerization of CI and especially of Cro, and from cooperative binding by CI, all of which result in sigmoid binding curves. Double-negative feedback is seen in the interplay between Cro and CI. From the standpoint of systems behavior, an adequate description of prophage induction would predict the threshold behavior and the effect of mutants on the value of the set point. Another important systems property is the stability of regulatory states. As noted above, the lysogenic state in wild-type hosts is reasonably stable, but sporadic SOS induction limits its stability. When the SOS system is blocked by a recA mutation, one can assess the intrinsic stability of the lysogenic state. We previously (Little et al., 1999) calculated this stability at about 4 s 10–7, a value describing the fraction of cells in a culture that have switched and will give rise to a burst of free phage. Subsequent work (our unpublished data, cited in (Aurell et al., 2002)) has shown that this value is a large overestimate. As noted above, cultures contain free phage, released from switched cells. Phage are detected by their ability to form plaques. Wild-type L forms cloudy or “turbid” plaques, which are turbid because lysogens arise in the growing plaque and continue to grow because they contain CI and hence are immune to superinfection by other L phages. Culture supernatants from recA– lysogens also contain substantial numbers of clear plaques, presumably arising by spontaneous mutation in cI. When such mutants arise, pre-existing CI is diluted out as the cells grow, and eventually the mutant prophages switch as CI levels drop below the switching threshold. Accordingly, to make an estimate of the switching rate for wild-type, one counts only the turbid plaques. Unexpectedly, almost all the turbid plaques observed in recA– (L+) lysogens are not wild-type but mutant (our unpublished data); indeed, nearly all are a PRM mutant we term prm240 that can form lysogens (hence the turbid plaques). However, these lysogens are extremely unstable; once again, they are “on the edge.” We surmise that when these mutants arise they come to predominate in the culture because most of the cells can divide but a substantial fraction switch. Subtracting the very large background arising from these mutants, we estimate that the wild-type switches at perhaps 2–5 s 10–9 per generation. This number is below the spontaneous mutation rate. It is plausible that these cells switch after the host has suffered a mutation that somehow destabilizes

|

265

266

|

Little

the lysogenic state. This model is hard to test because the cell that has switched is killed. Although no mutations are known that confer wholesale destabilization, they might be lethal. In a sense, the lysogenic state is so stable that the rate of switching cannot readily be determined experimentally. Another systems property is that of robustness. This term has various meanings; we use it to denote a relative insensitivity to changes in the quantitative values of the parameters of the system—values such as promoter strengths, affinity of Cro or CI for their operators, or dimer dissociation constants. Another term for this property is “parameter sensitivity” (Savageau, 1974; Savageau, 1971). Rather limited studies in L indicate that the circuitry is robust for certain properties but less so for others. We found that the differential affinities of Cro and CI for their operators could be eliminated, by making the sequences of OR1 and OR3 the same, without completely disrupting the regulatory circuitry (Little et al., 1999). On the other hand, the set point for threshold behavior is not very robust to changes in PRM strength, being roughly proportional to this strength (Michalowski and Little, 2005; Michalowski et al., 2004). Evolution of gene regulatory circuitry Work described above and elsewhere has led us to propose a two-stage model for the evolution of complex circuits (Little et al., 1999; Michalowski and Little, 2005; Michalowski et al., 2004; Atsumi and Little, 2004). In this model, a circuit arises initially in a strippeddown, basic form. For this initial circuit to persist, it must offer some kind of selective advantage—in the case of L, this would likely be the ability to maintain the lysogenic state. In the second stage, two kinds of changes occur that make the circuit progressively more fit. First, the values of the parameters can be refined. This might confer added stability or robustness. Second, qualitatively new features can be added to the circuit. We might term these features refinements or “bells and whistles”—they are not essential for the circuit to work, but they make it work better, or allow it to work under a wider range of conditions. Our work has provided examples of both stages of this circuit. One plausible pathway for creating an initial circuit is random assembly of pre-existing parts or modules, perhaps by non-homologous recombination. We use the term “module” to represent a regulatory protein, together with its cis-acting sites and the actions that occur when the protein binds these sites. Our work with semi-synthetic versions of the L circuit provides strong support for the proposition that modules can be combined to provide functional circuits. We replaced L Cro with Lac repressor and several lac operators, and were able to identify functional circuits (Atsumi and Little, 2004), some of which gave behavior resembling that of L. Importantly, this finding implies that L has retained a modular organization. In subsequent work (Atsumi and Little, 2006b) we have in addition replaced CI with Tet repressor, together with several cis-acting sites, and again have isolated functional circuits. Since Tet repressor lacks many of the features of CI, including cooperative DNA binding, positive autoregulation, and (in our circuit design) negative autoregulation, one might argue (see below) that these features are refinements but are not necessary for proper circuit operation. However, we find that the circuit has an altered wiring diagram, in that both Lac repressor and Tet repressor are required to maintain a stable lysogenic state. We surmise that this change has resulted as a way to compensate for the missing features provided

Prophage Induction of Phage L

by CI. Another approach to assembling circuits from simpler modules (Guet et al., 2002) has been to assemble synthetic circuits from simpler modules on a plasmid, followed by examination of their responses to various stimuli; many different wiring diagrams were identified giving a range of behavior. In the second stage of the evolutionary pathway, we postulate that refinements are added that improve circuit behavior. Several different studies with L provide support for this view. In each case, one of the features of the natural L circuit was removed by mutation and the effects on circuit behavior were analyzed. In each case, our criteria for normal behavior were the ability to grow lytically, to form stable lysogens, and to undergo prophage induction. The first case (Little et al., 1999), already described, is changing OR1 and/or OR3 so they have the same sequence, thereby removing the differential occupancy of the OR operators that was thought to play important roles in operation of the circuitry. The resulting three phage variants had qualitatively normal behavior. Each had quantitative defects, but prophage induction still had threshold behavior, albeit with a lower set point than wild-type. In the second study (Michalowski and Little, 2005), we removed positive autoregulation by mutation of cI, specifically a D38N mutation that removes a contact with the S70 subunit of RNA polymerase ( Jain et al., 2004). L cI D38N forms stable lysogens with an extremely low set point. We identified numerous alleles of PRM that restored the set point to values near that of wild-type; the resulting phage once again showed threshold behavior for prophage induction. We conclude that positive autoregulation is a refinement, rather than an essential feature of the regulatory circuitry; it likely plays an important role, at least in some contexts, since all lambdoid phage tested to date show this property. In a third study (Babić and Little, 2007), we removed cooperative DNA binding by either of two mutations, Y210H or Y210N, in cI. L cI Y210H could form unstable polylysogens (those with several tandem copies of the prophage), while L cI Y210N could not lysogenize. We isolated suppressors of this defect, and found that a variant with two changes in the OR region, a mild PRM up-promoter together with a mutation in OR2 that allows tighter CI binding, yielded a variant with largely normal behavior. Again, it gave threshold behavior for prophage induction, with a set point near that of wild-type. Lytic growth was somewhat impaired; we speculate that expression of CI from the stronger PRM may play a role. Again, we conclude that cooperative DNA binding is a refinement, albeit one that appears to make the circuitry work better under laboratory conditions. We surmise that cooperativity and positive autoregulation serve a somewhat redundant purpose in the L circuitry. Each provides an element of non-linearity or positive feedback to the circuitry. As discussed earlier, these features are important in allowing bistability and stable states of a regulatory circuit (Ferrell, Jr., 2002). Since each feature contributes, in its own way, to the L circuitry, they are redundant. The usual use of this term is to describe functions that serve biochemically related roles. Since the role here is to confer proper circuit behavior, by driving the circuit away from an “undecided” state, we term it “circuit-level redundancy” (Michalowski and Little, 2005) to distinguish this use of the term from the usual meaning. Two lines of evidence lend some further support to this suggestion. First, starting with phages carrying defects in both cooperativity and positive autoregulation, we have been

|

267

268

|

Little

unable, to date, to isolate variants that can form stable lysogens. This argument, based on negative evidence, is admittedly weak. Second, we described above synthetic L circuits in which Tet repressor replaces CI, removing both features. These circuits, as stated, have an altered wiring diagram, suggesting that the missing features are essential to give the behavior typical of L and its relatives. In any case, these studies lend support to the proposal that complex circuits can be built up in several stages from simple components to give a circuit with all the refinements we see in L. It will be of interest to see whether similar approaches in more complex circuits will be able to provide comparable evidence for plausible pathways for their evolution. General lessons from L Many widespread features of regulatory circuits that we now take for granted were first identified in studies on L. These include the following. First, the concept that regulatory states are set up in two stages, an establishment phase and a maintenance phase, was first developed in studies on regulation of cI (Reichardt and Kaiser, 1971), an analysis that separated the process of lysogenization into these two stages. Many, if not most, developmental processes in higher organisms are now recognized to follow a similar pattern—that is, one set of regulatory events establishes a regulatory state, and another set makes it permanent. In eukaryotes, the underlying mechanisms generally involve establishment and maintenance of particular states of chromatin (Ringrose and Paro, 2004). As best studied in Drosophila, the Polycomb and Trithorax groups of proteins perpetuate repressed and active chromatin states, respectively. In addition, a well-studied case that does not appear to involve chromatin states {Keyes, 1992 1575/id;Penalva, 2003 22475/id} is the expression of the Sex-lethal or Sxl gene in Drosophila. The Sxl protein is functional in females, but not in males, and is involved in regulating RNA splicing patterns. In the establishment phase, a female-specific promoter called Pe drives the expression of Sxl protein in females. At a later stage of development, the Sxl gene is expressed in both males and females from a different promoter, Pm (or Pl), but the Pm transcript requires pre-existing Sxl protein for proper splicing. Hence, in females, this protein is present and the mRNA is spliced to a form that can be translated to make more Sxl protein; in males, lacking functional Sxl protein, proper splicing cannot occur, and there is no way to make Sxl protein. Accordingly, in the maintenance phase the pattern set up early in development is perpetuated. Second, the concept of multiple promoters driving a particular gene was developed in the same study (Reichardt and Kaiser, 1971). This theme is now known to be very widespread in eukaryotes, although in this case it is often not multiple promoters, but multiple enhancers driving the same promoter. Now-classic examples are the “pair-rule” genes in Drosophila. These genes are expressed in a series of seven stripes along the anterior-posterior axis of the early embryo (Pankratz and Jäckle, 1990). In the best-analyzed case, expression of the eve gene is driven at the appropriate positions by particular combinations of transacting factors, operating at discrete enhancers. Some of these enhancers can be isolated from their context and shown to direct synthesis of a single stripe (Small et al., 1992). That is, enhancers with this property act largely autonomously to direct expression of the linked promoter, and enhancers devoted to particular stripes act independently of one another. Accordingly, the expression of the eve gene, and presumably other pair-rule genes, follows

Prophage Induction of Phage L

an “OR” logic in which the gene is expressed if condition A OR condition B OR…condition G is met. In S. cerevisiae, a well-studied case is expression of the HO gene, which is required for mating-type switching (Herskowitz, 1989). Again, multiple enhancers (termed UASs in yeast) are present upstream of the HO promoter. In this case, however, three different conditions must be met simultaneously, so that the logic is instead an “AND” logic. Conclusions Despite the wealth of knowledge about mechanistic aspects of L, and more recent forays into systems biology, several areas of L biology remain relatively poorly understood. First, it is known that Cro protein plays multiple essential roles in L development (Folkmanis et al., 1977; Eisen et al., 1975), but the details are not clear. Second, the impact of physiology on the lysis-lysogeny decision is not well known, and the mechanisms by which the FtsH protease activity is modulated are likewise obscure. Third, the details of the protein-DNA complexes formed upon CI-mediated long-range looping remain to be established. It is hoped that deeper inquiry into these areas will prove to be as fruitful for biology as a whole as the first 50 years of L research have been. Acknowledgments Work from my laboratory is supported by grants from the Office of Naval Research and the National Institutes of Health. References

Atsumi, S., and Little, J.W. (2004). Regulatory circuit design and evolution using phage L. Genes Dev. 18, 2086–2094. Atsumi, S., and Little, J.W. (2006a). Role of the lytic repressor in prophage induction of phage L as analyzed by a module-replacement approach. Proc. Natl. Acad. Sci. USA 103, 4558–4563. Atsumi, S., and Little, J.W. (2006b). A synthetic phage L regulatory circuit. Proc. Natl. Acad. Sci. USA 103, 19045–19050. Aurell, E., Brown, S., Johanson, J., and Sneppen, K. (2002). Stability puzzles in phage L. Phys. Rev. E 65, 051914. Babić, A.C., and Little, J.W. (2007). Cooperative DNA binding by CI repressor is a dispensable feature of the phage L gene regulatory circuitry. Proc. Natl. Acad. Sci. USA, in press. Bailone, A., Levine, A., and Devoret, R. (1979). Inactivation of prophage L repressor in vivo. J. Mol. Biol. 131, 553–572. Becskei, A., and Serrano, L. (2000). Engineering stability in gene networks by autoregulation. Nature 405, 590–593. Bell, C.E., Frescura, P., Hochschild, A., and Lewis, M. (2000). Crystal structure of the L repressor Cterminal domain provides a model for cooperative operator binding. Cell 101, 801–811. Bell, C.E., and Lewis, M. (2001). Crystal structure of the L repressor C-terminal domain octamer. J. Mol. Biol. 314, 1127–1136. Brooks, K., and Clark, A.J. (1967). Behavior of L bacteriophage in a recombination deficient strain of E. coli. Virology 1, 283–293. Burz, D.S., and Ackers, G.K. (1994). Single-site mutations in the C-terminal domain of bacteriophage L cI repressor alter cooperative interactions between dimers adjacently bound to OR. Biochemistry 33, 8406–8416. Calef, E., Avitabile, L.d.G., Marchelli, C., Menna, T., Neubauer, Z., and Soller, A. (1971). The genetics of the anti-immune phenotype of defective L lysogens. In: The Bacteriophage Lambda, A.D. Hershey, ed. (Cold Spring Harbor, NY: Cold Spring Harbor Laboratory), pp. 609–620. Cohen, S., Knoll, B.J., Little, J.W., and Mount, D.W. (1981). Preferential cleavage of phage L repressor monomers by recA protease. Nature 294, 182–184.

|

269

270

|

Little

Cox, M.M., Goodman, M.F., Kreuzer, K.N., Sherratt, D.J., Sandler, S.J., and Marians, K.J. (2000). The importance of repairing stalled replication forks. Nature 404, 37–41. Craig, N.L., and Roberts, J.W. (1980). E. coli recA protein-directed cleavage of phage L repressor requires polynucleotide. Nature 283, 26–30. Darling, P.J., Holt, J.M., and Ackers, G.K. (2000). Coupled energetics of L cro repressor self-assembly and site-specific DNA operator binding II: cooperative interactions of cro dimers. J. Mol. Biol. 302, 625–638. Das, A. (1992). How the phage lambda N gene product suppresses transcription termination: Communication of RNA polymerase with regulatory proteins mediated by signals in nascent RNA. J. Bacteriol. 174, 6711–6716. Dodd, I.B., Perkins, A.J., Tsemitsidis, D., and Egan, J.B. (2001). Octamerization of L CI repressor is needed for effective repression of PRM and efficient switching from lysogeny. Genes Dev. 15, 3013–3022. Dodd, I.B., Shearwin, K.E., Perkins, A.J., Burr, T., Hochschild, A., and Egan, J.B. (2004). Cooperativity in long-range gene regulation by the L CI repressor. Genes Dev 18, 344–354. Echols, H. (1986a). Bacteriophage L development: temporal switches and the choice of lysis or lysogeny. Annu. Rev. Genet. 2, 26–30. Echols, H. (1986b). Multiple DNA-protein interactions governing high precision DNA transactions. Science 233, 1050–1056. Echols, H., Green, L., Oppenheim, A.B., Oppenheim, A., and Honigman, A. (1973). Role of the cro gene in bacteriophage L development. J. Mol. Biol. 80, 203–216. Eisen, H., Brachet, P., Pereira da Silva, L., and Jacob, F. (1970). Regulation of repressor expression in L. Proc. Natl. Acad. Sci. USA 66, 855–862. Eisen, H., Georgiou, M., Georgopoulos, C.P., Selzer, G., Gussin, G., and Herskowitz, I. (1975). The role of gene cro in phage development. Virology 68, 266–269. Ferrell, J.E., Jr. (2002). Self-perpetuating states in signal transduction: positive feedback, double-negative feedback and bistability. Curr. Opin. Cell Biol. 14, 140–148. Folkmanis, A., Maltzman, W., Mellon, P., Skalka, A., and Echols, H. (1977). The essential role of the cro gene in lytic development by bacteriophage L. Virology 81, 352–362. Gimble, F.S., and Sauer, R.T. (1985). Mutations in bacteriophage L repressor that prevent RecA-mediated cleavage. J. Bacteriol. 162, 147–154. Gimble, F.S., and Sauer, R.T. (1989). Lambda repressor mutants that are better substrates for RecAmediated cleavage. J. Mol. Biol. 206, 29–39. Gottesman, S., Gottesman, M., Shaw, J.E., and Pearson, M.L. (1981). Protein degradation in E. coli: The lon mutation and bacteriophage lambda N and cII protein stability. Cell 24, 225–233. Guet, C.C., Elowitz, M.B., Hsing, W.H., and Leibler, S. (2002). Combinatorial synthesis of genetic networks. Science 296, 1466–1470. Heller, R.C., and Marians, K.J. (2006). Replication fork reactivation downstream of a blocked nascent leading strand. Nature 439, 557–562. Herman, C., Thévenet, D., D’Ari, R., and Bouloc, P. (1997). The HflB protease of Escherichia coli degrades its inhibitor LcIII. J. Bacteriol. 179, 358–363. Herskowitz, I. (1989). A regulatory hierarchy for cell specialization in yeast. Nature 342, 749–757. Herskowitz, I., and Hagen, D. (1980). The lysis-lysogeny decision of phage L: Explicit programming and responsiveness. Annu. Rev. Genet. 14, 399–445. Hochschild, A., Irwin, N., and Ptashne, M. (1983). Repressor structure and the mechanism of positive control. Cell 32, 319–325. Ito, K., and Akiyama, Y. (2005). Cellular functions, mechanism of action, and regulation of FtsH protease. Annu. Rev. Microbiol. 59, 211–231. Jain, D., Nickels, B.E., Hochschild, A., and Darst, S.A. (2004). Structure of a ternary transcription activation complex. Mol. Cell 13, 45–53. Johnson, A., Meyer, B.J., and Ptashne, M. (1978). Mechanism of action of the cro protein of bacteriophage lambda. Proc. Natl. Acad. Sci. USA 75, 1783–1787. Johnson, A.D., Meyer, B.J., and Ptashne, M. (1979). Interactions between DNA-bound repressors govern regulation by the L phage repressor. Proc. Natl. Acad. Sci. USA 76, 5061–5065. Keyes, L.N., Cline, T.W., and Schedl, P. (1992). The primary sex determination signal of Drosophila acts at the level of transcription. Cell 68, 933–943.

Prophage Induction of Phage L

Kobiler, O., Koby, S., Teff, D., Court, D., and Oppenheim, A.B. (2002). The phage lambda CII transcriptional activator carries a C-terminal domain signaling for rapid proteolysis. Proc. Natl. Acad. Sci. USA 99, 14964–14969. Kobiler, O., Rokney, A., Friedman, N., Court, D.L., Stavans, J., and Oppenheim, A.B. (2005). Quantitative kinetic analysis of the bacteriophage Lgenetic network. Proc. Natl. Acad. Sci. USA 102, 4470–4475. Koblan, K.S., and Ackers, G.K. (1992). Site-specific enthalpic regulation of DNA transcription at bacteriophage L OR. Biochemistry 31, 57–65. Koblan, K.S., and Ackers, G.K. (1991). Energetics of subunit dimerization in bacteriophage lambda cI repressor: linkage to protons, temperature, and KCl. Biochemistry 30, 7817–7821. Kourilsky, P. (1973). Lysogenization by bacteriophage lambda.I. Multiple infection and the lysogenic response. Mol. Gen. Genet. 122, 183–195. Kowalczykowski, S.C., Dixon, D.A., Eggleston, A.K., Lauder, S.D., and Rehrauer, W.M. (1994). Biochemistry of homologous recombination in Escherichia coli. Microbiol. Rev. 58, 401–465. Li, M., Moyle, H., and Susskind, M.M. (1994). Target of the transcriptional activation function of phage L cI protein. Science 263, 75–77. Little, J.W. (1984). Autodigestion of lexA and phage lambda repressors. Proc. Natl. Acad. Sci. USA 81, 1375–1379. Little, J.W. (2005). Threshold effects in gene regulation: When some is not enough. Proc. Natl. Acad. Sci. USA 102, 5310–5311. Little, J.W. (1993). LexA cleavage and other self-processing reactions. J. Bacteriol. 175, 4943–4950. Little, J.W. (1983). The SOS regulatory system: control of its state by the level of RecA protease. J. Mol. Biol. 167, 791–808. Little, J.W. (1991). Mechanism of specific LexA cleavage: Autodigestion and the role of RecA coprotease. Biochimie 73, 411–422. Little, J.W. (2006). Gene Regulatory Circuitry of Phage Lambda. In The Bacteriophages, R. Calendar, ed. New York, NY: Oxford University Press, pp. 74–82. Little, J.W., and Mount, D.W. (1982). The SOS regulatory system of Escherichia coli. Cell 29, 11–22. Little, J.W., Shepley, D.P., and Wert, D.W. (1999). Robustness of a gene regulatory circuit. EMBO J. 18, 4299–4307. Livny, J., and Friedman, D.I. (2004). Characterizing spontaneous induction of Stx encoding phages using a selectable reporter system. Mol. Microbiol. 51, 1691–1704. Luo, Y., pfuetzner, R.A., Mosimann, S., Paetzel, M., Frey, E.A., Cherney, M., Kim, B., Little, J.W., and Strynadka, N.C. J. (2001). Crystal structure of LexA: A conformational switch for regulation of selfcleavage. Cell 106, 585–594. McAdams, H.H., and Arkin, A. (1997). Stochastic mechanisms in gene expression. Proc. Natl. Acad. Sci. USA 94, 814–819. McCool, J.D., Long, E., Petrosino, J.F., Sandler, H.A., Rosenberg, S.M., and Sandler, S.J. (2004). Measurement of SOS expression in individual Escherichia coli K-12 cells using fluorescence microscopy. Mol. Microbiol. 53, 1343–1357. Meyer, B.J., Maurer, R., and Ptashne, M. (1980). Gene regulation at the right operator (OR) of bacteriophage L II. OR1, OR2, and OR3: Their roles in mediating the effects of repressor and cro. J. Mol. Biol. 139, 163–194. Meyer, B.J., and Ptashne, M. (1980). Gene regulation at the right operator (OR) of bacteriophage L III. L repressor directly activates gene transcription. J. Mol. Biol. 139, 195–205. Michalowski, C.B., and Little, J.W. (2005). Positive autoregulation of cI is a dispensable feature of the phage L gene regulatory circuitry. J. Bacteriol. 187, 6430–6442. Michalowski, C.B., Short, M.D., and Little, J.W. (2004). Sequence tolerance of the phage L PRM promoter: Implications for evolution of gene regulatory circuitry. J. Bacteriol. 186, 7988–7999. Neubauer, Z., and Calef, E. (1970). Immunity phase-shift in defective lysogens: non-mutational hereditary change of early regulation of L prophage. J. Mol. Biol. 51, 1–13. Pabo, C.O., Sauer, R.T., Sturtevant, J.M., and Ptashne, M. (1979). The L repressor contains two domains. Proc. Natl. Acad. Sci. USA 76, 1608–1612. Pankratz, M.J., and Jäckle, H. (1990). Making stripes in the Drosophila embryo. Trends Genet. 6, 287–292.

|

271

272

|

Little

Peat, T.S., Frank, E.G. McDonald, J.P., Levine, A.S., Woodgate, R., and Hendrickson, W.A. (1996). Structure of the UmuD’ protein and its regulation in response to DNA damage. Nature 380, 727–730. Penalva, L.O.F., and Sánchez, L. (2003). RNA binding protein Sex-lethal (Sxl) and control of Drosophila sex determination and dosage compensation. Microbiol. Mol. Biol. Rev. 67, 343–359. Perry, M.D., Li, W., Trent, C., Robertson, B., Fire, A., Hageman, J.M., and Wood, W.B. (1993). Molecular characterization of the her-1 gene suggests a direct role in cell signaling during Caenorhabditis elegans sex determination. Genes Dev. 7, 216–228. Peterson, K.R., and Mount, D.W. (1987). Differential repression of SOS genes by unstable LexA41 (Tsl1) protein causes a “split-phenotype” in Escherichia coli K-12. J. Mol. Biol. 193, 27–40. Pray, T.R., Burz, D.S., and Ackers, G.K. (1998). Cooperative non-specific DNA binding by octamerizing lambda cI repressors: A site-specific thermodynamic analysis. J. Mol. Biol. 282, 947–958. Ptashne, M. (2004). A Genetic Switch: Phage lambda revisited. (Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press). Ptashne, M., Backman, K., Humayun, M.Z., Jeffrey, A., Maurer, R., Meyer, B., and Sauer, R.T. (1976). Autoregulation and function of a repressor in bacteriophage lambda. Science 194, 156–161. Reichardt, L., and Kaiser, A.D. (1971). Control of L repressor synthesis. Proc. Natl. Acad. Sci. USA 68, 2185–2189. Reichardt, L.F. (1975). Control of bacteriophage lambda repressor synthesis after phage infection: the role of the N, cII, cIII and cro products. J. Mol. Biol. 93, 267–288. Révet, B., Von Wilcken-Bergmann, B., Bessert, H., Barker, A., and Müller-Hill, B. (1999). Four dimers of L repressor bound to two suitably spaced pairs of L operators form octamers and DNA loops over large distances. Curr. Biol. 9, 151–154. Ringrose, L., and Paro, R. (2004). Epigenetic regulation of cellular memory by the polycomb and trithorax group proteins. Annu. Rev. Genet. 38, 413–443. Roberts, J.W. (1993). RNA and protein elements of E. coli and L transcription antitermination complexes. Cell 72, 653–655. Roberts, J.W., and Devoret, R. (1983). Lysogenic induction. In: Lambda II, R.W. Hendrix, J.W. Roberts, F.W. Stahl, and R.A. Weisberg, eds. (NY: Cold Spring Harbor), pp. 123–144. Rupp, W.D., and Howard-Flanders, P. (1968). Discontinuities in the DNA synthesized in an excisiondefective strain of Escherichia coli following ultraviolet irradiation. J. Mol. Biol. 31, 291–304. Sassanfar, M., and Roberts, J.W. (1990). Nature of the SOS-inducing signal in Escherichia coli. The involvement of DNA replication. J. Mol. Biol. 212, 79–96. Savageau, M.A. (1971). Parameter sensitivity as a criterion for evaluating an comparing the performance of biochemical systems. Nature 229, 542–544. Savageau, M.A. (1974). Comparison of classical and autogenous systems of regulation in inducible operons. Nature 252, 546–549. Shean, C.S., and Gottesman, M.E. (1992). Translation of the prophage L cI transcript. Cell 70, 513– 522. Slilaty, S.N., and Little, J.W. (1987). Lysine-156 and serine-119 are required for LexA repressor cleavage: a possible mechanism. Proc. Natl. Acad. Sci. USA 84, 3987–3991. Small, S., Blair, A., and Levine, M. (1992). Regulation of even-skipped stripe 2 in the Drosophila embryo. EMBO J. 11, 4047–4057. Von Hippel, P.H., and Berg, O.G. (1986). On the specificity of DNA-protein interactions. Proc. Natl. Acad. Sci. USA 83, 1608–1612.

Phage F29: Membraneassociated DNA Replication and Mechanism of Alternative Infection Strategy

11

Wilfried J.J. Meijer, Daniel Muñoz-Espín, Virginia CastillaLlorente, and Margarita Salas Abstract Continuous research, spanning a period of more than three decades, has made the Bacillus bacteriophage F29 a paradigm for the study of several molecular mechanisms of general biological processes, including DNA replication and regulation of transcription. The genome of F29 consists of a linear double-stranded (ds) DNA, which has a terminal protein covalently linked to its 5a ends. Initiation of DNA replication, carried out by a protein-primed mechanism, has been studied in detail in vitro and is considered to be a model system that is also used by other linear genomes with a terminal protein linked to their DNA ends. Phage F29 has also been proven to be a versatile system to study in vitro transcription regulation in general and the switch from early to late phage transcription in particular. The detailed knowledge of in vitro phage F29 DNA replication and transcription regulation makes it an attractive model to study these processes in vivo. For many years it has been known that (i) phage F29 DNA replication, as well as that of other prokaryotic genomes, occurs at the cytosolic membrane, and (ii) the lytic F29 cycle is suppressed in early sporulating cells and under these conditions the infecting phage genome becomes trapped into the spore. The molecular mechanisms involved in these processes were largely unknown. We review here the advances recently obtained in our understanding of membrane-associated organization of F29 DNA replication and the mechanisms underlying the alternative infection strategy. General introduction The genus Bacillus incorporates many species of Gram-positive, aerobic, endospore-forming bacteria that normally inhabit the soil or decaying plant material. In these habitats, a large variety of phages have been isolated that infect Bacilli. All of these phages share some common features. First, they contain double-stranded (ds) DNA, and second, the virions have prolate icosohedral heads and are tailed. Modern phage taxonomy is based on properties of the virion and its nucleic acid (see Maniloff et al., 1999; Fauquet, 1999). The order of tailed phages, named Caudovirales, is classified into three families: Myoviridae (phages with contractile tails), Podoviridae (phages with short tails), and Siphoviridae (phages with long non-contractile tails). For a general review on tailed bacteriophages we refer to Ackermann (1998). In addition to taxonomy based on properties of the virion and its nucleic acid, phages can be divided into three groups based on their infection cycle. The

274

|

Meijer et al.

first group contains lytic phages that complete their life cycle within a well-defined period of time after infection and are unable to lysogenize their host. The second group is formed by the so-called pseudo-temperate phages. These are virulent phages with an extended and irregular latent period. Although this stage mimics lysogeny, it does not involve a stable prophage. The third group is constituted by the temperate phages. The genomes of these phages are able to integrate into the host genome and can be maintained in this lysogenic stage for many generations. Generally, during this stage, the cells are immune to infection with the same phage. Phage F29 belongs to a family of related phages which includes, in addition to F29, phages PZA, F15, BS32, B103, M2Y (M2), Nf and GA-1. These phages, which form part of the Podoviridae family, are the smallest Bacillus phages isolated to date and are among the smallest known dsDNA phages. Most of these phages infect Bacillus subtilis, but often they also infect other related species such as Bacillus amylolyquefaciens, Bacillus pumilus, and Bacillus licheniformis. Phages of this genus have been sub-classified into three groups based on serological properties, DNA physical maps, peptide maps and partial or complete DNA sequences (Yoshikawa et al., 1985; 1986; Pecenkova and Paces, 1999; Meijer et al., 2001a). The first group includes phages F29, PZA, F15 and BS32, and the second group includes B103, Nf and M2Y. The most distantly related phage of this family, GA-1, is unable to infect the standard B. subtilis strain 168 (Arwert and Venema, 1974) and has been placed in a third group. Sequence-analysis of the 16S-rRNA of the host strain of GA-1, G1R, showed that it is most closely related to B. pumilus (Horcajadas, 2000). The genomes of the F29-like phages consist of a linear dsDNA molecule of about 19 kb that have a phage-encoded protein, named terminal protein, covalently attached at each 5a DNA end. The DNA sequences of the complete genomes of F29 (19 285 bp) and PZA (19 366 bp) belonging to group I (Yoshikawa and Ito, 1982; Garvey et al., 1985; Vlcek and Paces, 1986; Paces et al., 1986), that of B103 (18 630 bp) belonging to group II (Pecenkova et al., 1997), and that of GA-1 (21 129 bp) belonging to group III (Meijer et al., 2001a) have been determined. The genetic and transcriptional maps of the genomes of F29 (group I), B103 (group II) and GA-1 (group III) are presented in Figure 11.1. This figure shows that, in most aspects, these genomes are similarly organized. In all three genomes the genes and ORFs are organized in operons. Depending on the time when they are first expressed during the infection cycle, these can be divided in early and late operons. In all three genomes the early-expressed operons are transcribed leftwards and the single lateexpressed operon is transcribed rightwards. The genes present in the late operon (genes 7 through 16), which is located in the central part of the genome, encode phage structural proteins, proteins involved in phage morphogenesis, and proteins required for lysis of the infected host. Note that the genome of phage GA-1 lacks gene 8.5 encoding the head fiber protein. The F29-related phage M2Y, belonging to group II, was also found to lack this gene (Yoshikawa et al., 1986). All three genomes contain an early expressed operon that is divergently transcribed with respect to the late operon. Genes 6, 5, 3 and 2 of this operon encode the four main proteins required for in vitro F29 DNA replication (see below). It also contains gene 4 that encodes the transcriptional regulator protein. In addition to its role in phage DNA replication, protein p6 also has a role in transcriptional regulation

F29 DNA Replication in Vegetative and Sporulating Cells

Figure 11.1 Genetic and transcriptional maps of F29 (group I, 19 285 bp), B103 (group II, 18 630 bp), and GA-1 (group III, 21 129 bp). The maps are aligned according to the A2b, A2c, A3 promoter region. The direction of transcription and length of the transcripts are indicated by arrows. The transcripts of late and early-expressed operons and the late and early promoters (boxed) are shown below and above the map, respectively. Positions of the various genes and ORFs are indicated between the two DNA strands. Genes are indicated with numbers, ORFs with letters (lowercase for B103 and uppercase for F29 and GA-1). Recently, it has been demonstrated that F29 ORFA (56 codons, located in the left-side early operon downstream of gene 1) constitutes a gene and that the encoded protein, p56, is an inhibitor of the hostencoded uracil DNA glycosylase (Serrano-Heras et al., 2006). The position of genes encoding proteins p17 and p16.7 that are conserved in all three phage genomes, located in the right early operon, are indicated. The positions of the F29 ORFs 16.9, 16.8, 16.6 and 16.5, and the B103 ORF 16.5, located at the right side of their genomes, are indicated with the numbers .9, .8, .6 and .5. Transcriptional terminators are indicated with hairpin structures. A light grey box indicates the DNA region encoding the pRNA, and a black box indicates the region spanning the early A2b, A2c and late A3 promoters. Adapted from Meijer et al. (2001a).

(Whiteley et al., 1986; Barthelemy et al., 1989; Elías-Arnanz and Salas, 1999; Camacho and Salas, 2000). Note that this early left-side located operon of GA-1 is smaller than the corresponding ones of F29 and B103. Another early expressed operon is located at the right side of the phage genomes. Only two genes of this operon, 17 and 16.7, are conserved in all three phage genomes. Finally, another feature shared by all three phages is the presence of a region located in the left part of the genome that encodes an RNA (pRNA), which is required for packaging of phage DNA. The genome of GA-1 is about 1.8 and 2.5 kb larger than those of F29 and B103, respectively. Although the structural organization of the GA-1 genome is similar to that of F29 and B103, it contains additional sequences, located at both genome ends, that may encode several proteins, counterparts of which are not present in the genomes of F29 and B103.

|

275

276

|

Meijer et al.

In vitro F29 transcription regulation and DNA replication Since most studies on F29-like phages have been performed using F29, the views presented are generally based on this phage. However, some interesting homologies and differences with other members of the F29 family of phages are also discussed. A brief overview of our current knowledge of in vitro F29 transcription regulation and DNA replication is given below. More extensive reviews on these topics have been published elsewhere (Salas, 1991; Salas et al., 1996; Blanco and Salas, 1996; Rojo et al., 1998; Salas, 1999; Meijer et al., 2001a). In vitro F29 transcription regulation Apart from the weak promoters A1IV and C1, the genome of F29 contains five strong promoters that are mainly responsible for expression of the three operons and that of the pRNA, required for DNA packaging (see Figure 11.1). Homologs of these promoters are present in the genomes of related phages of groups II and III. Note that the GA-1 genome contains two additional early operons at its left side that are expressed by promoters not present in the genomes of other related phages (Horcajadas et al., 2001). The four main early F29 promoters are recognized by the host-encoded vegetative RNA polymerase containing sigma factor A (SA-RNA polymerase), which, with the help of the F29 transcriptional regulator protein p4, also recognizes the single late A3 promoter. The A1 promoter, which does not appear to be regulated, drives expression of the pRNA. Expression of the right-side early operon, encoding proteins p17 and p16.7 and containing four additional open reading frames, is driven by the C2 promoter. In the case of F29 it has been demonstrated that the activity of the early promoter C2 decreases rapidly 10–15 min after infection (Kawamura and Ito, 1977; Holder and Whiteley, 1983; Monsalve et al., 1995; Camacho and Salas, 2000). Phage F29 protein p6 was shown to be responsible for in vivo and in vitro repression of promoter C2 by binding to the right side genome region including promoter C2 (Whiteley et al., 1986; Barthelemy et al., 1989; Camacho and Salas, 2001b). Protein p6 is an abundantly early expressed dsDNA binding protein that also plays an important role in initiation of phage DNA replication (see below). The equivalent C2 promoter of GA-1 also becomes repressed about 10 min after infection and the GA-1 p6 protein represses this C2 promoter in vitro (Horcajadas et al., 2001). Thus, due to protein p6-mediated repression, the C2 promoters are only expressed during the initial 10–15 min after infection, limiting the amount of proteins expressed by the downstream located genes/ORFs. The left-side early operon, encoding most of the DNA replication-involved proteins and the transcriptional regulator protein p4, is divergently transcribed with respect to the late operon encoding the structural and morphogenetic proteins as well as those responsible for lysis of the infected cell. The promoters that drive expression of these early and late genes are present in a short intergenic region located between these two operons. The early operon of F29 containing genes 6 to 1 is driven by the tandemly organized promoters named A2c and A2b, and the late F29 operon is transcribed from a single promoter named A3 (Sogo et al., 1979; Barthelemy et al., 1986; Mellado et al., 1986a, 1986b; Monsalve et al., 1995). The transition from early to late F29 transcription requires F29 protein p4, the product of the early gene 4. Protein p4, which is a dimer in solution, binds to its cognate

F29 DNA Replication in Vegetative and Sporulating Cells

DNA binding sites contacting only one side of the DNA helix (Rojo and Salas, 1991). The intergenic region comprising promoters A2c, A2b and A3 contains three major p4-binding sites. The centre of one of these, binding site 3, is located at position –82 relative to the transcription start site of the late promoter A3 (Barthelemy and Salas, 1989). Whereas this promoter contains a good consensus sequence at the –10 region for the vegetative B. subtilis SA-RNA polymerase, it lacks a typical –35 box. Therefore, the RNA polymerase alone does not bind to the A3 promoter, which explains why the downstream-located operon is not expressed during early infection times. Binding of protein p4 to its binding site 3 activates the late A3 promoter. The main role of protein p4 is to stabilize the binding of RNA polymerase to the A3 promoter as a closed complex and has little effect on the rest of the steps of the initiation process (Nuez et al., 1992). The F29 promoters A2c and A2b drive expression of the early operon containing genes 6 to 1. Of these, promoter A2b is the one located closest to the oppositely oriented late promoter A3; promoter A2c is located proximal to gene 6. Both early promoters are repressed by protein p4. The p4-binding site 3 located upstream of the late A3 promoter and which is required for activation of this promoter as described above, overlaps partially with the early A2b promoter. Binding of protein p4 to this site occupies the –35 region of the A2b promoter preventing expression of this promoter. Thus, protein p4 activation of the late promoter A3 is accompanied by an efficient repression of the A2b promoter (Rojo and Salas, 1991). Expression of the other early promoter, A2c, is also repressed by protein p4. Another p4-binding site, site 2, is located upstream of the A2c promoter (centered at position –72 relative to the transcription start site of A2c). Protein p4 binding to this site is stabilized in the presence of RNA polymerase indicating that the proteins bind cooperatively to the DNA. In this situation, the RNA polymerase can generate abortive initiation transcripts but is unable to escape from the A2c promoter (Monsalve et al., 1996a). Thus, repression of the A2c promoter occurs by overstabilization of the RNA polymerase to this promoter (Monsalve et al., 1997). Interestingly, repression at the A2c promoter and activation of the A3 promoter involves in both cases interaction between a region of protein p4 containing Arg120 and the carboxyl terminal domain (CTD) of the RNA polymerase A-subunit (Mencía et al., 1993; Mencía et al., 1996a, 1996b; Monsalve et al., 1996a, 1996b; Mencía et al., 1998). In addition to protein p4, also the F29 protein p6 plays a role in the switch from early to late F29 transcription (Elías-Arnanz and Salas, 1999; Camacho and Salas, 2001a; Calles et al., 2002). When both are present, proteins p4 and p6 bind cooperatively to the A2c, A2b, A3 promoter region in such a way that p4 is bound to its cognate binding sites 3 (upstream of the A3 promoter) and 1 (overlapping with the core region of the A2c promoter) and p6 to the intervening sequences. Since proteins p4 and p6 are both synthesized during infection, this probably reflects the mechanism of the early to late transcriptional switch in F29 infected cells. The main early and late in vivo transcription termination sites of F29 have been determined by S1-nuclease mapping (Barthelemy et al., 1987). Transcription of the convergently oriented late A3 and early C2 promoters terminate in the short intergenic region between gene 16 and ORF16.5 (see Figure 11.1). This DNA region contains an inverted repeat, and stem–loop structures with calculated free energies of –14.8 kcal and –16.8 kcal could be drawn for the early and late transcripts, respectively. In both directions, a uridine-rich

|

277

278

|

Meijer et al.

tail follows the stem–loop indicating that it functions as a Rho-independent bidirectional transcription terminator, named TD1. Another Rho-independent transcriptional terminator, named TA1, was found to be present within gene 4 of F29 (Barthelemy et al., 1987). It has been suggested that part of the transcripts initiated at the A2b and A2c promoters terminate at this terminator resulting in the synthesis of high levels of mRNA coding for proteins p6 and p5 (SSB) and lower levels of longer mRNA coding for proteins p6 to p1 (Barthelemy et al., 1987). Apart from possible differences in translation initiation efficiencies, this explains why p6 and p5 are synthesized in far higher amounts as compared to proteins p4, p3, p2 and p1 (Mellado et al., 1980; Abril et al., 1997; Gascón et al., 2000). Equivalent TA1 and TD1 transcriptional terminators are present in the genomes of B103 and GA-1. Note that the genome of GA-1 contains three additional putative Rho-independent terminators in the left part of its genome that would terminate transcription from its early promoters A2c, A2b, A1a, A1b and A1c (see Figure 11.1). In vitro F29 DNA replication Genes 6, 5, 3 and 2, located in the left-side early-expressed operon (see Figure 11.1), are indispensable for in vivo phage F29 DNA replication (Mellado et al., 1980). Gene 2 encodes the DNA polymerase, gene 3 the terminal protein, gene 5 the single-stranded DNA binding (SSB) protein, and gene 6 the double-stranded DNA binding protein. An in vitro F29 DNA replication system, based on these four purified proteins, has been established (Blanco et al., 1994). The availability of this system has allowed a detailed analysis of the in vitro F29 DNA replication mechanism and functional analysis of these four main replication proteins. DNA replication of F29 and that of the related phage genomes occurs via a so-called protein-primed mechanism, the basic features of which are outlined here for F29. A schematic representation of the in vitro F29 DNA replication mechanism is shown in Figure 11.2. Initiation of F29 DNA replication starts with recognition of the origins of replication, i.e. the terminal protein-containing DNA ends, by a terminal protein/DNA polymerase heterodimer. Binding of the viral-encoded protein p6 to the DNA end regions results in a nucleoprotein complex that highly activates initiation of DNA replication probably by opening the DNA ends (Serrano et al., 1994) thereby stimulating the formation of a covalent linkage between the first inserted nucleotide (dAMP) via a phosphoester bond to the hydroxyl group of Ser232 of the terminal protein, which is catalyzed by the F29 DNA polymerase (Blanco and Salas, 1984; Hermoso et al., 1985; Blanco et al., 1992). The formation of this first terminal protein-dAMP covalent complex is directed by the second nucleotide at the 3a-end of the template; then the terminal protein-dAMP complex slides-back one nucleotide to recover the information of the terminal nucleotide (Méndez et al., 1992). Next, the F29 DNA polymerase synthesizes a short elongation product before dissociating from the terminal protein (Méndez et al., 1997). Replication, which starts at both DNA ends, proceeds in a very processive way and is coupled to strand-displacement (Blanco et al., 1989). This results in the generation of so-called type I replication intermediates consisting of full-length double-stranded (ds) F29 DNA molecules with one or more single-stranded (ss) DNA branches of varying lengths. The ssDNA stretches generated are bound by the SSB protein p5. When the two converging DNA polymerases merge, a type I replication intermediate becomes physically separated into two type II replication

F29 DNA Replication in Vegetative and Sporulating Cells

initiation

Type I replication intermediate

termination

Type II replication intermediates

Figure 11.2 Schematic overview of in vitro F29 DNA replication. Replication starts by recognition of the p6-nucleoprotein complexed origins of replication by a terminal protein/DNA polymerase heterodimer. The DNA polymerase then catalyses the addition of the first dAMP to the terminal protein present in the heterodimer complex. Next, after a transition step (not shown), these two proteins dissociate and the DNA polymerase continues processive elongation until replication of the nascent DNA strand is completed. Replication is coupled to strand displacement. The F29-encoded SSB protein p5 binds to the displaced ssDNA and is removed by the DNA polymerase during later stages in the replication process. Continuous polymerization results in the generation of two fully replicated F29 genomes. Circles, terminal protein; triangles, DNA Figure 2 polymerase; striped ovals, replication initiator protein p6; grey diamonds, SSB protein p5; de novo synthesized DNA is shown as beads on a string.

intermediates. Each of these consists of a full-length F29 DNA molecule in which a portion of the DNA, starting from one end, is double-stranded and the portion spanning to the other end is single-stranded (Inciarte et al., 1980; Sogo et al., 1982; Gutiérrez et al., 1991). Continuous elongation by the F29 DNA polymerase completes replication of the parental strand (see Figure 11.2). In vivo F29 DNA replication The first experimental evidences supporting the view that prokaryotic DNA replication, including that of bacterial chromosomes, plasmids and viral genomes, occurs at the cytosolic membrane has been presented in the 1960s (for review see, Siegel and Schaechter, 1973). It is generally believed that the bacterial membrane not only provides a framework for the organization of DNA replication but that membrane association also functions to compartmentalize the proteins involved in DNA replication which would enhance its efficiency via surface catalysis. Moreover, there is evidence that, at least in Escherichia coli, the bacterial membrane plays an important role in the control of initiation of chromosomal DNA replication (for review see, Boeneman and Crooke, 2005). General aspects of compartmentalization of prokaryotic DNA replication via membrane association has been reviewed elsewhere (Bravo et al., 2005). Cell fractionation techniques, carried out in the 1960–1970s, showed that newly replicated DNA molecules and replication proteins were recovered in membrane fractions, and complementary studies using genetic and biochemical approaches added evidence that prokaryotic DNA replication is associated to the membrane (reviewed in e.g. Siegel and Schaechter, 1973; Firshein, 1989; Sueoka,

|

279

280

|

Meijer et al.

1998). Studies in this field were boosted upon the introduction of fluorescence microscopy methods in prokaryotic research, which allows the localization of proteins and/or DNA regions in time and space within single cells (for review see, Jensen and Shapiro, 2000). Besides adding further evidence that prokaryotic replication occurs at or near the membrane, these techniques revealed important insights into various cellular processes including those directly related with DNA replication such as chromosomal segregation, for example. One hallmark of these studies was the discovery that the DNA replication machinery is located at relatively static mid-cell positions within the cell, presumably via attachment to the membrane, leading to the factory model of replication in prokaryotes (Lemon and Grossman, 1998). Nevertheless, despite these novel insights, our knowledge about the organization of membrane-associated DNA replication and the proteins involved in this fundamental process is still rather poor. The B. subtilis phage F29 is an attractive model to study membrane-associated DNA replication for several reasons. First, detailed knowledge is available on in vitro F29 DNA replication. Second, F29 encodes most, if not all, proteins required for replication of its genome. And third, processes other than DNA replication that are probably involved in DNA-membrane interactions, such as DNA segregation, do not apply to the F29 life cycle. Convincing evidence that replication of F29 DNA occurs at the membrane was first provided by Ivarie and Pène (1973). Using ultracentrifugation techniques, these authors found that the membrane fractions contained parental F29 DNA and were enriched for newly synthesized phage DNA. Phage DNA-membrane complexes were detected near the onset of F29 DNA replication. Importantly, recovery of parental F29 DNA in the membrane fractions required the synthesis of early phage proteins, strongly indicating that membraneassociation is mediated by phage-encoded proteins. The observation that F29 DNA was not associated at the membrane in cells that were infected at the restrictive temperature with temperature-sensitive mutant phages in genes encoding the DNA polymerase or the terminal protein furthermore indicated that replication occurs at the membrane (McGuire et al., 1977). Membrane proteins p1 and p16.7, encoded by the F29 genes 1 and 16.7 which are present in the early expressed operons located at the left and right-side of the F29 genome, respectively (see Figure 11.1), have been suggested to be involved in the organization of membrane-associated F29 DNA replication. Currently known aspects of these two proteins are discussed below. Protein p1 Mutant F29 phage sus1(629) (Reilly et al., 1973), contains a point-mutation in gene 1 changing its seventh codon (CAA) into a nonsense TAA codon (Prieto et al., 1989). Phage DNA replication is affected in non-suppressor cells infected with mutant sus1(629) implying a role for protein p1 (85 residues) in in vivo F29 DNA replication (Carrascosa et al., 1976; Prieto et al., 1989; Bravo and Salas, 1997). Its presence appears to be critical for efficient in vivo F29 DNA replication when bacteria are growing at 37°C, rather than at 30°C (Bravo and Salas, 1998). Protein p1, which has an amphiphilic nature, was found to be associated with the membrane in cells infected with F29 as well as in cells in which p1 was expressed from a plasmid. These results indicate that protein p1 interacts directly with

F29 DNA Replication in Vegetative and Sporulating Cells

the membrane and that this interaction does not require other viral components (Bravo and Salas, 1997; Serrano-Heras et al., 2003). In addition, plasmid-expressed protein p1 supported efficient F29 DNA replication in non-suppressor B. subtilis cells infected with phage sus1(629) (Serrano-Heras et al., 2003). Phage F29-infected cells contain about 10 000 and 100 000 copies of protein p1 at early and late infection times, respectively (Bravo and Salas, 1997). The C-terminal region of p1 (residues 68–84) is highly hydrophobic suggesting that this region is involved in membrane association. This view is supported by the observation that a truncated version of p1 lacking its C-terminal 43 residues (p1∆C43) did not associate with membranes in vivo (Bravo and Salas, 1997). Besides being a membrane protein, three other features have been described for p1. Using different in vivo and in vitro approaches p1 has been demonstrated to form multimers. In addition, it was shown that p1 can interact in vitro with the F29 terminal protein (Bravo et al., 2000) and with RNA (Takeuchi et al., 1998). Possible functions for p1 in in vivo F29 DNA replication have been proposed based on these latter two features (see below). The first evidence that p1 forms multimers was the observation that a fusion protein of native p1 to maltose binding protein E (MalE-p1) formed long filamentous structures and that a purified variant of p1 lacking its N-terminal 33 residues (p1∆N33) assembled into long protofilaments that associated in a highly ordered, parallel array forming large two dimensional sheets (Bravo and Salas, 1998). Chemical cross-linking combined with cell fractionation techniques showed that native protein p1 also assembled into large multimeric structures at the membrane in vivo (Serrano-Heras et al., 2003). Analysis of protein p1 mutants has provided important insights in residues and/or regions involved in multimerization. Inspection of the p1 sequence revealed that the regions spanning residues 31 to 36 and 37 to 66 have a low and high probability of forming a coiled-coil structure, respectively. The extended putative coiled-coil region and the hydrophobic C-terminal region (residues 68–84) have been shown to be both involved in multimerization. Thus, residues of the putative coiled-coil region Leu46, Met53 and Leu60 are essential for p1∆N33 assembly into sheets (Bravo et al., 2001) and mutants K33E and K41E are severely affected in multimerization as assessed by in vitro cross-linking and glycerol gradient analysis (Hashiyama et al., 2005b). Regarding the hydrophobic C-terminal region it has been shown that p1 variants lacking either the 43 or 11 C-terminal residues were defective in self-association (Bravo et al., 2000), and also mutations W71R, K74E and F77S prevented protein p1 multimerization (Hashiyama et al., 2005a, 2005b). Moreover, it appeared that mutations in the hydrophobic C-terminal region affect the ability of protein p1 to multimerize more drastically than mutations in the predicted coiled-coil region (Hashiyama et al., 2005b). Heterologous expression of wild-type protein p1 in E. coli severely inhibits growth. This growth-inhibition phenomenon was used to select for mutants that have lost this phenomenon with the expectation that such mutants would be affected in authentic functions of p1. Using this approach, 31 different single substitution mutants covering 26 positions were obtained. Among these were mutants having substitutions in the C-terminal hydrophobic region shown to be affected in their multimerization properties (residues Trp71, Lys74 and Phe77). In addition, mutants with substitutions in the predicted coiled-coil region (residues Asp36, Glu49, Met53, Leu57, Glu59, Leu60) were obtained, which, based on their loss of growth inhibition, are possibly affected in multimerization (Takeuchi et al.,

|

281

282

|

Meijer et al.

2005). It should be mentioned that the studies on p1 and its variants published by the Japanese groups are based on versions of p1 containing a C-terminal histidine-tag. In vitro chemical cross-linking studies showed that p1∆C43 interacts with the F29 terminal protein. In addition, truncated p1 proteins having retained their N-terminal 42 residues interfere with the terminal protein-primed replication initiation reaction when present in excess (Bravo et al., 2000). These observations, together with the fact that p1 associates with the bacterial membrane (Bravo and Salas, 1997), led to propose that protein p1 is a component of a viral-encoded membrane-associated structure which would provide an anchoring site for the viral DNA replication machinery (Bravo and Salas, 1997, 1998; Bravo et al., 2000). Another function, which is not necessarily mutually exclusive with the anchoring model, has been attributed to F29 protein p1 by Takeuchi and coworkers (Takeuchi et al., 1995, 1998). They found that larger amounts of F29 DNA polymerase were produced in non-suppressor cells infected with mutant sus1(629) as compared to those infected with wild-type phage (Takeuchi et al., 1995). This indicated that protein p1 might downregulate the synthesis of the DNA polymerase. Experiments performed in E. coli showed that, in addition to the DNA polymerase, also the synthesis of terminal protein (gene 3) and the transcriptional regulator protein (gene 4) were diminished by protein p1. Because similar levels of mRNA encoding genes 4–1 were synthesized in the absence or presence of functional gene 1, protein p1 appeared not to affect transcription of these genes. The observation that protein p1 was able to bind mRNA of genes 1–4 suggested that the observed repression might be at the translational level (Takeuchi et al., 1998). Additional studies using the multimerization-defective p1 mutants W71R, K74E and F77S provided evidence that multimerization is important for the RNA binding and translational repression activities of protein p1 (Hashiyama et al., 2005b). Protein p16.7 Preamble A genetic map of F29 was constructed by mapping two collections of F29 reference mutants that had been obtained after different mutagenic treatments (Mellado et al., 1976). All the suppressor- and temperature-sensitive mutants could be assigned to 17 genes which were numbered sequentially from left to right (1 to 17) according to their relative map position. However, none of the mutants mapped to a ~2 kb region located between late gene 16 and early gene 17 at the right side of the F29 genome. Sequence analysis revealed that this region includes the major part of the right-side early-expressed operon that, in addition to gene 17 (166 codons), contains five additional open reading frames (ORFs) (Garvey et al., 1985). These ORFs were named (from right to left) 16.9 (108 codons), 16.8 (105 codons), 16.7 (130 codons), 16.6 (54 codons) and 16.5 (37 codons) (see Figure 11.1). Analyses of the deduced protein sequences revealed that the putative protein encoded by ORF16.7 displayed some interesting features, which, together with the fact that this ORF is present in an early-expressed operon, suggested that this putative protein might be involved in membrane-associated phage DNA replication (Meijer et al., 2001b). First, its N-terminal

F29 DNA Replication in Vegetative and Sporulating Cells

region (residues 1 to 22) has a very hydrophobic character and may constitute a transmembrane spanning domain. Membrane topology predictions indicated that the putative p16.7 protein may be a membrane protein with its N- and C-regions at the outside and inside of the cell, respectively. And second, the C-terminal part of the putative p16.7 protein sequence (residues 70 to 130) shows limited similarity to homeodomains which are typical DNA binding domains present in a large family of eukaryotic transcriptional regulators (Gehring et al., 1994). In addition, we found that the region spanning amino acids 19 to 60 has a high probability to form an A-helical coiled-coil structure (more than 90% according to the algorithm of Lupas et al. (1991)), suggesting that this region may function as a protein di- or multimerization domain. Thus, these data suggested that ORF16.7 might encode a di- or multimeric membrane protein with DNA binding activity. These predicted features, together with the fact that ORF16.7 is conserved in all F29-related phages known to date (Meijer et al., 2001a), prompted us to study the p16.7 protein. Protein p16.7 is early and abundantly expressed in F29-infected cells Our first objective was to determine if and to what levels p16.7 is expressed in F29-infected cells. Western blot analysis showed that p16.7 is synthesized in F29-infected cells, demonstrating that ORF16.7 constitutes a gene. Quantitative immunoblotting of samples taken at different times after infection revealed that p16.7 was detected 6 min after infection, that the protein level increased rapidly until 12–15 min postinfection and that these levels remained constant during the rest of the infection cycle (Meijer et al., 2001b). The amount of p16.7 protein present 15 min after infection was calculated between 65 000—130 000 molecules per infected cell. Genes 17 and 16.7 are the first and fourth gene of the right-side early operon, respectively. Simultaneous determination of the amount of p16.7 and p17 showed that they were expressed at very similar levels throughout the infection cycle in a 1:1 ratio. The operon containing genes 17 and 16.7 is strongly repressed by protein p6 about 10–15 min after infection (Whiteley et al., 1986; Camacho and Salas, 2000). This repression most probably explains the sudden halt in p16.7 accumulation 15 min after infection, which is further supported by the observation that p16.7 levels continued to increase in cells infected with a sus6 mutant phage F29 (our unpublished results). It also implies that p16.7 (and p17) are stable throughout the infection cycle, which is also supported by immunofluorescence studies (Meijer et al., 2000, our unpublished results). Protein p16.7 is a membrane protein Different approaches verified that p16.7 is a membrane protein and that its predicted Nterminal membrane spanning domain is responsible for membrane localization. Immunofluorescence microscopy of F29-infected cells showed that p16.7 localized in an irregular punctate pattern at the periphery of the cell, presumably the membrane (Meijer et al., 2000). Also an ectopically expressed p16.7 protein fused to the green fluorescent protein (p16.7-GFP) localized throughout the periphery of the B. subtilis cells in a pattern similar to that of cells stained with the membrane dye FM-64 (Meijer et al., 2001b). Moreover, cell fractionation studies of F29-infected cells showed that p16.7 was exclusively found in the membrane fraction (Meijer et al., 2001b).

|

283

284

|

Meijer et al.

Cell fractionation studies also demonstrated that the predicted N-terminal membrane spanning domain is responsible for membrane localization of p16.7. Thus, protein p16.7A, a variant of p16.7 in which the first 20 residues were replaced by a His(6)-tag and expressed in B. subtilis from a plasmid, was recovered mainly in the cytosolic fraction. As expected, control experiments showed that another variant, p16.7B having a C-terminal His(6)-tag, was exclusively recovered in the membrane fraction (Meijer et al., 2001b). Protein p16.7 is required for optimal F29 DNA replication in vivo by efficiently distributing F29 DNA replication from its initial to additional sites at the membrane To study if p16.7 is involved in in vivo F29 DNA replication we constructed a F29 mutant containing a suppressible mutation in gene 16.7 (codon 48; CAA to TAA). Genomic sus14(1242) F29 DNA was used as starting material. The resulting double mutant phage was named sus16.7(48)/sus14(1242) (Meijer et al., 2001b). Due to the mutation in gene 14, which encodes the holin protein that is required for lysis of the cell at the end of the infection cycle, in vivo F29 DNA replication can be analyzed at late infection times. As a first approach, possible effects of p16.7 on in vivo F29 DNA replication was studied by comparing the kinetics of accumulation of intracellular phage DNA in the absence or presence of p16.7. The absence of p16.7 resulted in accumulation of lower amounts of F29 DNA at all infection times and the time required to detect the first replicated F29 DNA molecules was delayed by about 20 min compared to the situation in which p16.7 is produced (Meijer et al., 2001b). These results showed that, although p16.7 is not essential under the laboratory conditions tested, its absence clearly affects efficient in vivo F29 DNA replication. To gain further insight in the role of p16.7 the in vivo localization of F29 DNA during the infection cycle was analyzed in the presence and absence of p16.7 using immunofluorescence techniques (Meijer et al., 2000). These experiments were performed by adding the thymine analog 5-bromodeoxy-uridine (BrdU) to F29-infected cultures and subsequent detection of BrdU incorporated in phage DNA by immunofluorescence. To prevent incorporation of BrdU into the host cell chromosome these experiments were carried out in the presence of 6-(p-hydroxyphenylazo)-uracil (HpUra), which is a selective inhibitor of DNA polymerase III holoenzyme in Gram-positive bacteria (Brown, 1970). In the wildtype situation, phage DNA replication at early infection times (~10 min post infection) localized to a single focus within the cell, nearly always towards one end of the host cell nucleoid. From then on, F29 DNA replication became rapidly distributed to multiple sites at the periphery of the nucleoid, just under the cell membrane. A highly similar pattern of the initial rounds of F29 DNA replication was observed in the absence of p16.7. However, under these conditions F29 DNA replication remained localized to the initial replication sites for prolonged times during the infection cycle. These results showed that p16.7 is required for efficient distribution of F29 DNA replication from its initial to additional sites at the membrane (Meijer et al., 2000). Impairment of this distribution, which permits phage F29 DNA replication to occur simultaneously at different sites at the membrane, explains the decreased efficiency of in vivo F29 DNA replication in the absence of p16.7.

F29 DNA Replication in Vegetative and Sporulating Cells

Protein p16.7 has single-stranded and double-stranded DNA binding activity and can interact with the F29 terminal protein Since the N-terminal membrane anchor prevented purification of native p16.7, the biochemical and structural properties of p16.7 were studied using variants lacking the membrane spanning domain. In one of these variants, p16.7A, the first 20 amino acids of p16.7 were replaced by a His(6)-tag. The observation that the C-terminal half of p16.7 has limited homology with DNA binding homeodomains present in many eukaryotic transcriptional regulators prompted us to analyze whether p16.7 has DNA binding activity. Gel retardation studies demonstrated that p16.7A can bind to both single-stranded (ss) and doublestranded (ds) DNA in an apparent non-specific way (Meijer et al., 2001b; Serna-Rico et al., 2002). The ssDNA binding activity was analyzed in more detail (Serna-Rico et al., 2002). On the one hand, electron microscopy analyses showed that p16.7A can bind to the displaced stretches of ssDNA present in F29 DNA replication intermediates. On the other hand, the presence of p16.7A can functionally substitute the F29 SSB protein (encoded by gene 5) in in vitro F29 DNA amplification assays in the sense that p16.7A, like SSB, prevented the F29 DNA polymerase to switch template during amplification. However, despite these features p16.7A did not display other characteristics found for F29 SSB and other SSB proteins. Thus, p16.7A has no helix-destabilizing activity and its presence did not stimulate F29 DNA replication in the minimal in vitro system. These latter results indicate that p16.7 is not a classic SSB protein, which is further supported by the fact that the F29 SSB protein is essential for in vivo F29 DNA replication (Mellado et al., 1980). Based on its role in in vivo F29 DNA replication we considered it plausible that p16.7 might have affinity for one or more proteins that are directly involved in this process. In vitro cross-linking as well as glycerol gradient analyses indicated that p16.7A can interact with the F29 terminal protein. Moreover, the fact that p16.7A-terminal protein complexes were observed at 50 but not at 500 mM salt concentration indicated that their affinity is based on electrostatic interactions (Serna-Rico et al., 2003). Protein p16.7A forms dimers in solution and p16.7A-dimers multimerize upon DNA binding Glycerol gradient analysis showed that p16.7A forms dimers in solution. Moreover, in addition to dimers, minor amounts of higher-order multimers were detected after in vitro p16.7A cross-linking (Meijer et al., 2001b). Cross-linking combined with Western blot analysis also revealed that p16.7 forms dimers and multimers in vivo (Meijer et al., 2001b). The following results provided evidence that p16.7A dimers can interact with each other and that this interaction is favored upon DNA binding (Serna-Rico et al., 2002, 2003). First, the nucleoprotein complexes formed in the presence of high levels of p16.7A did not enter the gels in retardation assays indicating that the complexes formed under these conditions are of high molecular weight. Second, analysis of DNA binding by footprinting showed that DNA fragments, either ds- or ssDNA, became fully protected from nuclease digestion in the presence of high p16.7A concentrations, suggesting that under these conditions the DNA fragments are covered by a continuous array of p16.7A molecules. Third, electron microscopy analysis showed that p16.7A can join individual short DNA

|

285

286

|

Meijer et al.

fragments forming long nucleoprotein filaments. And fourth, addition of DNA prior to in vitro cross-linking enhanced the formation of p16.7A multimers. The functional domain of p16.7 is constituted by the C-terminal half of the protein, p16.7C, which forms high-affinity dimers Various approaches were used to determine (i) the p16.7A region(s) involved in dimerization and (ii) the minimal functional domain of p16.7 (Muñoz-Espín et al., 2004). Glycerol gradient analyses showed that p16.7A forms dimers in solution (see above). This conclusion was confirmed by analytical gel filtration. The order of magnitude of the dissociation constant was estimated by this method to be about 20 nm at room temperature. As mentioned above, the protein p16.7 contains a single region (amino acids 19 to 60) that is predicted to have a high probability of forming an A-helical coiled-coil structure. A coiled-coil consists of at least two amphipathic A-helices that are wound into a superhelix having a hydrophobic interface, and these structures are typical elements involved in protein di- or multimer formation (Lupas, 1996). Therefore, this predicted coiled-coil region was identified as the prime candidate for p16.7 dimerization. As a first approach to test this hypothesis two p16.7 derivatives, p16.7N and p16.7N4, were characterized. Protein p16.7N contained the wild-type region of p16.7 spanning residues 21–68, i.e. the predicted coiled-coil region. Protein p16.7N4 contained the same region but included four Leu to Arg substitutions that would completely disrupt the hydrophobic face of the putative A-helix. After in vitro cross-linking, low amounts of dimers and no dimers at all were observed for p16.7N and p16.7N4, respectively. Next, the secondary structure of both proteins was analyzed by far UV circular dicroism (CD) spectroscopy. Whereas the CD spectrum of p16.7N4 was characteristic of a random coil at all conditions tested, that of p16.7N revealed a substantial helical content at high concentration (590 MM) at 25°C, which increased progressively at lower temperatures. These and additional results showed that the isolated predicted coiledcoil region of p16.7 is indeed able to form a dimeric coiled-coil of about 30 residues in a coupled folding and association process. However, the very low affinity of the isolated coiled-coil region (in the µM range at 4°C) was unlikely to be responsible for the high affinity observed for p16.7A, whose dissociation constant is in the nanomolar range at room temperature. This conclusion was supported by the fact that the dissociation constant of a p16.7A derivative, p16.7A4, containing the same four Leu to Arg substitutions in the coiled-coil region as those present in p16.7N4, was also in the nanomolar range. Together, these results indicated that a region other than the coiled-coil is responsible for the high dimerization affinity of p16.7A and led us to test whether the isolated C-terminal half of p16.7 forms high affinity dimers. For this, the region encoding p16.7 residues 63–130 was cloned in frame behind the His(6)-coding region of the E. coli pET-28b(+) expression vector and used to purify the resulting protein, named p16.7C. Analytical gel filtration indeed showed that p16.7C forms dimers in solution and that its dissociation constant is similar to that of p16.7A. Protein p16.7A was shown to have ssDNA and dsDNA binding activity and to have affinity for the F29 terminal protein (see above). Gel retardation and in vitro cross-linking assays showed that these functional properties are retained by p16.7C. Moreover, like

F29 DNA Replication in Vegetative and Sporulating Cells

p16.7A, p16.7C was able to form multimers, and multimerization was enhanced in the presence of DNA. From these results it was concluded that the dimeric p16.7C protein (p16.7 residues 63–130) constitutes the main functional domain of p16.7. Protein p16.7 has a modular organization As outlined above, cell fractionation and other experiments showed that the first 20 amino acids constitute a membrane anchor that is responsible for membrane localization of the native p16.7 protein in infected cells (Meijer et al., 2000, 2001b). Gel filtration, spectroscopic and functional analyses showed that the C-terminal half of p16.7 (residues 63–130; p16.7C) constitutes the functional dimeric domain of p16.7 (Muñoz-Espín et al., 2004). And finally, the isolated region corresponding to the p16.7 residues 21–68, present in p16.7N, formed a low-affinity coiled-coil (Muñoz-Espín et al., 2004). The latter analyses did not provide clues, however, about whether the coiled-coil is formed in the native p16.7 protein. It is likely that the mobility and orientation of the coiled-coil region will be restricted in the native p16.7 protein due to the high dimerization affinity of its C-terminal halves which might shift the equilibrium of these regions towards association in a parallel coiled-coil structure. In addition, membrane localization of the p16.7 trans-membrane domains might facilitate formation of the coiled-coil. Thus, the coiled-coil might be formed at physiological concentration and temperature in the context of the native protein. Since native p16.7 is insoluble we used p16.7A, which lacks the membrane anchor but contains the remaining p16.7 sequences, to address this question. Comparison of the A-helical content of p16.7A and p16.7A4 by CD spectroscopy demonstrated that the coiled-coil is indeed formed in p16.7A at low concentrations and at 25°C (Muñoz-Espín et al., 2004). Moreover, partial proteolysis of p16.7A with proteinase K produced two major products. Their analysis by trypsin digestion combined with MALDI-TOF mass spectrometry showed that they corresponded to the N-terminal region encompassing the coiled-coil region and the C-terminal region containing the high affinity dimerization domain. These analyses also showed that proteinase K had cleaved the protein preferentially in the short p16.7 region spanning residues 61–65. Together, these results provide strong evidence that the coiled-coil domain is formed in p16.7A at low concentrations and that it constitutes an independent structural module that is connected to the C-terminal functional domain by a short proteinase K-sensitive linker region (Muñoz-Espín et al., 2004). Altogether, the results obtained indicate that the dimeric p16.7 protein is composed of three modules: an N-terminal membrane anchor (residues 1–20), a coiled-coil region (residues 30–60) and a functional C-terminal domain (residues 63–130). Structural analyses of the apo and DNA complexed form of p16.7C The primary sequence of p16.7C shows similarity (around 20 and 40% identity and similarity, respectively) with DNA binding homeodomains, which are present in a large family of eukaryotic transcription factors (Gehring et al., 1994). In addition, analysis using secondary structure prediction programs predicted that the secondary structure of p16.7C is similar to that of homeodomains (Muñoz-Espín et al., 2004). Thus, p16.7C was predicted to contain three A-helices, with position and length corresponding to those present

|

287

288

|

Meijer et al.

in homeodomains. Based on the prediction programs, the helical content of p16.7C would be about 40%. This value was experimentally confirmed by CD spectroscopy analysis of p16.7C (Muñoz-Espín et al., 2004). To gain insight into the multiple features and functions of p16.7C we determined its solution and crystal structures (Asensio et al., 2005). The solution and crystallographic three dimensional (3D) structures show that p16.7C is organized as a symmetric dimer (Figure 11.3A). As predicted, each polypeptide chain contains three A helices (corresponding to p16.7 residues 72–81 (H1), 88–95 (H2) and 103–121 (H3)). The first two helices are connected by a six-residue loop (82–87, loop 1) and helices 2 and 3 are connected by a seven-residue loop (96–102, loop 2). The C-terminal region (residues 122–128) adopts an extended structure and residues 63–66 and 129–130 were disordered. The secondary and ternary structure of each monomer is stabilized by formation of a hydrophobic core resulting from the packing of the three helices. In addition, the side chains of residue Pro87 of each monomer, 4 Å apart at the dimer interfaces, pack against the indol rings of the Trp116 residue of each monomer and this particular arrangement is likely to contribute significantly to the stability of the dimer (Asensio et al., 2005).

A

NMR

C

X-Ray

C

C

L2a

H1a

H3a H3b L1a L1b

H2a

N

B

Figure 3

N

N

C

H1b

L2b

H2b

N

C

Figure 11.3 (see also Plate 11.3) Three dimensional structures of the apo and DNA complexed form of p16.7C. (A) Solution and crystal structure of p16.7C. Left, side-view of backbone atoms (N, CA, and C) of 25 superimposed NMR-derived structures of p16.7C. Right, ribbon representation of side view of p16.7C crystal structure. (B) Side view of a ribbon representation of the structure of p16.7C in complex with dsDNA. Individual p16.7C dimers are shown in different tones of gray. DNA is shown in the concave space above the tridimeric p16.7 unit. (C) Top view of electrostatic surface representation of a tridimeric p16.7C (dark gray, positive; light gray, negative). Figures reprinted with permission from the Journal of Biological Chemistry.

F29 DNA Replication in Vegetative and Sporulating Cells

Although the secondary structure of the p16.7C monomers is indeed very similar to that of homeodomains, the spatial organization of the three A-helices in p16.7C is fundamentally different from that of homeodomains. In fact, the structure of p16.7C defines a novel dimeric six-helical fold. The surface electrostatic distribution of the p16.7C dimer showed that the H3 helices and its following extended regions exhibit a moderate positive charge indicating that this surface could be involved in DNA binding, as was shown to be the case (see later). In the crystals obtained, p16.7C dimers form a fiber around a crystallographic 3-fold screw axis. Interactions between two neighboring p16.7C dimers in the crystal involve two salt bridges and six hydrogen bonds and cover 11% of the total solvent-accessible area. Of these interactions, the two salt bridges seemed to be most relevant for the stability of the complex. These two salt bridges, formed between Glu72 and Arg98 and the symmetry related pair, could reflect the multimerization mode of p16.7C multimers. To test this hypothesis p16.7C mutants pE72Q and pR98W were analyzed by in vitro cross-linking in the absence and presence of DNA for their ability to di- and multimerize. Both mutants displayed di- and multimerization activities similar to those of p16.7C indicating that these interactions are not essential for multimerization. Insights in the surface involved in multimerization of p16.7C dimers was obtained by NMR analysis by (i) comparing the spectra of p16.7C at relative low (350 µM) and high (3500 µM) concentrations, and (ii) by comparing the spectra at low p16.7C concentration (350 µM) in the absence and presence of dsDNA. These experiments indicated that the multimerization surface involves most of the lateral surfaces of the p16.7C dimer which is characterized by a striking self-complementarity. The 3D structure of the p16.7C dimer provided a detailed view of the interactions between the monomers. However, these structures afforded only limited information about their DNA binding and multimerization surfaces, and hardly contributed to the understanding of the mode by which p16.7C binds DNA. To gain insight in these features the crystal structure of p16.7C in complex with dsDNA was resolved at 2.9 Å resolution (Albert et al., 2005). Interestingly, three p16.7C dimers, arranged side by side defining a deep dsDNA binding cavity, form a functional dsDNA binding unit (see Figure 11.3B). As far as we know, the arrangement of three dimers forming a roughly half circular DNA binding site is the only one reported so far. The specific organization of the p16.7C dimers in the dsDNA binding unit is probably due to its particular self-complementary shape and involves the lateral surfaces of the dimers as suggested before (Asensio et al., 2005). The dsDNA fits remarkably well in the concave cavity formed by the three p16.7C dimers. In addition, this concave surface, which has a strong positive electrostatic potential (see Figure 11.3C), does not have edges or structural elements that can penetrate the DNA grooves. This indicates that dsDNA binding is purely driven by electrostatic forces, which is in agreement with the non-specific dsDNA binding activity observed for p16.7C. Another interesting feature is that the DNA binding cavity is lined by a striped pattern of positively charged and hydrogen bond donor side chains above which the DNA phosphate backbone lays. Whereas these interactions are likely to restrict a pure translational diffusion of the DNA, they may allow a screw-like displacement of the DNA helix, maintaining the DNA backbone at rather fixed positions within the complex. Taking into account that the

|

289

290

|

Meijer et al.

N-terminal extension present in native p16.7 is attached to the bacterial membrane, the only interface available for the formation of larger p16.7C oligomers would be that perpendicular to the DNA duplex. Possibly, p16.7 forms larger oligomers in vivo through this interface. Importantly, (i) the biochemical properties obtained for p16.7C can be explained by the configuration of p16.7C complexed with dsDNA and (ii) this configuration is compatible with the native protein being attached to the membrane and with its proposed role of organizing DNA at the membrane of the infected cell. Together, this indicates that the crystal structure of the tridimeric p16.7C unit bound to DNA reflects the principal mode of DNA binding by native p16.7 in infected cells. Future directions and an emerging analogy between membrane-organized DNA replication of F29 and that of the B. subtilis chromosome Although the biochemical, functional and structural analysis of F29 protein p16.7 carried out so far has improved our understanding of membrane-associated F29 DNA replication the picture is far from complete. A few of the many remaining questions to be answered are the following. First, do both ss- and dsDNA-binding play a role in membrane-associated F29 DNA replication? Second, what is the role of the coiled-coil structure? The coiled-coil could be important in separating the DNA binding domain at a certain distance from the membrane and/or it may be necessary to provide the flexibility required for formation of the tridimeric DNA binding unit. In addition, it is not unlikely that the coiled-coil serves to interact with other proteins involved in the F29 DNA replication process. Third, what surfaces are involved in the interaction between p16.7 and the terminal protein? And fourth, are there other proteins, either encoded by F29 or by B. subtilis, involved in membrane-associated F29 DNA replication? Answers to these and other questions may be obtained by ongoing and future studies. One of the major insights from the p16.7-related studies performed so far is that they have revealed that a membrane-associated protein helps to organize in vivo DNA replication by interacting non-specifically with DNA. As outlined below, recently published results might indicate that this also applies to membrane-associated replication of the B. subtilis chromosome. Initiation of DNA replication of the circular B. subtilis chromosome starts at a defined chromosomal site named origin of replication (oriC). Two replisomes are assembled at oriC which then replicate the chromosome in a bidirectional manner. The formation of a DnaAoriC complex, generated upon binding of DnaA to multiple cognate DnaA binding sites in the oriC region, induces local unwinding of a short region at oriC. Recruitment and loading of the hexameric replicative helicase on this ssDNA region occurs by a process called “priming” which, in addition to DnaA, requires the accessory proteins DnaB, DnaD and DnaI (Bruand et al., 1995, 2001; Ishigo-Oka et al., 2001; Noirot-Gros et al., 2002; Soultanas, 2002; Velten et al., 2003; Rokop et al., 2004; Bruand et al., 2005). Note that B. subtilis dnaB is not related to E. coli dnaB. The E. coli dnaB gene product is the replicative helicase which in B. subtilis is encoded by dnaC. Interestingly, both DnaD and DnaB exhibit non-specific ss and dsDNA binding activities (Marsin et al., 2001; Bruand et al., 2005). In addition, DnaB is a membrane-associated protein and membrane-associated DnaB has been suggested to form the membrane at-

F29 DNA Replication in Vegetative and Sporulating Cells

tachment site for chromosomal DNA replication initiation (Winston and Sueoka, 1980; Hoshino et al., 1987; Rokop et al., 2004). This view is supported by the observation that DnaB co-localizes with oriC during initiation of chromosome replication (Imai et al., 2000). Thus, like the F29 protein p16.7, the B. subtilis DnaB and DnaD proteins have non-specific ss and dsDNA binding activity and both p16.7 and DnaB are membrane-associated. Based on these shared features it is tempting to speculate that DnaB, perhaps in conjunction with DnaD, may play a role in membrane associated replication of the B. subtilis chromosome in a way similar to that of F29 protein p16.7. This speculation may be supported by the observation that DnaB, contrary to DnaA, was detected as foci during cell-division and also appeared as concrete cellular foci when DNA replication initiation was inhibited (Imai et al., 2000). Thus, the fact that the DnaB foci do not disassemble after the initiation of DNA replication suggests that it also plays a role during replication. In this respect, it is worth noting that it has been shown recently that DnaB and DnaD exert different global DNA remodeling activities in vitro (Turner et al., 2004; Zhang et al., 2005). Based on these findings it has been suggested that the primary roles of DnaB and DnaD would be to control and modulate the bacterial nucleoid structure (Zhang et al., 2005). Phage F29 uses spore formation of its host as a survival mechanism Bacillus subtilis, the host of phage F29 and all its related phages except GA-1, is a member of a large family of bacteria that respond to nutritional stress by forming highly resistant endospores that can remain dormant for virtually unlimited periods of time before germinating to resume growth. The endospore is formed by modified, extremely polar cell-division, which generates a tiny prespore compartment near one cell pole, and a much larger mother cell. The prespore develops with the cooperation of the mother cell, which eventually lyses to release the completed spore (reviewed in, Errington, 2003). Like its related phages, F29 is classified as a virulent phage that is optimized to complete its lytic cycle during the logarithmical phase of the infected cell. Indeed, when infected during this phase, the F29 lytic cycle is completed in about 50 min generating up to 1000 phage progeny. In addition, the F29 operons are driven by promoters that are recognized by the host-encoded RNA polymerase containing the vegetative sigma A subunit. Cells initiating sporulation in response to nutrient limitation are not optimally suited though for the production of large numbers of phage progeny. Under these conditions, it would be advantageous for the phage to suppress its lytic cycle and arrange that the infecting genome becomes incorporated into the developing spore thereby effectively postponing its usual virulent life cycle until the spore germinates. Such a strategy would restrict the infection cycle to the logarithmical phase of the cell that is optimal for the generation of large numbers of phage progeny. An additional advantage of such a strategy is that the spore is far more resistant than a phage particle; in fact, B. subtilis spores are among the most durable biological entities known. Therefore, by hiding in the spore the phage genome can withstand extremely harsh conditions. Finally, B. subtilis is mainly present as spores during most of the time in natural conditions. Thus, survival of F29 may be enhanced by exploiting its hosts’ ability to form spores under adverse environmental conditions. The observation, already reported more than 30 years ago, that the lytic cycle of F29 is

|

291

292

|

Meijer et al.

suppressed in early sporulating cells and that under these conditions its genome becomes incorporated into the developing prespore (referred to as phage genome trapping) strongly indicates that F29 indeed possesses such an adaptive infection strategy (Kawamura and Ito, 1974; Moreno, 1977). Moreover, this alternative strategy appears not to be unique for F29 because similar observations have been made for the F29-related phage F15 (Ito et al., 1973) as well as for various other phages that infect B. subtilis such as for example SP10, B3, Fe, PBS1and F105c30 (Takahashi, 1964; Yehle and Doi, 1967; Kawakami and Landman, 1968; Sonenshein and Roscoe, 1969; Buu and Sonenshein, 1975; Osburne and Sonenshein, 1976). Despite these early observations, nothing was known about the molecular mechanisms underlying this adaptive infection strategy of F29 or any of the other Bacillus phages. Recently, we have resolved, at least in part, these mechanisms for F29 (Meijer et al., 2005). Effective execution of this alternative strategy requires (i) that suppression of the lytic phage cycle is synchronized with initiation of sporulation and (ii) that the infecting phage genome is efficiently segregated into the small prespore compartment. As outlined below, host-encoded proteins are key players for both processes. The B. subtilis chromosomal segregation machinery is exploited for sporeentrapment of the infecting F29 genome During the last decade it has become clear that segregation of bacterial chromosomes and plasmids is an active process (for review see, Hiraga, 2000; Gordon and Wright, 2000; Møller-Jensen et al., 2000; Wu, 2004). In B. subtilis, the Spo0J protein plays an important role in chromosomal segregation, both in vegetative and sporulating cells (for review see, Draper and Gober, 2002; Wu, 2004). The spo0J gene forms an operon with soj. Soj and Spo0J are DNA binding proteins that are related to the ParA/ParB family of proteins, which are required for the stable maintenance of plasmids (reviewed in Gerdes et al., 2000). Spo0J binds preferentially to a 16 bp imperfectly inverted repeat sequence: 5a-TGTTCCACGTGAAACA-3a. This preferential Spo0J-binding site, called parS, is located within the B. subtilis spo0J gene near oriC (Lin and Grossman, 1998). The B. subtilis chromosome contains seven additional Spo0J binding sites, scattered over the ~1 Mbp oriC region, which have one or two bp differences with respect to the parS site in the spo0J gene (Lin and Grossman, 1998). Although Soj does not appear to be required for chromosome segregation in vegetative cells of B. subtilis, combination of soj, spo0J and parS can stabilize otherwise unstable plasmids in both B. subtilis (Lin and Grossman, 1998) and Escherichia coli (Yamaichi and Niki, 2000). Furthermore, soj and spo0J are both involved in segregation of the prespore chromosome during sporulation, together with at least two other proteins, RacA and DivIVA (Ben-Yehuda et al., 2003; Wu and Errington, 2003). Interestingly, we found that the genome of F29 contains five parS-like sites, four of which having a sequence identical to the one present in the spo0J gene (Murthy et al., 1998, see Figure 11.4). We considered it likely that the parS sites present in F29 DNA function by exploiting the hostencoded Spo0J chromosome segregation machinery to enhance spore entrapment of the F29 genome. In vivo binding of Spo0J to the F29 parS sites would be a prerequisite for this hypothesis to hold true. Chromatin immunoprecipitation (CHIP) assays indeed showed that all five F29 DNA regions containing a parS site sequence were 10- to 23-fold more

F29 DNA Replication in Vegetative and Sporulating Cells

Early

Early

0A-1 0A-2 0A-3

0A-4 0A-5

0A-6

TD1

Late A2b

A2c p4-1 -10

C2 p4-3

p4-2

-35

-10

-35

-10 -10

0A-1

0A-2

0A-3

A3

-35 0A-5

0A-6

Figure 11.4 Location of perfect 0A boxes and parS sites on the F29 genome. The direction of transcription and length of the transcripts are indicated by arrows. The positions of the perfect 0A-boxes and those of the parS sites are indicated with small horizontal and large vertical triangles, respectively. The parS site having two additional bp with respect to the other ones is hatched. The positions of the various genes are indicated with numbers. The bidirectional transcriptional terminator TD1 is indicated with a hairpin structure. Black circles represent the terminal protein. A black box indicates the region spanning the early promoters A2b and A2c, and the late A3 promoter. Blow-ups of the A2c-A3 and C2 promoter regions are shown in the lower part. Transcription start sites are indicated with bent arrows. The –35 and –10 boxes areFigure indicated 4 with grey filled boxes. Note that the late A3 promoter lacks a consensus –35 sequence. The positions of the main protein p4 binding sites are indicated with black boxes. Adapted from Meijer et al. (2005).

abundant in immunoprecipitates compared to several non-parS site F29 regions, demonstrating that Spo0J binds the F29 parS sites in vivo (Meijer et al., 2005). Four out of the five F29 parS sites are located in essential genes. The covalently attached terminal protein molecules at both F29 DNA ends, which are essential for F29 DNA replication, are the reason why the linear genome of F29 is not amenable to standard genetic manipulation. Therefore, the possibility to change the five parS sites by site-directed mutagenesis was not a realistic option and other strategies were chosen to analyze whether the F29 parS sites are functionally important for spore entrapment. As a first approach we cloned a F29 parS site onto a derivative of the low-copy number B. subtilis plasmid pLS20 (Meijer et al., 1995) and analyzed the efficiency of spore-entrapment of the plasmid with or without the parS site in wild-type B. subtilis cells. About 28% of the spores turned out to contain the plasmid lacking the parS site. However, this value increased to more than 88% when the plasmid contained the parS site (independent of its orientation). When the experiments were performed in an isogenic $soj-spo0J mutant background, ~30% of the spores harbored a plasmid, independent of whether or not it contained the parS site (spo0J mutants make very few spores because of the negative transcriptional regulator soj [(Ireton et al., 1994)], hence the use of a soj-spo0J double mutant). Thus, the presence of a F29 parS site enhances the segregation efficiency of a heterologous replicon into the prespore in a soj-spo0J-dependent manner. As a second approach, we compared the ratio between spore-entrapment of the F29 genome between wild-type and an isogenic $soj-spo0J strain. Spore entrapment efficiencies of the F29 genome were consistently higher in the wild-type

|

293

294

|

Meijer et al.

as compared to the $soj-spo0J strain (Meijer et al., 2005). The data from these two approaches, together with the observation that Spo0J binds the F29 parS sites in vivo, strongly indicate that the parS sites on the F29 genome are important for spore entrapment. The F29 lytic cycle is suppressed by Spo0A, the master regulator for entrance into sporulation In vivo F29 development is suppressed in a spo0A-mediated way The master regulator for initiation of sporulation is the B. subtilis Spo0A protein (Hoch, 1993). A multicomponent phosphorelay consisting of five histidine autokinases and two phosphorelay proteins (Spo0F and Spo0B) controls the activity of Spo0A. The activated form, Spo0A~P, binds to DNA sequences containing a so-called “0A-box,” where it exerts its role as a transcriptional regulator by activating the expression of certain genes, while repressing others (for review see, Grossman, 1995; Sonenshein, 2000; Piggot and Losick, 2002; Perego and Hoch, 2002). Based on its crucial role in the initiation of sporulation, Spo0A was an attractive candidate to mediate synchronization between the onset of sporulation and suppression of the lytic cycle of F29, either directly or indirectly. Early observations describing that F29 forms large plaques on certain early blocked sporulation mutants but not on the B. subtilis wildtype strain 168 (Ito and Spizizen, 1972) was in line with this assumption. Using isogenic B. subtilis strains we confirmed that F29 forms large plaques on spo0A and spo0B mutant strains but not on the wild-type B. subtilis strain. Given that the spo0B mutant strain is blocked for Spo0A activation, this indicated that activated Spo0A is responsible for the inhibition of phage F29 development (Meijer et al., 2005). The lytic F29 cycle is suppressed at the level of transcription Because Spo0A is a transcriptional regulator known to activate and repress B. subtilis genes, we studied whether the observed suppression of F29 development is due to repression of F29 transcription. In one approach, we analyzed the F29 expression profiles in cells infected at different times after initiation of sporulation (Meijer et al., 2005). In order to synchronize initiation of sporulation, cells were grown in rich medium up to the mid to late logarithmical growth phase and were then resuspended in poor medium to induce sporulation. As expected, relatively high levels of transcripts derived from the main F29 early promoters A2c, A2b and C2 and from its late A3 promoter were detected when cells were infected immediately after they were resuspended in sporulation medium. However, much lower or hardly detectable levels of the early A2c, A2b and C2-derived transcripts were observed in samples infected 30 or 45 min after resuspension in sporulation medium, respectively. Moreover, late promoter A3-derived transcripts were no longer detected when cells were infected 15 min or later after sporulation induction. In another approach single-copy transcriptional lacZ fusions were constructed for the early promoters containing their upstream regions and used to study the temporal expression in the wild-type strain and an isogenic spo0A background (Meijer et al., 2005). The results obtained showed that the B-galactosidase activities driven by these F29 promoters started to decline as soon as the cultures entered sporulation when the lacZ fusions were in the wild-type but not when present in the spo0A background. Together, these results

F29 DNA Replication in Vegetative and Sporulating Cells

indicated that the early F29 promoters become repressed during sporulation in a spo0Adependent way. Spo0A directly represses expression of F29 promoters To discriminate whether Spo0A represses the F29 promoters directly or indirectly we examined the F29 genome for the presence of 0A boxes, which were originally defined as 5a-TGTCGAA-3a (Strauch et al., 1990). Interestingly, this examination revealed that the genome of F29 contains six 0A boxes (see Figure 11.4). Whereas one of them, 0A-box 4 (0A-4) is present within gene 8.5 encoding the phage head fiber, the other five 0A-boxes are all located in the vicinity of promoters. Three 0A boxes (0A-1, 0A-2 and 0A-3) are present in the intergenic A2c-A3 promoter region, and two (0A-5 and 0A-6) are located upstream of promoter C2. DNase I footprinting was used to study whether the five F29 promoter-associated 0A boxes are bona fide Spo0A binding sites. For these and other in vitro experiments native Spo0A was purified from overexpressing E. coli cells as described (Muchová et al., 2004). Spo0A forms dimers upon phosphorylation and Spo0A-dimers constitute the active form of Spo0A (Asayama et al., 1995; Lewis et al., 2002; Ladds et al., 2003; Muchová et al., 2004). Ladds et al. (2003) showed that a portion of wild-type Spo0A purified from E. coli is in its active phosphorylated dimeric form. Indeed, gel filtration experiments showed that part of our purified Spo0A was in its dimeric form. The functionality of our purified Spo0A was further demonstrated by the facts that (i) it produced highly similar footprints on the B. subtilis promoter spoIIG-associated 0A boxes to those published (Satola et al., 1992), and (ii) it displayed in vitro promoter activating and repressing activity (see below). DNase I footprinting indeed showed that the F29 promoter-associated 0A boxes constitute bona fide Spo0A binding sites. The footprints showed that binding of Spo0A was not limited to the 0A-box sequence. Inspection of the protected sequences revealed that each 0A box analyzed is flanked with a 3 bp spacer by an imperfect 0A-box sequence that is located either upstream (0A boxes 1 and 2) or downstream (0A boxes 3, 5 and 6) of the consensus 0A box sequence. The dual 0A boxes 1 and 5 overlap partly the A2c and C2 promoter, respectively. In addition, Spo0A also bound an approximately 20 bp region flanking the dual 0A box 3, overlapping partly with the late A3 promoter. This latter region, which does not have an obvious 0A box sequence, has been shown to be intrinsically curved (Rojo and Salas, 1991) and this curvature may enhance binding of Spo0A to this region. The situation of dual 0A boxes with a 3 bp spacer is also found at the B. subtilis abrB promoter (Strauch et al., 1990). As a consequence of this organization, the 0A boxes are spaced exactly one helical turn from each other and hence are on the same face of the DNA helix. This is an ideal situation for the binding of a dimeric protein, which constitutes the active form of Spo0A (see above). Indeed, the DNA used for determination of the crystal structure of the C-terminal DNA binding domain of Spo0A (Spo0AC) in complex with DNA contains dual 0A boxes separated by 3 bp spacers, and Spo0AC binds these dual 0A boxes as tandem dimers that are located at the same face of the helix (Zhao et al., 2002). Thus, the typical organization of dual 0A boxes indicates that they are bound by Spo0Adimers, which is supported by the results of the DNase I footprints. In vitro transcription assays were used to examine whether Spo0A is directly responsible for repression of the early F29 promoters. The B. subtilis SA-RNAP-dependent spoIIG promoter is activated by Spo0A (Satola et al., 1992), and was included in these experiments

|

295

296

|

Meijer et al.

as a positive control. Whereas, as expected, Spo0A activated the spoIIG promoter, a Spo0A-dependent decrease of promoter activity was observed for the early F29 promoters demonstrating that Spo0A represses these promoters directly (Meijer et al., 2005). Spo0A represses the early F29 promoters by different mechanisms DNase I footprinting was used to gain insight in the mechanism by which Spo0A represses the early A2c, A2b and C2 promoters (Meijer et al., 2005). 0A box regions 1 and 2 overlap with part of the A2c core promoter and its upstream promoter region, respectively. Spo0A bound to these regions prevented RNA polymerase (RNAP) from binding to this promoter. Therefore, the A2c promoter is repressed by steric hindrance due to binding of Spo0A to the 0A box regions 1 and 2. Footprint analysis showed that binding of Spo0A to the 0A box region 3 prevented binding of RNAP to the A2b promoter. However, these dual 0A boxes are located rather far upstream of the core A2b promoter (positions –54 till –70 relative to the A2b transcription start site). The activity of the A2b promoter was shown to depend almost completely on the presence of an UP element (Meijer and Salas, 2004). Interestingly, the position of this UP element coincides with the dual 0A-box 3 region. Thus, binding of Spo0A to the UP element of the A2b promoter prevents docking of the CTD of the RNAP A subunit which is crucial for its activity. To our knowledge, repression by occupation of a promoter UP element had not been reported previously. Yet another situation was observed at the C2 promoter. As mentioned, two dual 0A box regions, 5 and 6, are located in the vicinity of the C2 promoter. 0A box region 5 partly overlaps with the core promoter and its direct upstream region. 0A box region 6 is located further upstream of the C2 promoter (positions –104 till –120 relative to the C2 transcription start site). Interestingly, Spo0A bound to 0A box regions 5 and 6 did not prevent RNAP from binding to the C2 promoter, but the simultaneous binding of Spo0A and RNAP caused some alterations in the RNAP-generated footprint. Thus, it led to (i) increased hypersensitivity of position –32, (ii) incomplete protection of position –7, and (iii) full protection of position –60. Binding of Spo0A to the 0A box 5 region is responsible for these alterations since these were also observed using a shorter DNA fragment that lacked the 0A box 6. In fact, the presence of only 0A box region 5 was sufficient for Spo0Amediated repression of the C2 promoter. Earlier we had shown that the C2 promoter contains an UP element and that binding of the CTD of the RNAP A subunit to the UP element results in full protection of the –46 to –50 region and partial protection of position –60 (Meijer and Salas, 2004). Hence, the 0A box 5 region overlaps with part of the C2 core promoter and part of its UP element. Enhanced binding of the CTD of the RNAP A-subunit to position –60 was responsible for the full protection of this position in the presence of both RNAP and Spo0A as shown by comparing the footprints generated by Spo0A in combination with either wild-type RNAP or RNAP containing a mutant A subunit lacking its 15 C-terminal amino acids (A$15-RNAP) (Mencía et al., 1996a). Moreover, these latter experiments also revealed (i) that position –32 became hypersensitive in both situations, and (ii) that full protection of the C2 core promoter by A$15-RNAP became lost in the presence of high Spo0A con-

F29 DNA Replication in Vegetative and Sporulating Cells

centrations. These results indicated therefore that the Spo0A-mediated hypersensitivity of position –32 does not require ACTD and that binding of ACTD to position –60 is required for stable binding of RNAP to the C2 promoter in the presence of Spo0A. The observation that binding of Spo0A to 0A box region 5 did not prevent wild-type RNAP from binding to the C2 promoter demonstrated that the Spo0A-mediated repression of this promoter should be exerted at a step after formation of the closed complex. Potassium permanganate footprinting showed that binding of Spo0A to 0A box region 5 affected open complex formation at this promoter. Together, these results strongly indicate that the Spo0A-mediated repression of the C2 promoter is due to overstabilization of the closed complex as a consequence of enhanced binding of ACTD at position –60 and possibly to direct interactions of Spo0A with the SA subunit of RNAP. Spo0A prevents activation of the late A3 promoter As mentioned above, the F29 late A3 promoter was not activated in vivo when cells were infected 15 min or later after induction of sporulation. Activation of promoter A3, which lacks a typical –35 box, occurs through p4-mediated recruitment of RNAP to the A3 promoter via contacts between protein p4 and the CTD of the RNAP A subunit (Mencía et al., 1996a). The p4-binding site 3, needed for promoter A3 activation, is located from positions –69 to –95 with respect to its transcription start site. Interestingly, 0A box 3 is positioned in between the A3 promoter and the p4-binding site 3 (see Figure 11.4). Binding of Spo0A to the 0A-box 3 region might therefore interfere with activation of the late A3 promoter. In vitro run-off assays indeed showed that activation of the late A3 promoter was prevented in a Spo0A-dependent way. The Spo0A-mediated prevention of promoter A3 activation was studied by DNase I footprinting. These analyses showed that protein p4-mediated recruitment of RNAP to the late A3 promoter was lost when Spo0A was bound to the extended 0A box region 3. Since some of the characteristic protein p4 footprint features were still observed under these latter conditions, it appears that p4 and Spo0A can bind simultaneously to their flanking cognate binding sites (Meijer et al., 2005). Protein p4-mediated activation of the late A3 promoter belongs to the so-called class I type of activation. Generally, activation of this class of promoters involves not only a surface domain of ACTD that interacts with the transcriptional activator, but also another ACTD surface domain that interacts with the DNA flanking the activator DNA binding site. Interactions of ACTD with both the activator and the DNA are normally required for class I promoter activation (for reviews see, Busby and Ebright, 1999; Browning and Busby, 2004). The late F29 A3 promoter seems to be no exception, since DNase I footprint analyses of the protein p4-A complex showed that the ACTD binds adjacent to the p4-binding site 3 (positions –58 and –65 relative to the A3 promoter transcription start site, Mencía et al., 1996a). Interestingly, the ACTD binding site of the A3 promoter coincides with the 0A box 3 region. As mentioned above, this Spo0A binding region also constitutes the docking site of ACTD crucial for binding of RNAP to the divergently oriented early A2b promoter (Meijer and Salas, 2004). Thus, binding of Spo0A to 0A box 3 region prevents activation of the late A3 promoter and represses the early A2b promoter by occupying, in both cases, the ACTD binding site.

|

297

298

|

Meijer et al.

Recapitulation and future directions Our studies showed that F29 has evolved at least two mechanisms that enable it to adapt and exploit the ability of its host to survive through the formation of endospores. By repressing transcription of genes required for the lytic cycle and providing cis-acting sites for segregation into the prespore compartment, the phage encapsulates its genome into one of the most resistant and durable structures in biology. It thereby postpones its replication and the destruction of its host until spore germination occurs, when conditions are likely to be much more favorable to its continued proliferation and spread. Thus, F29 ensures that the lytic cycle is restricted to the logarithmical growth phase. In addition, these studies have revealed new insights in the intriguing phage–host interactions. Finally, it is worth mentioning that it is known for several decades that, contrary to most other phages, infection of F29 hardly affects RNA and protein synthesis of the host (Schachtele et al., 1972). Non-interference with the major metabolic features of the host is required for successful formation of mature spores, which takes up to 7 hours, and hence for successful execution of the alternative infection strategy. So far it is not known whether the alternative infection strategy of F29, based on the presence of cis-acting binding sites for the host-encoded Spo0J and Spo0A proteins, is unique to F29. Analysis of the complete GA-1 sequence showed that it does not contain perfect 0A boxes nor parS sites. This indicates that GA-1, the only member of group III of the F29-family of phages, does not possess the alternative infection strategy, at least not based on the action of Spo0J and Spo0A. The presence of at least some 0A boxes and parS sites, however, in genomes of the F29-related phages belonging to groups I and II, as detected by inspection of their (in)complete genome sequences deposited in the nucleotide databases, suggests that these F29-related phages may possess this alternative strategy. Minor deviations in the 0A boxes with respect to their number, position and DNA sequence may modulate the Spo0A-mediated transcriptional regulation, which, in theory, may cause that they respond differentially to the physiological state of the host during infection. These and related issues are currently under study. Acknowledgments We thank Armando Albert, Juan Luis Asensio and Mauricio G. Mateu for their collaboration in the structural analyses studies of protein p16.7. We also want to acknowledge the Spanish Ministry of Science and Technology who supported W.J.J.M. and V.C-Ll by means of the “Ramón y Cajal” program and a predoctoral fellowship, respectively, and the Fondo de Investigación Sanitaria who supported D.M-E. These investigations were financed by grants BFU2005–00733, BFU2005–01878 from the Spanish Ministry of Education and Science to M.S. and W.J.J.M., respectively, and an Institutional grant from Fundación Ramón Areces to the Centro de Biología Molecular “Severo Ochoa.” References

Abril, A.M., Salas, M., Andreu, J.M., Hermoso, J.M., and Rivas, G. (1997). Phage F29 protein p6 is in a monomer-dimer equilibrium that shifts to higher association states at the millimolar concentrations found in vivo. Biochemistry 36, 11901–11908. Ackermann, H.-W. (1998). Tailed bacteriophages: the order Caudovirales. Adv. Virus Res. 51, 135–201.

F29 DNA Replication in Vegetative and Sporulating Cells

Albert, A., Muñoz-Espín, D., Jiménez, M., Asensio, J.L., Hermoso, J.A., Salas, M., and Meijer, W.J.J. (2005). Structural basis for membrane anchorage of viral F29 DNA during replication. J. Biol. Chem. 280, 4286–4288. Arwert, F., and Venema, G. (1974). Protease-sensitive transfection of Bacillus subtilis with bacteriophage GA-1 DNA: a probable case of heterologous transfection. J. Virol. 13, 584–589. Asayama, M., Yamamoto, A., and Kobayashi, Y. (1995). Dimer form of phosphorylated Spo0A, a transcriptional regulator, stimulates the spo0F transcription at the initiation of sporulation in Bacillus subtilis. J. Mol. Biol. 250, 11–23. Asensio, J.L., Albert, A., Muñoz-Espín, D., Gonzalez, C., Hermoso, J., Villar, L., Jiménez-Barbero, J., Salas, M., and Meijer, W.J.J. (2005). Structure of the functional domain of F29 replication organizer. Insights into oligomerization and DNA binding. J. Biol. Chem. 280, 20730–20739. Barthelemy, I., Mellado, R.P., and Salas, M. (1989). In vitro transcription of bacteriophage F29 DNA: inhibition of early promoters by the viral replication protein p6. J. Virol. 63, 460–462. Barthelemy, I., and Salas, M. (1989). Characterization of a new prokaryotic transcriptional activator and its DNA recognition site. J. Mol. Biol. 208, 225–232. Barthelemy, I., Salas, M., and Mellado, R.P. (1986). In vivo transcription of bacteriophage F29 DNA: transcription initiation sites. J. Virol. 60, 874–879. Barthelemy, I., Salas, M., and Mellado, R.P. (1987). In vivo transcription of bacteriophage F29 DNA: transcription termination. J. Virol. 61, 1751–1755. Ben-Yehuda, S., Rudner, D.Z., and Losick, R. (2003). RacA, a bacterial protein that anchors chromosomes to the cell poles. Science 299, 532–536. Blanco, L., Bernad, A., Esteban, J.A., and Salas, M. (1992). DNA-independent deoxynucleotidylation of the F29 terminal protein by the F29 DNA polymerase. J. Biol. Chem. 267, 1225–1230. Blanco, L., Bernad, A., Lázaro, J.M., Martín, G., Garmendia, C., and Salas, M. (1989). Highly efficient DNA synthesis by the phage F29 DNA polymerase. Symmetrical mode of DNA replication. J. Biol. Chem. 264, 8935–8940. Blanco, L., Lázaro, J.M., De Vega, M., Bonnin, A., and Salas, M. (1994). Terminal protein-primed DNA amplification. Proc. Natl. Acad. Sci. USA 91, 12198–12202. Blanco, L., and Salas, M. (1984). Characterization and purification of a phage F29-encoded DNA polymerase required for the initiation of replication. Proc. Natl. Acad. Sci. USA 81, 5325–5329. Blanco, L., and Salas, M. (1996). Relating structure to function in F29 DNA polymerase. J. Biol. Chem. 271, 8509–8512. Boeneman, K., and Crooke, E. (2005). Chromosomal replication and the cell membrane. Curr. Opin. Microbiol. 8, 143–148. Bravo, A., Illana, B., and Salas, M. (2000). Compartmentalization of phage F29 DNA replication: interaction between the primer terminal protein and the membrane-associated protein p1. EMBO J. 19, 5575–5584. Bravo, A., and Salas, M. (1997). Initiation of bacteriophage F29 DNA replication in vivo: assembly of a membrane-associated multiprotein complex. J. Mol. Biol. 269, 102–112. Bravo, A., and Salas, M. (1998). Polymerization of bacteriophage F29 replication protein p1 into protofilament sheets. EMBO J. 17, 6096–6105. Bravo, A., Serrano-Heras, G., and Salas, M. (2001). A single amino acid substitution within a coiled-coil motif changes the assembly of a 53-amino acid protein from two-dimensional sheets to filamentous structures. J. Biol. Chem. 276, 21250–21256. Bravo, A., Serrano-Heras, G., and Salas, M. (2005). Compartmentalization of prokaryotic DNA replication. FEMS Microbiol. Rev. 29, 25–47. Brown, N.C. (1970). 6-(-p-Hydroxyphenylazo)-uracil: a selective inhibitor of host DNA replication in phage-infected Bacillus subtilis. Proc. Natl. Acad. Sci. USA 67, 1454–1461. Browning, D.F., and Busby, S.J.W. (2004). The regulation of bacterial transcription initiation. Nature Rev. Microbiol. 2, 57–65. Bruand, C., Ehrlich, S.D., and Jannière, L. (1995). Primosome assembly site in Bacillus subtilis. EMBO J. 14, 2642–2650. Bruand, C., Farache, M., McGovern, S., Ehrlich, S.D., and Polard, P. (2001). DnaB, DnaD and DnaI proteins are components of the Bacillus subtilis replication restart primosome. Mol. Microbiol. 42, 245–255.

|

299

300

|

Meijer et al.

Bruand, C., Velten, M., McGovern, S., Marsin, S., Serena, C., Ehrlich, S.D., and Polard, P. (2005). Functional interplay between the Bacillus subtilis DnaD and DnaB proteins essential for initiation and re-initiation of DNA replication. Mol. Microbiol. 55, 1138–1150. Busby, S., and Ebright, R.H. (1999). Transcription activation by catabolite activator protein (CAP). J. Mol. Biol. 293, 199–213. Buu, A., and Sonenshein, A.L. (1975). Nucleic acid synthesis and ribonucleic acid polymerase specificity in germinating and outgrowing spores of Bacillus subtilis. J. Bacteriol. 124, 190–200. Calles, B., Salas, M., and Rojo, F. (2002). The F29 transcriptional regulator contacts the nucleoid protein p6 to organize a repression complex. EMBO J. 21, 6185–6194. Camacho, A., and Salas, M. (2000). Pleiotropic effect of protein p6 on the viral cycle of bacteriophage F29. J. Bacteriol. 182, 6927–6932. Camacho, A., and Salas, M. (2001a). Mechanism for the switch of F29 DNA early to late transcription by regulatory protein p4 and histone-like protein p6. EMBO J. 20, 6060–6070. Camacho, A., and Salas, M. (2001b). Repression of bacteriophage F29 early promoter C2 by viral protein p6 is due to impairment of closed complex. J. Biol. Chem. 276, 28927–28932. Carrascosa, J.L., Camacho, A., Moreno, F., Jiménez, F., Mellado, R.P., Viñuela, E., and Salas, M. (1976). Bacillus subtilis phage F29; characterization of gene products and functions. Eur. J. Biochem. 66, 229–241. Draper, G.C., and Gober, J.W. (2002). Bacterial chromosome segregation. Annu. Rev. Microbiol. 56, 567–597. Elías-Arnanz, M., and Salas, M. (1999). Functional interactions between a phage histone-like protein and a transcriptional factor in regulation of F29 early-late transcriptional switch. Genes Dev. 13, 2502–2513. Errington, J. (2003). Regulation of endospore formation in Bacillus subtilis. Nature Rev. Microbiol. 1, 117–126. Fauquet, C.M. (1999). Taxonomy, classification and nomenclature of viruses. In: Encyclopaedia of Virology, A. Granoff, and R.G. Webster, eds. (London: Academic Press, Limited), pp. 1730–1756. Firshein, W. (1989). Role of the DNA/membrane complex in prokaryotic DNA replication. Annu. Rev. Microbiol. 43, 89–120. Garvey, K.J., Yoshikawa, H., and Ito, J. (1985). The complete sequence of the Bacillus phage F29 right early region. Gene 40, 301–309. Gascón, I., Lázaro, J.M., and Salas, M. (2000). Differential functional behavior of viral F29, Nf, and GA-1 SSB proteins. Nucleic Acids Res. 28, 2034–2042. Gehring, W.J., Affolter, M., and Bürglin, T. (1994). Homeodomain proteins. Annu. Rev. Biochem. 63, 487–526. Gerdes, K., Møller-Jensen, J., and Jensen, R.B. (2000). Plasmid and chromosome partitioning: surprises from phylogeny. Mol. Microbiol. 37, 455–466. Gordon, G.S., and Wright, A. (2000). DNA segregation in bacteria. Annu. Rev. Microbiol. 54, 681–708. Grossman, A.D. (1995). Genetic networks controlling the initiation of sporulation and the development of genetic competence in Bacillus subtilis. Annu. Rev. Genet. 29, 477–508. Gutiérrez, C., Sogo, J.M., and Salas, M. (1991). Analysis of replicative intermediates produced during bacteriophage F29 DNA replication in vitro. J. Mol. Biol. 222, 983–994. Hashiyama, K., Takeuchi, A., and Makino, O. (2005a). A variant protein from F29 replication gene, gene 1, did not form homo-polymer due to a single amino acid substitution near the carboxyl terminus. Biosci. Biotech. Biochem. 69, 1045–1048. Hashiyama, K., Takeuchi, A., and Makino, O. (2005b). Effects of single amino acid substitutions at the predicted coiled-coil or hydrophobic region on the self-assembly of F29 replication protein, gp1. Biochem. Biophys. Res. Commun. 331, 1310–1316. Hermoso, J.M., Méndez, E., Soriano, F., and Salas, M. (1985). Location of the serine residue involved in the linkage between the terminal protein and the DNA of F29. Nucleic Acids Res. 13, 7715–7728. Hiraga, S. (2000). Dynamic localization of bacterial and plasmid chromosomes. Annu. Rev. Genet. 34, 21–59. Hoch, J.A. (1993). Regulation of the phosphorelay and the initiation of sporulation in Bacillus subtilis. Annu. Rev. Microbiol. 47, 441–465. Holder, R.D., and Whiteley, H.R. (1983). In vitro synthesis of late bacteriophage F29 RNA. J. Virol. 46, 690–702.

F29 DNA Replication in Vegetative and Sporulating Cells

Horcajadas, J.A. (2000). Transcriptional control of the Bacillus bacteriophage GA-1. Doctoral thesis. Universidad Autónoma, Madrid. Horcajadas, J.A., Meijer, W.J.J., Rojo, F., and Salas, M. (2001). Analysis of early promoters of the Bacillus bacteriophage GA-1. J. Bacteriol. 183, 6965–6970. Hoshino, T., McKenzie, T., Schmidt, S., Tanaka, T., and Sueoka, N. (1987). Nucleotide sequence of Bacillus subtilis dnaB: A gene essential for DNA replication initiation and membrane attachment. Proc. Natl. Acad. Sci. USA 84, 653–657. Imai, Y., Ogasawara, N., Ishigo-Oka, D., Kadoya, R., Daito, T., and Moriya, S. (2000). Subcellular localization of Dna-initiation proteins of Bacillus subtilis: evidence that chromosome replication begins at either edge of the nucleoids. Mol. Microbiol. 36, 1037–1048. Inciarte, M.R., Salas, M., and Sogo, J.M. (1980). The structure of replicating DNA molecules of Bacillus subtilis phage F29. J. Virol. 34, 187–190. Ireton, K., Gunther, N.W., and Grossman, A.D. (1994). spo0J is required for normal chromosome segregation as well as the initiation of sporulation in Bacillus subtilis. J. Bacteriol. 176, 5320–5329. Ishigo-Oka, D., Ogasawara, N., and Moriya, S. (2001). DnaD protein of Bacillus subtilis interacts with DnaA, the initiator protein of replication. J. Bacteriol. 183, 2148–2150. Ito, J., Meinke, W., Hathaway, G., and Spizizen, J. (1973). Studies on Bacillus subtilis bacteriophage F15. Virology 56, 110–122. Ito, J., and Spizizen, J. (1972). Early-blocked asporogenous mutants of Bacillus subtilis. In: Spores V, H.O. Halvorson, R. Hanson, and L.L. Campbell, eds. (Washington, DC: American Society for Microbiology), pp. 107–112. Ivarie, R.D., and Pène, J.J. (1973). DNA replication in bacteriophage F29: The requirement of a viralspecific product for association of F29 DNA with the cell membrane of Bacillus amyloliquefaciens. Virology 52, 351–362. Jensen, R.B., and Shapiro, L. (2000). Proteins on the move: dynamic protein localization in prokaryotes. Trends Cell Biol. 10, 483–488. Kawakami, M., and Landman, O.E. (1968). Nature of the carrier state of bacteriophage SP-10 in Bacillus subtilis. J. Bacteriol. 95, 1804–1812. Kawamura, F., and Ito, J. (1974). Bacteriophage gene expression in sporulating cells of Bacillus subtilis 168. Virology 62, 414–425. Kawamura, F., and Ito, J. (1977). Transcription of the genome of bacteriophage F29: Isolation and mapping of the major early mRNA synthesized in vivo and in vitro. J. Virol. 23, 562–577. Ladds, J.C., Muchová, K., Blaškovic, D., Lewis, R.J., Brannigan, J.A., Wilkinson, A.J., and Barák, I. (2003). The response regulator Spo0A from Bacillus subtilis is efficiently phosphorylated in Escherichia coli. FEMS Microbiol. Lett. 223, 153–157. Lemon, K.P., and Grossman, A.D. (1998). Localization of bacterial DNA polymerase: evidence for a factory model of replication. Science 282, 1516–1519. Lewis, R.J., Scott, D.J., Brannigan, J.A., Ladds, J.C., Cervin, M.A., Spiegelman, G.B., Hogget, J.G., Barák, I., and Wilkinson, A.J. (2002). Dimer formation and transcription activation in the sporulation response regulator Spo0A. J. Mol. Biol. 316, 235–245. Lin, D.C.H., and Grossman, A.D. (1998). Identification and characterization of a bacterial chromosome partitioning site. Cell 92, 675–685. Lupas, A. (1996). Coiled coils: new structures and new functions. Trends. Biochem. Sci. 21, 375–382. Lupas, A., van Dyke, M., and Stock, J. (1991). Predicting coiled-coils from protein sequences. Science 252, 1162–1164. Maniloff, J., Ackermann, H.-W., and Jarvis, A. (1999). Phage taxonomy and classification. In: Encyclopaedia of Virology, A. Granoff, and R.G. Webster, eds. (London: Academic Press, Limited), pp. 1221–1228. Marsin, S., McGovern, S., Ehrlich, S.D., Bruand, C., and Polard, P. (2001). Early steps of Bacillus subtilis primosome assembly. J. Biol. Chem. 276, 45818–45825. McGuire, J.C., Gilpatrick, M.W., and Pène, J.J. (1977). DNA replication of bacteriophage F29: Effect of two viral genes on the association of phage chromosomes with the host cell membrane. Virology 78, 234–240. Meijer, W.J.J., Castilla-Llorente, V., Villar, L., Murray, H., Errington, J., and Salas, M. (2005). Molecular basis for the exploitation of spore formation as survival mechanism by virulent phage F29. EMBO J. 24, 3647–3657.

|

301

302

|

Meijer et al.

Meijer, W.J.J., de Boer, A., van Tongeren, S., Venema, G., and Bron, S. (1995). Characterization of the replication region of the Bacillus subtilis plasmid pLS20: a novel type of replicon. Nucleic Acids Res. 23, 3214–3223. Meijer, W.J.J., Horcajadas, J.A., and Salas, M. (2001a). F29-family of phages. Microbiol. Mol. Biol. Rev. 65, 261–287. Meijer, W.J.J., Lewis, P.J., Errington, J., and Salas, M. (2000). Dynamic relocalization of phage F29 DNA during replication and the role of the viral protein p16.7. EMBO J. 19, 4182–4190. Meijer, W.J.J., and Salas, M. (2004). Relevance of UP elements for three strong Bacillus subtilis phage F29 promoters. Nucleic Acids Res. 32, 1166–1176. Meijer, W.J.J., Serna-Rico, A., and Salas, M. (2001b). Characterization of the bacteriophage F29-encoded protein p16.7: a membrane protein involved in phage DNA replication. Mol. Microbiol. 39, 731–746. Mellado, R.P., Barthelemy, I., and Salas, M. (1986a). In vitro transcription of bacteriophage F29 DNA. Correlation between in vitro and in vivo promoters. Nucleic Acids Res. 14, 4731–4741. Mellado, R.P., Barthelemy, I., and Salas, M. (1986b). In vivo transcription of bacteriophage F29 DNA early and late promoter sequences. J. Mol. Biol. 191, 191–197. Mellado, R.P., Moreno, F., Viñuela, E., Salas, M., Reilly, B.E., and Anderson, D.L. (1976). Genetic analysis of bacteriophage F29 of Bacillus subtilis: Integration and mapping of reference mutants of two collections. J. Virol. 19, 495–500. Mellado, R.P., Peñalva, M.A., Inciarte, M.R., and Salas, M. (1980). The protein covalently linked to the 5a termini of the DNA of Bacillus subtilis phage F29 is involved in the initiation of DNA replication. Virology 104, 96. Mencía, M., Monsalve, M., Rojo, F., and Salas, M. (1996a). Transcription activation by phage F29 protein p4 is mediated by interaction with the A subunit of Bacillus subtilis RNA polymerase. Proc. Natl. Acad. Sci. USA 93, 6616–6620. Mencía, M., Monsalve, M., Rojo, F., and Salas, M. (1998). Substitution of the C-terminal domain of the Escherichia coli RNA polymerase A subunit by that from Bacillus subtilis makes the enzyme responsive to a Bacillus subtilis transcriptional activator. J. Mol. Biol. 275, 177–185. Mencía, M., Monsalve, M., Salas, M., and Rojo, F. (1996b). Transcriptional activator of phage F29 late promoter: mapping of residues involved in interaction with RNA polymerase and in DNA bending. Mol. Microbiol. 20, 273–282. Mencía, M., Salas, M., and Rojo, F. (1993). Residues of the Bacillus subtilis phage F29 transcriptional activator required both to interact with RNA polymerase and to activate transcription. J. Mol. Biol. 233, 695–704. Méndez, J., Blanco, L., Esteban, J.A., Bernad, A., and Salas, M. (1992). Initiation of F29 DNA replication occurs at the second 3a nucleotide of the linear template: A sliding-back mechanism for protein-primed DNA replication. Proc. Natl. Acad. Sci. USA 89, 9579–9583. Méndez, J., Blanco, L., and Salas, M. (1997). Protein-primed DNA replication: a transition between two modes of priming by a unique DNA polymerase. EMBO J. 16, 2519–2527. Møller-Jensen, J., Jensen, R.B., and Gerdes, K. (2000). Plasmid and chromosome segregation in prokaryotes. Trends Microbiol. 8, 313–320. Monsalve, M., Calles, B., Mencía, M., Salas, M., and Rojo, F. (1997). Transcription activation or repression by phage F29 protein p4 depends on the strength of the RNA polymerase-promoter interactions. Mol. Cell 1, 99–107. Monsalve, M., Mencía, M., Rojo, F., and Salas, M. (1995). Transcription regulation in Bacillus subtilis phage F29: Expression of the viral promoters throughout the infection cycle. Virology 207, 23–31. Monsalve, M., Mencía, M., Rojo, F., and Salas, M. (1996a). Activation and repression of transcription at two different phage F29 promoters are mediated by interaction of the same residues of regulatory protein p4 with RNA polymerase. EMBO J. 15, 383–391. Monsalve, M., Mencía, M., Salas, M., and Rojo, F. (1996b). Protein p4 represses phage F29 A2c promoter by interacting with the A subunit of Bacillus subtilis RNA polymerase. Proc. Natl. Acad. Sci. USA 93, 8913–8918. Moreno, F. (1977). Development of bacteriophage F29 in sporulating and non-sporulating cells of Bacillus subtilis 168. Annales Microbiologie 128B, 3–18. Muchová, K., Lewis, R.J., Perecko, D., Brannigan, J.A., Ladds, J.C., Leech, A., Wilkinson, A.J., and Barák, I. (2004). Dimer-induced signal propagation in Spo0A. Mol. Microbiol. 53, 829–842.

F29 DNA Replication in Vegetative and Sporulating Cells

Muñoz-Espín, D., Mateu, M.G., Villar, L., Marina, A., Salas, M., and Meijer, W.J.J. (2004). Phage F29 DNA-replication organizer membrane protein p16.7 contains a coiled-coil and a dimeric, homeodomain-related, functional domain. J. Biol. Chem. 279, 50437–50445. Murthy, V., Meijer, W.J.J., Blanco, L., and Salas, M. (1998). DNA polymerase template switching at specific sites on the F29 genome causes the in vivo accumulation of subgenomic F29 DNA molecules. Mol. Microbiol. 29, 787–798. Noirot-Gros, M.F., Dervyn, E., Wu, L.J., Mervelet, P., Errington, J., Ehrlich, S.D., and Noirot, P. (2002). An expanded view of bacterial DNA replication. Proc. Natl. Acad. Sci. USA 99, 8342–8347. Nuez, B., Rojo, F., and Salas, M. (1992). Phage F29 regulatory protein p4 stabilizes the binding of the RNA polymerase to the late promoter in a process involving direct protein-protein contacts. Proc. Natl. Acad. Sci. USA 89, 11401–11405. Osburne, M.S., and Sonenshein, A.L. (1976). Behavior of a temperate bacteriophage in differentiating cells of Bacillus subtilis. J. Virol. 19, 26–35. Paces, V., Vlcek, C., Urbanek, P., and Hostomsky, Z. (1986). Nucleotide sequence of the right early region of Bacillus subtilis phage PZA completes the 19366-bp sequence of PZA genome. Comparison with the homologous sequence of phage F29. Gene 44, 115–120. Pecenkova, T., Benes, V., Paces, J., Vlcek, C., and Paces, V. (1997). Bacteriophage B103: complete DNA sequence of its genome and relationship to other Bacillus phages. Gene 199, 157–163. Pecenkova, T., and Paces, V. (1999). Molecular phylogeny of F like phages and their evolutionary relatedness to other protein-primed replicating phages and other phages hosted by Gram-positive bacteria. J. Mol. Evol. 48, 197–208. Perego, M., and Hoch, J.A. (2002). Two-component systems, phophorelays, and regulation of their activities by phosphatases. In: Bacillus subtilis and its Closest Relatives: From Genes to Cells, A.L. Sonenshein, J.A. Hoch, and R. Losick, eds. (Washington, DC: American society for microbiology press), pp. 473–481. Piggot, P.J., and Losick, R. (2002). Sporulation genes and intercompartmental regulation. In: Bacillus subtilis and its Closest Relatives: From Genes to Cells, A.L. Sonenshein, J.A. Hoch, and R. Losick, eds. (Washington, DC: American Society for Microbiology press), pp. 483–517. Prieto, I., Méndez, E., and Salas, M. (1989). Characterization, overproduction and purification of the product of gene 1 of Bacillus subtilis phage F29. Gene 77, 204. Reilly, B.E., Zeece, V.M., and Anderson, D.L. (1973). Genetic study of suppressor-sensitive mutants of the Bacillus subtilis bacteriophage F29. J. Virol. 11, 756–760. Rojo, F., Mencía, M., Monsalve, M., and Salas, M. (1998). Transcription activation and repression by interaction of a regulator with the A subunit of RNA polymerase: the model of phage F29 protein p4. Prog. Nucl. Acid Res. Mol. Biol. 60, 29–46. Rojo, F., and Salas, M. (1991). A DNA curvature can substitute phage F29 regulatory protein p4 when acting as a transcriptional repressor. EMBO J. 10, 3429–3438. Rokop, M.E., Auchtung, J.M., and Grossman, A.D. (2004). Control of DNA replication initiation by recruitment of an essential initiation protein to the membrane of Bacillus subtilis. Mol. Microbiol. 52, 1757–1767. Salas, M. (1991). Protein-priming of DNA replication. Annu. Rev. Biochem. 60, 39–71. Salas, M. (1999). Mechanism of initiation of linear DNA replication in prokaryotes. Genet. Eng. 21, 159–171. Salas, M., Miller, J.T., Leis, J., and DePamphilis, M.L. (1996). Mechanisms for priming DNA synthesis. In: DNA Replication in Eukaryotic Cells, M.L. DePamphilis, ed. (New York: Cold Spring Harbor Laboratory press), pp. 131–176. Satola, S., Baldus, J.M., and Moran, C.P.Jr. (1992). Binding of Spo0A stimulates spoIIG promoter activity in Bacillus subtilis. J. Bacteriol. 174, 1448–1453. Schachtele, C.F., DeSain, C.V., Hawley, L.A., and Anderson, D.L. (1972). Transcription during the development of bacteriophage F29: production of host and F29-specific ribonucleic acid. J. Virol. 10, 1170–1178. Serna-Rico, A., Muñoz-Espín, D., Villar, L., Salas, M., and Meijer, W.J.J. (2003). The integral membrane protein p16.7 organizes in vivo F29 DNA replication through interaction with both the terminal protein and ssDNA. EMBO J. 22, 2297–2306.

|

303

304

|

Meijer et al.

Serna-Rico, A., Salas, M., and Meijer, W.J. J. (2002). The Bacillus subtilis phage F29 protein p16.7, involved in F29 DNA replication, is a membrane-localized single-stranded DNA-binding protein. J. Biol. Chem. 277, 6733–6742. Serrano, M., Gutiérrez, C., Freire, R., Bravo, A., Salas, M., and Hermoso, J.M. (1994). Phage F29 protein p6: a viral histone-like protein. Biochimie 76, 981–991. Serrano-Heras, G., Salas, M., and Bravo, A. (2003). In vivo assembly of phage F29 replication protein p1 into membrane-associated multimeric structures. J. Biol. Chem. 278, 40771–40777. Serrano-Heras, G., Salas, M., and Bravo, A. (2006). A uracil-DNA glycosylase inhibitor encoded by a non-uracil containing viral DNA. J. Biol. Chem. 281, 7068–7074. Siegel, P.J., and Schaechter, M. (1973). The role of the host cell membrane in the replication and morphogenesis of bacteriophages. Annu. Rev. Microbiol. 27, 261–281. Sogo, J.M., Inciarte, M.R., Corral, J., Viñuela, E., and Salas, M. (1979). RNA polymerase binding sites and transcription map of the DNA of Bacillus subtilis phage F29. J. Mol. Biol. 127, 411–436. Sogo, J.M., Inciarte, M.R., Corral, J., Viñuela, E., and Salas, M. (1982). Structure of protein-containing replicative intermediates of Bacillus subtilis phage F29 DNA. Virology 116, 1–18. Sonenshein, A.L. (2000). Control of sporulation initiation in Bacillus subtilis. Curr. Opin. Microbiol. 3, 561–566. Sonenshein, A.L., and Roscoe, D.H. (1969). The course of phage Fe infection in sporulating cells of Bacillus subtilis strain 3610. Virology 39, 265–275. Soultanas, P. (2002). A functional interaction between the putative primosomal protein DnaI and the main replicative DNA helicase DnaB in Bacillus. Nucleic Acids Res. 30, 966–974. Strauch, M.A., Webb, V., Spiegelman, G.B., and Hoch, J.A. (1990). The Spo0A protein of Bacillus subtilis is a repressor of the abrB gene. Proc. Natl. Acad. Sci. USA 87, 1801–1805. Sueoka, N. (1998). Cell membrane and chromosome replication in Bacillus subtilis. Prog. Nucl. Acid Res. Mol. Biol. 59, 35–53. Takahashi, I. (1964). Incorporation of bacteriophage genome by spores of Bacillus subtilis. J. Bacteriol. 87, 1499–1502. Takeuchi, A., Hashiyama, K., Arai, R., and Makino, O. (2005). Isolation of a series of single missense mutants of a dna gene of phage F29, gene 1, utilizing their inhibitory effect on E. coli growth. Genes Genet. Syst. 80, 377–383. Takeuchi, A., Makino, O., and Hirokawa, H. (1998). Gene 1, one of the dna genes of bacteriophage F29, represses other dna genes through binding to mRNA. Biochem. Biophys. Res. Commun. 243, 566–571. Takeuchi, A., Makino, O., Matsumoto, K., and Hirokawa, H. (1995). Inhibition of protein synthesis from DNA replication genes by the product of gene 1, a dna gene of Bacillus subtilis phage F29. J. Gen. Appl. Microbiol. 41, 516. Turner, I.J., Scott, D.J., Allen, S., Roberts, C.J., and Soultanas, P. (2004). The Bacillus subtilis DnaD protein: a putative link between DNA remodelling and initiation of DNA replication. FEBS Lett. 577, 460–464. Velten, M., McGovern, S., Marsin, S., Ehrlich, S.D., Noirot, P., and Polard, P. (2003). A two-protein strategy for the functional loading of a cellular replicative DNA helicase. Mol. Cell 11, 1009–1020. Vlcek, C., and Paces, V. (1986). Nucleotide sequence of the late region of Bacillus phage F29 completes the 19285-bp sequence of F29. Comparison with the homologous sequence of phage PZA. Gene 46, 215–225. Whiteley, H.R., Ramey, W.D., Spiegelman, G.B., and Holder, R.D. (1986). Modulation of in vivo and in vitro transcription of bacteriophage F29 early genes. Virology 155, 392–401. Winston, S., and Sueoka, N. (1980). DNA membrane association is necessary for initiation of chromosomal and plasmid replication in Bacillus subtilis. Proc. Natl. Acad. Sci. USA 77, 2834–2838. Wu, L.J. (2004). Structure and segregation of the bacterial nucleoid. Curr. Opin. Genet. Dev. 14, 126–132. Wu, L.J., and Errington, J. (2003). RacA and the Soj-Spo0J system combine to effect polar chromosome segregation in sporulating Bacillus subtilis. Mol. Microbiol. 49, 1463–1475. Yamaichi, Y., and Niki, H. (2000). Active segregation by the Bacillus subtilis partitioning system in Escherichia coli. Proc. Natl. Acad. Sci. USA 97, 14656–14661. Yehle, C.O., and Doi, R.H. (1967). Differential expression of bacteriophage genomes in vegetative and sporulating cells of Bacillus subtilis. J. Virol. 1, 935–947.

F29 DNA Replication in Vegetative and Sporulating Cells

Yoshikawa, H., Elder, J.H., and Ito, J. (1986). Comparative studies on the small Bacillus bacteriophages. J. Gen. Appl. Microbiol. 33, 39–49. Yoshikawa, H., Garvey, K.J., and Ito, J. (1985). Nucleotide sequence analysis of DNA replication origins of the small bacteriophages: evolutionary relationships. Gene 37, 125–130. Yoshikawa, H., and Ito, J. (1982). Nucleotide sequence of the major early region of bacteriophage F29. Gene 17, 323–335. Zhang, W., Carneiro, M.J.V.M., Turner, I.J., Allen, S., Roberts, C.J., and Soultanas, P. (2005). The Bacillus subtilis DnaD and DnaB proteins exhibit different DNA remodelling activities. J. Mol. Biol. 351, 66–75. Zhao, H., Msadek, T., Zapf, J., Madhusudan, A., Hoch, J.A., and Varughese, K.I. (2002). DNA complexed structure of the key transcription factor initiating development in sporulating bacteria. Structure 10, 1041–1050.

|

305

Release of Progeny Phages from Infected Cells

12

Carlos São-José, João G. Nascimento, Ricardo Parreira, and Mário A. Santos

Abstract Progeny release from phage-infected cells can occur either by lysis of the host or by a singular secretion mechanism, which has been only documented so far for filamentous phages. All known double stranded DNA phages synthesize two lysis effectors, an endolysin and a holin, the first providing a muralytic function and the second a lysis timing device. Endolysins and holins from different phages can be structurally very diverse in spite of their functional similarities. In its export to the cell wall, the endolysin can either be dependent on holin-formed membrane lesions or use the general secretion pathway of the host. In several known cases an antiholin is also produced. This protein can be either soluble or membrane-bound. In T4, the antiholin is crucial in the response to superinfecting phage, in a process known as lysis inhibition (LIN). Phage members of the Microviridae and Leviviridae families are also bacteriolytic but use a single gene lysis strategy to release their progeny. The mechanism employed relies on the production of murein synthesis inhibitors and thus lysis by such phages is akin to lysis mediated by antibiotics which target the cell wall. The Inoviridae, filamentous phages, do not lyse their hosts. They are assembled during export, using transmembrane channels formed by at least one inner membrane phage-encoded protein and an outer membrane secretin. Introduction Basic strategies in phage release The last step of a phage vegetative cycle is the release of the newly assembled virions to the extracellular milieu. Perhaps not obvious at first glance, this final stage is of major importance with respect to phage survival and ecological fitness. An accurate time-defined and efficient release of phage progeny is crucial to maximize both the burst-size and the opportunity to infect new hosts. In double-stranded DNA (dsDNA) phages two complementary functions have independently evolved to achieve both rapid progeny release and optimal lysis time. One such function is provided by an endolysin, a peptidoglycan hydrolase, which is usually a soluble, non-structural component of the virion and which accumulates in the cytoplasm during phage development. The accumulated lytic activity is then suddenly released to the cell wall by the pore-forming ability of small proteins which, in the cases examined, seem

308

|

São-José et al.

to oligomerize into patches in the membrane to form sufficiently large lesions for endolysin delivery to its substrate. Structural alterations introduced in the holin may either prevent lysis or lead to premature or delayed lysis. Ecologically speaking, time-defined lysis is advantageous to the phage and it is a trait under high selective pressure depending on host fitness and abundance (Wang et al., 1996; Wang, 2006). The holin is thus the timing device or the molecular clock of a phage infection in dsDNA phages. As will be discussed further on, some endolysins can do without the export function of holins by being endowed with secretion signals, but even in these cases full activity of the enzymes seems to depend on the carefully programed timing of proton-motive force (pmf ) dissipation brought about by holin-mediated pore formation. Bacteriolytic viruses also comprise two groups of small icosahedral phages, which employ a single lytic function that seems to interfere with murein biosynthesis by affecting distinct steps in the process. Lysis timing in these cases does not appear to be a so-well regulated phenomenon as in dsDNA phages and lysis is more gradual by comparison. Finally, the filamentous phages with single-stranded (ss) DNA genomes are released from their hosts as part of their morphogenetic pathway, by a secretion-related mechanism which maintains the structural integrity of the bacterial cell. In the following sections, these three basic modes of phage progeny release will be described in more detail, with particular emphasis on selected examples from dsDNA phages. For further discussion on these topics the reader is referred to several recent reviews (Marvin, 1998; Bernhardt et al., 2002a; Young, 2002; São-José et al., 2003; Young and Wang, 2006). The holin/endolysin strategy of phages with dsDNA genomes: diversity of lysis players In most dsDNA phages, the holin and endolysin genes cluster together, in this order, as part of the late transcribed genes, although deviations from this spatial and temporal organization are found both in Gram-negative and Gram-positive systems. In addition to these basic lytic functions, other phage-encoded proteins may work as auxiliary lysis factors (see below and section on selected examples,). In general, phage-encoded peptidoglycan hydrolases (endolysins) are easily identified by simple analysis of phage genomic sequences, due to the relatively high amino acid conservation observed within their catalytic domains. Endolysins can present five major peptidoglycan degrading activities: N-acetyl-B-D-glucosaminidases, lytic transglycosylases and N-acetyl-B-D-muramidases (also known as “lysozymes”) attack the sugar moiety of the bacterial cell wall, N-acetylmuramoyl-L-alanine amidases degrade the amide bond connecting the glycan strand to the peptide stems, while endopeptidases act on the peptide moiety (see Figure 12.1 for a schematic representation of their sites of action in the peptidoglycan). Most endolysins display a single enzymatic activity, although certain lytic enzymes derived from phage infecting Gram-positive hosts may bear two distinct catalytic domains specifying different peptidoglycan hydrolase activities (Navarre et al., 1999; São-José et al., 2003; Pritchard et al., 2004). With few predicted exceptions, the endolysins encoded by phages of Gram-positive bacteria present a conserved modular architecture, in which the

Phage Release from Infected Cells

2 1

GlcNac

MurNac

GlcNac

3 4 GlcNac

5 MurNac

L-alanine D-alanine

D-glutamate

Meso -diaminopimelate , L-lysine or other

Peptide -bridges (when present )

Figure 12.1 Schematic structure of the bacterial cell wall peptidoglycan and points of attack of phage-encoded peptidoglycan hydrolases: 1 (N-acetyl-B-D-glucosaminidases), 2 (N-acetylB-D-muramidases or lytic transglycosylases), 3 (N-acetylmuramoyl-L-alanine amidases), 4 (L-alanoyl-D-glutamate peptidase), 5 (D-alanoyl-glycyl endopeptidase). GlcNAc indicates Nacetyl-glucosamine and MurNAc N-acetyl muramic acid.

N-terminus harbors the enzyme catalytic domain(s) and the C-teminus cell wall binding motifs (São-José et al., 2003; López and García, 2004; Loessner, 2005). These cell wallbinding domains (CWBD) may restrain the lytic action of the enzyme to a particular cell wall type (Loessner, 2005; Fischetti, 2005). The prototype endolysin exhibiting this modular design is the Cpl-1 lysozyme of the Streptococcus pneumoniae phage Cp-1, the only endolysin from a Gram-positive system whose three-dimensional structure has so far been determined (Hermoso et al., 2003). The polypeptide chain consists of an N-terminal N-acetylmuramidase domain linked to C- terminal choline-binding motifs. The work on pneumococcal phages also led to the notion of domain swapping, involving catalytic and binding domains, as an evolutionary process that generated enzymes with different specificities (Sheehan et al., 1996, 1997; López and García, 2004). Recent findings have also shown that some endolysins may be endowed with an N-terminal secretion signal, which targets them to the extra-cytoplasmic media through the host’s general secretion pathway (see below for selected examples). Based on the properties just described, the different types of phage or prophage-encoded endolysins can be represented by the eight prototypes shown in Figure 12.2. It should be noted that the type of enzymatic activity of most endolysins has been deduced from bioinformatics analysis, and has not been experimentally checked. This prevents, for instance, to distinguish between true lysozymes and lytic transglycosylases, which frequently share primary sequence identity and folding features (García et al., 1990; Blackburn and Clarke, 2001). The various endolysins displaying the same type of peptidoglycan hydrolase activity have been organized in different families in public databases, according to the different degrees of sequence relatedness. The same applies to the CWBDs (São-José et al., 2003).

|

309

310

|

São-José et al.

Enzyme architecture

Examples phage/prophage (host)

Acm/Ltg, Ami, Acg, Chi

λ (E. coli)

TMD SAR

Acm/Ltg

P1 (E. coli)

SP

Acm/Ltg

H-19B (E. coli)

SP

Acm/Ltg

CWBD

fOg44 (O. oeni)

Acm/Ltg, Ami, A-Epep

CWBD

φCp-1 (St. pneumoniae)

Acg, Ami, A-Gpep

Acg, Acm, Ami, A-Gpep, A-Epep

Ami, A-Gpep

Acg, Acm

Ami

CWBD

CWBD

φ11 (Sta. aureus)

B30 (St. agalactiae)

Acg

λsa2 (St. agalactiae)

Figure 12.2 General domain architecture of phage/prophage peptidoglycan hydrolases. The rectangles in white represent the endolysins’ catalytic region while CWBD indicates the cell wall binding domains. The possible catalytic domains displayed by each endolysin prototype are indicated: Acm (N-acetyl-B-D-muramidase), Ltg (Lytic transglycosylase), Ami (N-acetylmuramoyl-L-alanine amidase), Acg (N-acetyl-B-D-glucosaminidase), Chi (Chitinase), A-Epep (L-alanoyl-D-glutamate peptidase), A-Gpep (D-alanoyl-glycyl-endopeptidase). When experimentally confirmed, the enzymatic activity of the selected examples is highlighted in bold. The presence of signal peptides, transmembrane domains and signal-arrest-release sequences, either predicted using bioinformatics tools, and/or experimentally demonstrated, is indicated by SP, TMD and SAR, respectively.

Unique types of cell wall hydrolases seem to be produced by mycobacteriophages. While phages Ms6 and TM4 encode enzymes with a predicted amidase domain others, such as D29 or Bxb1, may employ chitinase-like hydrolases to bring about host cell lysis. These peculiar endolysins may reflect the unique composition and structural features of the mycobacterial cell wall. In fact, mycobacteriophages seem to encode a second lytic enzyme, such as LysB in Ms6 (Garcia et al., 2002), where in silico analysis indicated the presence of a cutinase-like (serine esterase) domain. Mycobacteriophages may therefore encode specific

Phage Release from Infected Cells

lytic functions to allow degradation of the cutin-like layer that covers the peptidoglycan/ arabinogalactan-rich mycobacterial cell wall. Unlike endolysins, holins are much more diverse and frequently unique with respect to their primary sequence (Wang et al., 2000). In many phage genomic studies, the simple identification of an orf, upstream of the endolysin-encoding gene, coding for a small (< 150 aa) putative product with at least one clear transmembrane domain (TMD) leads to its assignment as a putative holin gene. However, these predictions may be less obvious if deviations in gene arrangement are observed or if more than a single holin-like gene is found in the vicinity of the endolysin gene. The identification of the holin gene in prolatehead phages of L. lactis has been the matter of some controversy. Initial reports pointed to either a typical hol-lys organization, as reported for phage c2 ( Jarvis et al., 1995) or for an unusual localization of the holin coding sequence within the endolysin gene, as suggested for phage FvML3 (Shearman et al., 1994). However, whatever the role these small proteins may play in host lysis, they do not fulfill an important requisite of phage holins, i.e. the presence of at least one distinctive TMD. A third possibility emerged from the genome analysis of the related bacteriophage bIL67. A potential holin gene (orf37) was proposed to locate adjacent to the cos-site, being the last of the phage late-expressed genes (Schouler et al., 1994). In contrast to the first reports, the predicted product of orf37 displayed two putative membrane spanning A-helical domains, a short hydrophilic N-terminus and a highly charged C-terminus, features commonly found in phage holins. Most significantly, it contained a C-terminal region (> 60 amino acids) with high similarity to the characterized holin of B. subtilis phage F29 (Tedin et al., 1995). Lubbers et al. (1995) reported the presence of an analogous gene, in the same relative genome position, in a study where the complete genome sequence analysis of phage c2 was undertaken. For all these reasons, this gene adjacent to the phage cohesive ends seems to be the best candidate for the holin gene for prolate headed lactococcal phages. Holins have been classified in three different classes (I to III) according to the predicted or experimentally determined number of TMDs, three, two or one, respectively (Wang et al., 2000). The L S (class I) and the T4 T (class III) remain the sole cases where holin membrane topology has been experimentally addressed, showing an Nout-Cin and Nin-Cout configuration, respectively (Figure 12.3A). An Nin-Cin topology for the S21 holin of lambdoid phage 21, the prototype of class II holins has been suggested (Blasi and Young, 1996; Barenboim et al., 1999). However, some class II members may present alternative membrane topologies as suggested for the holin of the Clostridium perfringens phage F3626 holin, whose topology prediction suggests that both N- and C-termini face the outside of the membrane (Zimmer et al., 2002). Also, one can not exclude the possibility of the holin “functional unit” being built from a complex of different polypeptides, rather than from a single protein. This could be the case for the B. subtilis defective prophage PBSX (see below) and for other phages exhibiting more than one holin-like element. In fact, such unusual display of lysis genes was found in many phages of Streptococcus thermophilus. Sequence and Southern blot analysis revealed two holin-like sequences, with class I and class II features, respectively, encoded by genes located immediately upstream of the endolysin gene (Bruttin et al., 1997; Sheehan et al., 1999). Expression in E. coli of each of the

|

311

312

|

São-José et al.

A

Δpmf

S107

S105

C

N

Out Cytoplasmic membrane

N

++

N C

C

Type I (e.g. λ)

In C

Type II (e.g. P1)

N

Type III (e.g. T4)

B I

II

III

IV

+ + + +

+ + + + + + + + +

Out

- - - -

- - - + - - - + - - -

In

Figure 12.3 Topology of known holins and model for phage L S107/S105-regulated lysis mechanism. Holins are depicted in dark gray, the antiholin in light gray and the non-secreted endolysins are represented as gray ellipses. The rectangles represent A-helical transmembrane domains (TMD). A. The cytoplasmic membrane topology of type I (3 TMD), type II (2TMD) and type III (1TMD) holins is shown. For simplicity, only the two extra positive charges at the N-terminus of S107 antiholin are indicated. The antiholin S107 topology in an energized membrane is shown, where the first TMD is not translocated across the membrane, but whether it remains as a stabilized A-helix tethered to the membrane or not is unknown. A membrane proton motive force (pmf) depolarization event triggers the translocation of the first S107 TMD and a S105-like topology is achieved. B. S107 antiholin and S105 holin are located in the membrane at 1:2 ratio allowing homodimers and heterodimers to be formed (I). The oligomerization of S105 is impaired by interference of S107 but not totally prevented, and as one or few S105-enriched patches are formed, partial membrane depolarization events (thin straight arrows pointing inwards) occur (II). This depolarization induces the S107 peptides to assume a S105-like topology, resulting in an abrupt increase in the pool of functional holin which in turn causes a fast generalized membrane depolarization event (thick straight arrow; III). Finally, one or few massive holin-mediated lesions are formed capable of fast flowthrough of the cytosol-accumulated endolysin into the cell-wall moiety (IV).

putative holins of phage &O1205 from S. thermophilus was shown to cause cell death and leakage of an intracellular enzyme (isocitrate dehydrogenase), without apparent lysis of the cells (Sheehan et al., 1999). Although holins appear as the key factors determining lysis timing, fine tuning of lysis regulation may involve phage proteins with a holin antagonistic function. These have been generally designated as antiholins. In fact, Young and Blasi (1995) argued that all phages

Phage Release from Infected Cells

employing a holin-endolysin system would also need an antiholin function as a means to ensure the post-translational control of holin activity. As will be detailed in the examples below, the holin gene itself, by displaying two alternative start codons can encode two holinlike polypeptides, one corresponding to the holin effector and the other to the antiholin. In other cases these functions seem to be encoded by separate genes. In addition, some suggested holin antagonists are predicted to be cytoplasmic products, a fact that led to the awareness that proteins exhibiting antiholin function are potentially even more diverse than holins themselves (São-José et al., 2003; see also below). Selected examples of lysis mechanisms The best studied lysis system is the one of E. coli temperate phage L and, in fact, the absence of molecular studies in the past led to an archetypical or paradigmatic view of dsDNA phage release as variations on a L theme. Today however, there is an increasing awareness of diversity of phage-encoded lysis mechanisms, with at least subtle deviations from the paradigm being found in each case where a particular lysis system has been studied in some detail. Despite this, L-like lysis modules and analogous strategies are abundant and mechanistically conserved in dsDNA phages, an indication that this mode of phage release has strong ecological advantages. Some selected examples of lysis mechanisms from Gram-negative and Gram-positive phage/host systems are provided below, illustrating the current knowledge on the diversity of lysis mediated by dsDNA phages. The phage L paradigm E. coli phage L lysis module is composed of five genes translated as a polycistronic late mRNA, which encodes the S107 antiholin, the S105 holin, the R endolysin and lysis adjuvants Rz and Rz1 (Young, 2002). Another putative protein, ORF64 is encoded in the 5a of the mRNA but no role in lysis has been assigned to it. The Rz1 gene is nested in the Rz gene but in a different frame. So far nothing is known about the Rz function, whereas Rz1 is known to be a lipoprotein. They are likely involved in proteolytic cleavage of external membrane links to the cell wall thus facilitating outer membrane disruption. Both proteins are necessary to lysis only in conditions that stabilize the outer membrane, mainly the presence of millimolar concentrations of divalent cations, particularly Mg2+ (Zhang et al., 1999). The only essential genes for phage L-mediated lysis are the S105 holin and the R endolysin but the presence of S107 delays lysis onset, allowing for a larger burst-size. The R endolysin is a 158 aa murein lytic transglycosylase that is accumulated in the cytosol until a precisely timed sudden release to the periplasm, through the membrane-disrupting action of the S holin (Young, 2002). The holin is known to accumulate in the membrane in a non-lethal manner until the moment of lysis triggering, i.e. the infected host-cell remains physiologically undisturbed until a few seconds before the onset of lysis (Grundling et al. 2001). The S107 antiholin and the S105 holin are encoded in frame in the same S gene, and are present in an approximate 1:2 ratio, in a total of 1000–3000 S molecules per cell by the

|

313

314

|

São-José et al.

time of lysis onset (Wang et al., 2000). The differential expression of S107 and S105 is due to a Structure Directed Initiation loop (sdi) overlapping the Shine-Dalgarno sequence in the S mRNA which hinders initiation at the S107 start codon (Blasi and Young, 1996). The S105 holin and S107 antiholin share the same 105 aa sequence but S107 has an extra Met and Lys residues in the N-terminus. These extra residues in S107 confer two extra positive charges comparing to S105, one charge by the deformylation of Met1 and another by the Lys2 residue. These extra charges are known to hinder the translocation of the first TMD in an energized-membrane resulting in the altered topology of S107 compared to S105 (Figure 12.3A). The S105 holin has 3 TMD topology whereas in the antiholin the first hydrophobic segment is unlikely to span the membrane (Grundling et al., 2000a). It is this difference in topology that confers the phenotype of antiholin and holing, respectively, on such otherwise similar peptides. Quite remarkably, the dissipation of membrane pmf triggers the translocation of the first TMD of S107 which then becomes a topological homolog of S105 with similar hole/lesion-forming properties (Young, 2002 and references therein). Recent work provided evidence that S105 “holes” are in fact large caliber lesions several times bigger than needed for the extrusion of cytosol-accumulated R endolysin (Wang et al., 2003). This feature of S is in agreement with earlier observations of progressive oligomerization of S dimers in the membrane (Grundling et al., 2000b; 2000c). The discovery of membrane permeabilization to large molecules is also more concurrent with the perspective of one or few large patches of S oligomers capable of forming large lesions/rafts than with the formation of several discrete widespread holes (Wang et al., 2003). Heterodimers of S107 and S105 have been shown to form amongst the population of S105 homodimers (Grundling et al., 2000c), so the current view is that homodimer oligomerization is partially impaired and therefore the amount of holin functional oligomers is reduced in the presence of antiholin (see Figure 12.3B). Interestingly, the cytoplasmic chaperonin GroEL is capable of solubilization and delivery of megadalton complexes of S proteins. Quite remarkably, GroEL saturation is achieved with only 6 molecules of S107 approximately, whereas 350 molecules of S105 can be accommodated. This is yet another evidence of negative interference mediated by the antiholin on higher molecular weight complexes of the holin (Deaton et al., 2004). Listeria monocytogenes phage A118 and a subtle deviation from the paradigm: an antiholin embedded in frame, within the holin gene A118 is a temperate phage isolated from Listeria monocytogenes carrying a circularly permuted, terminally redundant, linear double-stranded DNA genome of approximately 41 kb (Loessner et al., 2000). At the end of its replication cycle, A118 causes host cell lysis through the combined action of a holin (Hol118) and an endolysin (Ply118), the latter displaying L-alanoyl-D-glutamate peptidase activity (Loessner et al., 1995). This highly active enzyme exhibits very stringent substrate specificity (Listeria serovars 1/2, 3 and 7), a marked preference for binding to a putative carbohydrate cell wall component located (or exclusively accessible) at the cell septal and pole regions (Loessner et al., 1995, 2002), and a modular organization (Loessner et al., 1995). Accordingly, the N-terminal 140 residues harbor the catalytic site while the C-terminus, although lacking any of the known cell surface-anchoring motifs, determines its cell wall-binding specificity (Loessner et al., 1995).

Phage Release from Infected Cells

Despite the apparent similarities (three transmembrane domains and a Nout–Cin topology) between Hol118 and LS (Loessner et al., 1995, 2002), complementation analyses using a L$Sthf background (Vukov et al., 2000), showed that Hol118 is a very poor substitute for S (Vukov et al., 2003). Moreover, the dual start motif in hol118, which is also similar to the one found in S, did not seem to be involved in the regulation of the phage holin (Vukov et al., 2003). However, DNA sequence, toeprinting and genetic analyses revealed the presence of a third, in-frame, internal translation start in hol118 (Vukov et al., 2003). Initiation at this ATG, encoding Met14, and located within the first TMD of the holin coding sequence, results in the production of an intragenic inhibitor of Hol118, designated Hol118(83), and explains the surprisingly poor function of the native hol118 allele in a L background. Apart from functioning simply as the start codon for the A118 antiholin, it seems to play an uncharacterized role in determining the exact moment for lysis induction by Hol118. In this regard, suppression of Hol118(83) synthesis by mutation of its translation start accelerated the onset of lysis, but the dominant effect on its timing was shown to depend on the amino acid residue substituting Met14 (Vukov et al., 2003). In contrast to LS, $9 did not play a membrane permeabilization inhibitory effect over Hol118, either in terms of pore formation or release of LR. In fact, due to the negative charge of Hol118 N-terminus (in contrast to the positively charged N-terminus of the L holin), the energized membrane could actually help it to acquire its Nout–Cin topology. Therefore, Hol118 seems to be clearly different from S in that the lysis effector inserts correctly, even in an energized membrane, but its assembly into “lesion forming” oligomers is compromised by Hol118(83). This has no pore-forming activity by itself, possibly due to the absence of the first TMD domain, and acts as a potent antiholin, super-imposing its effect on the intrinsic lysis clock of Hol118 (Vukov et al., 2003). Lysis by E. coli phage P2: a mysterious lysis modulator In P2, one of the best studied temperate E. coli phages, four late expressed genes, located within a cluster of genes encoding tail morphogenesis functions, are involved in the phage mediated host lysis (Ziermann et al., 1994). Two of them, designated Y and K, encode a class I holin and a putative transglycosylase, respectively, both of which are essential for cell lysis. This frequently observed hol-lys arrangement is, nevertheless, found immediately upstream a lysAB pair. Although non-essential, LysA and LysB affect the correct timing of lysis as lysA amber mutants cause early lysis, whereas analogous lysB mutants slightly delay it (Ziermann et al., 1994). The phenotype of lysA mutants and the hydrophobic nature and secondary structure of the putative LysA protein suggest that it may be a gpY antagonist (antiholin). However, despite the presence of an endolysin/holin/antiholin triad, a further deviation from other regulatory mechanisms implicating these functions results from the use of LysB as a lysis modulator. Regardless of the presence of a small number of homologs in the databases (LysB-like proteins in coliphage WPhi (NP_878209) and Yersinia pestis phage L-413C (NP_839860), gp11 of the phage PSP3 of Salmonella (NP_958066) or the enterophage 186 Orf27 (AAC34256)) its function as a lysis-control gene has not yet been elucidated. In all these phage genomes, the lysis operon is similarly organized, except for phage 186, where a lysA homolog is missing.

|

315

316

|

São-José et al.

Bacillus subtilis prophage PBSX: a two-component holin? The late operon composed by the xepA, xhlA, xhlB and xlyA genes, organized and transcribed in this order, was identified as the lysis module of the Bacillus subtilis defective prophage PBSX (Longchamp et al., 1994). The first orf in the operon encodes a protein previously shown to be exported (xep, for PBSX exported protein) during the phage lytic cycle (Mauël and Karamata, 1984). The putative product of xhlA is a small protein with a hydrophilic N-terminus and a putative transmembrane helix located at the C-terminus, suggesting that XhlA may be membrane-associated. It was anticipated that xhlB and xlyA would encode the typical holin-endolysin pair, in this case a predicted class II holin (XhlB) and an amidase (XlyA), respectively. Krogh et al. (1998) assessed the contribution of each of these proteins to host cell lysis by expressing the four genes in different combinations, under the control of their natural promoter, in B. subtilis. From the expression analysis the authors concluded that XepA does not seem to participate in lysis and that co-expression of XhlB (putative holin) and XlyA (endolysin) does not effect cell lysis unless they are co-expressed with XhlA. Apparently, neither XhlA nor XhlB could trigger effective lysis in pairwise combinations with the endolysin. Interestingly, XhlA was able to cause host cell lysis when expressed alone, although in a non-saltatory way, a feature that could not be attributed to XhlB. Saltatory lysis could only be achieved in the presence of the endolysin. Based on these results, it was proposed that XhlA and XhlB would associate in the membrane to form the functional holin that allows endolysin release (Krogh et al., 1998). In addition, the authors reported that co-expression of XhlA and XhlB also resulted in cell lysis, although with a delay of 30 min. A possible explanation for these data was that a second endolysin is produced upon induction of PBSX that would also use the XhlA/XhlB channel to reach the peptidoglycan (Krogh et al., 1998). Such endolysin (XlyB) was in fact reported by Dasilva et al. (1997). The Tectiviridae phage PRD1: a structural protein is the endolysin Bacteriophage PRD1 infects a broad range of Gram-negative hosts (such as Escherichia coli, Salmonella enterica, and Pseudomonas aeruginosa) as long as they carry an IncP-type multidrug-resistance plasmid encoding the phage receptor (Olsen et al., 1974; Grahan et al., 1997). This virus, which is included in the Tectiviridae family, is characterized by the presence of an internal membrane component lining the inside of an icosahedral capsid. In turn this encloses a double-stranded linear DNA genome with covalently linked proteins at both 5a ends, which are used as primers during phage DNA replication (Caldentey et al., 1993). Genetic analysis of nonsense phage mutants has clearly shown that two viral-encoded functions are involved in host cell lysis (Mindich, et al., 1982; Rydman and Bamford, 2003). Protein P15 is a peptidoglycan hydrolase with B-1,4-N-acetylmuramidase activity, whereas P35 is a class I lambdoid-like holin (Caldentey et al., 1994; Rydman and Bamford, 2003). Unlike most phages with holin/endolysin genes, which tend to locate close to one another in the phage DNA, the PRD1 lysis genes (XV and XXXV) are placed at opposite ends of the linear 15 kbp genome (Rydman and Bamford, 2003). Their expression is also temporally separated, as XV is expressed early in infection, whereas XXXV is expressed late in the phage cycle, along with the structural proteins (Grahn et al., 1994). In addition, and in

Phage Release from Infected Cells

contrast to other phage endolysins, P15 is a structural component of the viral particle. It is associated with the phage internal membrane component, where its incorporation in the virions has been shown to be modulated by the phage membrane-proteins P20 and P22 (Rydman and Bamford, 2002). Despite the association of P15 to the degradation of the host peptydoglycan, and the consequent release of intracellular phage particles (Mindich et al., 1982), PRD1 does encode a second protein with muralytic activity known as P7 (Bamford et al., 1991). Although the P7– mutants showed delayed and asynchronous DNA replication, this virion-carried putative transglycosylase is not essential for infectivity, and seems to play only an accessory role in cell wall degradation during phage DNA entry (Mindich et al., 1982; Rydman and Bamford, 2000). Expression of the PRD1 holin and endolysin, either during a phage infection or from plasmids, has shown that they are both essential for phage-mediated host cell lysis (Mindich, et al., 1982; Rydman and Bamford, 2003; Žiedaite et al., 2005). However, at about 35 min post-infection, PRD1 infected cells become susceptible to premature lysis induced by energy poisons, such as arsenate or cyanide, and start leaking K+ (Žiedaite et al., 2005). This phenomenon is not immediately followed by the disruption of the cell membrane-potential, or efflux of ATP. Nonetheless, it correlates with the depletion of the intracellular ATP pool, and underscores the accumulation of holin in the cytoplasmic membrane, which seems to be mediated by GroEL/GroES (Hänninen et al., 1997; Žiedaite et al., 2005). In spite of the “potential for lysis,” this state is repressed for another 20 min by a $9-dependent control mechanism, which supposedly prevents the holin from associating into complexes that allow the endolysin to cross the cell membrane and reach the cell wall (Žiedaite et al., 2005). Phage T4 and the LIN phenomenon The most remarkable feature of the T4 lysis system is its ability to delay lysis, upon reinfection of infected cells by homologous phage. This long-known phenomenon designated LIN (for lysis inhibition; Doermann, 1948) has been experimentally addressed in recent years (Paddison et al., 1998; Ramanculov and Young, 2001a; Ramanculov and Young, 2001b; Tran et al., 2005). As in “standard” cases, the lysis strategy followed by T4 obeys the holin/endolysin premise common to dsDNA phages, i.e. an endolysin (the E lysozyme; Streisinger et al., 1961) accumulates in the cytoplasm during morphogenesis and is targeted to the periplasm through holin-made conduits at a time which is T-allele specific (T being the T4 holin; Lu and Henning, 1992; Dressman and Drake, 1999). Also as in the lambda lysis system, the triggering of lysis by energy poisons is observed in T4, which together with recent results on t expression in a lambda context suggest that the mode of hole formation may not differ markedly from the model proposed for S functioning (Ramanculov and Young, 2001a, 2001c; Wang et al., 2003). Still, T presents unique structural features, being a bitopic protein with a large C-terminal periplasmic domain (TCTD) which contributes substantially to the comparatively large size of this holin with respect to others (Wang et al., 2000; Ramanculov and Young, 2001c; Tran et al., 2005). Large deletions in this domain do not completely abolish holin function (Ramanculov and Young, 2001c), suggesting that

|

317

318

|

São-José et al.

interactions between the transmembrane domains (TMDs) of T molecules are crucial events in oligomerization and pore formation, similar to the S situation (see above). Nevertheless, studies on the LIN phenotype additionally indicate that productive interactions between the periplasmic domains may be important for T-mediated triggering of pore formation. In fact, the T4 rI gene product appears to act as an antiholin in this system, by interacting with the TCTD (Ramanculov and Young, 2001b; Tran et al., 2005). This contrasts with the postulated role of other antiholins which are similar to holins in primary sequence and which likely prevent the conversion of holin oligomers into holes by interacting with holins through their TMDs (although the final outcome, i.e. delaying lysis is the same). As an antiholin, RI presents other peculiar features. It is predicted to be secreted to the periplasm (Paddison et al., 1998), although the nature of its N-terminal hydrophobic sequence as a cleavable signal peptide lacks experimental support (Tran et al., 2005). In any case, there is now clear evidence for an interaction between RI and the TCTD in the periplasm and it is proposed that such interactions prevent T from mediating pore formation, resulting in the LIN phenotype (Tran et al., 2005). Why does lysis inhibition only occur upon re-infection of an infected cell? The answer to this question seems to be connected to the very unstable nature of RI (Ramanculov and Young, 2001b; Tran et al., 2005), although why an invading phage will stabilize RI against proteolysis remains to be elucidated. Interestingly, T4wt but not T4rI exhibits delayed lysis in slow-growing cells cultured in chemostats (Los et al., 2003). This may reflect a stabilization of RI under such cellular conditions, suggesting that LIN can be both a response to host availability and host metabolism. Sec-dependent endolysins: novel types of lysis strategies The secreted endolysin of Oenococcus oeni phage fOg44 One of the fundamental concepts in lysis regulation that seemed to be well established until recently was that phage endolysins accumulated in the host cell cytoplasm until the formation of holin membrane lesions (Young et al., 2000; Wang et al., 2000). In this light, it was not easily conceivable that phage endolysins could be endowed with secretion signals. The first convincing data showing that phage endolysins can reach their substrate in a holin-independent way came from the studies on the Oenococcus oeni temperate phage fOg44. It could be established that the phage endolysin (Lys44, a predicted acetylmuramidase) was synthesized with a bona fide signal peptide (SP) that was recognized by the bacterial general secretion pathway (GSP). Expression of lys44 both in E. coli (São-José et al., 2000) and in Lactococcus lactis (São-José et al., 2004) led to the synthesis of a 46 kDa precursor and a 43 kDa mature form. The cleavage event occurred in E. coli between residues 26 and 27 of the primary product and required the activity of both the translocase-associated ATPase SecA and the LepB signal peptidase, two essential components of the GSP. Interestingly, only the Lys44 mature form displayed lytic activity against oenococcal cells, showing that the SP functions as a cis-inhibitory element that must be cleaved to generate an active enzyme (São-José et al., 2000). Study of Lys44 production during fOg44 infection of O. oeni showed that the mature product was already detectable half-way through the latent period, accumulating progressively until lysis (São-José et al., 2000). This indicated that the active, SP-processed form of the endolysin was targeted to the cell

Phage Release from Infected Cells

wall from the moment of its synthesis, raising the puzzling questions of how premature lysis is prevented and what role the holin plays is in this system (see below). This first report on Lys44 (São-José et al., 2000) also highlighted that endolysin secretion did not appear to be an exception to the rule. The homology observed between the Lys44 SP and the N-terminal region of endolysins from other phages of lactic acid bacteria led to the prediction of several SP-endowed enzymes, all belonging to the same family of acetylmuramidases. Experimental evidence for other secreted/processed endolysins has been obtained so far for the lytic enzymes of phages FAM2 (São-José et al., 2000; São-José, 2002) and Fg1e (Kakikawa et al., 2002) from Lactococcus lactis and Lactobacillus plantarum, respectively. Other putative SP-endolysins are those encoded by phages F10MC, fOg30 and fOgPSU-1 from O. oeni, Tuc2009, TP901-1, TPW22, FLC3 and ul36 from L. lactis, FJL-1 from Lb. plantarum, Lc-Nu from Lb. rhamnosus and FAT3 from Lb. casei (São-José et al., 2003, 2004). In addition, in a recent review São-José et al. (2003) highlighted that several lysozymes from phages infecting Gram-negative bacteria could be endowed with N-terminal secretion signals. Studies in the Young laboratory confirmed this prediction for the Lyz endolysin of the enterobacteriophage P1 (see below), where the molecular mechanism of this novel type of lysis strategy was further dissected. As mentioned above, the fact that endolysins can be targeted to the cell wall, through the bacterial GSP machinery, as they are being synthesized during phage lytic growth, raises the question of how lysis timing is regulated. An intriguing observation is that all phages which synthesize secreted endolysins or are proposed to do so, also appear to encode a holin-like protein. If endolysins can use host endogenous pathways to reach their substrate, a holin function would seem expendable. In the original phage where secreted endolysins have been first discovered (fOg44), a holin function was experimentally demonstrated for the product of hol117, a gene located immediately downstream lys44. Expression of this gene in E. coli, under the control of the L late promoter PR’ in a pBR322 derivative, resulted in a typical lysis curve about 65 min following induction of a LSam7 prophage (São-José et al., 2004). Additionally, transcription of lys44 and hol117 during fO44 lytic growth appears to result in a stable dicistronic mRNA (Parreira et al., 1999), suggesting a concomitant production of both lysis products during phage infection, as observed in L. However, the putative holin (Orf163) of fOg30, another oenophage producing an SP-endolysin failed to complement the LSam7 lysis defective phenotype, questioning the functionality of Orf163 as a holin (São-José et al., 2004). Two possible explanations for the complementation failure were proposed, in which either Orf163 represents a non-functional holin remnant, or it does not produce membrane lesions large enough to allow access of the L endolysin (R) to the E. coli periplasm (São-José et al., 2004). The model for lysis regulation in phages producing secreted endolysins and functional holins still proposes that the latter protein plays the crucial role of determining lysis timing (São-José et al., 2000, 2003, 2004). In this model it was proposed that, in their natural context, the activity of the targeted endolysins would be inhibited in the cell wall until dissipation of the membrane potential by the cognate holins. Thus, holins would activate (probably by collapsing the membrane potential) the exported endolysins rather than allowing their release. This mechanism also incorporates the membrane potential as a critical parameter in lysis regulation as in the L lysis mechanism, although there is still no clear

|

319

320

|

São-José et al.

explanation as to how the energized membrane would control, directly or indirectly endolysin activity (for further discussion on this issue please refer to the review by São-José et al., 2003). Control of lysis in P1-infected cells: an essential antiholin and a unique endolysin Analyses of amber and deletion mutants of the temperate enterobacteriophage P1 were used to assign functions in cells lysis to three late expressed genes (Walker and Walker, 1980). Mutations in gene 17 (or lyz) prevented host cell lysis, but not the phage lytic development, and its product, known as Lyz, was shown to correspond to the phage endolysin (Schmidt et al., 1996). On the other hand, while a lydAB deletion mutant showed a delayed and gradual lysis profile (although infected cells would promptly lyse in the presence of chloroform), a single lydB amber mutation caused early lysis with no plaque forming units generated (Walker and Walker, 1980). These results suggested that the products of the lydAB locus would correspond to the phage holin (LydA) and antiholin (LydB). Despite the absence of any LydB homologs in the databases, its role as LydA antagonist was further confirmed through plasmid-based expression of the phage lysis genes in E. coli (Schmidt et al., 1996). It is worth emphasizing that in contrast to the lambda and T4 antiholins, LydB is an essential product since no phages are produced in its absence. The P1 lysis system is also unusual in that lyz does not cluster with the overlapping holin and antiholin genes, but rather is located approximately 10 kbp away on the phage genome. Furthermore, and unlike most phages with a holin/endolysin pair, needing both proteins to accomplish host cell lysis, lydA mutants of phage P1 do form plaques, though their lysis phenotype is delayed when compared to that of the wild-type phage (Yarmolinsky and Sternberg, 1988). This apparent holin-independent Lyz-mediated mechanism of E. coli lysis by phage P1 has recently been addressed with surprising results. Lyz, a homolog of the phage T4 lysozyme, is different from most Gram-negative phage endolysins as it does not accumulate in the cytoplasm during phage intracellular multiplication, prior to its release through membrane lesions that result from holin oligomerization. Instead, Lyz is secreted by the Sec translocase using a novel signal-arrest-release (SAR) sequence, located at its N-terminus (Xu et al., 2004). This hydrophobic domain tethers Lyz to the membrane but allows its release to the periplasm. In the process, the enzyme is converted from a membrane-bound protein to a soluble state without proteolysis. This may occur slowly and spontaneously, accounting for the unprompted, and slow, lysis phenotype of lydA P1 mutants. However, triggering of programmed lysis by the phage holin will allow for a very rapid, and quantitatively significant, release of the SAR endolysin from the membrane, where it remains inactive or restrained from access to the peptidoglycan. This may result from a collapse of the membrane potential, as demonstrated by the fact that energy poisons trigger Lyz-mediated lysis or, alternatively, through a more profound disruption of the cell membrane architecture, which would facilitate the exit of the SAR sequence (Xu et al., 2004). Genetic analysis of P1 Lyz has shown that, apart from the catalytic Cys51, the protein has six other cysteine residues, one of which (Cys13) is imbedded in the SAR sequence (Xu et al., 2005). On the other hand, biochemical analysis of ammonium hydroxide cleavage products of cyanylated Lyz has shown that, soon after induction of a P1 lysogen, only

Phage Release from Infected Cells

the membrane-associated C13 is in a sulphydryl form, whereas all the other Cys residues are expected to form disulphide bonds. Still, after cell lysis has ensued, cleavage of Lyz at both C13 and C51 was detected, suggesting the presence of two isomeric forms of the protein, differing in disulphide-bond formation but not in their overall oxidation state (Xu et al., 2005). In fact, in the nascent form of Lyz, an enzyme-inactivating disulphide bond is formed between C44 and C51, which holds the N-terminal domain of the endolysin in a compact conformation, and precludes the opening of its active-site cleft. However, upon release of the C13 from the membrane, this residue provides a thiol group for the reduction of the C44–C51 bond, in turn inducing remodeling of discrete structural elements of the enzyme inactive form, and unlocking of its active-site (Figure 12.4). Therefore, the SAR domain of Lyz has two functions. First, it acts as a signal-arrest sequence and mediates the transport, and association, of the phage endolysin to the membrane. Secondly, once the protein is released, its C13 residue is involved in the isomerization reaction that relieves topological, covalent, and conformational constraints from Lyz. In turn, this directs the protein to assume an active conformation mediated by DsbA (the periplasmic primary oxidant involved in disulfide bond formation), and possibly facilitated by periplasmic foldases and chaperones (Xu et al., 2004; Xu et al., 2005). A single lysis gene: the Microviridae and Leviviridae phage strategy Unlike most of the well-studied phages, which have larger dsDNA genomes, the lysis strategy of icosahedral phages with small single-stranded genomes (ssDNA or ssRNA, under 6 kb) is poorly understood (Bernhardt et al., 2002a). These phages developed a lysis mechanism with a single effector protein unlike the multiple-protein strategies of phages with larger genomes (Young, 1992) probably because of the limited coding capacity of their small-size genomes.

Inactive Lyz

Active Lyz Cys 44

SH-Cys 51

=

S

active site cleft

Cys 44 S=S

out

S

Cys 51

Slow , spontaneous

out

CM

SH-Cys 13

!" CM Fast , holin induced

in

SAR

in

Cys 13

!" collapse

Figure 12.4 Model for the activation of Lyz of phage P1. Before the release of the SAR sequence from the fully energized membrane, the C13 residue is in a sulphydryl state, whereas C44 and C55 are in disulfide linkage. Upon release of the SAR sequence from the membrane, either spontaneously (slow) or by holin-induced collapse of the membrane potential (fast), Lyz undergoes refolding and disulfide bond isomerization. In the process, a new disulfide bond is formed (between C13 and C44), allowing for the formation of a correctly folded active site.

|

321

322

|

São-José et al.

The best studied of this type of phages is the Microviridae phage FX174 (circular, ssDNA), but recent and ongoing research efforts focusing on lysis are also being conducted on the linear ssRNA Leviviridae phages MS2 and QB by Young and co-workers (Bernhardt et al., 2001a, 2001b, 2002a). These three phages infect the Gram-negative bacterium E. coli and lysis is mediated by the proteins E, L and A2, respectively, none of which with cell-wall degrading activities (Bernhardt et al., 2002a). The absence of a phage-encoded muralytic activity is the most striking difference between these viruses and the dsDNA phages. The FX174-mediated lysis controversy Coliphage FX174 has been extensively studied since the early times of molecular biology because of the simplicity of its genome, a promising feature to unravel molecular mechanisms in their simplest form (Young, 1992). Early on, it was shown that E is the only essential gene for lysis induced by FX174 (Hutchinson et al., 1966). There are several models for FX174-mediated lysis: while Lubitz and co-workers defended E-mediated autolysis of E. coli and later a transmembrane model, the results by Young and co-workers support an E-mediated cell division interference model. The concept of E-mediated autolysis is rather controversial since it implies the induction of a multifactorial event, and therefore multiregulated, whose exact molecular mechanisms are still largely uncharacterized at the biochemical level in E. coli (Bernhardt et al., 2002a). The transmembrane model arose based on the observation that ghost cells are produced when E is overexpressed and that E is detected in membranes of E. coli (Witte et al., 1990). Lubitz and co-workers propose that these ghost cells can be useful in the development of novel vaccines (Mayr et al., 2005). However, the fact that these features are observed in E-overexpression conditions may suggest that the “tunnel” is an artifact rather than the actual FX174 release strategy. While the controversy on FX174-mediated lysis regarding the role of host autolysins remains to date (Bernhardt et al., 2002a), the overwhelming amount of genetic and biochemical evidence that support the cell division impairment model, rather than autolysis/ autolysin induction, will be taken in this review by discussing the mode of action of E, L and A2 lysis effectors in light of the former as outlined by Young and colleagues. The cell division interference model Young and co-workers have argued that the kinetic and morphological similarity between penicillin-induced and single-stranded genome phages-mediated lysis implies a common mechanism. Since penicillin is a murein synthesis inhibitor, E, L and A2 are most probably inhibitors but not necessarily with the same target. All the referred phages induce similar morphological changes in E. coli and have identical lysis kinetics and genetic features. This leads one to postulate that the lysis mechanism is analogous and, at least for FX174 and QB, data that substantiate this have been published, rendering the cell division interference model as the currently accepted mode of action by which non-filamentous single-stranded genome phages are released. This model postulates that dividing cells in the presence of phage-encoded murein synthesis inhibitors are unable to assemble new cell wall at the septum where it is needed the most. This void results in a weakened murein structure unable to sustain the osmotic

Phage Release from Infected Cells

pressure of the cytoplasm. A transient bulge is formed at mid-cell position and cell burst eventually occurs releasing progeny phages into the medium for a new infective cycle (Bernhardt et al., 2002a). FX174 E E is a 91aa polypeptide with a putative single TMD in its N-terminus (Figure 12.5) and is encoded within the gene of the morphogenesis scaffolding protein D. All E. coli clones that survive infection by FX174 carry mutations in genes encoding proteins known not to be directly involved in autolysis, which is considered a strong argument against the E-induced autolysis model (Bernhardt et al., 2002a). E. coli slyD mutants (sensitivity to lysis) were the first to be isolated and described (Roof et al., 1994). SlyD is a peptidyl-prolyl cis-trans isomerase (PPiase) or rotamase of the FKBP family and could be involved in the isomerization of one or more of the five prolines in E, probably facilitating its conformational stabilization to the active form and/or facilitating its integration in the cytoplasmic membrane (Roof et al., 1995, Bernhardt et al., 2002b). A second round of genetic screening was followed in order to isolate phage FX174Epos mutants (E plates on slyD) as plaque forming units in a lawn of E. coli slyD. Two single Murein

? Out

L

E

C

In

N

N

MurA

MraY

C

UDP-NAM L-Ala MurF D-Glu m-DAP UDP-NAM UDP-NAM D-Ala L-Ala MurB MurC D-Ala D-Glu MurD m-DAP MurE C

UDP-NAG

N

Lipid I

Cell wall synthetases

Lipid II

UDP-NAM UDP-NAM-NAG L-Ala MurG L-Ala D-Glu D-Glu m-DAP m-DAP D-Ala D-Ala D-Ala D-Ala

A2 Figure 12.5 Cell division interference model for icosahedral single-stranded genome phage release. As the cell divides, the cell wall synthesis pathway is fully functional and UDP-NAG (UDP-Nacetyl-glucosamine) is being catalyzed by MurA and then MurB to originate UDP-NAM (UDP-Nacetyl-muramic acid) which is then modified with a peptide segment of L-alanine (L-ala), D-glutamate (D-glu), meso-Diaminopimelic Acid (m-DAP) and two residues of D-alanine (Dala) sequentially added by MurC, MurD, MurE and MurF, respectively. Cytoplasmic membrane translocase MraY then catalyzes the formation of Lipid I (a bactoprenol lipophylic carrier transporting the murein precursor). Finally, MurG catalyzes the addition of a UDP-NAG residue to Lipid I and Lipid II is formed. Lipid II is flipped across the membrane and an array of cell wall syntethases including several Penicillin Binding Proteins (PBPs) carry on murein synthesis. The membrane topology of MraY is shown. TMDs where known E-resistance mutations are localized are depicted in white (see text). Phage FX174 E putative membrane topology is shown as well as phage MS2 L protein, but while E is an inhibitor of MraY, nothing is known about the target of L. On the other hand, phage QB A2 is a soluble inhibitor of MurA.

|

323

324

|

São-José et al.

missense mutations were identified in E and the most efficient Epos mutant contained them both (Bernhardt et al., 2000). The mutations lead to a replacement of an arginine at position 3 for an histidine (R3H) and a leucine for a phenylalanine at position 19 of the peptide (L19F). Because the presence of SlyD still improves the lysis phenotypes of these mutants, these amino acid substitutions account for partial gain-of-function phenotype, and so protein stability assays were performed. Unexpectedly it was found that Epos is in fact more abundant than E, and both proteins are equally unstable in a SlyD host. Therefore, it was postulated that the Epos mRNA is more stable or more efficiently translated than E’s and so Epos phenotype results simply from increased rate of protein accumulation (Bernhardt et al., 2002b). If lysis can occur in the absence of SlyD that means SlyD is not the target of E. In fact, this target was identified in another genetic screen as being the cytoplasmic membrane MraY translocase (Bernhardt et al., 2000). Eps (Epos sensitivity) mutants were isolated as spontaneous E. coli slyD survivors to overexpressed plasmid-borne Epos which were then assayed for resistance to FX174 (Bernhardt et al., 2000). Eps mutants were found to be allelic to mraY which encodes a 10 TMD cytoplasmic translocase catalyzing the synthesis of Lipid I and is the only membrane protein of the cytosolic pathway of murein precursor synthesis (Figure 12.5). Sequencing of the three E. coli Eps/mraY mutants fully resistant to FX174 revealed that mraY4 and mraY15 have the same F288L substitution located in TMD9 whereas mraY39 is $L172 in TMD5 (Bernhardt et al., 2000). Mutations in the presence of ethane methylsulfonate were also isolated and they were also shown to affect residues located in TMD5 (P170L and G186S) and in TMD9 (V291L) indicating that these two TMDs are the most likely targets of E (Bernhardt et al., 2002a). The spontaneous mraY mutants are dominant over the wild-type as determined by a merodiploidy assay and are not lethal, unlike a mraY null mutant. This means that mutant MraY essential function is not significantly affected, whereas its sensitivity to E is abolished (Bernhardt et al., 2000). Biochemical analyses were also conducted, showing that E affects the incorporation of radiolabeled diaminopimelic acid (DAP) in the cell wall and the in vitro activity of MraY. These effects can be replicated with a specific MraY inhibitor like mureidomycin (Bernhardt et al., 2001a). Although the mechanism by which E impairs MraY function is still to be defined, the fact that E lethality is restricted to its single TMD and MraY defects are located in the TMDs indicates that TMD-TMD interactions are probably involved. MS2 L and QB A2 The ssRNA phages MS2 and QB are far less studied than FX174 and recent data on their lysis mechanism has been essentially only provided by Young and co-workers. Similarly to FX174 E, phage MS2 L is a small 75 aa protein with a putative TMD, but in this case in the C-terminus (Figure 12.5), and is encoded within the coat protein and replicase genes of MS2. The regulation of L expression has been well studied and evidences indicate that interference with the stability of a large translation regulatory hairpin affects lysis timing. The target of L, however, remains unknown. Phage QB A2, unlike E and L, is not encoded within another gene, is not membraneassociated and is in fact a dual function protein. It is simultaneously the QB capsid protein

Phage Release from Infected Cells

responsible for adsorption to the host pilus and the lysis effector (Bernhardt et al., 2001b). Similarly to E, A2 has been subjected to genetical and biochemical characterization and the target has been identified as MurA, a soluble enzyme catalyzing the first step of murein precursor synthesis (Bernhardt et al., 2001b). The E. coli murArat1 (resistance to A-two) mutants are insensitive to QB virions and show a L138Q modification in MurA (Bernhardt et al., 2001b). Phage QB mutants overcoming rat mutations, named QB por (plates on rat), have been isolated and the mutations were mapped to the N-terminal 120 aminoacids of A2 (Young and Wang, 2005). Release without lysis: the filamentous phage strategy The filamentous phages constitute a large family of single-stranded DNA viruses that infect Gram-negative bacteria using pili as receptors. The Ff group of filamentous phage (M13, f1 and fd) which infect “male” E. coli bacteria via the conjugative F pilus are the best characterized members of this family (Model and Russel, 1988). In contrast to other bacterial viruses, filamentous phage particles do not accumulate in the cytoplasm and cell lysis does not occur during phage progeny release. Instead, filamentous phages are produced by a concerted mechanism of assembly and secretion across both membranes of the Gramnegative cell (for reviews see Model and Russel, 1988; Russel, 1995; Russel et al., 1997 and Marvin, 1998). It is of note that recently, a filamentous phage infecting the Gram-positive species Propionibacterium freudenreichii has been described, but the molecular details of its release remain to be investigated (Chopin et al., 2002). The virion of f1 (fd, or M13) is 880 nm long and 6–7 nm wide (Model and Russel, 1988). Its circular single-stranded DNA genome is wrapped in a tube composed of about 2700 copies of the major coat protein pVIII. Each end of the tube bears a set of different pairs of proteins each present in four to five copies per particle (Goldsmith and Konigsberg, 1977; Grant et al., 1980, Lin et al., 1980). pVII and pIX are at the end at which assembly initiates and pIII and pVI are at the end at which assembly terminates (Lopez and Webster, 1983). All five structural proteins insert in the inner membrane prior to assembly and phage secretion (Endemann and Model, 1995). Before assembly, the phage genome is sequestered in the cytoplasm as a rod-shaped complex with the phage-encoded ssDNA binding protein pV (Salstrom and Pratt, 1971; Gray, 1989). After assembly initiates the particle is elongated by removal of pV from the DNA and simultaneous replacement by the major coat protein as the phage is extruded from the cell. Finally, the proteins pIII and pVI are added to the virion and phage is released from the membrane and extruded into the medium (Lopez and Webster, 1983). Assembly sites and initiation of secretion Assembly and export of filamentous phage requires four non-capsid proteins: pIV, an outer-membrane protein which oligomerizes to form a channel, pI and pXI, both localized to the inner membrane, and a cytoplasmic host factor, thioredoxin (Russel, 1995; Feng et al., 1997). The phage assembly proteins are predicted to interact to form the phage assembly site, a region where the cytoplasmic and outer membranes appear to be in close contact (Lopez and Webster, 1985; Feng et al., 1999). pIV forms the outer membrane component of the assembly site. It can be isolated from the membrane as ring-like structures composed

|

325

326

|

São-José et al.

of 14 pIV molecules (Opalka et al., 2003) and it functions as an exit route for the emerging phage particle. This was elegantly demonstrated for phage f1, using a mutant pIV that is competent for phage export but whose channel opens in the absence of phage extrusion (Marciano et al., 2001). In E. coli lacking its native maltooligosaccharide transporter LamB, this pIV variant allowed oligosaccharide transport across the outer membrane. This entry was found to be decreased by phage production and still further decreased by mutant phages that cannot be released from the cell surface (see below). These experiments thus established that exiting phages occupy the lumen of pIV channels. The structure of pIV has been recently examined by cryo-electron microscopy associated with image analysis (Opalka et al., 2003). The structure revealed a barrel-like complex, 13.5 nm in diameter and 24 nm in length consisting of a dimer of unit multimers. A large pore, ranging in inner diameter from 6.0 nm to 8.8 nm is present in the middle of the multimer. This structural study further showed that a central domain gates this pore, explaining why expression of wild-type pIV does not increase the permeability of the outer membrane. Moreover, the pore diameter at the N-region was found to be smaller than the phage particle, implying that large conformational changes must occur to facilitate phage transport. pIV belongs to the secretin family of proteins which include outer membrane channels for type II and type III protein secretion (Russel et al., 1997). Interestingly, pIV from the E. coli phage f1 is more similar to the Klebsiella oxytoca homolog PulD than it is to pIV from the Pseudomonas aeruginosa phage pf3. Furthermore, although the gene order is largely preserved between f1 and pf3 their gene IVs are located in different positions. This suggests that filamentous phages may have obtained their gene IV from an ancestral bacterial host on separate occasions (Russel et al., 1997). Genomic analysis suggests that other filamentous phages lacking the gene for an outer membrane component may also exploit chromosomeencoded secretins for their release (Davis et al., 2000; Davis and Waldor, 2003). This is best illustrated by EpsD, the outer membrane component of the Vibrio cholerae type II protein secretion, which is used both for cholera toxin release and for extrusion of the filamentous phage CTX& (which encodes the toxin). pI and pXI are the inner membrane components of the assembly site. pXI is the result of an in-frame internal translation initiation event in gene I and is thus identical to the Cterminal third of pI in amino acid sequence (Rapoza and Webster, 1995). pI and pXI share a cytoplasmic domain predicted to be an amphipathic helix, a transmembrane domain and a periplasmic domain. However, the insertion of pI in the membrane results in the loss of membrane potential and cell death while insertion of pXI has no such effect (Horabin and Webster, 1988; Guy-Caffey and Webster, 1993; Haigh and Webster, 1999). One explanation is that the nature of the protein domain preceding residue 241 of the pI/pXI transmembrane region affects its structure. Although the role of pXI remains obscure, one of its functions could be to abolish the lethal effect of pI insertion alone by protein-protein interactions, as in the holin antiholin systems (see Section on dsDNA phages). Among other lines of evidence, support for an interaction between pI and pXI comes from the observation of their co-purification and the fact that pXI stabilizes pI against proteolysis (Feng et al., 1999). An assembly site transmembrane channel is most likely formed by interaction in the periplasm between pI/pXI and pIV. Indeed, a temperature-sensitive pIV mutation can be

Phage Release from Infected Cells

suppressed by a single amino acid replacement in the periplasmic portion of pI/pXI. Also, pIV from f1 and Ike are not functionally interchangeable but when both pI and pIV are exchanged some heterologous phage are assembled (Russel, 1993). In vivo, it may be the specific interaction of the periplasmic portion of pI with the corresponding portion of pIV that opens this pore to allow phage transit to the surrounding medium. The transmembrane complex consisting of pIV, pI and pXI appears to be formed even in the absence of substrate (Feng et al., 1999). Importantly, genetic analysis suggest that pI also acts to initiate assembly by recognizing the packaging signal in phage DNA, which protrudes from the pV-ssDNA complex, to bind thioredoxin and to interact with pVIII possibly to promote its incorporation into nascent phage (Russel et al., 1997). A speculative model has been proposed in which thioredoxin reductase, thioredoxin and pI protein interact to drive an engine for filamentous phage assembly (Russel and Model, 1985). In this model, pI was proposed to be the facilitator of assembly, undergoing conformational change at each round of assembly resulting in the oxidation of two thiol groups to a disulfide and being subsequently reduced by thioredoxin to enable it to enter the cycle again. In spite of what the model proposed it was shown subsequently that the redox activity of thioredoxin is not necessary for phage assembly (Russel and Model, 1986). More likely, thioredoxin could function as a processivity factor, as in the DNA polymerase of phage T7 (Tabor et al., 1987). Moreover, pI has a nucleotide binding domain in its cytoplasmic portion, suggesting that nucleotide binding and hydrolysis may facilitate removal of pV from the DNA and/or addition of coat protein. Release from the membrane Phage release from the inner membrane and capping of virions are mostly dependent on the pIII structural protein (Rakonjac and Model, 1998; Rakonjac et al., 1999). This important component plays also a role in early infection, mediating phage recognition of the tip of the pilus and then its co-receptor, the periplasmic portion of the inner membrane protein TolA (Karlsson et al., 2003). Interestingly, when pIII is missing, additional genomes are added to the growing filaments without release, thereby remaining associated with the cells as very long filaments (Rakonjac and Model, 1998). The interaction of the last ring of pVIII subunits with both the virion and the membrane may be what holds the phage to the cells and prevents it from detaching in the absence of pIII. pIII and pVI both interact with pVIII in the membrane prior to assembly as well as in the virion. The question as to how the filament made of pVIII is detached from the membrane by the addition of two proteins which are themselves anchored in the inner membrane has been addressed by Rakonjac et al. (1999). pIII is 406 residues long and contains a large hydrophilic portion (Beck et al., 1978). It is synthesized as a precursor with a signal sequence. In the inner membrane pIII is anchored via a transmembrane domain located at the C-terminus (379–401) and thus most of the protein is facing the periplasm. The N-terminal domain (1–218) mediates infection while the C-terminal domain (253–406) is involved in membrane release and capping the virion. To elucidate the requirements for release of phage from the membrane a set of deletions which progressively shortened the pIII C-domain from the N-terminal side were constructed and each was examined for its ability to mediate phage release and to be incorporated into phage filaments (Rakonjac et

|

327

328

|

São-José et al.

al., 1999). An 83-residue fragment did not release phage from the membrane but pVI was incorporated into assembling phage. A 93 residue polypeptide was sufficient to release the particles into culture supernatants although such particles showed an unstable phenotype in detergents. A fragment of 121 residues was needed to confer detergent resistance to such particles. Thus the C-domain could be subdivided into the C2 subdomain sufficient for release and the C1 region needed for virion stability. Significantly, pVI was not detected in cell-associated virions in the absence of pIII while it was detected in particles that contained all the tested pIII fragments. These results, combined with previous knowledge on the relative orientation and topology of pIII and pVIII in membranes and virions, led to a two-step mechanism of termination (Rakonjac et al., 1999). According to this proposal, a pre-termination complex would be first formed by the recognition of pIII and pVI by the assembly machinery (or the terminal ring of the pVIII tube) followed by their incorporation into the virion. The second step would be a conformational change in pIII to position the C2 subdomain so that it disrupts the interaction of pIII and pVI with the membrane. This would release the phage from the inner membrane and cap the virion. Release of the phage thus appears to be a process in which a hydrophilic domain of a membrane-anchored protein mediates excision of a protein complex consisting of itself (pIII) and other two integral membrane proteins (pVI and pVIII) from the lipid bilayer. A model depicting filamentous phage release is presented in Figure 12.6.

Figure 12.6 Model for export of Ff filamentous phage. (A) A transmembrane channel is formed from the phage products pIV, pI and pXI (linked to pI but not shown in the figure). Host thioredoxin (TrxA) forms a complex with pI at the cytoplasmic side. A nucleotide binding site in pI is shown. pI and pIV are shown unlinked but may form the assembly site by protein-protein interactions even in the absence of substrate (see text). (B) The ssDNA coated with pV is recruited to the assembly site and pV is progressively replaced (not shown) by pVIII, through the action of the pI/TrxA complex. (C) The last step of phage release is shown here, where pIII bends or flips to de-insert itself from the membrane in complex with pVIII and pVI. (D) Phage is released in the external milieu through the lumen of the pIV secretin.

Phage Release from Infected Cells

Concluding remarks In this review we have described the three known basic strategies by which virion progeny is released to the extracellular media at the end of the phage infection cycle. One paramount feature that is independent of the strategy used is that host cell physiology is not seriously compromised until the moment of phage release. Filamentous phages (Inoviridae) explore this feature to the limit since they are released from the host cells by a secretion mechanism, as part of the virion morphogenetic pathway, not even affecting cellular integrity. Small icosahedral phages of ssDNA and ssRNA genomes (Microviridae and Leviviridae, respectively) release their progeny by producing a single product that blocks specific steps of the cell wall biosynthesis, and therefore require actively dividing cells to achieve an efficient lysis of their hosts. Finally, the vast majority of the studied dsDNA phages (including Siphoviridae and Tectiviridae) escape from host cells by employing at least two lysis effectors, a peptidoglycan hydrolytic enzyme (endolysin) and a membrane-permeabilizing protein (holin), the coordinated action of which results in a sudden, and precisely scheduled cell burst at the end of the infection cycle. The holin is the lysis timing device and provides, at a sharply defined time, either the path for the delivery of endolysins to the cell wall or the means to activate those that are previously secreted and already positioned in this cell compartment. In both situations the cellular membrane potential is maintained until the moment that it is dissipated by the holin action, establishing the onset of cell lysis. Fine tuning of lysis timing in dsDNA phages may involve an antiholin function among other auxiliary lysis factors. Acknowledgments The financial support from the Fundação para a Ciência e a Tecnologia, Portugal, through grants SFRH/BPD/9429/2002 to C. São-José, SFRH/BD/13806/2003 to J.G. Nascimento, and POCTI/BIO/41872/2001 to M.A. Santos is acknowledged. References

Bamford, J.K.H., Hänninen, A.-L., Pakula, T.M., Ojala, P.M., Kalkkinen, N., Frilander, M., and Bamford, D.H. (1991). Genome organization of membrane-containing bacteriophage PRD1. Virology 183, 658–676. Barenboim, M., Chang, C.Y., F. dib Hajj and R. Young. (1999). Characterization of the dual start motif of a class II holin gene. Mol. Microbiol. 32, 715–727. Beck, E., Sommer, R., Auerswald, E.A., Kurz, C., Zink, B., Osterburg, G., Schaller, H., Sugimoto, K., Sugisaki, H., Okamoto, T., and Takanami, M. (1978). Nucleotide sequence of bacteriophage fd DNA. Nucleic Acids Res. 5, 4495–4503. Bernhardt, T.G., Roof, W.D., and Young, R. (2000). Genetic evidence that the bacteriophage FX174 lysis protein inhibits cell wall synthesis. Proc. Natl. Acad. Sci. USA 97, 4297–4302. Bernhardt, T.G., Struck, D.K., and Young, R. (2001a). The lysis protein E of FXl74 is a specific inhibitor of the MraY-catalyzed step in peptidoglycan synthesis, J. Biol. Chem. 276, 6093–6097. Bernhardt, T.G., Wang, I.-N., Struck, D.K., and Young, R. (2001b). A protein antibiotic in the phage QB virion: Diversity in lysis targets, Science 292, 2326–2329. Bernhardt, T.G., Wang, I.-N., Struck, D.K., and Young, R. (2002a). Breaking free: “protein antibiotics” and phage lysis. Res. Microbiol. 153, 493–501. Bernhardt, T.G., Roof, W.D., and Young, R. (2002b). The E. coli FKBP-type PPIase SlyD is required for the stabilization of the E lysis protein of bacteriophage FX174, Mol. Microbiol. 45, 99–108. Blackburn, N.T., and Clarke, A.J. (2001). Identification of four families of peptidoglycan lytic transglycosylases. J. Mol. Evol. 52, 78–84.

|

329

330

|

São-José et al.

Blasi, U., and Young, R. (1996). Two beginnings for a single purpose: the dual start holins in the regulation of phage lysis. Mol. Microbiol. 21, 675–682. Bruttin, A., Desiere, F., Lucchini, S., Foley, S., and Brüssow, H. (1997). Characterization of the lysogeny DNA module from temperate Streptococcus thermophilus bacteriophage phiSfi21. Virology 233, 136–148. Caldentey, J., Blaco, L., Bamford, D.H., and Salas, M. (1993). In vitro replication of bacteriophage PRD1 DNA. Characterization of protein-primed initiation site. Nucleic Acids Res. 21, 3725–3730. Caldentey, J., Hänninen, A.-L., and Bamford, D.H. (1994). Gene XV of bacteriophage PRD1 encodes a lytic enzyme with muramidase activity. Eur. J. Biochem. 225, 341–346. Chopin, M-C., Rouault, A., Ehrlich, S.D., and Gautier, M. (2002). Filamentous phage active on the Grampositive bacterium Propionibacterium freudenreichii. J. Bacteriol. 184, 2030–2033. Dasilva, E., Longchamp, P.S., and Karamata, D. (1997). Identification of XlyB, the second lytic enzyme of the prophage PBSX, abstr. B25, 9th International Conference on Bacilli, Lausanne, Switzerland, July 15–19. Davis, B.M., Lawson, E.H., Sandkwist, M., Ali, A., Sozhamannan, S., and Waldor, M.K. (2000). Convergence of the secretory pathways for cholera toxin and the filamentous phage, CTX&. Science 288, 333–335. Davis, B.M., and Waldor, M.K. (2003). Filamentous phages linked to virulence of Vibrio cholerae. Curr. Opin. Microbiol. 6, 35–42. Deaton, J., Savva, C.G., Sun, J., Holzenburg, A., Berry, J., and Young, R. (2004). Solubilization and delivery by GroEL of megadalton complexes of the L holin. Protein Sci. 13, 1778–1786. Doermann, A.H. (1948). Lysis and lysis inhibition with E. coli bacteriophage. J. Bacteriol. 55, 257–276. Dressman, H.K., and Drake, J.W. (1999). Lysis and lysis inhibition in bacteriophage T4: rV mutations reside in the holin t gene. J. Bacteriol. 181, 4391–4396. Endemann, H., and Model, P. (1995). Location of filamentous phage minor coat proteins in phage and in infected cells. J. Mol. Biol. 250, 496–506. Feng, J.N., Russel, M., and Model, P. (1997). A permeabilized cell system that assembles filamentous bacteriophage. Proc. Natl. Acad. Sci. USA 94, 4068–4073. Feng, J-N., Model, P., and Russel, M. (1999). A transenvelope protein complex needed for filamentous phage assembly and export. Mol. Microbiol. 34, 745–755. Fischetti, V.A. (2005). Bacteriophage lytic enzymes: novel anti-infectives. Trends. Microbiol. 13, 491– 496. García, P., Garcia, J.L., Garcia, E., Sánchez Puelles, J.M., and López, R. (1990). Modular organization of the lytic enzymes of Streptococcus pneumoniae and its bacteriophages. Gene 86, 81–88. Garcia, M., Pimentel, M., and Moniz-Pereira, J. (2002). Expression of Mycobacteriophage Ms6 lysis genes is driven by two sigma (70)-like promoters and is dependent on a transcription termination signal present in the leader RNA. J. Bacteriol. 184, 3034–3043. Goldsmith, M.E., and Konigsberg, W.H. (1977). Adsorption protein of bacteriophage fd- isolation, molecular properties and location in virus. Biochemistry, 16, 2686–2694. Grahn, A.M., Bamford, J.K.H., O’Neill, M.C., and Bamford, D.H. (1994). Functional organization of the bacteriophage PRD1 genome. J. Bacteriol. 176, 3062–3068. Grahn, A.M., Haase, J., Lanka, E., and Bamford, D.H. (1997). The assembly of a functional phage PRD1 receptor depends on 11 genes of the IncP plasmid mating pair formation complex. J. Bacteriol. 179, 4733–4740. Grant, R., Lin, T., Webster, R., and Konigsberg, W. (1980). Structure of filamentous bacteriophage: isolation, characterization and localization of the minor coat proteins and orientation of the DNA. In: Bacteriophage Assembly, M. DuBow, ed. (New York: Alan R. Liss, Inc.) vol 64, pp. 413–428. Gray, C.W. (1989). Three-dimensional structure of complexes of single-stranded DNA-binding proteins with DNA. Ike and fd gene 5 proteins form left-handed helices with single stranded DNA. J. Mol. Biol. 208, 57–64. Grundling, A., Blasi, U., and Young, R. (2000a). Biochemical and genetic evidence for three transmembrane domains in the class I holin, lambda S. J. Biol. Chem. 275, 769-776. Grundling, A., Blasi, U., Young, R. (2000b). Genetic and biochemical analysis of dimer and oligomer interactions of the lambda S holin. J. Bacteriol. 182, 6082–6090. Grundling, A., Smith, D.L., Blasi, U., and Young, R. (2000c). Dimerization between the holin and holin inhibitor of phage lambda. J. Bacteriol. 182, 6075–6081.

Phage Release from Infected Cells

Grundling, A., Manson, M., and Young, R. (2001). Holins kill without warning. Proc. Natl. Acad. Sci. USA 98, 9348–9352. Guy-Caffey J.K., and Webster R.E. (1993). The membrane domain of a bacteriophage assembly protein: membrane insertion and growth inhibition. J. Biol. Chem. 268, 5496–5503. Haigh, N.G., and Webster, R.E. (1999). The pI and pXI assembly proteins serve separate and essential roles in filamentous phage assembly. J. Mol. Biol. 293, 1017–1027. Hänninen, A.-L., Bamford, D.H., and Bamford, J.K.H. (1997). Assembly of membrane-containing bacteriophage PRD1 is dependent on GroEL and GroES. Virology 227, 207–210. Hermoso, J.A., Monterroso, B., Albert, A., Galán, B., Ahrazem, O., García, P., Martínez-Ripoll, M., García, J.L., and Menéndez, M. (2003). Structural basis for selective recognition of the pneumococcal cell wall by modular endolysin from phage Cp-1. Structure 11, 1239–1249. Horabin, J.I., and Webster, R.E. (1988). An amino-acid sequence which directs membrane insertion causes loss of membrane potential. J. Biol. Chem. 263, 11575–11583. Hutchison, C.A.III, and Sinsheimer, R.L. (1966). The process of infection with bacteriophage FX174. X. Mutations in a FX lysis gene. J. Mol. Biol. 18, 429–447. Jarvis, A.W., Lubbers, M.W., Beresford, T.P.J., Ward, L.J.H., Waterfield, N.R., Collins, L.J., and Jarvis, B.D.W. (1995). Molecular biology of lactococcal bacteriophage c2. Dev. Biol. Stand. 85, 561–567. Kakikawa, M., Yokoi, K.J., Kimoto, H., Nakano, M., Kawasaki, K., Taketo, A., and Kodaira, K. (2002). Molecular analysis of the lysis protein Lys encoded by Lactobacillus plantarum phage Fg1e. Gene 299, 227–234. Karlsson, F., Borrebaeck, C.A.K., Nilsson, N., and Malmborg-Hager, A.-C. (2003). The mechanism of bacterial infection by filamentous phages involves molecular interactions between TolA and phage protein 3 domains. J. Bacteriol. 185, 2628–2634. Krogh, S., JØrgensen, S.T., and Devine, K.M. (1998). Lysis genes of the Bacillus subtilis defective prophage PBSX. J. Bacteriol. 180, 2110–2117. Lin, T.C., Webster, R.E., and Konigsberg, W. (1980). Isolation and characterization of the C and D proteins coded by gene 9 and gene 6 in the filamentous bacteriophage f1 and fd. J. Biol. Chem. 255, 10331–10337. Loessner, M.J., Wendlinger, G., and Scherer, S. (1995). Heterogeneous endolysins in Listeria monocytogenes bacteriophages: a new class of enzymes and evidence for conserved holing genes within the siphoviridal lysis cassettes. Mol. Microbiol. 16, 1231–1241. Loessner, M.J., Inman, R.B., Lauer, P., and Calendar, R. (2000). Complete nucleotide sequence, molecular analysis and genome structure of bacteriophage A118 of Listeria monocytogenes: implications for phage evolution. Mol. Microbiol. 35, 324–340. Loessner, M.J., Kramer, K., Ebel, F., and Scherer, S. (2002). C-terminal domains of Listeria monocytogenes bacteriophage murein hydrolases determine specific recognition and high-affinity binding to bacterial cell wall carbohydrates. Mol. Microbiol. 44, 335–349. Loessner, M.J. (2005). Bacteriophage endolysins—current state of research and applications. Curr. Opin. Microbiol. 8, 480–487. Longchamp, P.F., Mauël, C., and Karamata, D. (1994). Lytic enzymes associated with defective prophages of Bacillus subtilis: sequencing and characterization of the region comprising the N-acetylmuramoyl-Lalanine amidase gene of prophage PBSX. Microbiology 140, 1855–1867. Lopez, J., and Webster, R.E. (1983). Morphogenesis of filamentous bacteriophage f1: orientation of extrusion and production of polyphage. Virology 127, 177–193. Lopez, J., and Webster, R.E. (1985). Assembly sites of bacteriophage f1 corresponds to adhesion zones between the inner and outer membranes of the host cell. J. Bacteriol. 163, 1270–1274. López, R., and García, E. (2004). Recent trends on the molecular biology of pneumococcal capsules, lytic enzymes, and bacteriophage. FEMS Microbiol. Rev. 28, 553–580. Los, M., Wegrzyn, G., and Neubauer, P. (2003). A role for bacteriophage T4rI gene function in the control of phage development during pseudolysogeny and in slowly growing host cells. Res. Microbiol. 154, 547–552. Lu, M-J., and Henning, U. (1992). Lysis protein T of bacteriophage T4. Mol. Gen. Genet. 235, 253–258. Lubbers, M.W., Waterfield, N.R., Beresford, T.P., Le Page, R.W., and Jarvis, A.W. (1995). Sequencing and analysis of the prolate-headed lactococcal bacteriophage c2 genome and identification of the structural genes. Appl. Environ. Microbiol. 61, 4348–4356.

|

331

332

|

São-José et al.

Marciano, D.K., Russel, M., and Simon, S.M. (2001). Assembling filamentous phage occlude pIV channels. Proc. Natl. Acad. Sci. USA 98, 9359–9364. Marvin D.A. (1998). Filamentous phage structure, infection and assembly. Curr. Opin. Struct. Biol. 8, 150–158. Mauël, C., and Karamata, D. (1984). Characterization of proteins induced by mitomycin C treatment of Bacillus subtilis. J. Virol. 49, 806–812. Mayr, U.B., Walcher, P., Chakameh, A., Riedmann, E., Haller, C., and Lubitz, W. (2005). Bacterial ghosts as antigen delivery vehicles. Adv. Drug Deliv. Rev. 57, 1381–91. Model, P., and Russel, M. (1988). Filamentous bacteriophage. In: The Bacteriophages, R. Calender, ed. (New York: Plenum Publishing) vol.2, pp. 375–456. Mindich, L., Bamford, D., Goldthwaite, C., Laverty, M., and Mackenzie, G. (1982). Isolation of nonsense mutants of lipid-containing bacteriophage PRD1. J. Bacteriol. 44, 1013–1020. Navarre, W.W., Ton-That, H., Faull, K.F., and Schneewind, O. (1999). Multiple enzymatic activities of the murein hydrolase from Staphylococcal phage F11. J. Biol. Chem. 274, 15847–15856. Opalka, N., Beckmann, R., Boisset, N., Simon, M.N., Russel, M., and Darst, S.A. (2003). Structure of the filamentous phage pIV multimer by cryo-electron microscopy. J. Mol. Biol. 325, 461–470. Olsen, R.H., Siak, J., and Gray, R.H. (1974). Characteristics of PRD1, a plasmid-dependent broad host range DNA bacteriophage. J. Virol. 14, 689–699. Paddison, P.S., Abedon, S.T., Dressman, H.K., Gailbreath, K., Tracy, J., Mosser, E., Neitzel, J., Guttman, B., and Kutter, E. (1998). The roles of bacteriophage T4 r genes in lysis inhibition and fine-structure genetics: a new perspective. Genetics 148, 1539–1550. Parreira, R., São-José, C., Isidro, A., Domingues, S., Vieira, G., and Santos, M.A. (1999). Gene organization in a central DNA fragment of Oenococcus oeni bacteriophage fOg44 encoding lytic, integrative and non-essential functions. Gene 226, 83–93. Pritchard, D.G., Dong, S., Baker, J.R., and Engler, J.A. (2004). The bifunctional peptidoglycan lysin of Streptococcus agalactiae bacteriophage B30. Microbiology 150, 2079–2087. Rakonjac, J., and Model, P. (1998). Roles of pIII in filamentous phage assembly. J. Mol. Biol. 282, 25–41. Rakonjac, J., Feng J-N., and Model, P. (1999). Filamentous phage are released from the bacterial membrane by a two-step mechanism involving a short C-terminal fragment of pIII. J. Mol. Biol. 289, 1253–1265. Ramanculov, E., and Young R. (2001a). Functional analysis of the T4 holin in a lambda context. Mol. Genet. Genomics 256, 345–353. Ramanculov, E., and Young, R. (2001b). An ancient player unmasked: T4 rI encodes a T-specific antiholin. Mol. Microbiol. 41, 575–583. Ramanculov, E., and Young, R. (2001c). Genetic analysis of the T4 holin: timing and topology. Gene 265, 25–36 Rapoza, M.P., and Webster, R.E. (1995). The products of gene I and the overlapping in-frame gene XI are required for filamentous phage assembly. J. Mol. Biol. 248, 627–638. Roof, W.D., Horne, S.M., Young, K.D., and Young, R. (1994). slyD, a host gene required for FX174 lysis, is related to the FK506-binding protein family of peptidyl-prolyl cis-trans-isomerases, J. Biol. Chem. 269, 2902–2910. Roof, W.D., and Young, R. (1995). FX174 lysis requires slyD, a host gene which is related to the FKBP family of peptidylprolyl cis-trans isomerases. FEMS Microbiol. Rev. 17, 213–216. Russel, M. (1993). Protein-protein interactions during filamentous phage assembly. J. Mol. Biol. 231, 689–697. Russel, M. (1995). Moving through the membrane with filamentous phages. Trends Microbiol. 3, 223–228. Russel, M., and Model, P. (1985). Thioredoxin is required for filamentous phage assembly. Proc. Natl. Acad. Sci. USA 82, 29–33. Russel, M., and Model, P. (1986). The role of thioredoxin in filamentous phage assembly. Construction, isolation, and characterization of mutant thioredoxins. J. Biol. Chem. 261, 14997–15005. Russel, M., Linderoth, N.A., and Săli, A. (1997). Filamentous phage assembly: variation on a protein export theme. Gene 192, 23–32. Rydman, P.S., and Bamford, D.H. (2000). Bacteriophage PRD1 DNA entry uses a viral membrane-associated transglycosylase activity. Mol. Microbiol. 37, 356:363.

Phage Release from Infected Cells

Rydman, P.S., and Bamford, D.H. (2002). The lytic enzyme of bacteriophage PRD1 is associated with the viral membrane. J. Bacteriol. 184, 104–110. Rydman, P.S., and Bamford, D.H. (2003). Identification and mutational analysis of bacteriophage PRD1 holin protein P35. J. Bacteriol. 185, 3795–3803. Salstrom, J.S., and Pratt, D. (1971). Role of coliphage M13 gene V in single-stranded DNA production. J. Mol. Biol. 61, 489–501. São-José, C., Parreira, R., Vieira, G., and Santos, M.A. (2000). The N-terminal region of the Oenococcus oeni bacteriophage fOg44 lysin behaves as a bona fide signal peptide in Escherichia coli and as a cisinhibitory element, preventing lytic activity on oenococcal cells. J. Bacteriol. 182, 5823–5831. São-José, C. (2002). Genome analysis and gene expression in oenophage fOg44—evidence for a new strategy of bacteriophage-induced host lysis. PhD thesis. University of Lisbon, Portugal. São-José, C., Parreira, R., and Santos, M.A. (2003). Triggering of host-cell lysis by double-stranded DNA bacteriophages: fundamental concepts, recent developments and emerging applications. In: Recent Research Developments in Bacteriology, S.G. Pandalai, ed. (Trivandrum, India: Research Signpost, Transworld Research Network) pp. 103–130. São-José, C., Santos, S., Nascimento, J., Brito-Maduro, A.G., Parreira, R., and Santos, M.A. (2004). Diversity in the lysis-integration region of oenophage genomes and evidence for multiple tRNA loci, as targets for prophage integration in Oenococcus oeni. Virology 325, 82–95. Schmidt, C., Velleman, M., and Arber, W. (1996). Three functions of bacteriophage P1 involved in cell lysis. J. Bacteriol. 178, 1099–1104. Schouler, C., Ehrlich, S.D., and Chopin, M.-C. (1994). Sequence and organization of the lactococcal prolate-headed bIL67 phage genome. Microbiology 140, 3061–3069. Shearman, C.A., Jury, K.L., and Gasson, M.J. (1994). Controlled expression and structural organization of a Lactococcus lactis bacteriophage lysin encoded by two overlapping genes. Appl. Environ. Microbiol. 60, 3063–3073. Sheehan, M.M., García, J.L., López, R., and García, P. (1996). Analysis of the catalytic domain of the lysin of the lactococcal bacteriophage Tuc2009 by chimeric gene assembling. FEMS Microbiol. Lett. 140, 23–28. Sheehan, M.M., García, J.L., López, R., and García, P. (1997). The lytic enzyme of the pneumococcal phage Dp-1: a chimeric lysin of intergeneric origin. Mol. Microbiol. 25, 717–725. Sheehan, M.M., Stanley, E., Fitzgerald, G., and van Sinderen, D. (1999). Identification and characterization of a lysis module present in a large proportion of bacteriophage infecting Streptococcus thermophilus. Appl. Environ. Microbiol. 65, 569–577. Streisinger, G., Mukai, F., Dreyer, D.J., Miller, B., and Horiuchi, S. (1961). Mutations affecting the lyzozyme of phage T4. Cold Spring Harbor Symp. Quant. Biol. 26, 25–30. Tabor, S., Huber, H.E., and Richardson, C.C. (1987). Escherichia coli thioredoxin confers processivity on the DNA polymerase activity of the gene 5 protein of bacteriophage T7. J. Biol. Chem. 262, 16212–16233. Tran, T.A.T., Struck, D.K., and Young, R. (2005). Periplasmic domains define holin-anti-holin interactions in T4 lysis inhibition. J. Bacteriol. 187, 6631–6640. Tedin, K., Resch, A., Steiner, M., and Bläsi, U. (1995). Dual translational start motif evolutionarily conserved in the holin gene of Bacillus subtilis phage F29. Virology 206, 479–484. Vukov, N., Scherer, S., Hibbert, E., and Loessner, M.J. (2000). Functional analysis of heterologous holing proteins in a L$S genetic background. FEMS Microbiol. Lett. 184, 179–186. Vukov, N., Moll, I., Bläsi, U., Scherer, S., and Loessner, M.J. (2003). Functional regulation of the Listeria monocytogenes bacteriophage A118 holin by an intragenic inhibitor lacking the first transmembrane domain. Mol. Microbiol. 48, 173–186. Walker, J.T., and Walker JR, D.H. (1980). Mutations in coliphage P1 affecting host cell lysis. J. Virol. 35; 519–530. Wang, I.-N., Smith, D.L., and Young, R. (2000). Holins: the protein clocks of bacteriophage infections. Annu. Rev. Microbiol. 54, 799–825. Wang, I.-N., Deaton, J., and Young, R. (2003). Sizing the holin lesion with an endolysin-B-galactosidase fusion. J. Bacteriol. 185, 779–787. Wang, I.-N. (2005). Lysis timing and bacteriophage fitness. Genetics, In Press. Witte, A., Wanner, G., Blasi, U., Halfmann, G., Szostak, M., and Lubitz, W. (1990). Endogenous transmembrane tunnel formation mediated by PhiX174 lysis protein E. J. Bacteriol. 172, 4109–4114.

|

333

334

|

São-José et al.

Xu, M., Struck, D.K., Deanton, J., Wang, I.-N., and Young, R. (2004). A signal-arrest-release sequence mediates export and control of the phage P1 endolysin. Proc. Natl. Acad. Sci. USA 101, 6415–6420. Xu, M., Arulandu, A., Struck, D.K., Swanson, S., Sacchettini, J.C., and Young, R. (2005). Dissulfide isomerization after membrane release of its SAR domain activates P1 lysozyme. Science 37, 113–117. Yarmolinsky, M.B., and Sternberg, M. (1988). Bacteriophage P1. In: The bacteriophages, R. Calendar, ed. vol.1 (New York, USA: Plenum Press), pp. 291–438. Young, R. (1992). Bacteriophage lysis: mechanism and regulation. Microbiol. Rev. 56, 430–481. Young, R., and Bläsi, U. (1995). Holins: form and function in bacteriophage lysis. FEMS Microbiol. Rev. 17, 191–205. Young, R., Wang, I.-N., and Roof, W.D. (2000). Phages will out: strategies of host cell lysis. Trends in Microbiol. 8, 120–128. Young, R. (2002). Bacteriophage holins: deadly diversity. J. Mol. Microbiol. Biotechnol. 4, 21–36. Young, R., and Wang, I.-N. (2006). Phage Lysis. In: The Bacteriophages. R. Calendar, ed. (New York, Oxford University Press). In Press. Zhang, N., and Young, R. (1999). Complementation and characterization of the nested Rz and Rz1 reading frames in the genome of bacteriophage L. Mol. Gen. Genet. 262, 659–667. Žiedaite, G., Daugelavičius, R., Bamford, J.K.H., and Bamford, D.H. (2005). The holing protein of bacteriophage PRD1 forms a pore for small-molecule and endolysin translocation. J. Bacteriol. 187, 5397–5405. Ziermann, R., Bartlett, B., Calendar, R., and Christie, G.E. (1994). Functions involved in bacteriophage P2-induced host cell lysis and identification of a new tail gene. J. Bacteriol. 176, 4974–4984. Zimmer, M., Scherer, S., and Loessner, M.J. (2002). Genomic analysis of Clostridium perfringens bacteriophage F3626, which integrates into guaA and possibly affects sporulation. J. Bacteriol. 184, 4359–436.

Index

A A118 phage 314–15 Abi systems see abortive infection abortive infection 112–14 Acetobacter spp. 99 Acetobacter europaeus 101 N-acetylmuramidases (lysozymes) 136 N-acetylmuramoyl-L-alanine amidases 19, 136 Acinetobacter spp. 3 Actinobacillus actinomycetemcomitans 244 adenosine deaminase deficiency 142 Adhya, Sankar 179 adsorption blocking 111 Aeromonas spp. 2 Aeromonas hydrophila 19 Aeromonas salmonicida 19 agricultural applications 167–71 aquaculture 169 biocontrol of plant pathogens 167–9 farm animals 169–71 calves 170–1 poultry 169–70 see also food industry AIDS 142 algal flocs 76 alimentary tract infections 127 amurins 138 amylases 28 anti-immune state 256 anti-phage antibodies 139–40 anti-phage humoral response 141–2 antibiotic resistance 126, 128, 179–80 transduction of 243–4 antiholins 313–14, 320–1 appendicularians 76 aquaculture 169 aquatic environments 13–15 Archaea 66–7, 93 viral influence on 80

archaephages 62 attachment sites 49

B Bacillus amylolyquefaciens 274 Bacillus anthracis 20, 185 Bacillus colistinus 99 Bacillus licheniformis 274 Bacillus polymyxa 99 Bacillus pumilus 274 Bacillus subtilis 98, 99, 213, 274 PBSX prophage 316 var. natto 98 Bacillus thuringensis 99 bacteria 66 evolution 10–12 intestinal 15–16 virulence 10–12 bacterial toxins 52, 237 see also individual toxins bacteriocins 75 Bacterioides fragilis, HSP40 64 bacteriophage-insensitive mutants 29, 110, 111 bacterioplankton 62, 63, 73, 76 lysogeny 75 Bacteroides fragilis 16 Basic Local Alignment Search Tool (BLAST) 49, 55 beef, phage treatment 163 benthic systems 66 Bermudez, Luiz 183 bIL170 phage 199 bIL286 phage 8 biocontrol foods 25–8 plant pathogens 30 biofilm eradication 30–1 biogeochemical cycles 75–7 bioinformatics 43–59 phage genome sequences 45–6

336

|

Index

bioluminescence 23 bistability 256 BK5-T phage 8, 201 Bläsi, Udo 183 BLAST see Basic Local Alignment Search Tool bone marrow transplantation 142 bor gene 173, 234 Botstein’s model 50 botulinum toxin 52, 237 Bradyrhizobium japonicum 30 Brevibacterium lactofermentum 99 Brochotrix thermosphacta 165 BS32 phage 274 Burkholderia thailandensis 15

C c2 phage 201–2, 213 calves, phage treatment 170–1 Campylobacter spp. 161, 165–7, 170 Campylobacter coli 147 Campylobacter jejuni 27–8, 147, 162 capsid 48, 95 carbon fixation 76 Caudovirales 1, 2, 43, 95, 194, 273 genome 44 cell division interference 322–3 chaperonins 314 Chase, Martha 4 Che8 phage 6, 7 chemiosmotic theory of DNA transfer 217 chemotaxis inhibitors protein (CHIPS) 12 Chlorella 72 chlorophyll 66 cholera 180–1 cholera toxin 11, 52, 237 chronic infection cycle 97 CI protein 251, 252–6, 263–4 organization and functions 254 CII protein 257 Clean In Place systems 108 Clostridium spp. 99 Clostridium botulinum 11, 229 Clostridium difficile 18, 179 Clostridium tetani 238 comparative genomics 5–7, 50–1 Corndog 7 Corticoviridae 2 Corynebacterium spp. 99 Corynebacterium diphtheriae 11, 229 CRITICA 48 Cro protein 256, 264 CTXJ phage 239 culture rotation 108–10 cyanobacteria 13–15, 62, 81 microhabitat 82–3 mortality 76

cyanophages 68–9, 71 Cystoviridae 2 cytotoxin 237

D dairy fermentation by-products 106 Delbrück, Max 4, 110 d’Herelle, Felix 3, 125, 172 diphtheria toxin 237 dissolved organic carbon 13, 75, 76 dissolved organic matter 13, 14 DNA amplification 23, 24 DNA damage/repair 71–2 DNA ejection blocking 110–11 DNA mismatch repair 242–3 DNA transfer 209–28 chemiosmotic theory 217 crossing host envelope 213–17 in vitro studies of DNA ejection 222–3 internal capsid pressure and DNA ejection 217–18 Myoviridae T4 phage 218–19 phage-host receptor recognition 211–13 Gram-negative bacteria 211–12 Gram-positive bacteria 212–13 Podoviridae T7 phage 219–20 Syphoviridae J29 phage 220 Syphoviridae T5 phage 221–2 domains 49 double-stranded DNA 209 double-stranded RNA 144–5

E ecology 13–16 aquatic environmenta 13–15 physiological environments 15–16 terrestrial environments 15 Edwardsiella tarda 19 efficiency of centers of infection 113 electron microscopy 103–4 Eliava, George 3 endo-B-N-acetylglucosaminidases 136 endolysins see lysins endopeptidases 19, 136 endotoxin 129 engineered defense systems 114–18 food industry 117–18 gene silencing 116 origin-derived phage-encoded resistance 115–16 phage structural proteins/antibodies 117 subunit poisoning 117 superinfection immunity 115 triggered suicide 116–17 Enterobacter spp. 131 Enterobacter cloacae 31

Index

Enterobacteriaceae 65, 196 Enterococcus faecalis 185 enterophagos 142 enterotoxin A 237 environment 61–91 bacteriophages in 66–70 extreme habitats 66, 80 phage ecology 70–4 phage identification isolation-based techniques 64–5 isolation-independent methods 65–6 Erwinia amylovora 30, 167 erythrogenic toxins 237 Escherichia spp. 131 Escherichia coli 11, 16, 18, 19, 52, 69, 99, 146, 160, 161, 196, 237 P2 phage 315 prophage integration sites 242 shiga toxin-producing 147 Escherichia coli diarrhea 171–9 host specificity and collateral damage 175–6 human trials 177–8 industrial phage production 174 mouse model 177 outlook 178–9 phage isolation 172 phage virulence factors 172–3 pharmacokinetics of phages 174–5 physiological aspects 176 resistance 176–7 safety issues 173–4 Escherichia faecium 128, 185 vancomycin-resistant 179 EU directive 2001/20/EC 148 Eubacteria 93 Eukarya 67 exotoxins 11 extreme habitats 66, 80

F F1Z15 phage 233 FASTA 49, 55 Felix phage 164 fib gene 237 filamentous phages 183, 238–9, 281, 307, 308 progeny release 325–8 assembly sites and initiation of secretion 325–7 release from membrane 327–8 Fischetti, Vincent 184 flow cytometry 62 fluorescent amplified fragment length polymorphism 147 fOG44 phage 318–20 food fermentations 28–9, 93–123

industrial incidence/significance of phages 98–100 monitoring of phages 100–4 phage contamination 104–7 phage control 108–18 phage life cycle 96–8 phage structure 95–6 sanitation processes 107–8 see also food industry, phase applications food industry, phage applications 161–7 animal feeding 161, 165–7 food spoilage 161, 164–5 fresh foods 161, 164 raw foods 161, 162–3 beef 163–4 poultry 162–3 food spoilage 161, 164–5 food webs 75–7 pelagic systems 67 FrameD 48 frequency of infected cells 62–3 fresh foods, phage treatment 161, 164 functional annotation 49–50 Fuselloviridae 2

G GA-1 phage 274 GALT 141 GB virus C 143 GenBank database 9 gene overlap 48 gene prediction 47–9 gene regulatory circuitry 266–8 gene silencing 116 gene transfer 69 virus-mediated 79–80 genetically modified organisms 115 genome annotation 46–50 functional annotation 49–50 gene prediction 47–9 identification of prophage sequences 46–7 genome sequencing 45–6 genomics 43–59 comparative 5–7, 50–1 prophages 44, 52–3 Gifsy-1 prophage 235 Gifsy-2 prophage 235 gipA gene 235 Glimmer software tool 47 Gluconobacter spp. 99 glucosamidases 19 Gram-negative bacteria phage-host receptor recognition 211–12 toxigenic phages 238–9 Gram-positive bacteria antibiotic-resistant 179–80

|

337

338

|

Index

phage-host receptor recognition 212–13 toxigenic phages 236–8 green fluorescent proteins 23 GroEL chaperonin 314 GtgE protein 235

H Hatfull, Graham 183 head maturation proteases 47 HECTOR sequence 68 HEPA filters 107 Hershey, Al 4 Hoc protein 139 holins 51, 96, 106, 135, 137, 184, 284, 308–14 topology 312 two-component 316 see also progeny release horizontal gene transfer 10, 46, 230–2 host gene expression 240–3 host invasiveness 234–6 host resistance 72, 129 host specificity 175–6 hydrothermal vents 66

I immunochemical detection 103 immunocompetence 141–2 immunomodulation 138–41 industrial fermentations 28–9 see also Food fermentations industrial phage production 174 industrial significance of phages 98–100 infections alimentary tract 127 tuberculosis 127 Inoviridae 2 integrases 47, 239–40 integrins 143 interferon 143, 144 internal phage reservoirs 106–7 International Committee on the Taxonomy of Viruses 53 interpolated Markov models 47 InterProScan 49 intestinal bacteria 15–16 intracellular bacteria 183

J JCL1032 phage 203

K “killing the winner” hypothesis 77–8, 81 Klebsiella spp. 131 Klebsiella bacilli 146

L lactic acid bacteria 193–4 host receptors 194–6 phage-host interaction 193–208 prophages 12 Lactobacillus spp. 6, 193 Lactobacillus acidophilus 98 Lactobacillus brevis 98, 99 Lactobacillus casei 98 Lactobacillus delbrueckii 196 subsp. bulgaricus 98 subsp. lactis 98 Lactobacillus fermentum 98 Lactobacillus helveticus 98 Lactobacillus johnsonii, NCC533 52 Lactobacillus plantarum 48, 52, 98, 99, 203 Lactobacillus salivarius, subsp. salivarius 52 Lactobacillus phages, receptor recognition 202–3 Lactococcus spp. 193 Lactococcus garvieae 19, 169 Lactococcus lactis 12, 15, 29, 99, 114, 115, 198, 212, 244 IL1403 52 Lactococcus phages, receptor recognition 198– 202 L phage 48, 50, 142, 143, 146 life cycle 252 lysis paradigm 313–14 prophage induction 251–72 CI protein 252–6 Cro protein 256 gene regulatory circuitry 266–8 lysis-lysogeny decision 256–8 lytic pathway 252 SOS regulatory system 258–61 systems behavior 265–6 threshold behavior 261–5 lambdoid coliphages 50 Legionella pneumophila 233 Leuconostoc spp. 193 Leuconostoc fallax 99 Leuconostoc mesenteroides 99 Leviviridae 2 progeny release 321–5 LexA protein 258, 259–60 LIN phenomenon 317–18 Lipothrixviridae 2 Listeria monocytogenes 15, 148 A118 phage 314–15 live animals, phage treatment 161, 165–7 calves 170–1 LP65 phage 203 Luciferase Reporter Phage 23, 24 Luria, Salvadore 4

Index

lysins 19–20, 47, 135–8, 308–13 immunogenicity of 137–8 medical applications 183–6 resistance to 137 sec-dependent 318–20 see also progeny release; and individual types lysis 266–8 from without 136, 184 products of 81 see also lytic lysis-deficient phages 129 lysogens 97 lysogenic conversion genes 10 lysogenic cycle 97–8 lysogenic starter cultures 105–6 lysogeny 10, 50, 62–4, 73, 74, 75, 256–8 lysozymes 136 lytic cycle 96–7 suppression by Spo-A 294–7 lytic enzymes 196–7 lytic pathway 252 lytic phages 73, 128 lytic proteins 135–8 lysins 19–20, 47, 135–8 lytic transglycosylases 136

microhabitat 82–3 microniches 83 Microviridae 2 progeny release 321–5 mitomycin C 63, 75 mode of antibacterial action 128 modular exchange 230–2 modular theory of phage evolution 50 monitoring of phages 100–4 biochemical methods 103 electron microscopy 103–4 microbiological methods 100–3 monoclonal antibodies 64 morons 230 motifs 49 MRSA 126, 128, 131, 132–3, 179 MS2 L phage 138, 324–5 muramidases 19 Mycobacterium avium 183 Mycobacterium bovis 24 Mycobacterium tuberculosis 23, 24 Mycoplasma mycoides 23 Myoviridae 2, 43, 51, 95, 194, 211, 273 DNA transfer 222–3 Myovirus T4 44

M

N

M protein 234 M2Y phage 274 major histocompatibility complex class II deficiency 142 maltose binding protein E 281 marine environments 69 marine/lake snow 76 medical applications 125–58, 171–86 anti-phage humoral response 141–2 antibiotic-resistant Gram-positive bacteria 179–80 bacteriophage lytic proteins 135–8 Escherichia coli diarrhea 171–9 immunomodulation 138–41 intracellular bacteria 183 non-replicating phages 183 pathogenic virus infection 142–6 phage lysins 183–6 phage typing 146–7 skin and wound infections 181–2 vibriophages 180–1 see also phage therapy Merril, Carl 179 metagenomics 8–9, 44, 45, 49 viral 54 MGAS315.3 phage 231 MGAS10394.6 phage 231

Nakae, Toshihiru 169 native defense mechanisms 110–14 abortive infection 112–14 adsorption blocking 111 DNA ejection blocking 110–11 restriction and modification 112 see also host natural phage occurrence 134 Neisseria meningitidis 239 Nf phage 274 nitrogen fixation 76 non-lytic phages 128 non-replicating phages 183 nuclear localization signal 143–4 nucleotide signatures 46 nutrient limitation 71

O Oenococcus oeni 99 fOG44 phage 318–20 open reading frames 46, 47 origin-derived phage-encoded resistance 115–16 ORPHEUS 48

P p1 protein 280–2 P2 phage 315

|

339

340

|

Index

p4 protein 276 p6 protein 276 p16.7 protein 276, 282–91 dimer formation 285–6 expression in J28-infected cells 283 functional domain 286–7 interaction with J29 terminal protein 285 as membrane protein 283–4 modular organization 287 role in DNA replication 284 structural analyses 287–90 p17 protein 276 P22 phage 212, 216 P335 phage 8, 200–1 Panton-Valentine leukocidins 234 particulate organic carbon 75 pasteurized fermentation substrates 104–5 pathogen detection 23–5 pathogenicity islands 10 PBSX prophage 316 pelagic systems 66 food web model 67 peptidoglycan 135, 196, 309 peptidoglycan hydrolases 310 Peyer’s patches 12 Pfam 49 PHACCS 65 phage abundance 66 phage classification 2 phage contamination 104–7 external sources 104–6 internal phage reservoirs 106–7 phage control 108–18 bacteriophage-insensitive mutants 110 engineered defense systems 114–18 native defense mechanisms 110–14 strain selection and culture rotation 108–10 phage conversion 80 phage display 20–1 phage diversity 67–9 phage ecology 70–4 DNA damage/repair 71–2 host range and resistance 72 nutrient limitation 71 phage life styles 73–4 phage forming units 64 phage ghosts 75 phage infection protein 195 phage life cycle 96–8 chronic infection cycle 97 lysogenic cycle 97–8 lytic cycle 96–7 phage life styles 73–4 phage lysates 75 phage particles 94 phage production 62–4

phage proteomic tree 7–8, 70 phage receptors 194–6 phage structural proteins/antibodies 117 phage structure 95–6 phage tails 75 phage therapy 16–19, 126–35, 159–92 agriculture 167–71 clinical studies 126–9 Escherichia coli diarrhea 171–9 food industry 161–7 historical aspects 159–61 medical applications 129–35, 171–86 antibiotic-resistant Gram-positive bacteria 179–80 Escherichia coli diarrhea 171–9 intracellular bacteria 183 non-replicating phages 183 phage lysins 183–6 skin and wound infections 181–2 vibriophages 180–1 phage typing 64, 146–7 phage-encoded resistance 115 phage-host interaction 193–208 phage-associated lytic enzymes 196–7 receptor recognition 197–203 phage-host receptor recognition 197–203, 211–13 Gram-negative bacteria 211–12 Gram-positive bacteria 212–13 Lactobacillus phages 202–3 Lactococcus phages 198–202 Streptococcus phages 198 phage-induced mortality 62–4 prokaryotes 74–5 phage-module 51 phage-resistant bacteria 72, 129 phagemids 128 phagicin 142–3 phagocytosis, resistance to 233–4 phagotyping 147 pharmacokinetics 174–5 J6 phage 144 J11de phage 243 J15 phage 274 J29 phage 214, 273–305 DNA transfer 220 genetic and transcriptional maps 275 host spore formation in survival 291–8 in vitro transcription regulation and DNA replication 276–80 protein p1 280–2 protein p16.7 282–91 J80 phage 143 JETA prophage 237 JMF2 phage 231 JSda phage 231

Index

JX174 phage 43, 69, 141 E protein 138, 322, 323–4 lysis mediation 322 photosynthesis genes 81 phyllosphere 167 phylogeny 53–4 Phytium ultimum 169 phytoplankton 62, 76 phytoplankton blooms 62 pip gene 195 plankton predation 62 plant pathogens, biocontrol 30, 167–9 plaque assay 100–3 Plasmaviridae 2 human applications 17–18 veterinary applications 18–19 platelet binding 139, 235 Plecoglossus altivelis 19, 169 podophages 15 Podoviridae 2, 43, 95, 194, 197, 273 DNA transfer 219–20 pokeweed mitogen 139 portal protein 47 poultry farming 161, 162–3, 168–70 PRD1 phage 316–17 PRINTS 49 Prochlorococcus 13 progeny release 307–34 basic strategies 307–8 filamentous phage strategy 325–8 assembly sites and initiation of secretion 325–7 release from membrane 327–8 holin/endolysin strategy 308–13 lysis mechanisms 313–21 Bacillus subtilis prophage PBSX 316 Escherichia coli phage P2 315 Listeria monocytogenes phage A118 314–15 phage L paradigm 313–14 sec-dependent endolysins 318–20 T4 phage and LIN phenomenon 317–18 Tectiviridae phage PRD1 316–17 Microviridae and Leviviridae phage strategy 321–5 cell division interference model 322–3 MS2 L and QB A2 324–5 JX174-mediated lysis 322 JX174E 323–4 prokaryotes 61, 62, 66 phage-induced mortality 74–5 phages and diversity 77–81 prophages 10, 97, 229–50 altered host gene expression 240–3 attachment sites 49 diversity 70 gene sequence identification 4607

genomics 44, 52–3 host invasiveness 234–6 induction 258–61 threshold behavior 261–5 integration targets 239–40 lactic acid bacteria 12 lifestyle 229–30 morons 230 resistance to serum killing and phagocytosis 233–4 toxigenic 236–9 Gram-negative species 238–9 Gram-positive species 236–8 transduction of antibiotic resistance 243–4 transmissibility, modular exchange and horizontal transfer 230–2 Propionibacterium freudenreichii 97, 99 proteases 28 Proteus spp. 131, 160 protistan grazing 78–9 protists 75 Pseudomonas spp. 131 Pseudomonas aeruginosa 71, 79, 99, 131, 146, 182, 237, 244 Pseudomonas fluorescens 15 Pseudomonas plecoglossicida 19 Pseudomonas putida 169 Pseudomonas tolaasii 30 PyoPhage 182 PZA phage 274

Q QB A2 phage 138, 324–5

R raw fermentation substrates 105 raw foods 161, 162–3 beef 162–3 poultry 162–3 reactive oxygen species 140 RecA protein 258–9, 260, 262–3 receptor binding protein 193, 212–13 recombinant DNA 4 regulatory states, stability of 265–6 replication 50 resistance 176–7 antibiotic 126, 128, 179–80 host 72, 129 to lysins 137 origin-derived phage-encoded 115–16 phage-encoded 115 to phagocytosis 233–4 to serum killing 233–4 respiratory syncytial virus 145 restriction and modification systems 112 robustness 266

|

341

342

|

Index

roseophages 71 Rudiviridae 2

S Salmonella spp. 16, 161, 164, 170 Salmonella choleraesuis 233 Salmonella enterica 28, 146 serovar Enteritidis 147 Salmonella enteritidis 27, 30 Salmonella gallinarum 18 Salmonella pullora 134 Salmonella typhi 146 Salmonella typhimurium 147 salmonella toxin 52 sanitation processes 107–8 Seriola quinqueradiata 19, 169 Serratia liquifaciens 15 serum killing, resistance to 233–4 set point 261, 262 SF370 phage 236 SF370.1 phage 231 Shiga toxin 11, 237, 262 Shigella spp. 133, 160, 172 Shigella flexneri 146, 238 Shigella sonnei 146 shotgun sequencing 44 Sie2009 protein 112 siphophages 15, 54 Siphoviridae 2, 6, 8, 43, 51, 70, 74, 95, 194, 211, 273 DNA transfer 220–2 sk1 phage 199 skin and wound infections 181–2 Smith-Waterman algorithm 49 Soothill, James 182 SopEJ prophage 234 SOS regulatory system 258–61 Sphingomonas spp. 68 Spo0A protein 294–7 Spo0J protein 292 spoilage bacteria 161, 164–5 spore formation and phage survival 291–8 Bacillus subtilis 292–4 suppression of lytic cycle by Spo0A 294–7 SPP1 phage 213 SPsP3 phage 231 SSB protein 278 staphage lysate 138 Staphylococcus spp. 160 Staphylococcus aureus 11, 131, 146, 147 methicillin-resistant see MRSA prophage integration sites 242 Staphylococcus enterica s. Typhimurium 237 sterile fermentation substrates 104 Sterilize In Place 108 stomatology 128

strain selection 108–10 Streptococcus spp. 160 Streptococcus imitis 11–12 Streptococcus phages, receptor recognition 198 Streptococcus pneumoniae 184 Streptococcus pyogenes 11, 20, 52, 184, 229, 232 prophage integration sites 242 prophages 231 Streptococcus thermophilus 6, 99, 114, 115, 193, 198, 212 Streptomyces aureofaciens 99 Streptomyces endus 99 Streptomyces griseus 99 Streptomyces kanamycetus 99 Streptomyces scabiei 30, 168 Streptomyces venezuelae 99 Stx1 prophage 238 substrates by-products of dairy fermentation 106 lysogenic starter cultures 105–6 pasteurized fermentation 104–5 raw fermentation 105 sterile fermentation 104 subunit poisoning 117 superinfection immunity 115 Sxl gene 268 Synechococcus 13 synteny 51 systems behavior 265–6

T T2 phage 142 T4 phage 5, 48, 133, 139, 143, 172, 212 DNA transfer 222–3 LIN phenomenon 317–18 T5 phage, DNA transfer 221–2 T7 phage, DNA transfer 219–20 T12 phage 236 tail fibers 47 tail spike proteins 213 tail tape meaure proteins 47 tail tip proteins 47 tailed phages 44, 66–7 tape-measure proteins 216 taxonomy 7–8, 53–4 Tectiviridae 2, 197 PRD1 phage 316–17 temperate phages 22, 97, 233 terminase 47 terrestrial environments 15 Tetragenococcus halophila 99 threshold behavior of prophage induction 261–5 toxic shock syndrome 232, 237 toxigenic phages 236–9 Gram-negative species 238–9 Gram-positive species 236–8

Index

toxins see bacterial toxins translational frameshifting 48 transmembrane domains 318 transmission electron microscopy 61, 67 triggered suicide 116–17 tRNA 48 tuberculosis 127 Twort, Frederick 3, 125

U urine, phage presence in 134

V vaccines 21–3 vancomycin-resistant Escherichia faecium 179 vesicular stomatitis virus 143 Vibrio spp. 3 Vibrio cholerae 11, 181 Vibrio natriegens 78 Vibrio parahaemolyticus 68, 69 Vibrio vulnificus 181 vibriophages 180–1 viral abundance 62–4, 66

viral loop see viral shunt viral metagenomics 54, 69–70 viral shunt 14, 75 virion assembly proteins 47 virulence factors 172–3 virus-mediated gene transfer 79–80 viruses 67 wastewater treatment 31 Wiskott-Aldrich syndrome 142

X X-linked agammaglobulinemia 142 X-linked hyper-IgM syndrome 142 Xanthomonas spp. 168 Xanthomonas campestris 99, 168 pv. Vesicatoria 30 XepA protein 316 XhlA protein 316 XhlB protein 316

Y YACOP 48 Yersinia ruckeri 19

|

343

Colour plates

A

NMR

C

X-Ray

C

C

L2a

H1a

H3a H3b L1a L1b

H2a

N

B

N

N

C

H1b

L2b

H2b

N

C

Figure 3 Plate 11.3 Three-dimensional structures of the apo and DNA complexed form of p16.7C. (A) Solution and crystal structure of p16.7C. Left, side-view of backbone atoms (N, CA, and C) of 25 superimposed NMR-derived structures of p16.7C. Right, ribbon representation of side view of p16.7C crystal structure. (B) Side view of a ribbon representation of the structure of p16.7C in complex with dsDNA. Individual p16.7C dimers are shown in red, yellow and blue. DNA is shown in green. (C) Top view of electrostatic surface representation of a tridimeric p16.7C (blue, positive; red, negative). Figures reprinted with permission from the Journal of Biological Chemistry.

|

A-1