Ecology and Physiology of Parasites: A Symposium 9781487595128

Table of contents :
Contents
Contributors
Preface
The development and ecology of coccidia and related intracellular parasites
Epidemiological considerations of the leishmanias with particular reference to the New World
Morphological and physiological considerations of extracellular blood protozoa
Physiological, morphological, and ecological considerations of some microsporidia and gregarines
Helminths as vectors of micro-organisms
Site-finding behaviour in helminths in intermediate and definitive hosts
The physiology and behaviour of the monogenean skin parasite Entobdella soleae in relation to its host (Solea solea)
The microcosm of intestinal helminths
The movement of nematodes in the external environment
The ecology of onchocerciasis in man and animals
Mosquito vector and vertebrate host interaction: The key to maintenance of certain arboviruses
The ecology of blood-sucking Diptera: An evolutionary perspective

Citation preview

Ecology and physiology of parasites A SYMPOSIUM Increasing wisdom and ingenuity are required if we are to master our environment and cope with the myriad of organisms that affect our existence. Not the least of these organisms are the parasites and pathogens which can be found in all animals. The ecological implications of parasitism are obvious, and the interrelationships among different organisms within the same host are fascinating, but more knowledge and understanding are needed. The symposium was held to stimulate discussion of the significance of ecological problems presented by parasites and to develop means of attacking some of these problems. The diversity of parasitism from protozoa to arthropods was emphasized and the speakers and topics were selected to interest those in various biological disciplines and professions. Organized by the Department of Parasitology in the School of Hygiene of the University of Toronto, and held at Toronto in February 1970, the symposium was an unqualified success. The enthusiastic interest, indicated by the attendance of over three hundred people from seven countries, and numerous requests for copies of the proceedings led to the publication in this volume of the twelve papers presented at the symposium. The opening remarks of the leader of the discussion which followed each paper have been included and a complete bibliography is provided for each topic. The contributors are leading specialists in their fields; their papers present the results of the most recent research and assemble and review the scattered literature on each topic. The text is illustrated throughout with diagrams and photographs. Parasitism and associated phenomena are excellent examples of problems requiring the interdisciplinary approach taken by the symposium. The results of such an approach are useful in a wide variety of disciplines: microbiology, invertebrate zoology, entomology, and tropical medicine, as well as parasitology. A. MURRAY FALLIS is Professor and Head of the Department of Parasitology in the School of Hygiene and Associate Dean of Division IV of the School of Graduate Studies of the University of Toronto.

This page intentionally left blank

Ecology and physiology of parasites A symposium held at University of Toronto 19 and 20 February 1970 EDITED BY A. M. FALLÍS Sponsors University of Toronto Medical Research Council The Wellcome Trust Organized by Department of Parasitology, School of Hygiene S. Desser A. M. Fallis R. S. Freeman D. F. Mettrick Susan Mclver K. A. Wright

University of Toronto Press

©University of Toronto Press 1971 Toronto and Buffalo Printed in Canada ISBN 0-8020-1730-4 Microfiche ISBN 0-8020-0061-4 LC 70-151365

Contents

Contributors

vii

A. M. FALLÍS

Preface

ix

DATUS M. HAMMOND

The development and ecology of coccidia and related intracellular parasites R. LAINSON and J. J. SHAW Epidemiological considerations of the leishmanias with particular reference to the New World

3

21

KEITH VICKERMAN

Morphological and physiological considerations of extracellular blood protozoa

58

JIRÍ VAVRA Physiological, morphological, and ecological considerations of some microsporidia and gregarines

92

D. L. LEE

Helminths as vectors of micro-organisms

104

MARTIN J. ULMER

Site-finding behaviour in helminths in intermediate and definitive hosts

123

G. C. KEARN

The physiology and behaviour of the monogenean skin parasite Entobdella soleae in relation to its host (Solea soled)

161

CLARK P. READ

The microcosm of intestinal helminths

188

H. R. WALLACE

The movement of nematodes in the external environment

201

VI

CONTENTS

B. O. L. DUKE

The ecology of onchocerciasis in man and animals

213

WILLIAM C. REEVES

Mosquito vector and vertebrate host interaction : The key to maintenance of certain arboviruses

223

j. A. DOWNES The ecology of blood-sucking Díptera: An evolutionary perspective

232

Contributors

J. A. DOWNES

Entomology Research Institute Canada Department of Agriculture, Ottawa B. O. L. DUKE

Helminthiasis Research Unit Centre de Recherches médicales Kumba, West Cameroon Federal Republic of Cameroon DATUS M. HAMMOND

Department of Zoology Utah State University, Logan, Utah G. C. KEARN

School of Biological Sciences University of East Anglia, Norwich, England R. LAINSON

The Wellcome Parasitology Unit Instituto Evandro Chagas, Belém, Para, Brazil D. L. LEE Department of Parasitology Houghton Poultry Research Station Houghton, Huntingdon, England CLARK P. READ

Department of Biology, Rice University Houston, Texas WILLIAM C. REEVES

School of Public Health University of California, Berkeley, California j. j. SHAW The Wellcome Parasitology Unit Instituto Evandro Chagas, Belém, Para, Brazil MARTIN J. ULMER

Department of Zoology and Entomology Iowa State University, Ames, Iowa KEITH VICKERMAN

Department of Zoology University of Glasgow, Glasgow, Scotland

Vlll

JIRÍ VAVRA

Institute of Parasitology Czechoslovak Academy of Sciences Prague, Czechoslovakia H. R. WALLACE

Division of Horticultural Research, C.S.I.R.O. Adelaide, South Australia

CONTRIBUTORS

Preface

The role of the infinitely small in nature is infinitely great. Pasteur as quoted by René Dubos in The Unseen World, Rockefeller Press, 1962

Increasing wisdom and ingenuity are required if we are to master our environment and cope with the myriad of organisms that affect our existence, not the least of which are the parasites and pathogens which occur in all types of animals. This is not to suggest that every parasite causes disease. On the contrary, as Theobald Smith pointed out fifty years ago, parasitism is a normal phenomenon and the interplay between host and parasite is continuous. The ecological implications of such a situation are obvious, but knowledge is scanty. The interrelationships among different organisms within the same host are fascinating although not well understood. Indeed, only recently the eminent parasitologist W. P. Rogers of Australia remarked, "there is no satisfactory explanation, in physiological terms, for the parasitic habits of any animal parasite." Parasitism and associated phenomena suggest, moreover, excellent examples of problems requiring interdisciplinary research and provide model systems which can be used to furnish exciting information. The possibilities for such uses increase as knowledge of the physiology and ecology of the parasites expands. A group of colleagues speculated on the significance of, and means of attacking, some of these ecological problems. "Why not," said one, "assemble a group of experts to discuss their diverse researches?" "Let us select topics and speakers," said another, "to emphasize the diversity of parasitism from protozoa to arthropods and interest those in various biological disciplines and professions." This seemed a commendable plan since frequently biologists, because of an understandable interest in life in water and on land, overlook the third type of environment, namely that within other animals. The enthusiasm of colleagues in the Department of Parasitology, and others with whom the idea was discussed, set in motion plans for a symposium. Their realization was made possible by the generous financial

X

support of the President's Fund, the School of Graduate Studies, the Medical Research Council, and the Wellcome Trust. Publication of the proceedings was facilitated by grants from the Wellcome Trust and the Canadian National Sportsmen's Show. The venture, as indicated by the papers that follow and an attendance of almost three hundred from seven countries, was an unqualified success. The planning committee is grateful for the encouragement and support of Dr A. J. Rhodes, Director, School of Hygiene, and Dr D. A. Chant, Chairman, Department of Zoology. We are especially indebted to Miss Ourom of the editorial staff of the University of Toronto Press. Her meticulous care and editorial skills are largely responsible for the assembly of the papers in this volume. Special gratitude is owing to Mrs N. Doughty and Mrs M. Staszak for their execution of the seemingly endless secretarial tasks and their conscientious attention to detail in planning the symposium. Finally the authors deserve our thanks for submitting illustrations and manuscripts requiring minimal editorial changes. We are grateful also to those who led the discussion for their pertinent comments. Possibly some who read the papers will be inspired to investigate problems that remain. The range is enormous. Answers to some, as in the past, will lead to improved health of man and his animals and could change the course of history. Solutions of others will provide knowledge for later application and satisfaction and encouragement to all who probe the unknown in search of it. A.M.F. Toronto, March 1970

PREFACE

Ecology and physiology of parasites A SYMPOSIUM

This page intentionally left blank

The development and ecology of coccidia and related intracellular parasites DATUS M. HAMMOND

In recent years, the use of the electron microscope and in vitro cultivation have resulted in important advances in our knowledge of the development and ecology of intracellular parasites, as well as of their interrelationships with the host. In this paper, I shall be concerned chiefly with the genera Eimeria and Isospora, and other related Sporozoa, especially Toxoplasma, Besnoitia, Sarcocystis, and Plasmodium. The stages beginning with the oocyst or sporozoites and ending with the first-generation merozoite will be emphasized. EXCYSTATION

Supported in part by research grants Ai-07488 from the National Institute of Allergy and Infectious Diseases, United States Public Health Service and GB-8252 from the National Science Foundation. Published as paper no. 997, Utah Agricultural Experiment Station.

In Eimeria species of ruminants in which excystation has been recently studied, two kinds of stimuli, acting in sequence, appear to be necessary to bring about excystation (Jackson 1962; Nyberg and Hammond 1964). The first stimulus, consisting of exposure of the oocysts to carbon dioxide, occurs normally in the rumen. Oocysts altered as a result of response to this stimulus will usually undergo excystation when exposed to the second stimulus, trypsin and bile (Lotze and Leek 1960) or bile salts (Hibbert, Hammond, and Simmons 1969). The carbon dioxide apparently stimulates the activation or the production of an enzyme, or an enzyme rate-limiting step, which causes a change in the permeability of the micropyle (Hibbert and Hammond 1968). Recently, evidence for the occurrence of such an enzyme or enzyme system was observed in an experiment in which E. bovis oocysts were incubated in supernatant fluid previously used for excystation of oocysts and then treated with a trypsin-bile salt mixture (Hibbert and Hammond, in preparation). Some excystation of the oocysts occurred and this took place also when trypsin inhibitor was added to the supernatant. Oocysts incubated in heated supernatant before treatment in trypsin and bile

4

salt did not excyst. Thus, the presence in the fluid of a heat-labile agent which affects excystation is suggested by these results. The oocysts of Eimeria species from chickens are broken mechanically in passing through the gizzard, and the sporozoites escape from the sporocysts when these are exposed to bile and trypsin in the small intestine (Doran and Farr 1962). Oocysts which remain intact apparently do not undergo excystation, although Lotze and Leek (1968, 1969) found that some such oocysts of E. tenella contained active sporozoites when recovered from the small intestine, large intestine, or faeces one to three hours after inoculation. Some of these oocysts had altered walls, possibly as a result of exposure to enzymes or carbon dioxide in the upper digestive tract. E. tenella oocysts responded to exposure to carbon dioxide in a manner similar to that of ruminant coccidia (Nyberg, Bauer, and Knapp 1968 ). Ninety per cent of the oocysts contained active sporozoites after 18 hours of exposure to carbon dioxide and 8 hours of incubation in trypsin and bile. However, we found that little or no complete excystation, i.e., escape of sporozoites from the oocyst as well as the sporocyst, occurred in oocysts of E. acervulina and E. necatrix from chickens, or in two species of Eimeria from ground squirrels, after the oocysts had been exposed to carbon dioxide and then trypsin and bile under the usual conditions (Hibbert and Hammond 1968). Moderate levels of excystation occurred in all four of these species when the percentage of carbon dioxide was increased from 50 to 90, and in the two species from chickens when the duration of exposure to carbon dioxide was extended from 10 to 30 hours or longer. Thus, it appears that oocysts of Eimeria species of ruminants differ from those of chickens and certain rodents in that, in the former, sporozoites evidently escape from the intact oocyst more readily. These species may differ little, if at all, in the conditions under which sporozoites escape from the sporocysts, however. Oocysts of different species are known to excyst at different rates, and in certain species of

DATUS M. HAMMOND

Eimeria from poultry (Farr and Doran 1962) and from ruminants (Hibbert and Hammond 1968) the rate of excystation is correlated with the distance which must be travelled in the digestive tract to reach the usual site of development. In these instances, species developing in the anterior portion or middle of the small intestine excyst faster than species developing in the posterior small intestine or large intestine. The most rapid rate of excystation of sporozoites from sporocysts in any species studied in our laboratory occurs in E. utahensis from the kangaroo rat, Dipodomys ordii. This species of Eimeria has sporocysts with relatively large substiedal bodies. These are plug-like structures lying immediately under the Stieda body, which is a thickened area of the sporocyst wall at one end of the sporocyst. During excystation, the Stieda body appears to swell and then dissolve or disintegrate. The substiedal body is pushed outward, slowly at first, and then explosively, leaving a relatively large opening through which the sporozoites immediately escape. Excystation in E. utahensis occurs within 15 seconds to 2 minutes after exposure of sporocysts to a trypsin-bile salt mixture (Hammond, Ernst, and Chobotar 1970). Such excystation in E. bovis usually requires 35 to 40 minutes. The substiedal body appears to play an important role in excystation in this species and in E. larimerensis and E. callospermophili from the Uinta ground squirrel. In the latter two species, bile or bile salts were observed to stimulate the motility of free sporozoites as well as sporozoites within free sporocysts, but excystation did not occur unless trypsin was present (Roberts, Speer, and Hammond 1970). Little is known about the ability of coccidia to survive as oocysts, free sporozoites, or intracellular stages within hosts such as ground squirrels over the hibernation season. In a recent study (Anderson and Hammond 1969) 20 captive Uinta ground squirrels were inoculated with one of three species of Eimeria 1 to 3 days before they became torpid as a result of their being placed in a chamber kept at 7° c. Oocysts, evidently from the inoculum, were discharged for 8

COCCIDIA AND RELATED INTRACELLULAR PARASITES

to 239 (mean, 182) days during the brief periods of activity which occurred in the torpid animals at intervals of 4 to 10 days. When the squirrels awakened after removal from the cold chamber in April, 16 of the 20 became infected, the intervals between awakening and the first discharge of oocysts being similar in the majority to the normal prépaient period. These findings indicate that the oocysts and/or early developmental stages survived in the torpid subjects from September through April, and suggest that such survival may occur in hibernating squirrels. PENETRATION OF HOST CELLS

Invasion of host cells by living sporozoites has been observed with the light microscope in E. bovis (Payer and Hammond 1967), E. auburnensis (Clark and Hammond 1969), E. ninakohlyakimovae (Kelley and Hammond 1970a), and E. callospermophili and E. bilamellata (Speer, Hammond, and Anderson 1970). In each of these observations, the sporozoite underwent a gliding movement immediately before penetrating the host cell, and began entering the host cell with its anterior end first, with no evident cessation or marked deceleration of the gliding movement. Usually, only a few seconds were required for invasion of the host cell. In E. bovis, when observations were made without a heating stage, sporozoites penetrated host cells more slowly, and sometimes stopped in the process of penetration. In E. ninakohlyakimovae (Kelley and Hammond 1970a) the anterior portion of the sporozoite narrowed to about one-third of the normal body width as it entered the host cell (Fig. 6), the remainder of the body being somewhat wider than usual at this time. The narrowed portion of the sporozoite evidently passed through a relatively small opening in the cell membrane of the host cell. Such an opening, however, can be observed only with the use of the electron microscope. In a similar study of the development of E. callospermophili in cultured cells, a sporozoite-shaped schizont (see below) was seen in

5

the process of penetrating a host cell whose surface membrane was interrupted at the site of entrance (Roberts, Hammond, and Speer 1970). This finding indicates that sporozoite-shaped schizonts and sporozoites of Eimeria enter host cells through a discontinuity in the cell membrane. Because it develops so rapidly, this opening is probably made chiefly by mechanical means. In a later stage of penetration of a host cell by a sporozoite of E. ninakohlyakimovae, the anterior portion of the sporozoite which was within the host cell expanded, but the part of the body passing through the opening in the cell membrane remained constricted (Fig. 7). When the sporozoite was about halfway within the host cell, a minute, sharply pointed protrusion appeared at the anterior end (Fig. 8 ). This protrusion may represent the conoid, which possibly plays a role in penetration. McLaren and Paget ( 1968 ) proposed that the conoid apparatus of E. tenella is extrudable and retractable or includes an extrudable and retractable element. Ryley (1969) found that extruded conoids occurred in negatively stained or shadowed sporozoites of E. tenella. In a scanning electron microscope study, Vetterling and Madden (1969) observed protruding conoids at the anterior ends of excysted sporozoites (which they termed "activated") and suggested that the conoid apparatus is used to penetrate cells. In electron microscope studies of the sporozoites of four Eimeria species, we have observed specimens of free, presumably "activated," sporozoites with conoids in the retracted position as well as in the protruded position (Roberts and Hammond 1970). This indicates that the location of the conoid in excysted sporozoites is more labile than implied by Vetterling and Madden. The paired or club-shaped organelles (also called rhoptries) may play a role in penetration, as suggested by McLaren and Paget ( 1968 ) for E. tenella, but in electron microscope studies of intracellular sporozoites of E. bovis in cell cultures no appreciable alteration in the appearance of these structures as compared with those of ex-

6

tracellular sporozoites was found (Sheffield and Hammond 1968). InBesnoitiajellisoni, organisms rapidly enter host cells in cell cultures (Payer et al. 1969). Sometimes a stylet-like tip was observed at the anterior end of pentrating organisms; the body underwent constriction while passing through the host cell membrane. Jadin and Creemers ( 1968), however, observed injury in the cell membrane and adjacent cytoplasm in a red cell apparently in an early stage of invasion by Toxoplasma, and suggested that a proteolytic substance escaping from the conoid region of the Toxoplasma was digesting a portion of the erythrocyte. Recently, it has been found in an electron microscope study that merozoites of Plasmodium berghei yoeli and of P. gallinaceum begin entering erythrocytes by contacting these with the conoid region (Ladda, Aikawa, and Sprinz 1969). The proper orientation of the Plasmodium merozoite towards the host cell was considered to be a passive event, occurring randomly, whereas the sporozoite of Eimeria species is moving anteriorly at the time penetration is begun, so that its proper orientation towards the host cell is an active event. A depression in the surface of the red cell occurs at the point of contact with the Plasmodium merozoite; the depression deepens to form a cavity as the merozoite enters the cell. When the merozoite is entirely within the cavity, the cell membrane of the host cell pinches together at the surface. The paired organelles decrease in size and density during the early stages of penetration, indicating that they may contain substances that assist in penetration. The host cell membrane remains intact throughout the invasion process, and becomes invaginated to form the lining of the vacuole surrounding the parasite. The manner of penetration of host cells by Plasmodium species might be expected to be different from that by Eimeria species because of differences in the organelles of the anterior region of the body, particularly the conoid. The origin of the membrane lining the parasitophorous vacuole surrounding the intracellular

DATUS M. HAMMOND

sporozoites of Eimeria species is unknown. McLaren (1969) reported that no limiting membrane other than that of the parasite could be observed around merozoites of E. tenella after these had penetrated host cells and assumed an oval shape. Scholtyseck (1969) observed that sporozoites of E. tenella which had just invaded cultured cells were located in a vacuole having an incomplete membrane. These findings indicate that the limiting membrane of the parasitophorous vacuole is not derived directly from the surface membrane of the host cell. DEVELOPMENT OF SPOROZOITES, TROPHOZOITES, AND EARLY SCHIZONTS

During the period immediately after penetration of a host cell by the sporozoite, the refractile bodies, which have also been called eosinophilic globules and paranuclear bodies, undergo marked changes. In E. bovis sporozoites during the first 24 hours in cultured cells, the anterior refractile body frequently appeared to move posteriorly. It usually became smaller as this occurred, and finally disappeared (Payer and Hammond 1969). In a cinemicrographic study, Payer (1969) found that in intracellular sporozoites of E. tenella, E. adenoeides, and E. meleagrimitis which had an anterior and a posterior refractile body, the anterior body moved posteriorly and merged with the posterior body. Finger-like projections which appeared along the anterior margin of the posterior refractile body later became detached, and were observed in the sporozoite cytoplasm. We have recently found that the anterior refractile body of E. callospermophili undergoes a marked decrease in size during the sporozoite's first 10 hours within the host cell and usually can no longer be seen after this time (Speer and Hammond 1970). A succession of small granules, which later become randomly distributed, is formed at the surface of the refractile body. The unusually large posterior refractile body also decreases in size during this time, and later forms several smaller spherical bodies. In a cinemicrographic study of the development of

COCCIDIA AND RELATED INTRACELLULAR PARASITES

this species (Speer and Hammond 1969), refractile bodies underwent frequent changes in shape, size, and location. In addition, the merging of the posterior refractile body with a somewhat smaller refractile body was observed. These changes in the refractile bodies are evidently associated with the use of their substance in the development of the parasite. In E. bovis, sporozoites developing in cell cultures transform into trophozoites during a period which extends from 3 days until 5 to 8 days after inoculation (Payer and Hammond 1967). The nucleus changes from vesicular to compact and approximately doubles in size. The nucleolus becomes greatly enlarged and the posterior refractile body changes from an ellipsoidal to a spheroidal shape. While these changes are being completed, the parasite is transformed from the elongate slender form characteristic of the sporozoite to the rounded form of the trophozoite. Similar nuclear changes were reported to be associated with this transformation in E. meleagrimitis (Doran and Vetterling 1968). E. ninakohlyakimovae sporozoites undergo a similar transformation to trophozoites but it occurs earlier than in E. bovis (Kelley and Hammond 1970a). The change in shape associated with the transformation takes the form of a gradual widening or a lateral outpocketing of the body, usually in the posterior portion. In E. auburnensis, E. callospermophili, and E. bilamellata, the early development follows a different course. The trophozoite stage is usually omitted, although some trophozoites of E. auburnensis were seen in cell cultures; the parasite undergoes considerable growth and completes several nuclear divisions while retaining the elongate form characteristic of the sporozoite. This stage, which we call the sporozoite-shaped schizont, later transforms into a spheroidal schizont by a rapid formation of a lateral outpocketing or by a gradual increase in width of the entire body. This omission of the trophozoite stage is not an abnormality associated with growth in cell cultures because similar development was seen in E. auburnensis (Chobotar, Hammond, and Miner

7

1969) and in E. callospermophili (Roberts, Hammond, Anderson, and Speer 1970) in the host animal. The sporozoite-shaped schizont of E. callospermophili retains the locomotor ability of the sporozoite, as demonstrated by the motility of specimens freed from their host cells by scraping off the monolayer from the coverglass (Speer, Hammond, and Anderson 1970). Such specimens flexed, glided, and entered host cells. A slender protuberance, probably representing the conoid apparatus, was observed with phase-contrast microscopy at the anterior end of the schizont during penetration. In an electron microscope study, sporozoite-shaped schizonts of E. callospermophili were found to have all of the cytoplasmic organelles, including pellicular structures, present in the sporozoite (Roberts, Hammond, Anderson, and Speer 1970). However, the anterior refractile body had disappeared or was represented by small granules. Dedifferentiation of the conoid apparatus, micronemes, and pellicular structures occurred only later, in the spheroidal schizont stage (Fig. 5 ), after the anlagen of merozoites had appeared. This course of development appears to differ considerably from that of merozoites of Plasmodium species, which undergo dedifferentiation of the specialized structures of the pellicle and anterior portion of the body while transforming into the trophozoite stage, so that the trophozoite and early schizont are relatively simple in morphology (Ladda 1969). Such a dedifferentiation accompanies transformation of the sporozoites of E. ninakohlyakimovae into trophozoites (Kelley and Hammond 1970b). NUCLEAR DIVISION

Division of the nucleus of the sporozoite of E. callospermophili has been observed in living specimens in cell cultures (Speer and Hammond 1970) and also with the electron microscope (Roberts and Hammond, in preparation). In living specimens, the first nuclear division of the sporozoite usually occurred 8 to 10 hours after

8

inoculation of sporozoites, and required about one hour for completion. A prominent nucleolus was present throughout the division process. In an early stage of division, the nucleus and nucleolus became elongate, with the latter appearing first spindle-shaped (Fig. 9) and then rodshaped (Fig. 10). In several specimens, the nucleolus and nucleus increased appreciably in length during a period of 3 to 4 minutes, and then assumed a dumb-bell shape. The daughter nucleoli separated and the nuclear membrane became infolded in the area between them (Fig. 11). Thus, the nuclear membrane apparently remains intact during division. However, Canning and Anwar (1969) reported that at the beginning of zygotic meiosis in E. tenella the nucleolus and nuclear membrane disappeared and a spindle was formed. In electron micrographs, a nucleolus could not be seen in sporozoites immediately after they had entered cells, but a prominent nucleolus was present in the enlarged nucleus of sporozoites which apparently had been within the host cell for 6 to 8 hours. A pair of centrioles was seen adjacent to the nucleus in some specimens. Dividing nuclei were greatly elongated, with a narrow middle region, having the Golgi complex in a depression at one side and a rod-shaped nucleolus (Fig. 1 ). In some specimens, a centriole was seen at one pole, and microtubules were present at the lateral margin of the nucleolus. This suggests the possible occurrence of an arrangement of spindle fibres similar to that reported for trypanosomes by Rudzinska and Vickerman (1968 ). In a stage in which division was nearly complete, the daughter nucleoli were visible and the nuclear membrane had an infolding which incompletely separated them (Fig. 2). Clumps of granules, possibly representing chromatin, appeared to have a random distribution in the nucleoplasm. In microgametocytes of E. auburnensis, nuclei in an early stage of division were observed to have an intranuclear spindle apparatus, with fibres radiating from opposite poles of the nucleus (Hammond, Scholtyseck, and Chobotar

DATUS M. HAMMOND

1969). In the later stages of division, during the separation of the two daughter nuclei, the inner membrane of the nucleus became infolded before the outer one. According to Ladda ( 1969 ), nuclear division in Plasmodium species occurs by a form of endomitosis in which chromosomal replication occurs independently, and attachment of chromosomes to nuclear membranes may serve for segregation of chromosomes into daughter nuclei. The nuclear membrane remains intact; the nucleus elongates and divides into two equal parts. Elements interpreted as microtubules and chromosomes are seen before and independent of nuclear division, although chromosomes have not yet been adequately demonstrated. Much work remains to be done before an understanding of nuclear division in this group of parasites can be obtained. LATE DEVELOPMENT OF SCHIZONTS AND FORMATION OF MEROZOITES

In the unusually large first-generation schizonts of E. bovis, E. auburnensis, and E. ninakohlyakimovae, which have a maximum diameter of 200 to 300 /¿m when mature, the nuclei often become arranged near the surface as development proceeds (Hammond, Ernst, and Miner 1966; Chobotar, Hammond, and Miner 1969; Wacha and Hammond 1970). Invaginations or infoldings of the peripheral layer of nuclei then occur. Frequently, these later form spheroidal, ellipsoidal, or tabulated masses having a single layer of nuclei at the periphery. These masses, which have been termed blastophores, are probably comparable to the pseudocytomeres of the schizonts of certain Plasmodium species (Garnham 1951 ; Garnham, Bird, Baker, and Killick-Kendrick 1969). The blastophores have a surface membrane consisting of a single unit membrane derived from that of the sporozoites. The first indications of merozoite formation in E. bovis are thickenings of the surface membrane of the blastophore in the areas overlying nuclei (Sheffield and Hammond 1967). The thickened area of the cell membrane then becomes elevated

COCCIDIA AND RELATED INTRACELLULAR PARASITES

into a conical protuberance and an inner membrane becomes separated from the surface membrane. As the merozoite forms, the inner membrane extends posteriorly, and the conoid and subpellicular microtubules, as well as the spheroidal anlagen of the club-shaped organelles, appear. A nucleus and other organelles are incorporated into the forming merozoite and it finally becomes separated from the residual body. A similar process of merozoite development has been observed in E. nieschulzi by Colley (1968), in E. tenella by McLaren (1969) and Sénaud and Cerna (1968, 1969), in E. magna and E. pragensis by the latter authors (1968, 1969), and in an Isospora species by Schmidt, Johnston, and Stehbens (1967). Scholtyseck ( 1965 ) has described a different kind of merozoite formation in E. perjorans and E. stiedae. In these, the merozoites were formed internally, and were separated from each other and from the remaining cytoplasm of the schizont by spaces developing in the endoplasmic reticulum. In E. callospermophili, the early development of the merozoites in schizonts, presumably of the second generation, resembled that of E. bovis, except that blastophores were not formed (Scholtyseck, Hammond, andTodd 1968). A highly developed endoplasmic reticulum played some part in the separation of the individual schizonts. Recently, we have found that the merozoites in first-generation schizonts of E. callospermophili begin to form internally, instead of at the surface (Roberts, Hammond, Anderson, and Speer 1970). The early stages of formation occur in sporozoite-shaped schizonts having from four to six nuclei. The anlagen of merozoites are observed in association with nuclei having microtubular spindle apparatuses, which are located near the outer margin rather than the centre of the nucleus. Often one or two centrioles are seen adjacent to each pole of the spindle apparatus (Fig. 3 ), near which the anlage of the merozoite later appears (Fig. 4). The anlage consists of the inner membrane of the anterior portion of the merozoite and the immature conoid. No pre-

9

cursor of these structures could be seen. The subpellicular microtubules then appear, arranged in pairs. Somewhat later, the spheroidal anlagen of the club-shaped organelles and the forming refractile body are seen. A Golgi complex is located immediately to one side of the pole of the spindle apparatus, and associated flattened cisternae are present at the base of the forming merozoite. Possibly the Golgi apparatus participates in the formation of certain of the organelles of the merozoite. At a later stage of development, the nucleus divides and each daughter is incorporated into a merozoite (Fig. 5 ). As the merozoites develop, the sporozoite-shaped schizont transforms into a spheroidal schizont. The inner pellicular membrane and anterior end organelles gradually undergo dedifferentiation. The immature merozoites are usually oriented so that they point obliquely towards the surface of the schizont. As the merozoites become larger, they assume a more peripheral position, reach the surface of the schizont, and then grow out as radial protuberances (Fig. 5). In this process, an outer surface membrane is evidently acquired by the merozoite from that of the schizont. The later stages of merozoite formation resemble those of E. bovis, with the merozoites being attached at their posterior ends to a residual body. Two refractile bodies are present in each merozoite (Hammond, Speer, and Roberts 1970). Internal formation of merozoites such as occurs in E. callospermophili has not been reported in any other coccidian. However, the occurrence of a centrocône similar to the pole of the spindle apparatus in E. callospermophili and a centriole associated with merozoite formation was reported in E. pragensis, E. magna, and E. tenella by Sénaud and Cerná (1968,1969). These authors pointed out the resemblance of the centrocône to similar structures in endodyogeny of Toxoplasma gondii (Sénaud 1967; Sheffield and Melton 1968). Merozoite formation in E. callospermophili resembles endodyogeny in T. gondii in that each merozoite begins to form internally, in association with a nucleus in an early stage of division.

10

DATUS M. HAMMOND

The nuclear divisions associated with merozoite formation appear to differ from those occurring during the early development of the schizont with respect to the location of the spindle apparatus; this is peripheral in the former and central, if observed at all, in the latter. Centrioles, sometimes appearing in pairs, occur adjacent to each pole of the peripheral spindle apparatus. Nuclear divisions similar to that associated with merozoite formation have been reported to occur in association with microgamete formation in E. auburnensis (Hammond, Scholtyseck, and Chobotar 1969) and in E. intestinalis and E. magna (Snigirevskaya 1969). In the latter two species, a mitotic spindle, with a peripheral location, was seen in dividing nuclei of microgametocytes, and a centriole was located adjacent to each of the poles. In E. auburnensis, pairs of centrioles were sometimes seen in association with the poles of the peripherally located spindle apparatus. Thus, it appears that a similar pattern of nuclear division is associated in Eimeria species with the development of two different kinds

of daughter individuals, merozoites and microgametes. The poles of the spindle apparatus may play a role in inducing the formation of the merozoites and microgametes. Such an association has been suggested for the formation of sporozoites in Plasmodium gallinaceum (Terzakis, Sprinz, and Ward 1967). Peripheral nuclear fibres associated with sporozoite formation were reported in P. berghei by Vandenberg and Rhodin (1967). Bradbury and Trager (1968) suggested that substances diffusing through nuclear pores from an intranuclear spindle apparatus, or specialized areas of the nuclear membrane itself, may play a role in organizing axonemes during microgametogenesis of Haemoproteus columbae. The centrioles observed in the vicinity of the poles of the spindle apparatus in the Eimeria species may play an intermediate role in inducing the formation of merozoites and microgametes. Kinetosomes were observed in the microgametocytes of P. gallinaceum and P. cathemerium by Aikawa, Huff, and Sprinz ( 1969). Since these were seen

ABBREVIATIONS, ALL FIGURES: ARE, anterior refractile body; c, conoid anlage; CE, centriole; co, clubshaped organelle anlage; G, Golgi complex; IF, intravacuolar folds of the vacuolar membrane; IM, inner pellicular membrane of merozoite anlage; M, mitochondrion; MN, microneme; N, nucleus, NU, nucleolus; PRB, posterior refractile body; PV, parasitophorous vacuole; RB, developing anterior refractile body of merozoite; SA, spindle apparatus.

1 Sporozoite with dividing nucleus. Note elongate shape of nucleus and nucleolus, and intact nuclear membrane. From culture fixed 12 hours after inoculation. X I 1,000 2 Portion of sporozoite with nucleus almost completely divided. Note infolding of nuclear membrane between the two nucleoli. From culture fixed 14 hours after inoculation. X 11,000 3 Portion of sporozoite-shaped schizont, showing nucleus with spindle apparatus near its outer margin and three longitudinally sectioned centrioles (two adjacent to one pole and one adjacent to the other). From culture fixed 16 hours after inoculation. X I 5,000 4 Portion of sporozoite-shaped schizont, showing anlagen of two merozoites, each adjacent to a pole of a spindle apparatus. Note centrioles in anlage at left and Golgi complex at base of anlage at right. From culture fixed 16 hours after inoculation; fixative, 2.5 per cent glutaraldehyde and 2.5 per cent osmium tetroxide together. X 15,000

FIGURES 1-4. Electron micrographs of Eimeria callospermophili sporozoites and schizonts grown in embryonic bovine intestinal cells, fixed according to the method of Karnovsky ( 1965), embedded in Epon, sectioned with a diamond knife and Sorvall ultramicrotome, and stained with uranyl acetate and lead citrate, unless otherwise stated. Prepared by William L. Roberts with a Zeiss 9A electron microscope.

12

in proximity to darkened areas in the nucleus, it was suggested that the nucleus may play some role in their formation. Structures resembling centrioles were seen in young oocysts of P. berghei yoeli and microtubules thought to be spindle fibrils were observed at the periphery of the nucleus in this stage and in ookinetes of this species by Garnham et al (1969).

DATUS M. HAMMOND

Usually, mature merozoites show little or no motility while retained within the parasitophorous vacuole of the host cell in which they developed. However, moving merozoites were seen within a schizont of E. auburnensis in a cell culture (Clark and Hammond 1969). Occasionally, gliding movements and motion of the anterior ends of merozoites in a lateral direction were seen within schizonts of E. callospermophili.

Merozoites of this species (Speer, Hammond, and Anderson 1970) and of E. alabamensis (Sampson, Hammond, and Ernst 1971) were observed leaving the parasitophorous vacuole by their own motility. In the latter, merozoites often left in twos. In an electron micrograph showing a merozoite of E. callospermophili in the process of escaping, the body of the merozoite was constricted at the point of exit, in a manner similar to that of sporozoites while entering cells (Roberts, Hammond, Anderson, and Speer 1970). In E. ninakohlyakimovae, merozoites which had penetrated new host cells were markedly constricted as they left these cells (Kelley and Hammond 1970a). These findings suggest that merozoites enter and leave cells in a similar manner to sporozoites. Recently, we have found that merozoites in mature schizonts, as well as free merozoites, of E. callospermophili and other species in cell cultures become motile when stimulated by bile or

FIGURES 5, 9—11. E. callospermophili, continued. FIGURES 6—8. E. ninakohlyakimovae. 5 Electron micrograph of spheroidal schizont, showing outgrowth of two merozoites. Note that inner pellicular membrane of schizont has almost completely dedifferentiated by this stage. From a ligated intestinal segment of a ground squirrel fixed 16 hours after introduction of sporozoites; prepared as for Figure 4. X 12,000 6-8 Photomicrographs by G. L. Kelley of a living specimen, phase-contrast microscopy, in process of penetrating host cell, with constriction of sporozoite at point of passage through cell membrane (arrow). In each figure, anterior end of sporozoite is oriented towards top of page. From embryonic lamb thymus cell culture, two days after inoculation. XI,500 6 Sporozoite shortly after beginning penetration of cell, with constricted anterior portion of body. 7 Anterior fourth of sporozoite within cell; this portion has become wider, so that constriction at point of entrance is evident.

8 Penetration about half completed; posterior ref ractile body is included in the constricted portion of the body; note proturberance near anterior end of body. 9-11 Photomicrographs by C. A. Speer of intracellular sporozoites undergoing nuclear division, phase contrast microscopy. X 1,600 9 Sporozoite with relatively wide sporozoite body, elongated nucleus, and spindle-shaped nucleolus (arrow). From culture of third-passage embryonic bovine kidney cells, 8 hours after inoculation. 10 Sporozoite with nucleus in stage comparable to that of Figure 1, having rod-shaped nucleolus (arrow) lying in the narrow middle region of the elongated nucleus. From culture of third-passage embryonic bovine kidney cells, 8^2 hours after inoculation. 11 Sporozoite with nucleus in stage comparable to that of Figure 2, showing invagination of nuclear membrane (arrow) between the two nucleoli. From culture of fourth-passage embryonic bovine intestinal cells, 9 hours after inoculation.

MOTILITY AND FURTHER DEVELOPMENT OF MEROZOITES

14

bile salts (Speer, Hammond, and Kelley 1970). The merozoites within mature schizonts usually undergo a marked increase in motility when bile or a bile salt is added to the coverslip preparation, and soon begin leaving the host cell. Motility of free merozoites, previously released by rupture of the host cells, is observed with much greater frequency than normal in preparations to which bile or bile salts have been added. The stimulation of motility in merozoites in this way has also been observed in E. bilamellata and E. larimerensis, also from the Uinta ground squirrel, as well as E. nieschulzi from the rat and E. ninakohlyakimovae from the sheep. If this is found to hold true more widely in the coccidia, it may represent an important ecological aspect of the host-parasite relationship. In the species of Eimeria from ruminants and rodents in which in vitro cultivation has been attempted in our laboratory, the merozoites apparently have much less capacity to develop than do the sporozoites. In all of these species, sporozoites have developed to mature first-generation schizonts, but so far we have been unsuccessful in obtaining development of merozoites to mature second-generation schizonts in cell cultures. This is also true of results obtained when merozoites of E. bovis collected from experimentally infected calves were used (Hammond, Payer, and Miner 1969 ). Except for a single instance of a binucleate second-generation schizont of E. bovis (Hammond and Fayer 1968), further development of merozoites has been observed only in E. callospermophili. Small numbers of multinucleate second-generation schizonts of this species, some of which were in the early stages of merozoite formation, were seen (Speer, Hammond, and Anderson 1970). These results differ from those of Bedrnik ( 1969), who found that merozoites of E. tenella obtained from chickens developed, in general, in tissue culture much better than did sporozoites. He noted, however, that merozoites obtained from chickens on the sixth day of infection grew best, indicating that later generations of merozoites may have better in vitro growth potential than those of the first

DATUS M. HAMMOND

generation. Bedrnik (1967) reported the development of gametocytes and oocysts of E. tenella from second-generation merozoites, and Strout and Ouellette (1969) obtained mature microgametocytes and macrogametes of this species in cell cultures inoculated with sporozoites. HOST-PARASITE RELATIONSHIPS

Each intracellular stage of Eimeria species lies in a vacuole which has been termed the parasitophore Vakuole (Scholtyseck and Piekarski 1965) or "parasitophorous vacuole" (Hammond, Scholtyseck, and Miner 1967) and "periparasitic vacuole" (Stehbens 1966). The membrane of the host cell lining this vacuole is usually smooth, but in some species it has numerous fine folds or villus-like structures protruding into the vacuole. In E. auburnensis macrogametes, these intravacuolar folds evidently disintegrate, forming particulate material, which may be taken into the parasite (Hammond, Scholtyseck, and Chobotar 1967). Similar, but somewhat smaller folds or villus-like structures are found along the membrane lining the parasitophorous vacuole of schizonts of E. callospermophili (Fig. 1), E. nieschulzi (Colley 1968), andE. miyairii (Andreassen and Behnke 1968). In Adelina tribolii, the vacuolar membrane protrudes into the vacuole, forming vesicles and membranous structures, which may assist in the nutrition of the parasite (Zizka 1969). Similar vesicles and membranes were found in the vacuoles of Lankesterella hylae (Stehbens 1966) andinMyno sporides amphiglenae (Henneré 1967). In the former, this material was evidently taken into the parasite by engulfment or surface activity of the pellicle. Stereocilia, at least some of which are outgrowths of the vacuolar membrane, were observed in the vacuoles surrounding T. gondii organisms (Sheffield and Melton 1968). The vacuolar membrane surrounding the large first-generation schizont of E. bovis has numerous blebs, formed apparently in the process of pinching off to form free vesicles, which evidently

COCCIDIA AND RELATED INTRACELLULAR PARASITES

disintegrate in the vacuole (Sheffield and Hammond 1966). This observation suggests the transfer of material from the host cell into the vacuole, from which it is probably taken in for use by the parasite. Andreassen and Behnke (1968) observed small vesicles in the cytoplasm of host cells harbouring E. miyairii schizonts. These were often near the vacuolar membrane, and sometimes protruded into it. The suggestion was made that these might empty into the vacuole, providing a means of nourishment for the parasite. Other structures occurring at the surface of the vacuolar membrane or at the surface of the parasite have been discussed by Scholtyseck (1968) and by Scholtyseck, Volkmann, and Hammond (1966). The mechanism of ingestion of nutrients by the parasite is still incompletely known. The occurrence of the invaginations in the surface of the parasite, variously called micropores, micropyles, or cytostomes, is widespread among Eimeria species and in Toxoplasma, Sarcocystis, Besnoitia, Plasmodium, and other genera. Evidence of ingestion of nutrients through these structures has been obtained in only a few Eimeria species. Snigirevskaya and Cheissin (1968) reported the occurrence of active micropores in all of the endogenous stages with the exception of the microgamete in E. intestinalis. Scholtyseck (1969) observed micropores with food vacuoles forming at their bases in sporozoites of E. tenella within cultured cells. He found that the cytoplasm of the host cell in the vicinity of the parasite undergoes disintegration soon after invasion of the host cell by the parasite, and that this material is included in the parasitophorous vacuole and ingested through the micropores. Sénaud and Cerná (1968) reported the occurrence of active micropores in the merozoites of E. pragensis. We have seen such micropores also in the intracellular sporozoites of E. alabamensis (Sampson and Hammond 1970). In each of these instances, the micropores evidently function as cytostomes. Ingestion of nutrients through a cytostome has also been reported for Besnoitia ¡ellisoni (Sheffield 1967) andiorPlasmodium

15

species of birds and mammals other than rodents (Ladda 1969). A branched cytostome was observed in Sarcocystis by Sénaud (1966). In other species, the micropores may be vestigial or may function in some manner as yet unknown. In the microgametocytes of E. auburnensis, micropores were seen to occur in clusters of as many as nine (Hammond, Scholtyseck, and Chobotar 1969). Ingestion of nutrient materials may occur by pinocytosis in the macrogametes of E. auburnensis, usually in association with vshaped invaginations at the surface (Hammond, Scholtyseck, and Chobotar 1967). The factors affecting susceptibility of hosts to coccidia are incompletely understood. It is known, however, that older chickens are more susceptible than young ones to certain Eimeria species (Long 1967; Rose 1967), that the breed or strain of chickens also influences susceptibility to Eimeria infections (Long 1968), and that strains of parasites differ in pathogenicity ( Joyner and Norton 1969). Changes which have been observed in host cells harbouring stages of Eimeria species include increases in the number of mitochondria in cells parasitized by macrogametes of E. perforans (Scholtyseck 1963). This indicates a higher than normal rate of metabolism in infected cells. The mitochondria later undergo degeneration. Host cells parasitized by several diferent Eimeria species undergo marked changes, including enlargement of the nucleus, a marked increase in size of the nucleolus, and rearrangement of the chromatin into finer masses than normal. The changes are most pronounced in host cells harbouring large schizonts such as those of E. bovis, E. auburnensis, and E. ninakohlyakimovae\ in the two former species the cytoplasm of the host cell also increases markedly in volume. Evidently, substances originating in the parasite stimulate the host to undergo these changes, which may be associated with production by the host cell of materials used in the growth of the parasite. Thus, parasites of this kind are evidently able to cause modifications in the physiology and growth of the host cell. Such

16

changes presumably result in a more favourable environment for the development of the parasite.

DATUS M. HAMMOND

Doran, D. J., and Vetterling, J. M. 1968. Survival and development of Eimeria meleagrimitis Tyzzer, 1929, in bovine kidney and turkey intestine cell cultures. /. Protozool. 15:796-802 Farr, M. M., and Doran, D. J. 1962. Comparative excystation of 4 species of poultry coccidia. /. REFERENCES Protozool. 9:403-7 Aikawa, M., Huff, C. G., and Sprinz, H. 1969. Payer, R. 1969. Refractile body changes in sporoComparative fine structure study of the gametozoites of poultry coccidia in cell culture. Proc. cytes of avian, reptilian, and mammalian malarial Helminthol. Soc. Wash., D.C. 36:224-31 parasites. J. Vltrastruct. Research 26:316-31 Fayer, R., and Hammond D. M. 1967. DevelopAnderson, L. C., and Hammond, D. M. 1969. Exment of first-generation schizonts of Eimeria perimental coccidian infections in captive Uinta bovis in cultured bovine cells. /. Protozool. ground squirrels, Spermophilus armatus, during 14:764-72 the hibernating season. /. Protozool. 16(Suppl.) : - 1969. Morphological changes in Eimeria bovis 15-16 sporozoites during their first day in cultured Andreassen, J., and Behnke, O. 1968. Fine structure mammalian cells. /. Parasitai. 55:398-401 of merozoites of a rat coccidian Eimeria miyairii, Fayer, R., Hammond, D. M., Chobotar, B., and with a comparison of the fine structure of other Eisner, Y. Y. 1969. Cultivation of Besnoitia jelsporozoa. J. Parasitai. 54:150-63 lisoni in bovine cell cultures. /. Parasitai. 55: Bedrnik, P. 1967. Development of sexual stages 645-53 and oocysts from the second generation of EimGarnham, P. C. C. 1951. The mosquito transmiseria tenella merozoites in tissue cultures. Folia sion of Plasmodium inui Halberstaedter and Parasitai 14:364 Prowazek, and its pre-erythrocytic development - 1969. Some results and problems of cultivation in the liver of the rhesus monkey. Trans. Roy. of Eimeria tenella in tissue cultures. Acta Vet. Soc. Trop. Med. Hyg. 45:45-52 38:31-5 Garnham, P. C. C., Bird, R. G., Baker, J. R., Bradbury, P. C., and Trager, W. 1968. The fine Desser, S. S., and El-Nahal, H. M. S. 1969. structure of microgametogenesis in HaemoproElectron microscope studies on motile stages of teus columbae Kruse. /. Protozool. 15:700-12 malaria parasites, vi. The ookinete of PlasmoCanning, E. U., and Anwar, M. 1969. Nuclear dium berghei yoelii and its transformation into studies on sporozoan oocysts. Progr. in Protothe early oocyst. Trans. Roy. Soc. Trop. Med. zool., Proc. HI Intern. Congr. Protozool., LeninHyg. 63:187-94 grad, p. 23 Garnham, P. C. C., Bird, R. G., Baker, J. R., and Chobotar, B., Hammond, D. M., and Miner, M. L. Killick-Kendrick, R. 1969. Electron microscope 1969. Development of first-generation schizonts studies on the motile stages of malaria parasites. of Eimeria auburnensis. J. Parasitai. 55:385-97 vu. The fine structure of the merozoites of exoClark, W. N., and Hammond, D. M. 1969. Develerythrocytic schizonts of Plasmodium berghei opment of Eimeria auburnensis in cell cultures. yoelii. Trans. Roy. Soc. Trap. Med. Hyg. 63 : /. Protozool. 16:646-54 328-32 Colley, F. C. 1968. Fine structure of schizonts and Hammond, D. M., Ernst, J. V., and Chobotar, B. merozoites of Eimeria nieschulzi. J. Protozool. 1970. Composition and function of the substiedal 15:374-82 body in the sporocysts of Eimeria utahensis. J. Doran, D. J., and Farr, M. M. 1962. Excystation Parasitai. 56:618-19 of the poultry coccidium, Eimeria acervulina. Hammond, D. M., Ernst, J. V., and Miner, M. L. J. Protozool. 9:154-61 1966. The development of first-generation schi-

COCCIDIA AND RELATED INTRACELLULAR PARASITES

zonts of Eimeria bovis. J. Protozool. 13:559-64 Hammond, D. M., and Payer, R. 1968. Cultivation of Eimeria bovis in three established cell lines and in bovine trachéal cell line cultures. /. Parasitai. 54:559-68 Hammond, D. M., Payer, R., and Miner, M. L. 1969. Further studies on in vitro development of Eimeria bovis and attempts to obtain secondgeneration schizonts. /. Protozool. 16:298-302 Hammond, D. M., Scholtyseck, E., and Chobotar, B. 1967. Fine structures associated with nutrition of the intracellular parasite Eimeria auburnensis. J. Protozool. 14:678-83 - 1969. Fine structural study of the microgametogenesis of Eimeria auburnensis. Z. Parasitenk. 33:65-84 Hammond, D. M., Scholtyseck, E., and Miner, M. L. 1967. The fine structure of microgametocytes of Eimeria perforans, E. stiedae, E. bovis, and E. auburnensis. J. ParasitoL 53:235—47 Hammond, D. M., Speer, C. A., and Roberts, W. 1970. Occurrence of refractile bodies in merozoites of Eimeria species. /. ParasitoL 56:189-91 Henneré, E. 1967. Etude cytologique des premiers stades du développement d'une coccidie: Myriosporides amphiglenae. J. Protozool. 14:27—39 Hibbert, L. E., and Hammond, D. M. 1968. Effects of temperature on in vitro excystation of various Eimeria species. Exptl. ParasitoL 23:161-70 Hibbert, L. E., Hammond, D. M., and Simmons, J. R. 1969. The effects of pH, buffers, bile and bile acids on excystation of sporozoites of various Eimeria species. J. Protozool. 16:441—4 Jackson, A. R. B. 1962. Excystation of Eimeria arloingi (Marotel, 1905) ; stimuli from the host sheep. Nature 194:847-9 Jadin, J., and Creemers, J. 1968. Ultrastructure et biologie des toxoplasmes. m. Observations de toxoplasmes intraerythrocytaires chez un mammifère. Acta Trop. 25:267-70 Joyner, L. P., and Norton, C. C. 1969. A comparison of two laboratory strains of Eimeria tenella. Parasitology 59:907-13 Karnovsky, M. J. 1965. A formaldehyde-glutaraldehyde fixative of high osmolality for use in electron microscopy. /. Cell BioL 27:137A-8A

17

Kelley, G. L., and Hammond, D. M. 1970a. Development of Eimeria ninakohlyakimovae in cell cultures. /. Protozool. 17:340-9 - 1970b. The fine structure of the early stages of Eimeria ninakohlyakimovae in cell cultures. /. Protozool. 17(Suppl.):17 Ladda, R. L. 1969. New insights into the fine structure of rodent malarial parasites. Military Med. 134:825-65 Ladda, R. L., Aikawa, M., and Sprinz, H. 1969. Penetration of erythrocytes by merozoites of mammalian and avian malarial parasites. J. ParasitoL 55:633-44 Long, P. L. 1967. Studies on Eimeria praecox Johnson, 1930, in the chicken. Parasitology 57:351-61 — 1968. The effect of breed of chickens on resistance to Eimeria infections. Brit. Poult. Sci. 9:71-8 Lotze, J. C., and Leek, R. G. 1960. Some factors involved in excystation of the sporozoites of three species of sheep coccidia. /. ParasitoL 46(Suppl.):46-7 - 1968. Excystation of the sporozoites of Eimeria tenella in apparently unbroken oocysts in the chicken. /. Protozool. 15:693-7 - 1969. Observations on Eimeria tenella in feces of inoculated chickens. /. Protozool. 16:496-8 McLaren, D. J. 1969. Observations on the fine structural changes associated with schizogony and gametogony in Eimeria tenella. Parasitology 59:563-74 McLaren, D. J., and Paget, G. E, 1968. A fine structural study on the merozoite of Eimeria tenella with special reference to the conoid apparatus . Parasitology 58:561-71 Nyberg, P. A., Bauer, D. H., and Knapp, S. E. 1968. Carbon dioxide as the initial stimulus for excystation of Eimeria tenella oocysts. /. Protozool. 15:144-8 Nyberg, P. A., and Hammond, D. M. 1964. Excystation of Eimeria bovis and other species of bovine coccidia. /. Protozool. 11:474-80 Roberts, W. L., and Hammond, D. M. 1970. Ultrastructural and cytological studies of sporozoites of four Eimeria species. /. Protozool. 17:76-86

18 Roberts, W. L., Hammond, D. M., Anderson, L. C., and Speer, C. A. 1970. Ultrastructural study of schizogony in Eimeria callospermophili. J. Protozool 17:584-92 Roberts, W. L., Hammond, D. M., and Speer, C. A. 1970. Ultrastructural study of the intra- and extracellular sporozoites of Eimeria callospermophili. J. Parasitai. 56:907-17 Roberts, W. L., Speer, C. A., and Hammond, D. M. 1970. Electron and light microscope studies of the oocyst walls, sporocysts and excysting sporozoites of Eimeria callospermophili and E. larimerensis. J. Parasitai. 56:918-26 Rose, M. E. 1967. The influence of age of host on infection with Eimeria tenella. J. Parasitai. 53 : 924-9 Rudzinska, M. A., and Vickerman, K. 1968. The fine structure. In Infectious Blood Diseases of Man and Animals, ed. D. Weinman and M. Ristic, 1:217-306. New York: Academic Press Ryley, John F. 1969. Ultrastructural studies on the sporozoite of Eimeria tenella. Parasitology 59: 67-72 Sampson, J. R., and Hammond, D. M. 1970. The fine structure of the early stages of Eimeria alabamensis. J. Protozool. 17(Suppl.) :17 Sampson, J. R., Hammond, D. M., and Ernst, J. V. 1971. Development of Eimeria alabamensis from cattle in mammalian cell cultures. /. Protozool., in press Schmidt, K., Johnston, M. R. L., and Stehbens, W. E. 1967. Fine structure of the schizont and merozoite of Isospora sp. (Sporozoa: Eimeriidae) parasitic in Gehyra varié gata (Dumeril and Bibron, 1836) (Reptilia: Gekkonidae). /. Protozool. 14:602-8 Scholtyseck, E. 1963. Elektronenmikroskopische Untersuchungen über die Wechselwirkung zwischen dem Zellparasiten Eimeria perforans und seiner Wirtszelle. Z. Zellforsch. 61:220-30 - 1965. Elektronenmikroskopische Untersuchungen über die Schizogonie bei Coccidien (Eimeria perforans und E. stiedae). Z. Parasitenk. 26:50— 62 - 1968. Neue Einblicke in die Parasit-Wirt-Beziehungen mit Hilfe der Elektronenmikroskopie. Z. Parasitenk. 31:67-84

DATUS M. HAMMOND

- 1969. Electron microscope studies of the effect upon the host cell of various developmental stages of Eimeria tenella in the natural chicken host and in tissue cultures. Acta Vet. 38:153-6 Scholtyseck, E., Hammond, D. M., and Todd, K. S., Jr. 1968. Electron microscope studies of the schizonts and merozoites of an Eimeria species from Uinta ground squirrels, Citellus armatus. J. Protozool. 13(Suppl.):18 Scholtyseck, E., and Piekarski, G. 1965. Elektronenmikroskopische Untersuchungen an Merozoiten von Eimerien (Eimeria perforans und E. stiedae) und Toxoplasma gondii. Z. Parasitenk. 26:91-115 Scholtyseck, E., Volkmann, B., and Hammond, D. M. 1966. Spezifische Feinstrukturen bei Parasit und Wirt als Ausdruck ihrer Wechselwirkungen am Beispiel von Coccidien. Z. Parasitenk. 28 : 78-94 Sénaud, J. 1966. L'ultrastructure du micropyle de Toxoplasmasida. Compt. rend. 262:119-21 - 1967. Contribution à l'étude des sarcosporidies et des toxoplasmes (Toxoplasmea). Protistologica 3:167-232 Sénaud, J., and Cerna, 2. 1968. Etude en microscopic électronique des merozoites et de la merogonie chez Eimeria pragensis (Cerna et Sénaud 1968 ), coccidie parasite de l'intestin de la souris (Mus musculus). Ann. Sta. Biol. Besse-enChandesse 3:221-42 - 1969. Etude ultrastructurale des merozoites et de la schizogonie des coccidies (Eimeriina) : Eimeria magna (Perard 1925) de l'intestin des lapins et E. tenella (Railliet et Lucet, 1891) des coecums des poulets. /. Protozool. 16:155-65 Sheffield, H. G. 1967. The function of the micropyle in the cyst organisms of Besnoitia jellisoni. J. Parasitai. 53:888 Sheffield, H. G., and Hammond, D. M. 1966. Fine structure of first-generation merozoites of Eimeria bovis. J. Parasitai. 52:595-606 - 1967. Electron microscope observations on the development of first-generation merozoites of Eimeria bovis. J. Parasitai. 53:831-40 - 1968. Electron microscope observations on sporozoites of Eimeria bovis in cultured bovine kidney cells. /. Protozool. 15(Suppl.) : 18

COCCIDIA AND RELATED INTRACELLULAR PARASITES

Sheffield, H. G., and Melton, M. L. 1968. The fine structure and reproduction of Toxoplasma gondii. J. Parasitai. 54:209-26 Snigirevskaya, E. S. 1969. Changes in some ultrastructures during microgametogenesis in the rabbit coccidia Eimeria intestinalis and E. magna. Tsitólogiya 11:382-5 (In Russian) Snigirevskaya, E. S., and Cheissin, E. M. 1968. The role of the micropore in the nutrition of endogenous developmental stages of Eimeria intestinalis (Sporozoa, Coccidia). Tsitologiya 10: 940-4 (In Russian) Speer, C. A., and Hammond, D. M. 1969. Cinemicrographic observations on the development of Eimeria callospermophili in cultured cells. /. Protozool. 16(Suppl.):16 - 1970. Nuclear divisions and refractile body changes in sporozoites and schizonts of Eimeria callospermophili in cultured cells. /. Parasitol. 56:461-7 Speer, C. A., Hammond, D. M., and Anderson, L. C. 1970. Development of Eimeria callospermophili and E. bilamellata from the Uinta ground squirrel, Spermophilus armatus, in cultured cells. J. Protozool. 17:274-84 Speer, C. A., Hammond, D. M., and Kelley, G. L. 1970. Stimulation of motility in merozoites of five Eimeria species by bile salts. /. Parasitol. 56:927-9 Stehbens, W. E. 1966. The ultrastructure of Lankesterella hylae. J. Protozool. 13:63-73 Strout, R. G., and Ouellette, C. A. 1969. Gametogony of Eimeria tenella (coccidia) in cell cultures. Science 163:695-6 Terzakis, J. A., Sprinz, H., and Ward, R. A. 1967. The transformation of the Plasmodium gallinaceum oocyst in Aedes aegypti mosquitoes. /. CellBiol 34:311-26 Vandenberg, J., and Rhodin, J. 1967. Differentiation of nuclear and cytoplasmic fine structure during sporogonic development of Plasmodium berghei. J. CellBiol. 32:C7-C10 Vetterling, J. M., and Madden, P. A. 1969. Ultrastructure of the conoid apparatus of dormant and activated sporozoites of Eimeria. Progr. and Abstr., 44th Ann. Meet. Am. Soc. Parasitol., pp. 73-4

19

Wacha, R. S., and Hammond, D. M. 1970. The development of the endogenous stages of Eimeria ninakohlyakimovae in domestic sheep. /. Parasitol. 56(4, sec. 2):355 Zizka, Z. 1969. The fine structure of the macrogametocytes of Adelina tribolii Bhatia, 1937 (Eucoccidia, Telosporea) from the fat body of the beetle Tribolium castaneum Hbst. /. Protozool. 16:111-20

Discussion W. C. MARQUARDT

Department of Zoology, Colorado State University, Fort Collins, Col., USA Two aspects of the ecology and the physiology of coccidia justify consideration. The oocyst is passed from a host with a uninucleated sporoplasm. Development occurs over a period of a day to a week, depending on species and temperature. The temperature at which development occurs is approximately 16 to 30° c; molecular oxygen is required also. Other environmental factors have little effect. For example, oocysts are routinely allowed to sporulate in 2 per cent potassium dichromate, which is hardly a non-toxic solution; they will develop also in 2 per cent formalin or 1 per cent sulphuric acid. They are susceptible to relatively low concentrations of mercuric chloride or iodine. We know little about the mechanisms involved in either the entrance or exclusion of these substances. The respiration of the oocysts has been studied by Wilson and Fairbairn (/. Protozool. 8 [1961], 41 CM 6) and more recently by Wagenbach and Burns (/. Protozool. 16 [1969], 257-63), and it is well established that respiration drops to unmeasureable levels after sporulation. In three species of poultry coccidia it was found that the respiration rate of the organisms increases dramatically upon receipt of the stimuli which Dr Hammond has described. Substances which act on the sporozoite have been described, but we have no information on their action at the cellular level. These organisms may lie dormant for as long as four years; yet when

20

they receive the stimulus the respiration immediately increases and excystation occurs. It is of interest, too, that excystation will occur in almost any host, as shown by Justin Andrews in 1927, although development is limited to a narrow spectrum of hosts. In addition, the organisms are specific to certain cells within the animals which are suitable hosts. Eimeria may use, for example, epithelial, endothelial, and epithelial cells in that order in completing the life cycle. Eimeria bovis, which Dr Hammond has discussed, uses first endothelial cells and then epithelial cells for the rest of its life cycle. Such highly refined tastes seem to be lost when these organisms are placed in tissue culture. The preference, for example, of certain of the bovine coccidia for bovine embryonic trachea cells is difficult to understand, considering the specificity that the organisms have. Eimeria meleagrimitis of turkeys seems to develop better in bovine kidney than in turkey intestinal cell cultures. It has been said that tissue culture cells dedifferentiate and become more generalized and therefore better able to serve as hosts for the coccidia. For me, this is no answer, and it makes things more confusing in the light of what we know about in vivo development. Dr Hammond and other investigators have done fine work on coccidial immunity. There is, for example, a species-specific immunity of a high order. Immunization of one species of coccidia provides little cross-immunity. There are a few humoral antibodies effective against merozoites in vzirabut they are ineffective in vivo. Lastly, in young animals the clinical symptoms of coccidiosis are directly related to the size of the inoculum. However, coccidiosis is a problem in older animals where apparently the contamination of the area is not related to the severity of the clinical disease. We know little about the interreactions that occur here.

DATUS M. HAMMOND

Epidemiological considerations of the leishmanias with particular reference to the New World R. LAINSON and J. J. SHAW

A recent bibliography (Anon. 1967b) lists some 4,500 references to Leishmania and leishmaniasis since the parasite was first seen (Cunningham 1885), and the literature is expanding rapidly with recent renewed interest in the subject. Whether regarded as comprising different species, subspecies, or mere biological races, the genus Leishmania enjoys a wide geographical range in most tropical and subtropical areas and extends from Central and South America through Europe, Africa, Egypt, the Sudan, southern Russia, India, and China. Broadly speaking, the disease is divisible into visceral and cutaneous infections, although this division is by no means always maintained. The visceral disease (kala-azar ) is preceded by an initial cutaneous lesion and may later evolve into extensive cutaneous involvement (e.g. "post kala-azar" in man and the generalized skin infections of dogs and foxes). Conversely, dogs naturally infected with L. trópica may show parasites in the internal organs; and, in experimental animals, infections with other parasites normally producing only cutaneous lesions may spread to the viscera (L. mexicana and L. braziliensis sensu latu in mice and hamsters). It is clearly impossible to cover the epidemiology of leishmaniasis in both the Old and the New World in a short survey. A brief review of the more important features in the Old World is necessary, however, to make the undoubtedly more complex situation which exists in the Americas more understandable.

"KALA-AZAR" OR VISCERAL LEISHMANIASIS (L. donovani) Following the first description of the disease by Leishman (1903, 1904) in India, kala-azar was soon found to have a remarkably wide distribution in the Old World; to give but the major areas, it is now well known in China, Asiatic

22

Russia, the whole of the Mediterranean, the Sudan, and East Africa. Morphologically the parasite responsible for the disease in these areas appears identical and, with few exceptions, the general pathology in man is the same. There are, however, quite considerable differences in the epidemiology of kala-azar in some endemic regions and this has led to disagreement as to whether the name L. donovani should be used in all instances. Indian kala-azar It is appropriate to start our review with the situation in India, where the disease has exacted such a heavy toll of life in the past and where much of the pioneer study was carried out. Here, kala-azar appears to be essentially a human, "domestic" disease. There are two important explanations for this. First, the parasite is readily demonstrable in the peripheral blood of the febrile patient and is thus available for the sandfly vector, Phlebotomus argentipes. Secondly, P. argenupes has developed a close association with man, breeding in moist soil in the cracks of floors, walls, and surrounds of houses, stables, etc. Under such circumstances kala-azar passes easily from man to man and there is little "need" for reservoir-hosts. Indeed, no other source of infection has been found in spite of exhaustive examination of dogs, cats, and other animals, but whether or not there once existed such a zoonosis is still a matter for further investigation. Mediterranean kala-azar The situation regarding kala-azar in the Mediterranean and the Caspian regions is so different from that in India that the parasite is frequently referred to by the separate name of L. infantum (as the name implies, the disease is particularly prevalent in young children although by no means confined to them). First, the parasite is not commonly found in the peripheral blood and the infection rate of sandflies feeding on man is usually very low. Secondly, and in striking contrast with the situation in India, Mediterranean

R. LAINSON AND J. J. SHAW

kala-azar might well be regarded as a disease principally of dogs : indeed the incidence is much higher in these animals than in man. Unlike man, the dog may frequently harbour a chronic infection and, furthermore, tends to develop a generalized skin infection with abundant amastigotes scattered throughout the dermis. As a result, the sandfly vectors may show infection rates almost to the 100 per cent level after feeding on infected dogs. The dog, then, is the major host and the source of the disease for man, who probably plays little part in the maintenance of the parasite in nature. Dogs act as similar reservoirs of kalaazar in other geographic areas, including East China, Central Asia, and Transcaucasia. East African and Sudanese kala-azar The wide distribution of the genus Leishmania and the existence of certain species infecting reptiles suggest an extremely ancient origin and indicate that we should look further than the relatively modern man-dog relationship in the evolution of leishmaniasis. L. donovani has, in fact, been isolated from jackals and foxes in the Old World (Central Asia, France, and Transcaucasia) and, even more interestingly, there is increasing evidence of the role of rodents in the epidemiology of kalaazar. Thus, in Kenya, infected gerbils and ground squirrels have been found in the apparent absence of the canine disease (Heisch 1957, 1963; Heisch, Grainger, and Harvey 1959). In the Sudan L. donovani has been isolated from the rodents Rattus rattus, Acomys albigena, and Arvicanthis niloticus, again without evidence that the dog is at all concerned in the epidemiology (Hoogstraal and Dietlein 1964), although isolated infections were also recorded in two other carnivores, Genetta genetta and Felis serval. It may be, then, that wild rodents represent the more primitive source of L. donovani, one that is still maintained in some areas such as East Africa and the Sudan, and that the Canidae-man relationship is a relatively recent association. In

FIGURES 1-4. Visceral leishmaniasis in Brazil. 1 2

Endemic area in Ceará State, northeastern Brazil. Naturally infected dogs.

3 Typical riverside forest in Abaetetuba region, Para State, in which human cases of kala-azar have occurred. 4 The fox, Cerdocyon thous, found naturally infected with L. donovani in Belém, Para (from Lainson, Shaw, and Lins 1969).

24

support of this theory it may be noted that wild rodents appear to have acquired a well-established tolerance to L. donovani, whereas infection in the canine (and human) hosts is more frequently associated with acute and often fatal disease, an indication that this might be a relatively new and poor host-parasite relationship. New World kala-azar The existence of kala-azar in the Americas had been suspected, but it was not until 1913 that the first real evidence was forthcoming when Migone described a case from Paraguay. The patient, an Italian immigrant, had spent some time in Brazil and had probably contracted the disease there. In 1926, Mazza and Cornejo recorded undoubted indigenous Latin-American cases after the spleen puncture of two Argentinian children who had never been out of that country; and in 1934 the introduction of viscerotome studies in Brazil finally indicated the extent of the disease when 41 cases were reported from eight different states (Penna 1934). Since that time the disease in humans has also been reported in Bolivia, Colombia, Venezuela, Salvador, Guatemala, and Mexico. The important role of the dog in the epidemiology of Old World kala-azar naturally stimulated an interest in these animals in Brazil, and infected dogs (and one cat) were finally found in Para State, North Brazil (Chagas et al 1937, 1938). As pointed out by Deane and Deane (1962), however, "It was unfortunate that Chagas and co-workers made their studies in a region (the Amazon) which ... was far from typical" (for kala-azar); later observations in the drier northeast, where kala-azar is much more common, revealed a dual reservoir: the dog (Fig. 2) and the wild fox Lycalopex vetulus (Deane, L. M. and Deane, M. P., 1954). Once again, a "semi-domesticated" species of sandfly, Lutzomyia longipalpis, was incriminated as the vector. When discussing kala-azar in Brazil it is important to bear in mind the vastness of the

R. LAINSON AND J. J. SHAW

country and the very different environments in which the disease occurs. Thus, in the northeast it is associated with semi-arid areas which often experience long droughts (Fig. 1 ), where the undergrowth is sparse and composed of low trees and shrubby undergrowth. In the endemic areas of the Amazon region, Para, however, the terrain is completely different - it is a dense, high rainforest with extreme humidity (Fig. 3 ). The early investigators of this sylvatic kalaazar in Para were convinced of the existence of some wild animal reservoir of L. donovani which would be the logical explanation for the sporadic nature of the disease in the area - but all their examinations of wild animals proved negative. It was not until recently that any further light was thrown on the problem, when workers in Belém finally isolated L. donovani from another type of fox, Cerdocyon thous (Fig. 4) (Lainson, Shaw, and Lins 1969). To date, two out of nine of these animals have been found to be infected and further studies are in progress. Many authorities have considered L. donovani not to be indigenous to the New World but to have been introduced relatively recently into this hemisphere by man and his domestic animals. Such a hypothesis seemed sound when only a handful of kala-azar cases had been reported, but it becomes more difficult to accept now that the disease is known in almost every country in South and Central America, from Argentina to Mexico. It seems more likely that visceral leishmaniasis in the Americas is an indigenous disease originating from a hitherto unsuspected animal source, with foxes, dogs, and man representing virtually new and "accidental" hosts. In this connection it is interesting to note that Deane (1948) found amastigotes in the viscera of seven specimens of the sloth (Choloepus didactylus). The animals were, interestingly enough, from the Abaetetuba region of northern Brazil, where sporadic cases of kala-azar were already known to occur. Unfortunately, the all-important inoculation of these parasites into hamsters was not attempted. Deane (1956), in fact, felt that the bodies in the viscera of the sloths were more

EPIDEMIOLÓGICA!. CONSIDERATIONS OF THE LEISHMANIAS

likely to be the amastigote form of another organism, the endoerythrocytic flagellate Endotrypanum schaudinni, since six out of seven of the animals harboured this parasite. Although Shaw ( 1969 ) has shown that Endotrypanum can indeed give rise to an amastigote stage within the erythrocyte, there is no other firm evidence to support Deane's theory, and the possibility cannot be excluded that he was dealing with natural infections of sloths with L. donovani (see the discussion on Leishmania isolated from sloths in Panama). Alencar, Pessôa, and Fontenele (1960) examined the blood of 153 domestic rats (Rattus norvégiens alexandñnus), trapped in "um foco de leishmaniose tegumentar endémica" in the State of Ceará, by NNN culture. They isolated a single strain of Leishmania from one of these animals and suggested that it was probably L. braziliensis. As the areas of cutaneous and visceral leishmaniasis in that region clearly overlap, and in the light of our present knowledge on the cutaneous nature of L. braziliensis infection in other wild rodents it seems much more likely that the parasite was in fact L. donovani. CUTANEOUS LEISHMANIASIS

Leishmania trópica (Old World cutaneous leishmaniasis or "oriental sore") Oriental sore has a range equal to that of kalaazar in the Old World but, apart from some overlapping of the diseases, tends to have a distinct geographic distribution. This is to be expected for a different parasite transmitted by different sandfly vectors. Thus, in India, it is prevalent in the northwest, but does not extend into the kalaazar areas in the east. While kala-azar has gone far in becoming an anthroponosis (apparently completely so in India) , oriental sore remains very much a zoonosis. In India it would appear that the dog is the only reservoir of infection for man, rodent hosts not having been incriminated so far. Elsewhere, however, there is overwhelming evidence of the important role played by rodents in its epidemi-

25

ology, to the extent that man and dog might well be regarded as accidental, or incidental, hosts. The first major studies of this problem were carried out in Turkmenistan (Latyshev and Kryukova 1941 ; Sekhanov and Suvorova 1960), where a beautifully clear-cut zoonosis was disclosed. The large gerbil Rhombomys opimas is the principal host of L. trópica in the deserts and semi-deserts of Turkmenia, with additional sources in other gerbils or ground squirrels, including Meriones, Spermophilopsis, and Hemiechinus spp. (Anon. 1968). All are burrowing animals, infection being maintained by the sandfly vectors in the warm, moist burrows. Up to 25.3 per cent of R. opimus may be infected. The disease derived from this rodent source in the rural areas of Turkmenistan (the "wet sore") appears to differ from that occurring in the urban districts ("dry sore"), where the only reservoirs so far indicated are dogs and man himself. These distinct epidemiologic features, together with clinical and immunological differences, have led the Russians to recognize two nosologically independent parasites, L. trópica major and L. trópica minor, respectively. Evidence is fast accumulating that rodents are important as reservoir-hosts of oriental sore in other, geographically similar regions. Thus, R. opimus has been found naturally infected in Iran (Ansari and Faguih 1953) and the "sand-rats," Psammomys obesus and Meriones sp., in Judea (Gunders, Foner, and Montilio 1968; Gunders etal 1968). The Leishmania braziliensis complex (New World cutaneous leishmaniasis) If the epidemiology of Old World leishmaniasis appears complex, that of the New World is infinitely more so. As with kala-azar, there has been much disagreement as to whether or not cutaneous leishmaniasis is an indigenous disease of Latin-America. Some have favoured the view that the disease has evolved from L. trópica brought to the Americas by the conquistadores or immigrants from endemic areas of the Old

FIGURES 5-10. American cutaneous leishmaniasis. (Figures 9 and 10, Lainson and Strangways-Dixon 1962b, 1964) 5-7 Uta, in Peru. 5 Endemic area of uta in the Peruvian Andes showing a typical breeding-place of the sandfly vectors in a dry stone wall. 6 Lesion on the nose of a naturally infected dog.

7 Lesion on the nose of a young girl. 8-10 Chiclerós ulcer, in British Honduras. 8 Tropical rain-forest in British Honduras, Central America - endemic area for chiclero's ulcer. 9 Lesion due to L. mexicana on the tail of a naturally infected forest rodent, Ototylomys phyllofis. 10 Typical erosion of the ear by L. mexicana.

EPIDEMIOLÓGICA!. CONSIDERATIONS OF THE LEISHMANIAS

World; indeed, even recent editions of textbooks sometimes include all of Central America in figuring the geographical distribution of L. trópica (Manson-Bahr 1966). The relatively recent discovery of a huge reservoir of infection in the wild rodents of remote, uninhabited areas of Central and South America, however, has made this idea of evolution from L. trópica untenable, as has the disclosure of variations noted in a study of the parasite from man and wild animals, which suggest a wide degree of speciation over a long evolutionary period. The leishmanial origin of "úlcera de Baurú" in Brazil was first demonstrated in 1909 by Lindenberg and by Carini and Paranhos. Two years later both Splendore and Carini found the parasite in nasopharyngeal lesions of typical mucocutaneous cases. From the appearance of the uncomplicated skin lesion it was understandable that the American disease was at first regarded as oriental sore and the parasite as L. trópica "identificaçâo das úlceras de Baurú ao botâo do Oriente" (Carini and Paranhos 1909). In 1911, however, Vianna proposed the name Leishmania braziliensis because of supposed morphological differences between the two parasites. Dermal leishmaniasis occurs from Yucatan in the north to Argentina in the south. The early tendency was to ascribe all infections to L. brauliensis, but the need for differentiation soon became apparent in order to distinguish the different forms of the disease found in different parts of the Americas. Perhaps the most clear-cut form is that of "uta," in Peru. No person who has worked on leishmaniasis in the lush forests of the New World can fail to be impressed by the stark contrast presented by the environment in the endemic areas of uta on the barren slopes of the Andes (Figs. 5 and 8) and the completely different ecology. Herrer (1948, 195la, b) described the disease as occurring up to an altitude of almost 3,000 metres, with transmission taking place under extremely arid conditions. No infections could be found in wild animals, but parasites occurred in up to 50 per cent of dogs (Fig.

27

6). These animals often showed no definite ulcers but amastigotes could be demonstrated in apparently normal skin, especially that of the nose and lips. The disease in man is relatively mild and usually manifests itself as single skin lesions (Fig. 7). Infection among children was at one time considered almost inevitable; recovery is fairly rapid, however, and, as in oriental sore, results in immunity during later life. The semi-domestic nature of the disease is unique among the New World forms of cutaneous leishmaniasis and has enabled relatively good control by the peridomestic use of DDT. The situation in regard to the possible existence of wild animal hosts, now or in the past, remains obscure. We must go to the opposite, northern limits of American cutaneous leishmaniasis to study the next most clearly defined form of the disease, "chiclero's ulcer" of the Yucatan, British Honduras, and Guatemala. This again is a relatively mild infection in man and, like uta, is generally a single-sore infection with a high recovery rate giving firm immunity and no nasopharyngeal involvement (Fig. 13). Unlike uta, however, chiclero's ulcer is essentially a disease of the lowaltitude rain-forest (Fig. 8), where it is transmitted by strictly sylvatic sandfly species. It is highly enzootic among wild forest rodents and is never associated with urban areas. Infected dogs have never been found. As the name suggests, it is an occupational disease, occurring principally among the men who spend long periods in the forests collecting chewing-gum latex (chicle), mahogany, and other forest products. Clinically the disease is interesting in view of the high percentage of lesions of the ear (up to 40 per cent). These ear lesions tend to be persistently chronic, lasting for many years and usually resulting in partial or complete loss of the pinna (Fig. 10). Uta and chiclero's ulcer present such distinct geographical, epidemiological, and clinical differences that they are generally attributed to separate species: L. peruviana Vêlez, 1913 and L. mexicana Biagi; emend. Garnham, 1962. The situation regarding other forms of Latin-American cutaneous leishmaniasis is less clear, al-

28

though some workers have differentiated (at least on clinical grounds) between "costa-rican," "Panamanian," "Colombian," "pian-bois" (Guyanas), and "mucocutaneous" (Brazil, Venezuela, East Peru, etc. ). It is not the object of this paper to discuss the taxonomic status of the leishmanias responsible for these diseases. The epidemiology and biology of the organism(s) concerned are still too little known to ascribe any more specific name than "L. braziliensis sensu latu" As Pífano (1960) cautiously concluded, L. braziliensis probably represents an "etiological complex integrated by species, varieties or biologically different races." A few words must be said at this point, however, about the condition known as "mucocutaneous" leishmaniasis (Fig. 16), if only because it is particularly prevalent in the country in which we are working, Brazil, and because it is the only known fatal form of cutaneous leishmaniasis. Opinions are divided as to whether or not the parasite responsible differs from another, or others, which give rise to skin lesions only (Fig. 15). Although the mucocutaneous disease appears more common in certain regions, such as Sao Paulo and Mato Grosso states, Brazil, it is known to occur less frequently in other parts of Brazil and in other Latin American countries where the simple cutaneous form is common. The problem has always been complicated by the fact that nasopharyngeal involvement follows an initial skin lesion elsewhere on the body (the site of the infective insect bite), and that the spread to the nose and palate tissues may not take place until many months or even years after the primary lesion has healed. It has remained difficult to determine, therefore, whether nasopharyngeal lesions are the outcome of infection only in certain individuals, or whether this area is a site of predilection on the part of a particular Leishmania strain or species. The only sure way of resolving this problem is by the intensive comparative study of strains isolated from both man and wild animals in areas endemic for mucocutaneous and appar-

R. LAINSON AND J. J. SHAW

ently "simple" cutaneous leishmaniasis. Ideally, such studies should include observations on the pathology in the human host and experimental animals, the morphology of both amastigotes and promastigotes, the behaviour in various local sandfly species, the development in various culture media, serology, and immunology. Animal reservoirs of New World cutaneous leishmaniasis Perhaps with the exception of uta, the various forms of New World cutaneous leishmaniasis are clearly zoonoses involving wild animals : human infection is almost always associated with penetration of forested areas and the disease is frequently acquired in regions devoid of human habitation. Surprisingly enough, definite proof of the role of wild animals as reservoirs was not forthcoming until relatively recently, largely due to the erroneous assumption by earlier workers that the infections would be largely visceral, with a para-

FIGURES 11-16. American cutaneous leishmaniasis. (Figures 11-13, Lainson and Strangways-Dixon 1962b, 1964; Figure 14, Shaw 1969) 11 Multiple lesions on the tail of O. phyllotis naturally infected with L. mexicana. 12 Smear showing amastigotes from a tail lesion. 13 Chiclero's ulcer on a hand. 14 Neck lesions due to Panamanian leishmaniasis. 15 Arm lesion due to L. braziliensis, sensu latu, Para State, Brazil. 16 Typical mucocutaneous leishmaniasis.

30

sitaemia easily detected by the NNN culture of heart-blood. The assumption at first seemed justified, and workers in Panama (Anon. 1958) finally reported the isolation of Leishmania from the heart-blood of "spiny rats" after the examination of many hundreds of animals by NNN culture. They found a 10 per cent infection rate in the blood of these rodents (Proechimys and Hoplomys spp. ), but were unable to demonstrate parasites in the viscera of either the natural host or subinoculated hamsters. Inoculation of the culture forms into the skin of human volunteers, however, did produce cutaneous lesions containing Leishmania. In subsequent years the infection could not be found in many hundreds of further spiny rats, and the significance of the blood infections found earlier still remains obscure. Following the same system, in Brazil, Forattini (1960) examined 928 wild mammals by NNN culture, including 881 rodents of nine different genera. One "forest-rat" (Kannabateomys amblyonyx) and an "agouti" (Dasyproctaazarae) both showed bodies resembling amastigotes of Leishmania in smears from skin lesions, and promastigotes were isolated from the heartblood of a single "paca" (Cuniculus paca). As far as we know, material was not inoculated into either laboratory animals or human volunteers. Later in the same year, Alencar et al. (1960) reported negative results from the NNN culture of heart-blood from 39 wild rodents in Gear a State, Brazil. One successful isolation was, however, recorded from the blood of a domestic rat (Rattus rattus alexandrinus). As discussed previously, this parasite could have been L. donovani\ again, we do not know of any subsequent report on its behaviour in experimental animals or man. In spite of these relatively disappointing results, other investigators still pursued the same line of approach, and Lainson and StrangwaysDixon (1962,1964) recorded completely negative results from the NNN culture of heart-blood, liver, and spleen from 272 wild animals during

R. LAINSON AND J. J. SHAW

their work on the epidemiology of L. mexicana in British Honduras. Following their studies on experimental infections in a variety of wild and laboratory animals, however, they were struck by the much greater susceptibility of the smaller rodents, and accordingly devoted their attention to the smaller forest species which were trapped in an area in which the human disease was particularly prevalent at that time. Eventually, natural infections were found in three different rodents : Ototylomys phyllotis (8 out of 20), Heteromys desmarestianus (6 out of 58), and Nyctomys sumichrasti ( 1 out of 8). It is easy to see how such infections might have been missed in the previous screening programs using NNN culture techniques, since they were purely cutaneous and in no case could parasites be isolated from heartblood or viscera. Furthermore, the lesions, although usually rich in parasites (Fig. 12), were inconspicuous : they were restricted to the skin of the tail and usually appeared as "whitish areas, rarely ulcerated, and occasionally showing some dermal thickening" (Figs. 9 and 11). Typical L. mexicana lesions developed in volunteers inoculated intradermally with parasites from the three different hosts, and studies by subsequent workers have confirmed the predominant role of both Ototylomys and Heteromys in the epidemiology of chiclero's ulcer (Disney 1968). The observations in British Honduras provided a stimulus which quickly produced results in other parts of Central and South America. Thus, in 1963, Dr Otis Causey discussed with us similar tail lesions that he had noted among rodents being examined in his virus program at the Instituto Evandro Chagas, Belém, Brazil. A remarkably heavy focus of leishmaniasis was uncovered among the wild animals there. Infected animals noted to date include the rodents Oryzomys capita and Proechimys guyannensis and the marsupial Marmosa murina (Nery-Guirnaraes and Azevedo 1964; Nery-Guimaraes, Azevedo, and Damasceno 1968; Lainson and Shaw 1968, 1969b; Shaw and Lainson 1968) (Figs. 17 and 19). We believe that up to 70 per

EPIDEMIOLÓGICA!. CONSIDERATIONS OF THE LEISHMANIAS

cent of the O. capiio population may acquire infection, and that this little rodent is clearly the major host. In Venezuela, Leishmania was encountered in a tail lesion of a specimen of Zygodontomys microtinus captured in Barinas State, where mucocutaneous leishmaniasis is highly endemic (Kerdel-Vegas and Essenfeld-Yahr 1966). Subsequent studies in Venezuela have revealed infections in dogs and donkeys (Pons 1968), although the author was of the opinion that the disease does not usually come from man or domestic animals, but is a zoonosis derived from wild animals. Recent studies have been made on the epidemiology of leishmaniasis in a newly opened area of the Serra do Roncador, Mato Grosso State, Brazil, where cutaneous and mucocutaneous cases were occurring among new immigrants to that area (Lainson and Shaw 1969a, 1970). Among 107 animals examined, infections were recorded in fourteen Oryzomys capiio, two O. concolor, one O. macconnelli, one Neacomys spinosus, one Nectomys squamipes (all rodents), and a single opossum, Marmosa m urina. The lesions were restricted to the tail in all instances (Figs. 18 and 21 ) except the Nectomys, which showed small lesions of one ear. The parasites were isolated in hamsters and, with one exception, appear similar to those previously isolated from wild rodents in the forests near Belém. Finally, in Trinidad, Tikasingh (1969) has reported leishmanial lesions, resembling those described above, on the tails of Oryzomys laticeps, Marmosa fuscata, and M. mitis. The parasites have been isolated in hamsters and NNN culture. Clearly an immense reservoir of leishmaniasis must exist among the wild rodent populations of the Latin-American forests, probably with a number of secondary host-reservoirs among other animals such as the marsupials. However, many questions remain concerning the importance of such animals to the human disease.

31

Proof that man can be infected by these animal parasites is of obvious importance since we already know species of Leishmania which are apparently completely or partially host-specific (L. enriettii of guinea-pigs in Brazil and the lizard leishmanias of the Old World). The demonstration of infectivity to man is, however, of use only if the infection can run a sufficiently long course to reproduce the naturally acquired disease, for it has been shown that even some lizard leishmanias may produce a transient skin infection in man (Manson-Bahr and Heisch 1961 ). In relatively mild forms of leishmaniasis such as chiclero's ulcer (L. mexicana) this has been possible (Lainson and Strangways-Dixon 1962, 1964), so there is little doubt as to the nature of the parasite in the rodent hosts. In some parts of Latin America, however, particularly in Brazil, where such experiments are hazardous because of the possible development of the mucocutaneous disease, a comparison of Leishmania "strains" is best limited to experimental hosts such as the hamster, mouse, and monkey. It has been suggested that the various clinical forms of leishmaniasis in Latin America might be due to infection with distinct forms of Leishmania which are normally resident in different wild animal hosts (Lainson and StrangwaysDixon 1964). The comparative study of large numbers of parasites from both man and animals in given endemic areas is consequently of the utmost importance. Perhaps the most obvious problem is the much debated one of "cutaneous" and mucocutaneous" leishmaniasis, as briefly discussed in the preceding pages. Are these simply different clinical manifestations of a disease caused by the same organism or are different parasites involved? Our laboratory has been particularly interested in this subject and has recently published some preliminary observations based on approximately five years of study of Leishmania isolates from both of these types of human infection and others from wild animal hosts in the same endemic areas of Brazil. We have noted two

32

distinct behaviour patterns among the strains of Leishmania isolated from similar, newly acquired, single-sore infections of man. These patterns are so strikingly different in both hamsters and NNN that it is tempting to consider that there are two different parasites. It is interesting, too, that wild rodents have provided leishmanias which exactly match both these "types" of parasites found in man. Although more comparative study of these parasites is necessary, we have the following distinct impression (Lainson and Shaw 1970): In certain areas of the Brazilian forests there exist at least 2 forms of Leishmania which are normally cutaneous parasites of different species of small mammals, in particular the rodents and marsupials. The 2 types of parasites may be found in the same ecological region. Both principally attack the skin of the tail, more rarely the ears and feet: the infection is usually seen as small white plaques or nodules, sometimes as extensive ulcers. The 2 parasites behave in a strikingly different way in the skin of inoculated hamsters and in NNN medium. One, the fast growing parasite, produces large histiocytomata at the site of intradermal inoculation [Figs. 22—241. The lesions take but a few months to develop and are packed with enormous numbers of amastigotes. NNN cultures produce such a rich growth of flagellates that the parasites are visible macroscopically as a white scum on the surface of the culture fluid ... The second slow growing leishmanias multiply much less rapidly in the hamster's skin. They usually produce a small, fleshy nodule which histologically shows much tissue-reaction but few parasites. In extreme cases it may be 6-12 months before the lesions become apparent. In others, a more ulcerative lesion may cause considerable destruction of tissue: these lesions are again characterized, however, by extensive inflammatory reaction evoked by small numbers of parasites. NNN cultures of these leishmanias are usually difficult to maintain and, after repeated attempts, we have still failed to grow some strains, which initially produce a promising

R. LAINSON AND J. J. SHAW

growth, but then die out after a few transfers ... Man, intruding into the enzootic areas, may become infected with either type of parasite. Initially both will produce similar skin lesions at the site of the insect-bite, but the parasites maintain the characteristics defined above. The slow growing organism appears to be that which predominantly gives rise to the naso-pharyngeal involvement of the classic mucocutaneous leishmaniasis of Brazil. In man, as in the hamster, the lesions are notable for their very protracted development and extreme host-cell response to very few parasites. It remains difficult to be sure which of these two "types" characterized the parasite originally described as L. braziliensis by Vianna, (1911). The name Leishmania braziliensis, can only be used in sensu lato for what is probably an extensive complex of species, sub-species or races of Leishmania which may occur not only in different geographical regions but also in similar hosts in the same forest areas.

FIGURES 17—27. American cutaneous leishmaniasis in Brazil. (Figures 17-19, 21-27, Lainson and Shaw, 1968, 1969c) 17 Ear lesion in a naturally infected rodent, Proechimys guyannensis (Belém, Para). 18 Tail lesion in a naturally infected rodent, Ne acomys spinosus (Mato Grosso). 19 Tail lesion in a naturally infected rodent, Oryzomys capito (Belém, Para). 20 Amastigotes in a smear from the tail lesion shown on the left. 21 Inconspicuous lesion on the base of the tail of a wild rodent, Oryzomys concolor (Mato Grosso). 22-24 Histiocytomata produced in hamsters by "fast-growing" strains of L. braziliensis, sensu latu, showing metastatic spread (Fig. 24). 25 Smear from a foot, above. 26-27 Small, "slow-growing" lesions produced in hamsters by other strains of L. braziliensis, sensu latu.

34

Panamanian leishmaniasis Although the human disease apparently presents no significant differences from that found in neighbouring countries (Fig, 14), the situation regarding the reservoir-hosts in Panama presents so many fascinating complications that it warrants separate consideration. We have seen that in other parts of Latin America there appears to be emerging a common factor in the epidemiology of the various forms of cutaneous leishmaniasis, namely the intensive foci of infection in the common species of small rodents. The parallel is made even closer by the similarity of the lesions, which are largely restricted to the skin of the tail. In spite of many years of research in Panama, however, no similar rodent foci have been encountered and we are faced, instead, with infections in a wider range of unrelated mammals which show a different type of disease. Thus, the spiny rats Proechimys and Hoplomys have shown parasites only in heart-blood. In porcupines (Coendú), however, Panamanian workers have described a generalized infection of apparently normal skin, with one animal giving a positive culture from bone-marrow (Herrer, Thatcher, and Johnson 1966). Similar skin infections were noted in sloths (Bradypus injuscatus and Choloepus hoffmannï), with parasites also isolated from heartblood, liver, and spleen (Anon. 1969). Parasites were also isolated from apparently normal skin of a single marmoset (Saguinus geoffroyi), and infection of an unspecified nature is reported in one specimen of the "olingo," Bassaricyon gabMi, Procyonidae (Anon. 1967a). The closest approach to the type of skin lesion seen in rodents elsewhere in the Americas was a single small lesion on the ear of a kinkajou, Potos flavus, Procyonidae (Thatcher, Eisenmann, and Hertig 1965). It has been suggested by some authors that all these might represent "accidental" infections of unusual hosts and that the "natural host is perhaps yet to be discovered" (Anon. 1968). This

R. LAINSON AND J. J. SHAW

implies that there exists in Panama a major undiscovered source (small rodents?) from which stem all these "accidental" infections, including that of man (Fig. 34). From the observations so far, however, this seems unlikely. The record of 10 out of 13 porcupines infected, for example, hardly suggests "accidental" infection and we feel that the most likely explanation is the existence of different leishmanial parasites in animals in different ecological niches (Fig. 35). Whether or not all such parasites can, or do, infect man is debatable, but it is interesting to note that the Gorgas Memorial Laboratory team wrote a few years ago: "From the data to hand [serological comparison of Leishmania from different human cases] it seems certain that in Panama at least 2 distinct forms of Leishmania exist which are capable of causing human disease. In addition, other immunological groups may appear among those parasites which cannot be distinguished morphologically from the human pathogens but which do not produce dis-

FIGURES 28—33. American cutaneous leishmaniasis, in sandfly vectors. (Figure 31, Strangways-Dixon and Lainson 1966; Figures 30 and 32, Lainson and Shaw 1968; Figure 33, Shaw 1969) 28 Disney-trap for catching sandflies attracted to rodents. 29 Lutzomyia sp., biting man. 30 Dissected Lu. flaviscutellata, promastigotes breaking out of ruptured mid-gut (natural infection). 31 Section of Lu. cruciata experimentally infected with L. mexicana, promastigotes blocking cardia and passing into oesophagus. 32 Smear of promastigotes from naturally infectedLw. flaviscutellata (Belém). 33 Developmental stages of Endotrypanum schaudinni in the stomach of Lu. sanguinaria (Panama).

36

FIGURE 34. Leishmaniasis in Panama, previously suggested situation. An undiscovered source of Leishmania from which arise "accidental" infections in a variety of arboreal and terrestrial animals (porcupines, sloths, marmosets, kinkajous, spiny rats, and man).

R. LAINSON AND J. J. SHAW

ease" (Anon. 1965; Schneider and Hertig 1966). This conclusion forms a most interesting parallel with that reached by our own laboratory on the subject of cutaneous and mucocutaneous leishmaniasis in Brazil, as discussed previously. Clearly it is of the utmost importance to pursue these lines of approach further. We have commenced an examination of porcupines and sloths in Brazil for evidence of infections similar to those described in Panama, with the hope that our colleagues in the latter country might continue in their endeavour to uncover the type of rodent leishmaniasis with which we are so familiar. In conclusion, one further consideration might be given to the type of infections seen in some of the wild animals of Panama. It will be remembered that, in our discussion on New

EPIDEMIOLÓGICA!. CONSIDERATIONS OF THE LEISHMANIAS

World kala-azar, it was suggested that there may exist an indigenous source of infection among the wild animals (other than the foxes). Could some of the infections of Panamanian animals in fact be due to L. donovanil When considering this question, we might bear in mind the following: (a) Although kala-azar has not yet been reported in Panama, it is known in virtually every country to the north and south of the Isthmus. (b) Lutzomyia longipalpis, the vector, has been recorded in Panama, and a specimen of this sandfly has been found there, heavily infected with promastigotes (Johnson, McConnel, and Hertig 1963 ). The authors mentioned the interest of this finding because Lu. longipalpis is "typically a dry-land species ... associated with neotropical kala-azar in Guatemala, El Salvador,

37

FIGURE 35. Leishmaniasis in Panama, present suggestion of the existence of different leishmanial parasites in animals in different ecological niches. Man as an accidental host of one or more of these leishmanias.

38

and several South American countries." It is as well to remember the sylvatic habitat of Lu. longipalpis in the lower Amazon region, however, and its association there with sporadic cases of kala-azar. (c) The infections described in the porcupine and sloths in Panama call to mind those produced by L. donovani in dogs and foxes (i.e. generalized skin infection and visceral involvement) more than the sort of infection produced in wild animals by parasites of the L. braziliensis complex (i.e. discrete localized skin lesions and apparently no visceral involvement). Deane (1948 ) has described Leishmania-like parasites in the viscera of sloths from areas of endemic sylvatic kala-azar in Brazil. (d) The Panamanian workers have stressed that the inoculation of isolates from sloths into hamsters has produced an "appearance and behaviour ... indistinguishable from L. braziliensis" (Anon. 1969). L. donovani will readily produce skin lesions when inoculated intradermally into a variety of experimental animals, including hamsters, and these lesions may be indistinguishable from those caused by L. braziliensis (sensu latu) (Wenyon 1926; and personal observations). ( e ) It has been shown by immunodiffusion tests that the porcupine strains of Leishmania "could not easily be identified with ... most of the human strains [L. braziliensis s.L] isolated in Panama" (Schneider 1968).

VECTORS AND TRANSMISSION

The problems of vectors and transmission have been left as a separate and final part of our paper. In this way we feel that at least some degree of order and continuity of text can be achieved for both parasitological and entomological aspects, while remembering, of course, that parasite and vector are inseparably linked in the study of any insect-borne disease. In 1904, Rogers noted the transformation of Leishman-Donovan bodies (amastigotes) into

R. LAINSON AND J. J. SHAW

leptomonads (promastigotes) and other workers immediately compared these flagellates with the naturally occurring Leptomonas spp. found in certain insects. As a result, bedbugs, fleas, lice, mosquitoes, hippoboscids, houseflies, ticks, and even phytophagous bugs in turn were suspected as the vectors of leishmaniasis (Wenyon 1926). Although the French workers in North Africa astutely suggested that phlebotomine sandflies were likely vectors (Pressât 1905; Sergent and Sergent, 1905 ), 46 years were to elapse between the discovery of Leishmania and the first experimental transmission by the true vectors - almost certainly due to the trail of entomological "red herrings" which bedevilled the early studies. The first observation which supported the hypothesis of the French workers was that of Wenyon (1911), who found that 6 per cent of the sandflies he examined in Aleppo, Syria, were infected with flagellates. Later, Leishmania was isolated from wild sandflies in North Africa (Sergent et al 1921) and Israel (Adler and Theodor 1926a), while Sinton (1925) noted that the distribution of both cutaneous and visceral leishmaniasis in India coincided with that of two species of Phlebotomus. Various sandflies were experimentally infected with Leishmania by numerous workers (Knowles, Napier, and Smith 1924; Christophers, Shortt, and Barrand 1925; Young and Hertig 1926; Adler and Theodor 1927; Parrot and Donatien 1927) and the course of the infection was followed carefully. The flagellates were shown to undergo a period of rapid division in the stomach and to migrate forward to the pharynx and buccal cavity, supporting the idea that the parasite was transmitted by the bite of the sandfly. Shortt et al. (1931) finally transmitted Indian L. donovani to hamsters by feeding infected P. argentipes on them, but the transmission rates were poor. It would seem that a supplementary diet of sugar or fruit juices assists in providing a suitable medium for Leishmania to grow within the gut of the sandfly and it was not until Smith, Haider, and Ahmed ( 1941 ) introduced the idea of letting infected flies feed on

EPIDEMIOLÓGICA!. CONSIDERATIONS OF THE LEISHMANIAS

raisins that transmission rates were improved, resulting in the successful transmission of L. trópica to a number of volunteers (Adler and Berl941). By 1941, therefore, the role of sandflies as vectors of Leishmania was relatively firmly established. They had been found harbouring the parasite in nature, and they could be infected experimentally. Also transmission to both man and animal was proved in the laboratory. It is fascinating, nevertheless, that ticks have recently been suspected once more of being vectors of canine visceral leishmaniasis in the Mediterranean (Giraud, Ranque, and Cabassu 1954) and in Brazil (Sherlock 1964). Until more evidence is forthcoming we can only point out that the distribution of the dog tick (Rhipicephalus sanguineus) is so much wider than that of canine kala-azar that its role, if any, can only be a minor or local one. Transmission of kala-azar in the Old World We have already mentioned that kala-azar in the Old World has different reservoir-hosts in different geographic areas, and, as might be expected, different species of sandflies have been incriminated as vectors in these regions. The vector of Indian kala-azar is P. argentipes, a domestic sandfly that transmits the disease directly from man to man. Although it appears to bite hamsters (Shortt et al. 1931 ), no mention is made as to whether or not it bites dogs, which, it may be remembered, are the reservoirs of kala-azar in the Mediterranean area and China. Indian L. donovani also develops well in P. perfiliewi, P. major syriocus, and P. chinensis simici (Adler and Theodor 1935a), species which apparently do not occur in India. Kindle (1931) showed that, unlike Chinese L. donovani, the promastigotes of Indian strains did not become attached to the mid-gut epithelium of P. chinensis. Adler and Theodor (1957) considered the existence of this attachment to be a strong indication that the parasite is in a suitable host-vector and that P. chinensis would probably

39

not, therefore, be an efficient vector of Indian L. donovani. P. chinensis, which is responsible for the transmission of Chinese kala-azar, is also a domestic species, but feeds readily on both man and dogs. When fed on infected hamsters and dogs the infection rate is high (Young and Hertig 1926; Feng and Chung 1938), but on humans it is much lower (Hindle 1928). This would suggest that transmission of Chinese kala-azar from dog to man is more likely than from man to man. Mediterranean kala-azar appears to be transmitted largely by P. perniciosus, P. longicuspis, and P. major. When fed on infected dogs P. perniciosus may show almost a 100 per cent infection rate (Parrot, Donation, and Lestoguard 1930), although a much lower rate was obtained by feeding these insects on severely infected humans (Adler and Theodor 1931 ). In Southern France epidemiological evidence has led Rioux et al (1968 ) to consider P. ariasi as the major vector among man and dogs and possibly wild foxes, even though P. perniciosus was present in the endemic area. P. major is more susceptible to Mediterranean L. donovani than P. perniciosus (Adler and Theodor 1935a) and has been indicated as the local vector in Turkestan (Chodukin 1929). Recently, P. chinensis and P. caucásicas have been incriminated as likely vectors in the sandy deserts of southeastern KaraKum, while P. mongolensis appears to be involved in the more mountainous regions (Ponirovsky 1969). Sudanese and East African kala-azar are contracted in rural or uninhabited areas and are not urban diseases. It would seem that transmission differs in the two countries. In the Sudan the predominant man-biter, P. orientalis, is readily infected with local L. donovani strains and has been found naturally infected (Hoogstraal, Dietlein, and Heyneman 1962; Heyneman 1963). Up to 51 per cent of the P. orientalis fed on a kala-azar patient became infected (McConnel 1964). In Kenya P. orientalis is comparatively rare, and members of the "Synphlebotomus" complex (P. martini, P. vansomerence, and P. celiae) are the suspected vectors (Heisch,

40

Wijers, and Minier 1962; Minier and Wijers 1963 ). A strain of L. donovani isolated from P. martini (Heisch et al. 1962) was found to be serologically indistinguishable from strains isolated from man (Adler 1963). Minier and Wijers (1963 ) were able to infect 76 per cent of P. martini fed on an infected human and noted that the flagellates were attached by their flagella to the intestinal wall. Transmission of American kala-azar Phlebotomine sandflies were clearly the first suspects as vectors of leishmaniasis in the New World and were first indicated as such by the Brazilian Kala-Azar Commission in their studies on a small focus of kala-azar in the lower Amazon region (Chagas et al 1937, 1938). They found that the commonest blood-sucking insect in and around the houses of the kala-azar patients was a sandfly, Lutzomyia longipalpis. Further observations showed that both visceral leishmaniasis and Lu. longipalpis were limited to the drier areas ( "terra firme" ). Studies on the geographical distribution of the phlebotomine sandflies in the Amazon Valley (Damasceno, Arouk, and Causey 1949) have since shown that Lu. longipalpis is limited to river margins and coastal regions of the lower Amazon as is human visceral leishmaniasis. Experimental feedings of Lu. longipalpis on dogs suffering from visceral leishmaniasis also showed that this particular sandfly was susceptible (Ferreira, Deane, and Mangabeira Fo. 1938; Chagas 1940). The true importance of Lu. longipalpis as the vector in the lower Amazon region, however, remained inconclusive, and even as recently as 1966 a single indigenous case of kala-azar was reported on the island of Mar ajó in the apparent absence of Lu. longipalpis (Costa 1966). The search for the vector moved to the semiarid northeastern region of Brazil (Fig. 1 ) where many human cases were being diagnosed, and preliminary observations (Ponde, Mangabeira Fo., and Jansen 1942) in the new-found endemic areas of Bahía, Ceará, and Pernambuco

R. LAINSON AND J. J. SHAW

again showed Lu. longipalpis to be a common domestic species. It was shown to feed avidly on man and the known reservoirs, dogs and foxes (Deane 1956). As in the Amazon region, the distribution of Lu. longipalpis coincided closely with that of human kala-azar. In the valleys and dried river floodplains, where most of the human cases of kala-azar occur, Lu. longipalpis was the predominant man-biter. In the neighbouring hills, however, where kala-azar is rarer, Lu. whitmani and Lu. migonei were the major anthropophilic species. In endemic areas of Ceará and Bahia Lu. longipalpis has been found naturally infected with promastigotes (Deane, M. P., and Deane, L. M., 1954a; Lopes 1956). Although the organisms were not isolated in culture or hamsters, it is highly possible that they were L. donovani, as they were located in the anterior part of the gut. The development was studied experimentally in flies fed on naturally infected dogs, foxes, and humans (Deane, M. P., and Deane, L. M., 1954b, 1954c, 1955). In the flies fed on dogs the infection rate was lower than in similar experiments with Mediterranean kala-azar but up to 100 per cent of the flies fed on foxes became infected. The infections were located in the midgut and fore-gut and, in one fly fed on a fox, promastigotes were seen in the distal portion of the epipharynx. The infection rate in the flies fed on humans was higher than that obtained by other workers with Mediterranean kala-azar but lower than that recorded for East African kalaazar. Lu. longipalpis has also been captured in the endemic kala-azar districts of Guatemala and Mexico (Leon and Figueroa 1959; Biagi, Lopez, and Biagi 1965 ). Although not encountered in great numbers in Guatemala (only eight males were captured, in chicken runs), it was nevertheless considered to be the vector. Human kala-azar has not been recorded in Panama but Lu. longipalpis is present (Fairchild and Hertig 1959). We have already drawn attention to the possibility that some of the Leishmania infections found in animals there might be

EPIDEMIOLÓGICA!, CONSIDERATIONS OF THE LEISHMANIAS

due to L. donovani and that promastigotes have been found in a single specimen of Lu. longipalpis (Johnson et al 1963). In Venezuela Lu. longipalpis has been found associated with outbreaks of kala-azar in the semi-arid western and central plains (Amaral, Torrealba, and Henriquez 1961 ; Pifano, Alvarez, and Ortiz 1962; Torrealba 1964). It was collected both inside and outside human habitations and animal houses, and was observed to feed on both man and a wide range of domestic animals. Torrealba (1964) was of the opinion that it only bit man in significant numbers when there was a marked increase in the sandfly population. As in Brazil, this species has been infected experimentally (Pifano, Romero, and Alvarez 1962), although of 30 laboratory-reared flies fed on an infected dog only one became infected. At this point it is interesting to mention that, as far as we know, there has been no successful experimental transmission of American kala-azar by the bite of Lu. longipalpis, so that absolute proof of the role of this sandfly as the vector is still lacking. Nevertheless, it seems fairly certain that Lu. longipalpis is a major vector of L. donovani from the reservoir-hosts to man and perhaps, on occasions when the sandfly population is high, from man to man. It is less certain that Lu. longipalpis is the only vector of South American kala-azar. Pifano and Romero (1964), studied the epidemiology of two cases from Sucre State, Venezuela, and noted that Lu. evansi and Lu. gomezi were the major man-biters. In the apparent absence of Lu. longipalpis they suggested Lu. evansi as the most likely vector. In Brazil, Sherlock (1964) drew attention to the existence of canine kala-azar in areas where Lu. longipalpis was rare or absent, and in Belém, Brazil, we found visceral leishmaniasis in foxes but no Lu. longipalpis or any cases of human kala-azar. Probably other zoophilic species of sandflies transmit the disease among the wild reservoirhosts; when infected reservoir-hosts, such as the fox, come into a rural or urban area where Lu. longipalpis exists, transmission to dogs and man may take place. This would explain partially the

41

sporadic nature of human and canine visceral leishmaniasis in the lower Amazon region, Venezuela, and parts of Central America. In the New World, canine and human cases of visceral leishmaniasis seem to be relatively poor sources of infection for Lu. longipalpis, although man is a somewhat better source of parasites than in Mediterranean kala-azar. In most respects American visceral leishmaniasis is epidemiologically closest to kala-azar of Central Asia and Transcaucasia, where jackals and foxes appear to be the reservoir-hosts. The wide geographical distribution of kala-azar throughout Latin America suggests that the disease, if imported recently from the Old World, has had remarkably efficient dissemination. Considering the delicate nature and short flight range of the sandfly vectors, however, it seems unlikely that they could have been responsible for such a dramatically rapid spread of the disease. As discussed earlier, kala-azar is much more likely to be an indigenous disease in the neotropics. Transmission of Old World cutaneous leishmaniasis We have already mentioned the historical facts that led to the incrimination of phlebotomine sandflies as vectors of L. trópica. The role of P. papatasii in the transmission of L. trópica in Jericho was studied very carefully by Adler and Theodor (1926b), who showed conclusively that specimens of P. papatasii taken from endemic areas were naturally infected with promastigotes which were serologically identical with L. trópica isolated from naturally acquired human lesions. Later, proof was obtained by experimentally transmitting L. trópica by the bite of P. papatasii (Adler and Ber 1941). Sinton ( 1925 ) noted that the distribution of P. sergenti in India coincided with the distribution of L. trópica, and subsequently P. sergenti was found to be more susceptible to L. trópica than P. papatasii (Adler and Theodor 1929a). In Baghdad and Mosul the prevalence of oriental sore has been thought to be linked with the high

42

percentage of P. sergenti caught in houses ( Adler and Theodor 1928). Various workers have listed the species of sandflies of North Africa, Asia, and India (Sinton 1925; Adler and Theodor 1929b; Theodor 1938; Pringle 1956; Theodor and Mesghali 1964) and it seems generally agreed that P. papatasii and P. sergenti are the major vectors of the urban type of cutaneous leishmaniasis. The transmission of rural or zoonotic dermal leishmaniasis is more complicated as it involves transmission among burrowing rodents. The existence of sandflies in rodent burrows has been known since 1929 (Sergent and Parrot 1929; Vlasov 1932). Some species such as P. papatasii and P. caucasiens occur in both burrows and the domestic environment of villages and towns, while others appear to be restricted to animal burrows. In general it would seem that P. papatasii is the main vector between rodents and man although within the burrows the vectors among the rodents have been listed as P. papatasii, P. caucasiens, P. mongolensis, P. alexandri, Sergentomyia arpaklensis, and S. clydei (Anon. 1968). There is, however, some doubt as to the importance of at least the Sergentomyia species, for elsewhere it is implied that members of this genus normally feed on lizards and are frequently infected with lizard Leishmania (Adler and Theodor 1935b, 1957; Anon. 1968; Heisch 1954; Nadim, Seyedi Rashti, and Mesghali 1968; Safyanova 1969). In Africa, however, S. clydei appears to feed on both reptiles and mammals (Adler and Theodor 1957) and infections found in this species could thus be of either mammalian or reptilian origin! Transmission of New World cutaneous and mucocutaneous leishmaniasis The transmission of cutaneous leishmaniasis is even more complicated in the New World than in the Old. The late Professor Adler (1964) summarized the situation admirably when he wrote: "The leishmanias of mammals have speciated more in the New than in the Old World

R. LAINSON AND J. J. SHAW

and each species may well have its own vectors, reservoir and spectrum of host infectivity. Speciation in the genus Phlebotomus has also been more extensive and more complex ... The bionomics of the forest species ... will not be easy to elucidate. Some 30 species of sandflies have been found associated with Cuniculus paca; this illustrates the complexity of the problems in South America." As in the Old World, a variety of haematophagus arthropods such as ticks, mosquitoes, fleas, mites, and tabanid flies were all suggested as possible vectors. One of the more interesting suggestions was made by Townsend (1915), who considered that species of ceratoponids of the genus Forcipomyia were probably the vectors of uta (these are mostly predacious and not haematophagus insects). He found amastigotes in the hind-gut of some specimens of these flies and produced a small lesion, apparently containing amastigotes, by inoculating them into the skin of a guinea pig. In spite of these suggestions, Neiva and Barbará (1917) felt that the most likely vectors of mucocutaneous leishmaniasis were phlebotomine sandflies (Fig. 29), and observed that the distribution of mucocutaneous leishmaniasis in Brazil coincided with that of sylvatic species. Many other workers (listed by Pessôa and Barreto 1948) also noted that, with the exception of uta in Peru, Latin-American dermal leishmaniasis is a disease of forest workers. These observations, in conjunction with the incrimination of phlebotomine sandflies in the Old World, led most workers to concentrate also on members of this same family in the New World. Research on the transmission of dermal leishmaniasis in the Americas can be considered under two broad headings : laboratory experiments, incriminating Lutzomyia sandflies as capable vectors ; and field studies to determine the specific vectors in different geographical areas. Obviously the two are closely related, but it must be remembered that in epidemiological studies it is the vector in nature and not in the laboratory that is important.

EPIDEMIOLÓGICA!. CONSIDERATIONS OF THE LEISHMANIAS

Laboratory experiments Many American sandflies have been successfully infected by a variety of techniques with different strains of Leishmania from cases of cutaneous leishmaniasis (Aragâo 1922: Lu. intermedia', Pessôa and Coutinho 1941 : Lu. fischeri and Lu. whit maní', Pif ano 1940: Lu. panamensis;Herúg and McConnel 1963 : Lu. sanguinaria, Lu. gomezi, Lu. panamensis, Lu. trapidoi, Lu. ylephiletrix; Strangways-Dixon and Lainson 1966: Lu. ylephiletrix, Lu. cruciata, Lu. geniculata, Lu. bispinosa, Lu. pessoana, Lu. shannoni, Lu. ovallesi, and Lu. apicalis-, Coelho, Falcâo, and Falcâo 1967: Lu. longipalpis, Lu. renei, Lu. intermedia, Lu. whitmani, Lu. sallesi, Lu. shannoni, Lu. arthuri, Lu. montícola, Lu. cavernícola, and Lu. coelhoï). In such sandflies thé respective strains developed readily, 49 to 100 per cent becoming infected, and the site of development was generally in the anterior region of the gut (Fig. 31). Some of these workers successfully infected animals by inoculating triturated flies while others did not; the reasons for the failures are not clear. The first successful experimental transmission by the bite of an infected New World phlebotomine was that of Strangways-Dixon and Lainson (1962). These authors fed a variety of wildcaught sandfly species on hamsters infected with L. mexicana and subsequently transmitted the infection to man, on one occasion, by the bite of Lu. pessoana. In the same year, in Brazil, Coelho and Falcâo (1962) transmitted L. mexicana by the bite of experimentally infected Lu. longipalpis and Lu. renei. Although these represented the first recorded transmissions by the bite of sandflies in Latin America, the species of sandflies used can, of course, only be considered as potential vectors in nature until epidemiological studies confirm otherwise. More recently Williams (1966b) has transmitted L. mexicana with Lu. cruciata, which has in the past been considered a possible vector on epidemiological grounds (Biagi 1953; Biagi and Biagi 1953). Undoubtedly many phlebotomine species of

43

the New World can support the development of Leishmania, and probably can also transmit the parasite by their bite. This in conjunction with the coincidental distribution of cutaneous leishmaniasis and sylvatic species of Lutzomyia clearly suggests that they are the natural vectors. Field studies The only known form of dermal leishmaniasis not associated with forested areas in the New World is uta in Peru. Epidemiological evidence suggests that Lu. verrucarum and Lu. peruensis are probably the major vectors (Herrer 195 Ib). It is interesting that the life history of Lu. verrucarum is similar to that of P. papatasii, the vector of L. trópica, for it rests and breeds in rock crevices, stone walls, and under boulders (Fig. 5). It readily enters houses, where it can be found resting during the day, and transmission from dog to dog and to man can presumably take place indoors or outdoors. The distribution of cutaneous leishmaniasis in this mountainous area does not, however, completely coincide with that of P. verrucarum and presumably other species are also involved (Hertig 1964). In epidemiological studies on the sylvatic forms of cutaneous leishmaniasis it is obviously of importance to know something of the transmission among forest animals. Most of the work in the past, however, has been concerned with attempts to determine the important vectors to man. This is not surprising, of course, as it is only recently that the major reservoirs have been, or are being, discovered. The search for natural infections of promastigotes of Leishmania in known man-biters has been carried out extensively throughout Latin America. Unhappily, most of the evidence is circumstantial, for although some 504 natural infections have been described in wild-caught Lutzomyia spp., only 14 have been proved to belong to the genus Leishmania. In an area of endemic mucocutaneous leishmaniasis in Sao Paulo State, Brazil, promastigotes were found in Lu. migonei (0.21 per cent),

44

LU. whitmani (0.20 per cent), and Lu. pessoai (0.29 per cent) (Pessôa and Pestaña 1940; Pessôa and Coutinho 1940, 1941; Coutinho 1941 ). In another endemic area of Paraná State, a single infection has been found in Lu. intermedia (Forattini and Santos 1952). Although all are known man-biters (Barreto 1943; Forattini 1954), in no instance was it demonstrated that the flagellates were Leishmania. In Venezuela promastigotes have also been found in wild-caught Lu. migonei and Lu. longipalpis (Pifano 1940, 1943), and in a single specimen tentatively identified as Lu. anduzei (Pifano in Forattini 1959a). All these sandflies were caught in endemic cutaneous leishmaniasis areas, but the nature of the flagellates was apparently not determined. The Venezuelan workers (Pifano 1940; Pifano, Alvarez, and Ortiz 1962) have suggested that in some areas of Venezuela Lu. panamensis is the major vector of cutaneous leishmaniasis to man. The evidence upon which this suggestion is based is not conclusive, however, and further detailed studies are required to confirm the view. The most extensive study on natural flagellate infections of wild-caught phlebotomine sandflies was performed by Johnson, McConnel, and Hertig (1963). These workers found promastigotes in 416 common man-biting Panamanian Lutzomyia sandflies (Lu. panamensis, 1.9 per cent; Lu. sanguinaria, 4.7 per cent; Lu. gomezi, 5.0 per cent; Lu. ylephiletrix, 9.4 per cent; Lu. trapidoi, 15.4 per cent; and Lu. shannoni, 5.4 per cent). From serological tests and morphological studies on a few of these natural infections it would seem that Panamanian sandflies are infected with a number of different flagellates (McConnel 1963; Schneider and Hertig 1966; Wallace and Hertig 1968). Five strains of sandfly promastigotes (two from Lu. trapidoi, one from Lu. sanguinaria, one from Lu. gomezi, and one from Lu. ylephiletrix) were identified as belonging to two serologically distinct groups of human Leishmania (Schneider and Hertig 1966) ; of these five, those from Lu. trapidoi, Lu. ylephiletrix, and Lu. gomezi all produced amasti-

R. LAINSON AND J. J. SHAW

gotes in the skin of hamsters (McConnel 1963; Schneider and Hertig 1966). Another isolate from Lu. trapidoi infected hamsters but did not belong to either of the two human groups of Leishmania. One strain from Lu. sanguinaria has been identified as a Crithidia species (Wallace and Todd 1965; Wallace and Hertig 1968), while the identity of the other natural infections remains obscure. It has recently been suggested that some of these infections could have been Endotrypanum schaudinni, which develops as promastigotes (Fig. 33) in phlebotomine sandflies (Shaw 1964, 1969). The remarkable frequency of occurrence of promastigotes in wild-caught Panamanian sandflies shows how much care is needed in interpreting the significance of natural flagellate infections. Any phlebotomine fauna in the South American forests is obviously exposed to a variety of trypanosomatids of which only some will be Leishmania that are infective to man. In this respect it is interesting to note the recent description of flagellates in specimens of Lu. micropyga captured in tree-trunks and animal burrows in Bahia, Brazil (Sherlock and Pessôa 1966). The parasites were described as leptomonads (promastigotes) but, from the illustrations, they appear to be the developmental stages of a trypanosome. With the development of a trap (Fig. 28) to catch sandflies that bite rodents (Disney 1966) it became possible for the first time to identify and dissect the possible vectors among the rodent reservoirs. Disney (1968) showed that Lu. olmeca* was the predominant rodent-biter in *This species was first identified in British Honduras as Lu. apicalis by Garnham and Lewis (1959) and was later considered identical with Lu. flaviscutellata by Disney (1968). Fairchild (see Disney 1968) considered that Panamanian material appeared to be intermediate between Mexican and South American material, and that Lu. olmeca was a northern race or possibly a subspecies of flaviscutellata. From our own comparison of material from British Honduras and Belém, Brazil, however, we prefer to follow Vargas and Nájera (1959) and Theodor (1965) and retain the name Lu. olmeca for the British Honduras species.

EPIDEMIOLÓGICA!. CONSIDERATIONS OF THE LEISHMANIAS

British Honduras and found it naturally infected with promastigotes; on three occasions these were shown to be Leishmania. He also found one promastigote infection in an aboreal species, Lu. permira, but was not able to identify the parasite. Disney considered that Lu. olmeca was the principal vector among the rodents but thought that other species might be secondary vectors, such as Lu. cruciata, Lu. permira, Lu. panamensis. Biagi, Biagi, and Beltran (1965 ) confirmed the importance of Lu. olmeca as a vector of L. mexicana and, after a successful transmission to man with wild-caught flies, considered it an important vector to man. In this same region, however, other workers (Garnham and Lewis 1959; Williams 1966a; Williams, Lewis, and Garnham 1965; Strangways-Dixon and Lainson - see Shaw and Lainson 1968) found that only 0.1 to 1.2 per cent of the man-biters were Lu. olmeca, while Biagi et al. ( 1965 ) recorded a figure of 8.5 per cent. The long-term observations of Williams (1966a) generally confirmed that Lu. olmeca is not a common man-biter. This fact, added to its apparent absence as an anthropophilic species in some endemic areas of Mexico (Biagi and Biagi 1953 ), makes it difficult to accept Lu. olmeca as the principal vector of L. mexicana to man. More recently, however, Williams (1969), after dissecting over 13,000 sandflies, considered Lu. olmeca to be the only insect host of L. mexicana in British Honduras. He again noted that Lu. olmeca was rarely attracted to man, but explained transmission to man on the basis that there was "quite a high degree of contact between the fly and man in the hour or so after dawn" (5.49 flies per 10 hours of catching)Thatcher (1968), in Panama, used an oiled trap baited with animals similar to that of Disney. He showed that the common man-biters Lu. panamensis, Lu. trapidoi, and Lu. sanguinaria fed on a number of wild animals (opossums, kinkajous, and porcupines), some of which are known to be naturally infected with Leishmania. He found that Lu. trapidoi and Lu. sanguinaria preferred to feed in the trees, but drew attention

45

to the fact that both are common man-biters at ground level (Johnson et al. 1963 ). Other species such as Lu. panamensis, Lu. pessoana, and Lu. flaviscutellata fed near the ground. The Gorgas Memorial Laboratory team's results over the past 10 years have indicated that Lu. trapidoi is the major vector in Panama and that it becomes infected in the forest canopy. Its importance as a vector to man would seem to depend on its vertical movement within the forest. Recently Lu. panamensis has been found infected with promastigotes of Leishmania (Christensen, Herrer, and Telford 1969) ; it avidly bites both animals and man, and is almost exclusively a ground level feeder, suggesting that it becomes infected at ground level. Further work on the cycle of transmission at ground level may help to clarify the picture in Panama. Recent work in Surinam (Wijers and Linger 1966) showed Lu. squamiventris to be the commonest man-biting sandfly. A large number were dissected but none was found infected. Twelve natural infections of promastigotes were, however, found in the anterior part of the gut of Lu. anduzei. This species was found to bite man in the early morning, but it is not a common manbiter and the infected specimens were caught while they were resting in tree buttresses; however, it was concluded "that of all the man-biting species of sandflies P. anduzei is the most likely species to be the main vector of cutaneous leishmaniasis to man." Unfortunately, inoculation of the flagellates into animals produced no conclusive results, and it can only be presumed that perhaps some of these parasites were L. braziliensis. In the neighbouring Amapá region of Brazil the predominant man-biter is also Lu. squamiventris and, again, Lu. anduzei appears rarely to bite man (Forattini 1959b). With the aid of Disney-traps, we discovered that the major vector concerned in the focus of rodent leishmaniasis in the Utinga forest, Belém, Brazil, was Lu. flaviscutellata (Lainson and Shaw 1968). The overwhelming predominance of this species among the sandflies caught with rodent bait convinced us that this was likely to

46

be the case, so our early dissections were limited to this species. Promastigotes were found in the anterior part of the gut (Figs. 30 and 32) in eight out of 2,706 flies, and when inoculated into hamsters, six of these isolates produced infections which were indistinguishable from that caused by the rodent Leishmania (Fig. 22). The failure of two strains to infect these animals was due in one case to the death of the flagellates prior to inoculation, and in the second probably to a scarcity of organisms in the inoculum. The infection rate of 0.29 per cent might be considered low for an efficient vector. Based on the average nightly catch of 14.87 Lu. ftaviscutellata off a single rodent, however, one might expect an infected fly to bite a wild rodent every 23 days (Shaw and Lainson 1968). This clearly shows the high potential of Lu. ftaviscutellata as a vector among the wild rodents, though it is equally clear that this insect is most unlikely to be concerned in the onward transmission of the disease to man. In the Utinga forest, Belém, for example, it represented only 0.69 per cent of all the sandflies caught off human bait and thus cannot be considered an anthropophilic species. With this biting-rate, one can expect an infected Lu. ftaviscutellata to bite man only every 445 days! If other factors such as the percentage of infected flies that are actually infective, the lifespan of the individual insects, the average number of blood meals, flight range, etc. are taken into account, man would stand even less chance of being infected by the bite of this species. After four years of observations on the rodent- and man-biting sandflies in this same area we have come to the conclusion that there is, in fact, no common sandfly species here which can seriously be considered as a likely vector of leishmaniasis to man. Dissection of numerous sandflies of other species (Lu. infraspinosa, Lu. rooti, Lu. saulensis, Lu. monstruosa} which are attracted in small numbers to rodents has revealed no Leishmania infections (although some were found with developmental stages of trypanosomes). All these observations fit well with the fact

R. LAINSON AND J. J. SHAW

that, in spite of the enormous numbers of infected rodents in this area, we have been unable to find a single record of human cutaneous leishmaniasis in the Utinga area during the past 10 to 15 years. Of interest in this connection is a recently published report on the existence of rodent leishmaniasis in Trinidad (Tikasingh 1969), which records Lu. ftaviscutellata from the enzootic area. This species has also been found biting man in small numbers in Trinidad (Aitken, Worth, and Tikasingh 1968). Although human leishmaniasis has been recorded from some of the West Indian islands, it is unknown on the island of Trinidad. Although regarded as different species, Lu. ftaviscutellata and Lu. olmeca are clearly closely related sandflies, sharing several biological features which doubtless contribute to the fact that they are of prime importance in the maintenance of American rodent leishmaniasis. Both are ground-level feeders and avid rodent-biters but, interestingly enough, both are disinclined to bite man and seem unlikely to be effective vectors of the disease to man. This latter role is presumably left to other species which bite both man and rodents equally readily, information about which is sadly lacking. Since our original discussion on this subject (Shaw and Lainson 1968) further pertinent observations have come to hand which might be dealt with briefly here. It has been implied that Lu. olmeca is caught biting man more often in the early morning (Disney 1968; Williams 1969), but this does not seem to be the case with Lu. ftaviscutellata in Surinam or Brazil ( Wijers and Linger 1966; personal observations). It has also been felt that certain unusual conditions, such as dramatic climatic changes or the disturbance of resting flies, might stimulate normally non-anthropophilic species to bite man. From our experience in Brazil and the situation in Trinidad (see above) it would seem that such conditions do not appear to alter the biting habits of Lu. ftaviscutellata sufficiently for it to become a vector to man. This view is supported further by recent observations made by us in Mato

EPIDEMIOLOGICAL CONSIDERATIONS OF THE LEISHMANIAS

47

Grosso State, Brazil. The "gallery" forest immediately adjoining the expedition base-camp contained rodents infected with Leishmania, but no case of the human disease has been reported among non-immune personnel who have been working regularly in the area for the past two years. In neighbouring undisturbed forest, however, transmission to man was regularly taking place. Lu. ftaviscutellata was common in both areas but only in the undisturbed forest did we find substantial numbers of other species which fed readily on rodents and man.

One can postulate an infinite variety of hypothetical situations in the epidemiology of cutaneous leishmaniasis in Latin America, depending on the biting habits of different sandfly species. We have attempted to summarize diagrammatically only those for which there is some epidemiological evidence (Fig. 36). In the simple enzootic (Fig. 36 A) the vector, such as Lu. ftaviscutellata, rarely bites man and leishmaniasis remains a disease of the smaller cricetid rodents. Secondary hosts include other animals (e.g. marsupials) on which this species

FIGURE 36. Cutaneous leishmaniasis, suggested epidemiology in the Americas, based on present available evidence.

A,C Sylvatic endemic. Zoophilic and anthropophilic vectors. Transmission to man c. Many parts of Central and South America. D Sylvatic endemic. Anthropophilic vectors. Transmission to man and other secondary hosts B2. Panama, for example. E Rural endemic. Anthropophilic vectors among dogs with transmission to man. Example: Uta, Peru.

A Simple sylvatic enzootic. Zoophilic vectors among rodents with occasional transmission to other animals B!. Man rarely infected. Examples: Belém forests, Para, Brazil; Trinidad.

48

feeds less readily (Fig. 36 B! ). This appears to be the present situation in the forests around Belém, Para, Brazil and in Trinidad. In other endemic areas the enzootic is maintained by such rodent-biters as Lu. flaviscutellata and Lu. olmeca, but in conjunction with other species of sandflies which commonly bite both rodents and man (Fig, 36 c). Any human intruder may become infected by the bite of the latter anthropophilic species and we thus have a zoonosis. This is the situation in endemic areas we have studied in parts of Brazil, where Lu. flaviscutellata is the main vector among the rodents. We feel that this is probably the situation, too, in the forests of British Honduras, Yucatan, and Guatemala, where L. mexicana is maintained among the rodents largely by Lu. olmeca. A simpler zoonosis may exist, where the vectors appear to be anthropophilic species (Fig. 36 D) . This appears, at least in part, to be the situation in Panama where such species as Lu. trapidoi and Lu. sanguinaria bite infected arboreal animals such as sloths. Presumably because of their vertical flight movements they may then transmit to man at ground level. Secondary hosts are other arboreal animals such as monkeys and kinkajous (Fig. 36 B2). Lastly, in Peru, we have the only known form of New World cutaneous leishmaniasis that is not associated with the forest and apparently has no wild animal reservoir. The disease, uta, is transmitted among dogs and man by peridomestic species of sandflies, probably Lu. verrucarum and Lu. peruensis (Fig. 36 E) . In an area of enzootic leishmaniasis the risk of human infection will thus depend to a great extent on the behaviour of the local sandfly species; it is of obvious importance to secure as much information as possible on the biology of all the common species, particularly on their bitinghabits. Thus, a high proportion of a given species may readily be infected in the laboratory and even transmit Leishmania', but if this species does not readily bite the wild animal hosts, or does not occur where they live, it can have little

R. LAINSON AND J. J. SHAW

importance in the epidemiology of leishmaniasis in that area. Not all sandfly species have equal capacity to transmit Leishmania. Some may develop hindand mid-gut infections, but only those with high infection-rates in the fore-gut and mouthparts will be efficient vectors. This must be borne in mind in attempts to pin-point local vectors by the dissection of wild-caught sandflies. We must stress the occurrence in sandflies of flagellates other than those of the Leishmania causing human disease. These may be other leishmanias not concerned in human infection, monoxenous insect flagellates, or stages in the life cycles of other parasitic trypanosomatids (Fig. 33). Wherever possible, inoculation of susceptible animals (hamsters) or man should be used to establish their identity. Finally, although it is generally agreed that mechanical transmission plays little part in the general epidemiology of leishmaniasis, there are some workers who feel that its importance has been underestimated (Berberian 1966). There are, indeed, records of the disease being contracted in countries where there are no sandflies. The source of these infections has always been traced to some form of mechanical transmission such as blood transfusion or direct contact. Perhaps more important is the fact that both L. trópica and L. mexicana have been transmitted experimentally by the bite of Stomoxys calcitrans, following interrupted feeding (Berberian 1938; Lainson and Southgate 1965). The occasional role of this and similar biting flies in natural, mechanical transmission should not be overlooked. There remains a great need for more comparative studies on the different leishmanias: their morphology and behaviour in the sandfly vectors, serology, immunology, metabolism, etc. In the past some workers have used strains designated simply as "L. braziliensis" giving little or no indication of the parasite's history. We have seen, however, that this name probably includes a complex of different parasites, even when they

EPIDEMIOLÓGICA!. CONSIDERATIONS OF THE LEISHMANIAS

are derived from the same geographic area. This in itself indicates the care needed in interpreting the results of individual investigators who may be using biologically different organisms. The possibility that continued laboratory passage of a given Leishmania may change its biological characteristics superimposes yet another problem. To prevent this, it is important to avoid the use of old laboratory strains, especially those which have passed through many hands. In conclusion, we are only too aware of the omissions that space and time will not permit us to cover in this paper. Perhaps, however, we have indicated something of the complexity of the subject and the problems still facing the worker in this field. ACKNOWLEDGMENTS

This work was prepared under the auspices of the Wellcome Trust, London, in conjunction with the World Health Organization, Geneva, and the Instituto Evandro Chagas da Fundaçâo Serviço Especial de Saúde Pública, Brazil. We are much indebted to Dr J. E. de Alencar and Dr A. Herrer who extended so much hospitality to R. Lainson during his visit to their laboratories in Fortaleza, Brazil, and Lima, Peru, in 1963; in particular for the welcome opportunity of photographing the kala-azar and uta endemic areas. The Royal Society of Tropical Medicine and Hygiene kindly granted us permission to reproduce Figures 4, 9-11, 13, 17, 19,21-27,30-32. We are grateful to the London School of Hygiene and Tropical Medicine for permission to use Figures 14 and 33. REFERENCES Adler, S. 1963. A serological comparison between a Leishmania from Phlebotomus martini and strains from kala-azar acquired in Kenya. Israel J. Exptl.Med. 11:31-4 - 1964. Leishmania. Advances in Parasitai. 2:3596

49

Adler, S., and Ber, M. 1941. The transmission of Leishmania trópica by the bite of Phlebotomus papatasii. Indian J. Med. Research 29:803—9 Adler, S., and Theodor, O. 1926a. Further observations on the transmission of cutaneous leishmaniasis to man from Phlebotomus papatasii. Ann. Trop. Med. Parasitol. 20:175-95 - 1926b. The identity of Leishmania trópica Wright, 1903 and Herpetomonas papatasii, Adler, 1925. Ann. Trop. Med. Parasitol. 20:355-64 - 1927. The behaviour of cultures of Leishmania trópica, L. infantum and L. brasiliensis in the sandfly Phlebotomus papatasii. Nature 119:48-9 - 1928. Infection of Phlebotomus sergenti with Leishmania trópica. Nature 122:278 - 1929a. Attempts to transmit Leishmania trópica by bite; the transmission of L. trópica by Phlebotomus sergenti. Ann. Trop. Med. Parasitol. 23: 1-18 - 1929b. The distribution of sandflies and leishmaniasis in Palestine, Syria, and Mesopotamia. Ann. Trop. Med. Parasitol. 23:269-306 - 1931. Investigations on Mediterranean kala-azar. i. Introduction and epidemiology. Proc. Roy. Soc. (London), B 108:447-59 - 1935a. Investigations on Mediterranean kalaazar. vu. Further observations on canine visceral leishmaniasis. Proc. Roy. Soc. (London), B 116: 494-504 - 1935b. Investigations on Mediterranean kalaazar. x. A note on Trypanosoma platydactyli and Leishmania tarentolae. Proc. Roy. Soc. (London)^ 116:543-4 - 1957. Transmission of disease agents by Phlebotomine sand flies. Ann. Rev. Entomol. 2:203-26 Aitken, T. H. G., Worth, C. B., and Tikasingh, E. S. 1968. Arbovirus studies in Bush Forest, Trinidad, W.L, September 1959-December 1964. Am. J. Trop. Med. Hyg. 17:253-68 Alencar, J. E., Pessôa, E. P., and Fontenele, Z. F. 1960. Infecçâo natural de Rattus rattus alexandrinus por Leishmania (provavelmente L. braziliensis) em zona endémica de leishmaniose tegumentar do estado do Ceará, Brasil. Rev. Inst. med. trop., Sâo Paulo 2:347-8

50 Amaral, A. D. F., Torrealba, J. W., and Henriquez, C. F. 1961. Phlebotomus longipalpis Lutz y Neiva, 1912, probable transmisor de la leishmaniasis visceral en Venezuela. Gac. méd. Caracas 70:389-408 Anon. 1958. Kept. Gorgas Mem. Lab. 1957 - 1965. Kept. Gorgas Mem. Lab. 1964 - 1967a. Kept. Gorgas Mem. Lab. 1966 - 1967b. Leishmaniasis Reference Lists 18901966. WHO/Leish/67.3; WHO/Leish/67.4 - 1968. WHO Inter-regional Travelling Seminar on Leishmaniasis. USSR, 22 May-10 June, 1967 - 1969. Kept. Gorgas Mem. Lab. 1968 Ansari, N., and Faguih, M. 1953. Cutaneous leishmaniasis from L. trópica in Rhombomys opimus. Ann. parasitol. humaine et comparée 28:241—6 Aragao, H. B. 1922. Transmissâo de leishmaniose no Brasil pelo Phlebotomus intermedius. Brasilméd. 36:129-30 Barreto, M. P. 1943. Observaçoes sobre a biología, em condiçoes naturals, dos flebotomus do Estado de Sao Paulo. Tese conc. Docencia Livre Fac. Med. Univ. Sao Paulo. Sao Paulo: Tipographia Rossolillo Berberian, D. A. 1938. Successful transmission of cutaneous leishmaniasis by the bite of Stomoxys calcitrans. Proc. Soc. Exptl. BioL, N.Y. 38:2546 - 1966. Mechanical transmission of Leishmania. Trans. Roy. Soc. Trop. Med. Hyg. 60:277-8 Biagi, F. F. 1953. Algunos comentarios sobre las leishmaniasis y sus agentes etiológicos, Leishmania trópica mexicana, neuva subespecie. Med. rev. mex. 33:401-6 Biagi, F. F., and Biagi, A. M. de B. 1953. Algunos Flebotomus del área endémica de leishmaniasis tegumentaria americana del Estado de Campeche (Méx.). Medicina (Mex.) 33:315-19 Biagi, F. F., Biagi, A. M. de B., and Beltran, H. F. 1965. Phlebotomus flaviscutellatus, transmissor natural de Leishmania mexicana. Prensa méd. mex. 30:267-72 Biagi, F. F., López, R., and Biagi, A. M. de B. 1965. El kala-azar en México; problema ecológico por estudiar. Rev. Inst. salub. y enferm. trop. 25:3-12

R. LAINSON AND J. J. SHAW

Carini, A. 1911. Leishmaniose de la muqueuse rhino-bucco-pharyngée. Rev. méd., Sao Paulo 14:364-6 Carini, A., and Paranhos, U. 1909. Identificaçâo des úlceras de Baurú ao botao do Oriente. Rev. méd., Sao Paulo 12:111-16 Chagas, A. W. 1940. Criacao de flebotomus e transmissâo experimental da leishmaniose Americana. Mem. inst. Oswaldo Cruz 35:327-33 Chagas, E., Cunha, A. M., Castro, G. O., Ferreira, L. C., and Romana, C. 1937. Leishmaniose visceral americana (Nova entidade mórbida do homem na América do Sul). Relatório dos trabalhos realizados pela comissâo encarregada do estudo da Leishmaniose visceral americana em 1936. Mem. inst. Oswaldo Cruz 32:321-89 Chagas, E., Cunha, A. M., Ferreira, L. C., Deane, L., Deane, G., Guimarâes, F. N., Paumgartten, M. J., and Sá, B. 1938. Leishmaniose visceral americana (Relatório dos trabalhos realizados pela comissâo encarregada do estudo da Leishmaniose visceral americana em 1937). Mem. inst. Oswaldo Cruz 33:89-229 Chodukin, N. I. 1929. Fundamental problems of the epidemiology of kala-azar in relation to the epidemiology of canine leishmaniasis in Middle Asia. Suppl. to Pensée Med. d'Usbekistane et de Turquménistane, p. 146 Christensen, H. A., Herrer, A., and Telford, S. R. 1969. Leishmania braziliensis s. lat., isolated from Lutzomyia panamensis in Panama. /. Parasitol. 55:1090-1 Christophers, S. R., Shortt, H. E., and Barraud, P. J. 1925. The development of the parasite of Indian kala-azar in the sandfly Phlebotomus argentipes Annandale and Brunetti. Indian J. Med. Research 12:605-8 Coelho, M., and Falcâo, A. R. 1962. Transmissâo experimental de Leishmania braziliensis. 2. Transmissâo da amostra mexicana por picada de Phlebotomus longipalpis e de Phlebotomus renei. Rev. Inst. méd. trop., Sao Paulo 4:220-4 Coelho, M., Falcâo, A. R., and Falcâo, A. L. 1967. Desenvolvimento de especies do género Leishmania em especies brasileiras de flebótomos do género Lutzomyia Franca, (1924). i. Evoluçâo

EPIDEMIOLÓGICA!. CONSIDERATIONS OF THE LEISHMANIAS

de L. brasiliensis em flebótomos. Rev. Inst. med. trop., Sâo Paulo 9:177-91 Costa, O. R. 1966. Calazar no Municipio de Cachoeira do Arari, Para. Rev. Serv. especial Saúde publ 12:91-8 Coutinho, J. O. 1941. Localizaçâo de formas em leptomonas, possivelmente de Leishmania brasiliensis, no faringo do Phlebotomus pessoai naturalmente infectado. Arq. Hig. Saúde públ. 6: 23-9 Cunningham, D. D. 1885. On the presence of peculiar parasitic organisms in the tissue of a specimen of Delhi boil. S ci. Mem. Med. Offrs. Army India 1:21 Damasceno, R. G., Arouk, R., and Causey, O. R. 1949. Estudos sobre Flebotomus na Amazónica. Parte vi - Contribuiçao ao conhecimento da distribuiçâo geográfica e da incidencia por tipo de captura, de 64 especies identificadas. Rev. Serv. especial Saúde públ. 2:817-43 Deane, L. M. 1948. Comentario ao trabalho de C. A. Hoare, "The relationship of the Haemoflagellates." Proc. 4th Intern. Congr. Trop. Med. and Malaria ( Washington ), 2:1117 - 1956. Leishmaniose visceral no Brasil; estudos sobre reservatórios e transmissores realizados no Estado do Ceará. Thesis, Fac. Med. Univ. Sao Paulo, Rio de Janeiro: Ed. S.N.E.S. Deane, L. M., and Deane, M. P. 1954. Encontró de leishmanias ñas visceras e na pele de urna raposa na zona endémica de Calazar, nos arredores de Sobral, Ceará. Hospital, Rio de Janeiro 45:41921 - 1962. Visceral leishmaniasis in Brazil: Geographical distribution and transmission. Rev. Inst. med. trop., Sâo Paulo 4:198-212 Deane, M. P., and Deane, L. M. 1954a. Infecçâo natural de Phlebotomus longipalpis por leptomonas, provàvelmente de Leishmania donovani, em um foco de calazar, no Ceará. Hospital, Rio de Janeiro 45:697-701 - 1954b. Infecçâo experimental de Phlebotomus longipalpis em caso humano de leishmaniose visceral. Hospital, Rio de Janeiro 46:487-9 - 1954e. Infecçâo expérimental de Phlebotomus longipalpis em raposa (Lycalopex vetulus) na-

51

turalmente parasitada pela Leishmania donovani. Hospital, Rio de Janeiro 46:651-3 - 1955. Observaçôes sobre a transmissao de leishmaniose visceral no Ceará. Hospital, Rio de Janeiro 48:347-64 Disney, R. H. L. 1966, A trap for phlebotomine sandflies attracted to rats. Bull. Entomol. Research 56:445-51 - 1968. Observations on a zoonosis: Leishmaniasis in British Honduras. J. Appl. Ecol. 5:1-59 Fairchild, G. B., and Hertig, M. 1959. Geographic distribution of the Phlebotomus sandflies of Central America (Diptera, Psychodidae). Ann. Entomol. Soc. Am. 52:121-4 Feng, L. C., and Chung, H. L. 1938. The development of Leishmania in Chinese sandflies fed on dogs with canine Leishmaniasis. China Med. J. 56:35-46 Ferreira, L. C., Deane, L., and Mangabeira Fo., O. 1938. Infecçâo de "Flebotomus longipalpis" pela "Leishmania chagasi." Hospital, Rio de Janeiro 14:2-3 Forattini, O. P. 1954. Algumas observaçôes sobre biología de flebotomus (Diptera: Psychodidae) em regiao da Bacia do Rio Paraná (Brasil). Arq. Fac. Hig. Saúde públ., Univ. Sâo Paulo 8:15136 - 1959a. Flebotomus transmissores das Leishmanioses na regiao neotropical. Arch, venezolanos med. trop. 3:193-205 - 1959b. Sobre osflebotomusdo territorio do Amapá, Brasil. Arq. Fac. Hig. Saúde públ., Univ. Sâo Paulo 13:159-64 - 1960. Sobre os reservatórios naturais da leishmaniose tegumentar americana. Rev. Inst. med. trop., Sâo Paulo 2:195-203 Forattini, O. P., and Santos, M. R. 1952. Nota sobre infecçao natural de Phlebotomus intermédius Lutz e Neiva 1912, por formas em leptomonas, em um foco de leishmaniose tegumentar americana. Arq. Hig. Saúde públ. 17:171-4 Garnham, P. C. C. 1962. Cutaneous leishmaniasis in the new world with special reference to Leishmania mexicana. Sci. Repts. 1st Super. Sanita 2:76-82 Garnham, P. C. C., and Lewis, D. J. 1959. Parasites

52 of British Honduras with special reference to leishmaniasis. Trans. Roy. Soc. Trop. Med. Hyg. 53:12-40 Giraud, P., Ranque, J., and Cabassu, H. 1954. Epidemiologie de la leishmaniose viscérale humaine méditerranéene en particulier dans ses rapports avec la leishmaniose canine. Arch, franc, pédiat. 11:337-53 Gunders, A. E., Foner, A., and Montilio, B. 1968. Identification of Leishmania species isolated from rodents in Israel. Nature 219:85-6 Gunders, A. E., Lidror, R., Montilio, B., and Amiti, P. 1968. Isolation of Leishmania species from Psammomys obesus in Judea. Trans. Roy. Soc. Trop. Med. Hyg. 62:465 Heisch, R. B. 1954. Studies on leishmaniasis in East Africa. I. The epidemiology of an outbreak of kala-azar in Kenya. Trans. Roy. Soc. Trop. Med. Hyg. 48:449-69 - 1957. The isolation of Leishmania from a ground squirrel in Kenya. E. African Med. J. 34:183 - 1963. Is there an animal reservoir of kala-azar in Kenya? E. African Med. J. 40:359-62 Heisch, R. B., Grainger, W. E., and Harvey, A. E. C. 1959. Isolation of a Leishmania from gerbils in Kenya. /. Trop. Med. Hyg. 62:158-9 Heisch, R. B., Wijers, D. J. B., and Minier, D. M. 1962. In pursuit of the vector of kala-azar in Kenya. Brit. Med. J. 1:1456-8 Herrer, A. 1948. Nota preliminar sobre leishmaniasis natural em perros. Rev. med. exptl. (Lima) 7:62-9 - 195la. Estudios sobre Leishmaniasis Tegumentaria en el Perú. Rev. med. exptl. (Lima) 8:45 - 195Ib. Estudios sobre Leishmaniasis Tegumentaria en el Perú. VI. Relación entre Phlebotomus y Leishmaniasis Tegumentaria. Rev. med. exptl. (Lima) 8:119-37 Herrer, A., Thatcher, V. E., and Johnson, C. M. 1966. Natural infections of Leishmania and trypanosomes demonstrated by skin culture. /. Parasitol. 52:954-7 Hertig, M. 1964. Bartonellosis. In Preventive Medicine in World War II, éd. Ebbe Curtis Hoff, vn:23-8. Washington, D.C.: Office of SurgeonGeneral, Dept. of Army

R. LAINSON AND J. J. SHAW

Hertig, M., and McConnel, E. 1963. Experimental infection of Panamanian Phlebotomus sandflies with Leishmania. Exptl. Parásito!. 14:92106 Heyneman, D. 1963. Leishmaniasis in the Sudan Republic. 12. Comparison between infections of Leishmania donovani in Phlebotomus papatasii (Díptera: Psychodidae) and the natural infections found in P. orientalis with its implications as to the normal vector of Sudanese kala-azar. Am. J. Trop. Med. Hyg. 12:725-40 Hindle, E. 1928. Further observations on Chinese kala-azar. Proc. Roy. Soc. (London), B 103: 599-619 - 1931. The development of various strains of Leishmania in Chinese sandflies. Proc. Roy. Soc. (London), B 108:366-83 Hoogstraal, H., and Dietlein, D. R. 1964. Leishmaniasis in the Sudan Republic: Recent results. Bull. Wld. Hlth. Org. 31:137-43 Hoogstraal, H., Dietlein, D. R., and Heyneman, D. 1962. Leishmaniasis in the Sudan Republic. 4. Preliminary observations on man-biting sandflies (Psychodidae: Phlebotomus) in certain upper Nile endemic areas. Trans. Roy. Soc. Trop. Med. Hyg. 56:411-22 Johnson, P. T., McConnel, E., and Hertig, M. 1963. Natural infections of leptomonad flagellates in Panamanian Phlebotomus sandflies. Exptl. Parasitol. 14:107-22 Kerdel-Vegas, F., and Essenfeld-Yahr, E. 1966. American leishmaniasis in a field rodent. Trans. Roy. Soc. Trop. Med. Hyg. 60:563 Knowles, R., Napier, L. E., and Smith, R. O. A. 1924. On a herpetomonas found in the gut of the sandfly Phlebotomus argentipes, fed on kalaazar patients. Indian Med. Gaz. 59:593-697 Lainson, R., and Shaw, J. J. 1968. Leishmaniasis in Brazil: I. Observations on enzootic rodent leishmaniasis - incrimination of Lutzomyia flaviscutellata (Mangabeira) as the vector in the lower Amazonian basin. Trans. Roy. Soc. Trop. Med. Hyg. 62:385-95 — 1969a. Some reservoir-hosts of Leishmania in wild animals of Mato Grosso State, Brazil. Two distinct strains of parasites isolated from man

EPIDEMIOLÓGICA!, CONSIDERATIONS OF THE LEISHMANIAS

and rodents. Trans. Roy. Soc. Trop. Med. Hyg. 63:408-9 - 1969b. Leishmaniasis in Brazil: in. Cutaneous leishmaniasis in an opossum, Marmosa murina (Marsupialia, Didelphidae) from the lower Amazon region. Trans. Roy. Soc. Trop. Med. Hyg. 63:738-40 - 1970. Leishmaniasis in Brazil: v. Studies on the epidemiology of cutaneous leishmaniasis in Mato Grosso State, and observations on two distinct strains of Leishmania isolated from man and forest animals. Trans. Roy. Soc. Trop. Med. Hyg. 64:654-67 Lainson, R., Shaw, J. J., and Lins, Z. C. 1969. Leishmaniasis in Brazil: iv. The fox, Cerdocyon thons (L) as a reservoir of Leishmania donovani in Para State, Brazil. Trans. Roy. Soc. Trop. Med. Hyg. 63:741-5 Lainson, R., and Southgate, B. A. 1965. Mechanical transmission of Leishmania mexicana by Stomoxys calcitrans. Trans. Roy. Soc. Trop. Med. Hyg. 59:716 Lainson, R., and Strangways-Dixon, J. 1962a. Dermal leishmaniasis in British Honduras: Some host-reservoirs of L. braziliensis mexicana. Bri. Med.]. 1:1596-8 - 1962b. Leishmania mexicana'. The epidemiology of dermal leishmaniasis in British Honduras. Trans. Roy. Soc. Trop. Med. Hyg. 57:242-65 — 1964. The epidemiology of dermal leishmaniasis in British Honduras: Part n. Reservoir-hosts of Leishmania mexicana among the forest rodents. Trans. Roy. Soc. Trop. Med. Hyg. 58:136-53 Latyshev, N. L, and Kryukova, A. P. 1941. On the epidemiology of cutaneous leishmaniasis. The cutaneous leishmaniasis as a zoonotic disease of wild rodents in Turkmenia. Trudy Voenno-med. Akad. RKKA 25:229-42 (In Russian) Leishman, W. B. 1903. On the possibility of the occurrence of trypanosomiasis in India. Brit. Med. J. 1:1252 - 1904. Note on the nature of the parasitic bodies found in tropical splenomegaly. Brit. Med. J. 1:303 Leon, J. R., and Figueroa, L. N. 1959.1. Descubrimento de la primeira zona endémica de kala-

53

azar guatemalense y sus condiciones epidemiológicas. Rev. Col. méd., Guatemala 10:240-5 Lindenberg, A. 1909. A úlcera de Baurú e seu microbio. Rev. méd., Sao Paulo 12:116-20 Lopes, J. A. S. 1956. Phlebotomus longipalpis naturalmente infectados comformas em leptomonas na cidade de Jocobina, Estado da Bahia. Rev. med. Paraná, Brasil 25:57-8 Manson-Bahr, P. E. C., and Heisch, R. B. 1961. Transient infection of man with a Leishmania (L. adíen) of lizards. Ann. Trop. Med. Parásito!. 55:381-2 Manson-Bahr, P. H. 1966. Manson's Tropical Diseases — A Manual of Diseases of Warm Climates. London: Bailliére, Tindall & Cassell Mazza, S., and Cornejo, A. J. 1926. Primeros casos autoctones de kala-azar infantil comprovados en el Norte de la República. Bol. Inst. clin. Quir. 11:140-4 McConnel, E. 1963. Leptomonads of wild-caught Panamanian Phlebotomus: Culture and animal inoculation. Exptl. Parasitai. 14:123-8 - 1964. Leishmaniasis in the Sudan Republic: 17: Infections in Phlebotomus orientalis Parrot (Díptera: Psychodidae) after feeding on a human kala-azar patient. /. Trop. Med. Hyg. 67 : 88-9 Migone, L. E. 1913. Un caso de kala-azar a Assuncion (Paraguay). Bull. soc. pathol. exotique 6: 118-20 Minter, D. M., and Wijers, D. J. B. 1963. Studies on the vector of kala-azar in Kenya, iv. Experimental evidence. Ann. Trop. Med. Parasitai. 57:24-31 Nadim, A., Seyedi Rashti, M. A., and Mesghali, A. 1968. On the nature of leptomonads found in Sergentomyia sintoni in Knorassan, Iran and their relation to lizard leishmaniasis. /. Trop. Med. Hyg. 71:240 Neiva, A., and Barbará, B. 1917. Leishmaniasis tegumentaria americana. Numerosos casos autoctones en la República Argentina. 1° Conf. SudAmer. Soc. Microbiol. Patol., Buenos Aires, pp. 311-72 Nery-Guimarâes, F., and Azevedo, M. 1964. [Wild rodents (Oryzomys goeldi) of Amazonia with

54

natural Leishmania infection.] Hospital, Rio de Janeiro 66:279-85 Nery-Guimarâes, F., Azevedo, M., and Damasceno, R. 1968. Leishmaniose tegumentar — zoonose de roedores silvestres na Amazonia. Mem. inst. Oswaldo Cruz 66:151-68 Parrot, L., and Donatien, A. 1927. Le parasite du bouton d'Orient chez le phlébotome. Infection naturelle et expérimentale de P. papatasii (Scop.). Arch. inst. Pasteur Algérie 5:9-21 Parrot, L., Donatien, A., and Lestoquard, F. 1930. Sur le développement du parasite de la leishmaniose canine viscérale chez Phlebotomus major var. perniciosus Newstead. Bull. soc. pathol. exotique 23:724—6 Penna, N. A. 1934. Leishmaniose viscéral no Brasil. Brasil-méd. 48:949-52 Pessôa, S. B., and Barreto, M. P. 1948. Leishmaniose tegumentar americana. R. de Janeiro, Minist. Educ. e Saúde, Service de Documentaçao, Imprensa Nacional Pessôa, S. B., and Coutinho, J. O. 1940. Infecçâo natural do Phlebotomus pessoai por formas em leptomonas, provàvelmente da Leishmania brasiliensis. Rev. biol. hig. 10:139-42 - 1941. Infecçâo natural e experimental nos flebotomus pela Leishmania brasiliensis, no Estado de Sâo Paulo. Hospital, Rio de Janeiro 20:25-35 Pessôa, S. B., and Pestaña, B. R. 1940. Infecçâo natural de Phlebotomus migonei por formas em leptomonas provàvelment da L. brasiliensis. Acta Mea. Chir. Brasil 5:106-11 Pif ano, C. F. 1940. Sobre el desarrollo de Leishmania trópica brasiliensis en Phlebotomus de Venezuela. Gac. méd. Caracas 48:114-16 - 1943. Notas sobre entemologia médica venezolana. I. Flebótomus transmissores de leishmaniasis tegumentaria en Valle de Yaracury. Bol ent. venez. 2:99-102 - 1960. Algunos aspectos de la patología comparada geográfica de la leishmaniasis tegumentaria en el trópico Americano. Gac. méd. Caracas 68:89-98 Pifano, C. F., Alvarez, A., and Ortiz, I. 1962. Investigaciones sobre la leishmaniasis tegumentaria Americana en los Valles del Tuy, Venezuela.

R. LAINSON AND J. J. SHAW

Arch, venezolanos med. trop. 4:149-55 Pifano, C. F., and Romero, M. J. 1964. Comprobación de um nuevo foco de leishmaniasis en Venezuela, Valle de Cumanaco, Edo., Sucre. Gac. méd. Caracas 72:473-9 Pifano, C. F., Romero, M. J., and Alvarez, A. 1962. Algunas observaciones sobre la ecología en condiciones naturales y de laboratorio, del Phlebotomus longipalpis Lutz y Neiva, 1912 : Vector del Kala-Azar en la región neotrópica. Arch, venezolanos med. trop. 4:17—23 Pondé, R., Mangabeira Fo., O., and Jansen, G. 1942. Alguns dados sobre a Leishmaniose visceral americana e doença de Chagas no Nordeste Brasileiro. Mem. inst. Oswaldo Cruz 37:333-52 Ponirovsky, E. N. 1969. [Epidemiological importance of sandflies (Phlebotomidae) in foci of visceral leishmaniasis in the Turkmen SSR.] Medskaya Parazitol. 38:62-6 Pons, A. R. 1968. Leishmaniasis tegumentaria americana en el Asentamiento Campesino de Zipayare. Aspectos epidemiológicos, clinicose e inmunológicos. Su importancia en la Reforma Agraria. Kasmera 3:5-59 Pressât, A. 1905. Le paludisme et les moustiques (prophylaxie). Paris : Masson et Cie Pringle, G. 1956. Kala-azar in Iraq. Preliminary epidemiológica! considerations. Bull. Endem. Dis. 1:275-94 Rioux, J. A., Goluan, Y., Croset, H., Houin, R., and Tour, S. 1968. Enquête écologique sur les leishmanioses dans le sud de la France. Bull. Inst. natl. santé rech. méd. 23:1125-64 Rogers, L. 1904. Preliminary note on the development of Trypanosoma in cultures of the Cunningham-Leishman-Donovan bodies of cachexial fever and kala-azar. Lancet 2:215-16 Safyanova, V. M. 1969. Leptomonads of naturally infected sandflies in the foci of zoonotic cutaneous leishmaniasis in Turkmenia. In Progress in Protozoology. Abstracts of Papers read at the 3rd Intern. Congr. on Protozool., Leningrad, 2nd-10th July 1969. Published by "Nauk," Leningrad Schneider, C. R. 1968. Immunodiffusion studies on a skin-inhabiting Leishmania from the tropical

EPIDEMIOLÓGICA!. CONSIDERATIONS OF THE LEISHMANIAS

porcupine Coendú rothschildi Thomas. /. Parásito!. 54:638-9 Schneider, C. R., and Hertig, M. 1966. Immunodiffusion reactions of Panamanian Leishmania. Exptl Parasitol 18:25-34 Sekhanov, M. V., and Suvorova, L. G. 1960. [Natural foci of cutaneous leishmaniasis in southwest Turkmenistan.] Medskaya Parazitol. 29: 524-8 Sergent, Ed., and Sergent, Et. 1905. Sur un culucide nouveau très commum à Biskra (Grabhamia subtilis). Compt. rend. soc. biol. 57:673 Sergent, Ed., Sergent, Et., Parrot, L., Donatien, A., and Béguet, M. 1921. Transmission du clou de Biskra par le phlébotome (Phlebotomus papatasi Scop.). Compt. rend. (Paris), 173:1030 Sergent, Et., and Parrot, L. 1929. Sur l'existence de Phlebotomus papatasi (Scop.) et de Phlebotomus minutus Rondani, en rae campagne. Bull, soc. paîhol. exotique 22:544 Shaw, J. J. 1964. A possible vector of Endotrypanum schaudinni of the sloth Choloepus hoffmanni in Panama. Nature 201:417-48 - 1969. The Haemoflagellâte s of Sloths. London School Trop. Med. Hyg. Mem. 13. London: H. K. Lewis & Co. Ltd. Shaw, J. J., and Lainson, R. 1968. Leishmaniasis in Brazil: n. Observations on enzootic rodent leishmaniasis in the lower Amazon region - the feeding habits of the vector, Lutzomyia flaviscutellata in reference to man, rodents and other animals. Trans. Roy. Soc. Trop. Med. Hyg. 62: 396-405 Sherlock, I. A. 1964. Notas sobre a transmissao da leishmaniose visceral no Brasil. Rev. bras, malariol 16:19-26 Sherlock, I. A., and Pessôa, S. B. 1966. Leptomonas infectando naturalmente Phlebotomus em Salvador (Bahia, Brasil). Rev. lat.-amer. microbiol. parasitai. 8:47-8 Shortt, H. E., Smith, R. O. A., Swaminath, C. S., and Krishnan, K. V. 1931. Transmission of Indian kala-azar by the bite of Phlebotomus argénupes. Indian J. Med. Research 18:1373—5 Sinton, J. A. 1925. Notes on some Indian species of the genus Phlebotomus. xi. The role of insects

55

of the genus Phlebotomus as carriers of the disease, with special reference to India. Indian J. Med. Research 12:701 Smith, R. O. A., Haider, K. C., and Ahmed, I. 1941. Further investigations on the transmission of kala-azar. Part vi. A second series of transmissions of L. donovani by P. argentipes. Indian J. Med. Research 29:799-802 Splendore, A. 1911. Bouba, blastomicose, leishmaniose ... Impr. méd. 19:1-7 Strangways-Dixon, J., and Lainson, R. 1962. Dermal leishmaniasis in British Honduras: Transmission of L. brasiliensis by Phlebotomus species. Brit. Med. J. 1:297-9 - 1966. The epidemiology of dermal leishmaniasis in British Honduras. Part 3. The transmission of Leishmania mexicana to man by Phlebotomus pessoanus, with observations on the development of the parasite in different species of Phlebotomus. Trans. Roy. Soc. Trop. Med. Hyg. 60:192201 Thatcher, V. E. 1968. Studies on phlebotomine sandflies using castor oil traps baited with Panamanian animals. J. Med. Entomol. 5:293-7 Thatcher, V. E., Eisenmann, C., and Hertig, M. 1965. A natural infection of Leishmania in the kinkajou Potus flavus, in Panama. /. Parasitol. 51:1022-3 Theodor, O. 1938. On African sandflies. m. Bull. Entomol. Research 29:165-73 - 1965. On the classification of American Phlebotominae. /. Med. Entomol. 2:137-9 Theodor, O., and Mesghali, A. 1964. On the Phlebotaminae of Iran. /. Med. Entomol. 1:285-300 Tikasingh, E. S. 1969. Leishmaniasis in Trinidad. A preliminary report. Trans. Roy. Soc. Trop. Med. Hyg. 63:411 Torrealba, J. W. 1964. Consideraciones sobre epidemiología de la leishmaniasis viscéral en Venezuela. Gac. méd. Caracas 72:99-115 Townsend, C. H. T. 1915. The insect vector of uta, a peruvian disease. J. Parasitol. 2:67-73 Vargas, L., and Nájera, D. A. 1959. Phlebotomus farilli n.sp., Ph. humbolti n.sp., y Ph. olmecus n.sp., de Mexico (Díptera, Psychodidae). Rev. Inst. salub. y enferm. trop. 19:141—53

56 Velez, L. R. 1913. Uta et espundia. Bull. soc. pathol. exotique 6:545 Vianna, G. 1911. Sobre urna nova especie de Leishmania (Nota preliminar). Brasil-méd. 25:411 Vlasov, V. P. 1932. On the finding of sandflies in the environs of Szchkad in the burrows of rodents Rhombomys opimus, Licht and Spermophilopsis leptodactylus Licht. Mag. Parasitol. Inst. Zool. Acad. Sci. USS 3:89-102 Wallace, F. G., and Hertig, M. 1968. Ultrastructural comparison of promastigote flagellates (Leptomonads) of wild-caught Panamanian Phlebotomus. J. Parasitol. 54:606-12 Wallace, F. G., and Todd, S. R. 1965. The comparison of leptomonads from Phlebotomus by electron microscopy. In Progress in Protozoology. Abstracts of Papers read at the 2nd Intern. Congr. on Protozool., London, 29th July-5th Aug., Excerpta Med. Intern. Congr. Ser. 91:133 Wenyon, C. M. 1911. Report of six months' work of the expedition to Baghdad on the subject of oriental sore. J. Trop. Med. Hyg. 14:103-9 - 1926. Protozoology. London: Bailliére, Tindall & Cassell Wijers, D. J. B., and Linger, R. 1966. Man-biting sandflies in Surinam (Dutch Guiana) : Phlebotomus anduzei as a possible vector of Leishmania braziliensis. Ann. Trop. Med. Parasitol. 60:501-8 Williams, P. 1966a. The biting rhythms of some anthropophilic phlebotomine sandflies in British Honduras. Ann. Trop. Med. Parasitol. 60:35764 - 1966b. Experimental transmission of Leishmania mexicana by Lutzomyia cruciaia. Ann. Trop. Med. Parasitol. 60:365-72 - 1969. On the transmission of Leishmania mexicana to man. Standing Advisory Committee for Medical Research in the British Caribbean, 14th Scientific Session, 17th-21st April, St. Augustine, Trinidad, pp. 16-17 Williams, P., Lewis, D. J., and Garnham, P. C. C. 1965. On dermal leishmaniasis in British Honduras. Trans. Roy. Soc. Trop. Med. Hyg. 59: 64-71 Young, C. W., and Hertig, M. 1926. The develop-

R. LAINSON AND J. J. SHAW

ment of flagellates in Chinese sandflies (Phlebotomus) fed on hamsters infected with L. donovani. Proc. Soc. Exptl. Biol. 23:611-15

Discussion LESLIE A. STAUBER

Department of Zoology and Bureau of Biological Research, Rutgers University, New Brunswick, N.J., USA Drs Lainson and Shaw's fine presentation of the current status of thinking on the epidemiology of the leishmaniases has shown so many points of concurrence among us that the task of discussion is made rather difficult. As a result, the issues raised are somewhat peripheral to the epidemiological problems. The first point deals with the finding, or not finding, of parasites and the distribution of parasites in the bodies of infected vertebrate hosts. I like to deal with quantitative model systems for their conceptual values but recognize the need to find explanations for apparent deviations occurring in naturally infected hosts. Variations in site of inoculation, source and strain of parasite, numbers of parasites inoculated, and age, sex, and condition of host animal often yield strikingly different results. Important also is the sensitivity of the method used to detect parasitization. This is well illustrated by the paper just heard and by the work of the Panama group. Both have been especially productive in disclosing infections in an increasing number of possible wild-animal reservoirs for cutaneous leishmaniasis. Kala-azar in the Sudan is another case. As the search progressed there (H. Hoogstraal and D. Heyneman, Am. J. Trop. Med. Hyg. 18:10911210 [1969]) and methods improved, the list of animals found naturally infected continued to grow also. Besides, in spite of the oft-repeated claim that parasites were not found in the peripheral blood of human patients in the Sudan, L. C. Rohrs (Am. J. Trop. Med. Hyg. 13:265-71 [1966]), by persistent search of blood films, did find them regularly, and E. McConnell (/. Trop. Med. Hyg. 67:88-9

EPIDEMIOLÓGICA!, CONSIDERATIONS OF THE LEISHMANIAS

[1964]) in a single trial was highly successful in infecting Phlebotomus orientalis from a patient. The methods used earlier must have been less than optimal since the blood smear examination method used by Rohrs is well known to be an inefficient procedure. N. S. Mansour, L. A. Stauber, and J. R. McCoy (/. Parasitol. 56:468-72 [1970]) in my laboratory studied kala-azar in dogs whose genetic and infectious history was known and which were in excellent nutritional condition. Using standardized procedures, the courses of infection of isolates of Leishmania donovani from the Mediterranean, Kenya, and the Sudan were followed for up to 512 days after inoculation. Biopsy material from blood and several organs obtained at various times was examined in culture and by animal inoculation. The blood and skin proved to have parasites only at irregular and unpredictable times even from the dogs infected with the Mediterranean strain of parasites. The evidence for lymphatic and haematogenous spread of leishmania or leishmania-infected macrophages seems well supported to me. Visceralization of parasites or the appearance of metastatic lesions are best explained on this basis. In the work on dogs, therefore, the possibility that the positive skin samples were obtained from parasites in the blood trapped in those skin samples cannot be overlooked. Under natural conditions traumatic, intercurrentinfectious, and allergic assaults on the host animal and the host's responses to these assaults during the weeks, months, or years of infection may greatly alter the local abundance of parasites and complicate the resulting clinical picture. Drs Lainson and Shaw accordingly searched for skin lesions to sample for parasites, but the wild animals with visceral leishmaniasis from the Sudan or the Panamanian porcupines without superficial lesions as a guide had to be searched more blindly. Such factors need to be considered in the search for reservoirs and in evaluating the importance of a particular positive or negative finding in such a survey. Several points about cutaneous leishmaniasis in Panama need to be added. New human cases are appearing regularly wherever deforestation is occurring. These new cases are from widely different geographical localities. The cases reported by B. C.

57

Walton, D. A. Person, and R. Bernstein (Am. J. Trop. Med. Hyg. 17:19-24 [1968]) in American military in-jungle training showed a very broad spectrum of lesion type, including even one described as "mucocutaneous." Apparently no differences have been noted in the hamster infections produced by Panamanian isolates so they are presumed to be closely similar. The epidemiology there is becoming more complicated and an ever-increasing list of wild-animal infections is appearing. At last count, 11 species of mammals from four orders (Carnivora, Edentata, Primata, and Rodentia) had proven infections. There are differences between these isolates in morphology of promastigotes in culture, in infectivity to hamsters, in behaviour in the sandfly gut, and in antigenic relationships. However, at least one species of mammal from each of the orders noted above has yielded a promastigote isolate producing typical lesions in hamsters. Some isolates from wild-caught sandflies were antigenically distinct, and that from the skin of the porcupine (Coendou rothschildi) was not only antigenically distinct but non-infective to hamsters on repeated trials. The sloth infections in Panama are the most interesting at the moment. Undoubtedly more than one type of flagellate has been isolated from these animals. There is no question that more than one type of promastigote (presumably of the genus Leishmania) has been found. Clearly some of the isolates are identical with the isolates from human infections, in hamster nose lesions, in morphology, and in shared antigenicity. These parasites are most unlikely to be L. donovani in my opinion. The high prevalence of leishmanial isolates from sloths and the high incidence of cutaneous infections in man are inconsistent with such an idea and, as Lainson and Shaw point out, kala-azar has not yet been reported from Panama. Also, the sloth isolates produce typical cutaneous lesions in hamsters with no greater tendency to visceralize than the other Panamanian cutaneous isolates. However, whether the sloth is a zoonotic reservoir in Panama is, as Dr Lainson has already stated, not yet proved although A. Herrer and S. R. Telford, Jr. (Science 164:141920 [1969]) consider it a prime suspect of the moment.

Morphological and physiological considerations of extracellular blood protozoa KEITH VICKERMAN

The bloodstream of vertebrates offers parasites a rich supply of nutrients and oxygen and facilities for rapid removal of waste products even if these are produced in abundance. The ubiquity of blood-sucking arthropods on land and leeches in water, moreover, ensures opportunities for transmission to other vertebrates by cyclical or mechanicals means. But bloodstream-dwelling has a serious disadvantage for, in adopting it, the intruder inevitably provokes the host's immune reaction to a far greater extent than if it had chosen the gut or even the tissues of its host. The intracellular niche of the sporozoan blood parasites might be expected to provide some protection from the host's humoural defences, but extracellular blood parasites must equip themselves with some means of countering host antibodies if their stay in the blood is to be more than brief. There are hazards in cyclical transmission by vector too, for the luxuries of circulating blood must be relinquished for the austerities of life in the vector's gut if the parasite is to survive a cycle of development which will lead to infection of a new vertebrate host. In this paper I shall be concerned almost wholly with the trypanosomes, and with the ways in which they react to these challenges. I shall also discuss briefly how the host responds to the trypanosome's reaction. Above all, I shall try to show how the physiological changes implicit in the trypanosome's responses are reflected in its fine structure, for it is mostly with the correlation of structure and function in the life cycles of these parasites that I have been concerned in my own researches. Although trypanosomes are known from all classes of vertebrates, we know most about the mammalian parasites. The dichotomy of this group into trypanosomes which are transmitted through the posterior station (hind-gut) and those that are transmitted via the anterior station

EXTRACELLULAR BLOOD PROTOZOA

(mouthparts) of the vector has long been recognized and Hoare (1964) has now dignified these categories taxonomically by naming them the Stercoraria and Salivaria respectively. The bulk of what I have to say will be about the Salivaria, the tsetse-borne trypanosomes of Africa and their emigré descendants, but I shall append relevant information on the stercorarian species and those parasitizing lower vertebrates. Among the salivarían trypanosomes I shall be preoccupied with Trypanosoma brucei. IMMUNOLOGICAL ASPECTS

Salivarían trypanosomes The bloodstream phase of the life cycle of T. brucei The human sleeping-sickness trypanosomes and their morphologically indistinguishable counterparts in game animals are perhaps best regarded as genetic variants or subspecies of T. brucei (Hoare 1966). T. brucei brucei of game animals and cattle will not infect man, but T. b. rhodesiense and T. b. gambiense will, causing, respectively, acute sleeping-sickness in East Africa and a more chronic form of the disease in West and Central Africa. T. b. rhodesiense is known to have a reservoir in antelopes and domestic cattle and is transmitted cyclically by the tsetse fly Glossina morsitans and related savanna flies, whereas T. b. gambiense has no known reservoir-host and is borne by Glossina palpalis and other riverine tsetse. The three T. brucei subspecies are morphologically indistinguishable at all stages in their life cycle. In the mammalian bloodstream they are seen as trypomastigote forms (nomenclature of Hoare and Wallace 1966) which vary considerably in size and shape, depending on the phase of the infection, and range from long, thin, slender forms to short, fat, stumpy forms. If we plot the number of T. brucei present in the blood of an infected mammal, we observe considerable fluctuation from day to day, and

59

definite peaks and troughs can be recognized (Fig. 1 ). We have believed for some time (Massaglia 1907) that each trough or remission represents the destruction of most of the trypanosomes by host antibodies. Each recrudescence is thought to be due to proliferation of trypanosomes which have an antigenic constitution different from their predecessors and so have managed to escape the host's immune assault (Levaditi and Mutermilch 1909). The trypanosome population's rise and fall have characteristic morphological accompaniments : in the ascending parasitaemia the dividing trypanosomes are predominantly long and slender, but as the crisis approaches many trypanosomes become short and stumpy, and during remission stumpy forms are usually predominant (Fig. 1 ). This morphological variation (pleomorphism) is continuous and occurs in clone strains of trypanosomes (Oehler 1914) so there is no question of there being a genetic basis to the multiplicity of forms. The tacit assumption is that the dividing slender trypanosomes transform into non-dividing stumpy ones, but the physiological significance of the transformation and the nature of the stimulus inducing it are not clear. As the advent of stumpy forms heralds the crisis brought on by host antibodies, however, it has long been suspected that the host's immune response induces the transformation from slender to stumpy forms (Bevan and MacGregor 1910; Ashcroft 1957). There is evidence that it is the stumpy forms which initiate cyclical development in the tsetse fly mid-gut (Robertson 1912;Reichenow 1921; Wijers and Willett 1960). In this context the behaviour of pleomorphic trypanosomes when they are maintained by syringe passage in laboratory rodents is relevant. After several passages the trypanosomes become monomorphic, slenderlike forms only being found in the blood (Fairbairn and Culwick 1947): this loss of pleomorphism is accompanied by loss of ability to develop cyclically in tsetse flies (Ashcroft 1960). A parallel loss is the ability of the trypanosomes to establish themselves in culture on blood agar

60

KEITH VICKERMAN

B

A

DAYS

PATENT

FIGURE 1. Relationship between the course of parasitaemia and variations in trypanosome morphology for a strain of T. brucei rhodesiense in an experimentally infected rat. Long, slender trypanosomes are predominant in the rising parasitaemia, and

short, stumpy forms during the remission. The populations at A and B would belong to different serotypes, those from B being unaffected by the antibody which had produced remission of the A population. (Based on data from Luckins 1969)

media (Reichenow 1932).* This is not surprising as the forms assumed in culture are morphologically and physiologically identical with those found in the fly mid-gut (Thomson and Sinton

1912), i.e. trypomastigotes in which the kinetoplast lies some distance from the posterior extremity of the body (Fig. 2). Loss of pleomorphism does not interfere with antigenic variation (Gray 1962) so it is unlikely that the stumpy forms are necessary for antigenic change and hence for continuing the bloodstream infection at a relapse. Slender trypanosomes are present at all stages in the infection and they might well be the forms which give rise to the relapse popula-

*T. evansi represents the natural counterpart of these non-transmissible and non-cultivable strains. Hoare (1940) has suggested that T. evansi arises from T. brucei when this trypanosome is carried outside the tsetse belt by camels and transferred from host to host by biting flies acting like syringes.

EXTRACELLULAR BLOOD PROTOZOA

tion. The contention that the stumpy forms have lost the ability to divide and instigate new bloodstream populations is supported by work on the relation between infectivity (iD 03 ) andtrypanosome numbers for samples taken at different stages in the population cycle. Slender trypanosomes of the early ascendancy have a high infectivity which is progressively reduced as stumpy forms become predominant in the blood (Cunningham, van Hoeve, and Lumsden 1963). The host's immune response and trypanosome antigenic change How host antibodies bring about the procession of trypanosome antigenic types (serotypes) observed in chronic infections is not known for certain, but we can offer two possible explanations. One is that these antibodies eliminate the predominant antigenic population at each crisis, but that genetic mutants of different serotype remain unaffected and through selection proliferate to give rise to the recrudescence (Watkins 1964; Seed and Gam 1966b). The other explanation is that each individual trypanosome carries in its genotype the full range of variant antigens, and that during a remission some trypanosomes manage to avoid host antibodies by switching to an alternative serotype (Inoki, Osaki, and Nakabayashi 1956). Genetical studies on Paramecium aurelia (Beale 1954) have shown that this sort of adaptation at the individual level underlies change of the immobilization antigens in this free-living protozoan. Unfortunately trypanosomes have no reliably inducible sexual process and so similar studies are not possible with them; but certain observations seem to weight the argument in favour of adaptation (rather than mutation) as the cause of the trypanosome population's repeated revival: I shall present these briefly. The fact that antigenic variation can occur in clone infections (Gray 1965a) has been used to discredit the mutation hypothesis, but Watkins (1964) claimed that the mutation rate required to account for the appearance of variants at 3-

61

day intervals in the mouse is less than 1 in 10G trypanosomes, which is not abnormally high. What is more significant, however, is Gray's ( 1965b) finding that in cyclically transmitted T. brucei infections, the antigens appear in a predictable sequence (A —» B —» c -> D etc.) : on passing through the tsetse fly vector all variants revert to the "basic antigen" (A) to start the sequence all over again. In the antibody-free environment of a non-immune host the basic antigen may replace other variants. It is difficult to account for such predictability in terms of mutation. The metacyclic trypanosome infection appears to have its antigenic variation programmed in the individual in the same way that the antigenic development of an embryo is programmed in the nucleus of the fertilized egg. The random appearance of antigenic variants reported by some workers (e.g. Seed and Gam 1966b) might be attributed to the fact that they studied old syringe-passaged isolates. I have already referred to the morphological changes which result from continuous passage of these trypanosomes through rodents and, as I shall describe later, these are linked with physiological changes. Although such strains have proved invaluable in the elaboration of techniques for studying trypanosome serology, we may find eventually that antigenic variation in these strains is as relevant to the fly-initiated infection as studying the antigens of a tumour is relevant to a study of antigenic change in a developing mammal ! Like the variable antigens of Paramecium, those of T. brucei have been identified as unconjugated proteins (Brown and Williamson 1962; Seed and Weinman 1963; Williamson and Brown 1964). After trypsin digestion of the major variant antigens, Le Page (1968 ) could not recognize common pep tides. This suggests that the antigens are products of different genes and so provides more evidence for the adaptation (as opposed to the mutation) theory. The variable antigens have been characterized largely by agglutination reactions using living trypanosomes (Inoki, Kitaura, Makaboyosi, and

62

KEITH VICKERMAN

Kurogochi 1952; Inoki, Osaki, and Nakabayashi 1956; Soltys 1957b; Cunningham and Vickerman 1962; Gray 1962, 1965a, b, 1966; Seed 1963; Watkins 1964; Seed and Gam 1966a) and to a lesser extent by neutralization of inf ectivity reactions (Soltys 1957a). T. brucei enhances its immunogenicity by shedding its variable antigens into the host's blood as "exoantigen" (Weitz 1960a, b; Miller 1965). This is

shown by the fact that cell-free serum from infected rats will induce the formation of variantspecific agglutinating antibodies if injected into other rats : indeed the exoantigen is highly immunogenic. In addition to the variant antigens, trypanosomes of successive parasitaemic waves contain antigens in common with one another. These common antigens can be demonstrated by precipitin (Seed 1963; Williamson and Brown

Salivary gland stages

Bloodstream trypomastigotes metacyclic

dividing epimastigote

(antigenic changes) (basic antiger

slender

reacquired) epimastigote

dividing slender dividing intermediate

trypomastigote

trypomastigote stumpy

(antigenic identity lost) Mid-gut & cardia stages ( = culture forms) IN

TSETSE

FLY

FIGURE 2. Diagram of stages in the life cycle of T. brucei to show cyclical changes in surface properties in relation to antigenic identity and gross morphological characters (size; position of nucleus and kinetoplast). Forms with a surface coat (more positively charged) are shaded. The letters A, B, c, etc., refer to the sequence of antigenic variants characterizing successive relapse populations in the

IN

MAMMAL

blood. The transformation from slender to stumpy trypanosomes takes place in each relapse population. On entering the fly all variants lose their antigenic identity (i.e. assume common surface antigen x) as they lose the coat and then eventually revert to the basic antigen A as they reacquire the coat to become metacyclic forms. (Based on Vickerman 1969b)

63

EXTRACELLULAR BLOOD PROTOZOA

1964), immunofluorescence (Williams, Duxbury, Anderson, and Sadun 1963 ), and complement fixation techniques.* Seed (1963) found that bloodstream trypanosomes lose their variant antigenic identity when they are put into culture in vitro, and a similar loss might be assumed to occur on entering the fly. Moreover, different culture derivatives of T. brucei will cross-agglutinate with heterologous antisera (Seed 1964), so the culture forms do not display the antigenic diversity of their bloodstream predecessors. The reversion to basic antigen envisaged by Gray ( 1965b) as taking place in the fly must occur before the metacyclic stage is reached : the infectivity of metacyclics can be neutralized by antiserum to the first bloodstream parasitaemia arising after infection with similar metacyclics (Cunningham 1966), showing that metacyclics have the basic antigen. The variable exoantigens have been interpreted as metabolic by-products of the trypanosomes, possibly isozymic forms of enzymes secreted by the flagellates and adhering to them (Njogu 1966; Rudzinska and Vickerman 1968) : the secretion of glucose-6-phosphatase, for example, has been demonstrated for T. brucei gambiense (Seed, Byram, and Gam, 1967), though the exoantigen has not been identified with this or any other enzyme as yet. Weitz ( 1960a, b) observed that extensive washing of trypanosomes destroyed their agglutinability by homologous antiserum, and he attributed this to removal of exoantigen from their surfaces. He noted, however, that further exoantigen could be released from such washed trypanosomes on disruption. This discussion has been concerned almost entirely with T. brucei, but a similar story of antigenic variation is emerging for T. (Nannomonas)

* According to Seed and colleagues (1969) the agglutinating antibody is at first a macroglobulin (IgM) but later is also found in the microglobulin (IgG) fraction along with the antibodies to the common antigens. The persistent macroglobulinaemia in trypanosomiasis they attribute partly to the continual production of IgM antibodies to new antigenic variants.

congolense (Wilson 1968) and T. (Duttonella) vivax (Clarkson and Awan 1968). The structural basis of antigenic change By piecing together these observations on the variable antigens of trypanosomes with recent electron microscope studies on fine structural changes which occur during the life cycle of these flagellates, we are beginning to get a clearer picture of the nature of antigenic variation. These structural changes concern the trypanosome surface (Fig. 2). Electron micrographs of all bloodstream forms of T. brucei, T. congolense, and T. vivax show that the trypanosomes are bounded by the usual three-ply surface membrane but this is enveloped by a 15 nm-thick surface coat covering the entire body and flagellum (Fig. 3). This coat can be removed by extensive washing with saline. The coat is not present in culture forms of either T. brucei (Fig. 4) or T. congolense', it is also missing from developmental stages of T. brucei in the tsetse fly, but it is regained as the flagellates transform into metacyclics in the fly's salivary glands (Vickerman 1968,1969b). These structural changes parallel changes in the surface charge on the trypanosomes. Broom and Brown ( 1937), utilizing the erythrocyte's permanent negative charge in a simple cell adhesion test, noticed that positively charged T. brucei gambiense became negatively charged with the onset of development in the vector. They later (Broom and Brown 1939) observed that reversion to a positively charged surface took place in the salivary glands. More recent comparisons of the charge on bloodstream and culture trypanosomes (Hollingshead, Pethica, and Ryley 1963; Lanham 1968) using the more sophisticated techniques of electrophoresis and absorption on anión exchange resins have shown that the electrophoretic mobility of bloodstream T. brucei has an isoelectric point at about pH 7 (i.e. the net surface charge is zero under physiological conditions) whereas T. congolense and T. vivax are more negatively charged, though not as negative as cultured T. brucei with its isoelectric

EXTRACELLULAR BLOOD PROTOZOA

NOTE ON ELECTRON MICROGRAPHS I Unless Otherwise stated material was fixed in phosphate-buffered glutaraldehyde and stained with uranyl acetate and lead citrate; the scale on each micrograph represents 1 jam. ABBREVIATIONS: ar, agranular (smooth) membranes; ax, axoneme of flagellum; bb, basal body; cr, cristae of mitochondrion; cv, coated vesicle; db, dense body (microbody, peroxisome?) ; e, endosóme (nucleolus) ; f, flagellum; far, flagellum-associated reticulum; fd, flagellar desmosome; fp, flagellar pocket; G, Golgi apparatus; gr, granular reticulum; k, kinetoplast; m, mitochondrion; mvb, multivesicular body; n, nucleus; pmt, pellicular microtubules; pr, paraxial rod of flagellum; r, ribosomes; sm, surface membrane. Other abbreviations are given under individual figures. FIGURES 3 and 4. Electron micrographs of T. brucei rhodesiense to show difference in surface character between bloodstream and culture forms (from Vickerman 1969b). X 130,000 3 Bloodstream form. Transverse section of flagellum (viewed from base) and adjacent pellicle to show adhesion zone (between large arrowheads). Note the coat overlying the surface membrane of both body and flagellum, also the pellicular microtubules. The region of the desmosome-like attachment of the flagellum to the pellicle is indicated by small arrowheads. A reduced microtubule (rmt) and branch of the granular endoplasmic reticulum (far) flank this attachment. 4 Culture form. A section comparable to that shown in Figure 3 but the attached flagellum (f) is viewed from the tip : adhesion is between arrowheads, attachment zone is at arrows. The flagellum to the left (ff) belongs to another trypanosome. Note the absence of a surface coat.

65

point of pH 3. Extensive washing of bloodstream trypanosomes (which would remove the surface coat) leads to a shift in the isoelectric point to this low value. It seems probable that the high isoelectric point is linked to the presence, the low value to the absence, of the surface coat. We can relate the cycle of coating and uncoating to antigenic behaviour in the following way. The loss of surface antigenic character which bloodstream trypanosomes undergo in culture or in the fly mid-gut or after extensive washing in saline may be due to loss of the surface coat observed by electron microscopy. The antigenic sameness of metacyclic trypanosomes and the trypanosomes of the first bloodstream population may be due to the fact that both metacyclics and bloodstream forms carry a surface coat. The cross-agglutination found with different culture derivatives of T. brucei may be accounted for by antigenic similarity of their uncoated membranes. It seems likely that the antigenic uniqueness of each bloodstream variant resides in the surface coat and that the exoantigen may be discarded coat. There is evidence that the variable antigens are located in the surface coat from studies with ferritin-conjugated antibodies (Vickerman and Luckins 1969). Clone stabilates of two relapse variants of a single strain of T. brucei were injected into rabbits and these animals bled six days later for antisera with maximum variantspecific antibody titre (Gray 1965a). Ferritin was conjugated to the globulins extracted from the sera and trypanosomes treated with homologous and heterologous conjugate were examined as sections by electron microscopy. The ferritin conjugate showed a marked affinity for the trypanosome surface in homologous reactions only (Fig. 7 ), but trypanosomes which had had their coats removed by extensive washing did not bind even homologous conjugate (Fig. 8 ). These experiments suggest that the antibody produced by the rabbits was directed largely against surface coat components and that the coats were variantspecific in their antigenic character. As yet little information is available on the

Golgi apparatus

flagellar pocket

subtending granular reticulum

1st basal body

2nd (barren) basal body kinetoplast sac of secretion secretory reticulum

nucleus anterior granular reticulum

mitochondrial, canal flagellum-associàted granular reticulum pellicular microtubules flagellum

FIGURE 5. Diagram to show principal structures revealed by the electron microscope in the bloodstream trypomastigote form of a salivarian trypanosome (T. congolense). The flagellate is seen cut in sagittal section except for most of the shaft

of the flagellum and the anterior extremity of the body. Pellicular microtubules which underly the whole body surface are shown only in this anterior portion. (Based on Vickerman 1969a)

EXTRACELLULAR BLOOD PROTOZOA

composition of the trypanosome surface coat. Using fixed trypanosomes it is possible to digest away the surface coat with pronase (Vickerman 1969b), suggesting that proteins enter into its make-up. Moreover, the pH dependence of the electrophoretic mobility of bloodstream T. brucei suggests the presence of both acid and basic groups on the surface of the trypanosomes and the pK values for these groups suggest that they are carboxyl and amino respectively (Hollingshead et al. 1963 ), i.e. the surface is protein in nature. It will be recalled that the variable antigens have been identified as unconjugated proteins. It seems unlikely that the coat is an extraneous accretion of host serum proteins because an antigenically and morphologically similar coat is present in the metacyclic trypanosome. I do not want to pretend, however, that bloodstream trypanosomes do not bind host proteins, for we have evidence that they do. Washed T. vivax from mice will agglutinate with rabbit antimouse blood antiserum but not with normal rabbit serum (Ketteridge 1970). Desowitz (1954) believed that such bound serum components were important in adapting the trypanosome to life in the bloodstream of a particular host species, as for example in adapting sheep-passaged T. vivax to rodents. If the surface coat contains the variant antigens, how is the coat produced and how is antigenie change effected, for if we adopt the adaptation theory of antigenic change we must explain how one antigen is replaced by another? Perhaps the most likely mode of origin of the surface coat is by a process of secretion (Vickerman 1969a). Between the nucleus and the flagellar pocket the salivarían trypanosomes (Fig. 5) house all the apparatus of cellular secretion as known in higher organisms : there is abundant granular endoplasmic reticulum for the synthesis of protein to be secreted, there is a well-developed Golgi apparatus for packaging the secretion, and there is an elaborate system of smoothmembraned tubules and vesicles to export the secretory product to the flagellar pocket (Fig.

67

6). The secretory product could well be coated membrane, for the secretory apparatus is less prominent in culture forms. The exoantigen released from disrupted washed trypanosomes may represent unsecreted surface coat material. Autoradiographic studies with labelled amino acids, similar to those conducted on glandular tissue cells (Leblond and Warren 1965), are needed to confirm this hypothesis. If the coated membrane is secreted, a switch in coat antigens would mean a switch in the coat protein being secreted. But how would the trypanosome relieve itself of an old coat in order to replace it with a new one? In rapidly dividing trypanosomes the simplest answer would be that the daughter trypanosomes are progressively clothed in the new coat, the old one becoming diluted out. But another method of shedding old coat has been suggested by electron microscope studies on negatively stained trypanosomes. These reveal the presence of long, fine threads of cytoplasm extending from the anterior and posterior extremities of the bloodstream flagellates (Wright, Lumsden, and Hales 1970; MacAdam and Herbert 1970). These cytoplasmic streamers or plasmanemes* appear to be deciduous for they are found free in the serum of infected animals. Sectioned material (Figs. 7, 8) shows that they consist of coated membrane surrounding a cytoplasmic core. If the trypanosome surface is constantly being shed by way of these streamers, then we have here an obvious source of exoantigen and a means of getting rid of unwanted coat. The existence of these streamers on trypanosomes taken directly from the bloodstream and not subjected to any washing procedure, however, remains to be demonstrated. In summary, the surface coat of the salivarían trypanosomes appears to be an adaptation to bloodstream life for it is absent from the developmental cycle in the vector but is acquired by *I have used the term plasmaneme in preference to the term filopodia used by the cited authors, as there is no evidence that these structures are in any way analogous to fine pseudopodia.

68

KEITH VICKERMAN

EXTRACELLULAR BLOOD PROTOZOA

69

the metacyclics preparatory to infecting the mammal again. The coat may constitute a replaceable surface which is highly immunogenic, the different antigenic nature of the replacement coat allowing the trypanosome to avoid the host's antibodies and survive to give rise to the relapse population. The relationship between the host's immune response and changes in trypanosome morphology

FIGURE 6. Longitudinal (above) and transverse sections of bloodstream T. congolense. The longitudinal section shows the structures probably involved in secretion: subtending granular reticulum with blebs (arrowed) directed towards the Golgi apparatus, and nearby secretory reticulum (ar) with sacs of secretion (sar) ; the secretion may be conveyed from these sacs to the flagellar pocket by smooth-membraned tubules (st) and vesicles. The lower transverse section shows a coated vesicle forming from the flagellar pocket (black arrow) and one already formed in the cytoplasm. For other labelling see Figure 4. X 30,000 FIGURE 7. Part of transverse section of trypanosome (T. brucei) treated with homologous ferritin-conjugated antiserum. The ferritin conjugate particles (arrowed) adhere to the surface coat covering the body, flagellum, and plasmanemes (pi). X 120,000 (From Vickerman and Luckins 1969) FIGURE 8. Similar section to Figure 7 but of a trypanosome well washed in saline before treatment with homologous ferritin-antibody conjugate. The conjugate particles (arrowed) adhere abundantly to the plasmaneme, which retains its surface coat, but not to the surface membrane, from which the coat has been removed. X 120,000 (From Vickerman and Luckins 1969)

We have noted that the transformation from slender to stumpy forms accompanies the remission of parasitaemia in T. brucei, and so it is possible that the two events have a common cause. As the decline in parasite numbers is undoubtedly brought about by host antibodies, it has been widely assumed that the morphological change is also brought on by the host's antibody response. It is surprising, therefore, that there is no direct experimental evidence supporting a role for antibody in inducing the transformation; indeed, circumstantial evidence from infected animals in which the immune response has been suppressed implies that the transformation can proceed in the absence of an immune response. Ashcroft (1957) found that rats infected with a pleomorphic strain of T. brucei rhodesïense and given daily doses of cortisone acetate (5 mg/kg) had enhanced parasitaemias devoid of stumpy forms. Petana (1964) confirmed this observation. Luckins (1969), however, employing strains of different virulence, found that using even higher doses of th? drug, elimination of stumpy forms could never be achieved. He obtained similar results after using x-irradiation (Fig. 9) and splenectomy to bring about demonstrable suppression of agglutinin production. However, in chick embryos T. brucei produces stumpy forms in the absence of antibody production by the host (Wijers and Goedbloed 1967; Goedbloed 1968). It is of course possible that a variety of unfavourable environments can bring about the morphological change (Wijers 1960) ; indeed the nature of the physio^gical changes which accompany this transformation

Rat A

RatB

DAYS PATENT

FIGURE 9. Relationship between variation in morphology and course of parasitaemia for a clone strain of T. b. rhodesiense in two rats. Rat A was subjected to 600 rads of total body x-irradiation 24 hours before infection in order to suppress the immune response. Rat B was an untreated control.

Stumpy trypanosomes were present in the enhanced parasitaemia of the irradiated rat and the proportion of these to slender forms varied somewhat, if not so markedly as in the control. (From Luckins 1969)

EXTRACELLULAR BLOOD PROTOZOA

makes this appear very plausible (I shall describe these changes later). But the idea that antibodies are the usual agents provoking the genesis of stumpy forms cannot be abandoned lightly. Stercorarian trypanosomes and haemo flagellât e s of lower vertebrates We probably know more about the host-parasite relations of Trypanosoma (Herpetosoma) lewisi than of any other Stercorarian trypanosome.* Following contaminative infection of a rat by metacyclic trypanosomes, this trypanosome undergoes multiplication in the epimastigote form in visceral blood capillaries, appearing in various shapes and sizes in the peripheral blood after about the fifth day. About this time a crisis occurs in which most of the trypanosomes are killed by a trypanocidal antibody (Coventry 1930), more recently identified as a microglobulin (D'Alesandro 1959, 1966). The survivors are resistant to this antibody and are said to be prevented from multiplying further by a second antibody, "ablastin" (Taliaferro 1932). They become trypomastigote (adult) forms of uniform size, and these can initiate cyclical development in rat fleas (Nosopsyllus jasciatus, Xenopsylla cheopis}. If they fail to escape the rat circulation in this way the adult trypanosomes are cleared from the blood by a second trypanocidal antibody which is formed a few weeks after the first. This second trypanocidal antibody is a macroglobulin (D'Alesandro 1966). Not all workers see the necessity to invoke the intervention of "ablastin" (Augustine 1943; Ormerod 1963), and the dispute has been reviewed by D'Alesandro (1966). Ablastin is comparable to a microglobulin in its electrophoretic and ultracentrifugal behaviour (D'Alesandro 1959) but it is not absorbable by living trypanosomes and trypanosomes are not sensitized by exposure to it. Passive transfer of ablas*The title of my contribution precludes emphasis on T. cruzi, which is an intracellular parasite in its multiplicative phase.

71

tin activity has been demonstrated. Cortisone (200-400 mg/kg) treatment results in enhanced T. lewisi parasitaemias with a prolonged period of multiplication (Sherman and Ruble 1967), and similar results have been obtained after xirradiation (Tempelis and Lysenko 1965), suggesting that ablastin production can be suppressed. If adult trypanosomes are freed from the effect of ablastin by transfer to a non-immune host, nucleic acid and protein synthesis increases (Taliaferro and Pizzi 1960). D'Alesandro (1966) found a 60 per cent reduction in the incorporation of adenine in ablastin-inhibited parasites. Immunogel diffusion studies with extracts of adult T. lewisi by Entner (1968b) have shown the presence of a host serum component tightly bound to the trypanosome. The binding sites appear to be present in trace amounts only in the multiplicative forms. Whether this serum component is ablastin is not known. T. lewisi appears to have limited antigenic variation; one antigenic change helps to evade the first trypanolytic antibody but the second cannot be evaded. A surface coat is present in both the multiplicative (Molyneux 1970) and adult (Vickerman 1969b) bloodstream forms of this trypanosome but not in culture forms (Vickerman 1969b). The coat is diffuse and filamentous (Fig. 10) rather than compact as in the salivarian trypanosomes. A similar fuzzy coat is apparent in electron micrographs of other Stercorarian and lower vertebrate trypanosomes as well as blood-dwelling trypanoplasms (Fig. 12). The pronounced negative charge of both adult bloodstream and culture forms of T. lewisi (Hollingshead^/a/. 1963) suggests that the coat is different from that of the salivarian trypanosomes in its chemical composition as well as its structure. "Metabolic products" released by the multiplicative forms of T. lewisi (Thillet and Chandler 1957) will immunize rats against homologous challenge, so these "products" appear to be comparable to the exoantigen of T. brucei. The limited ability of T. lewisi to escape its host's trypanocidal antibodies might prove

FIGURE 10. Transverse section of anterior tip of adult bloodstream T. lewisi to show that the mitochondrion extends to this extremity. Note the fuzzy coat on the surface membrane and the gap in the pellicular microtubules where the flagellar desmosome attaches the flagellum to the body. X 84,000 (From Vickerman 1969b)

FIGURE 11. Longitudinal section of adult bloodstream T. lewisi showing principal structures. Note the profiles of the mitochondrial network (seen branching at arrow) with its plate-like cristae. A transverse section of a similar trypanosome in the nuclear region and blood platelets (P) lie to the right. X15,000

FIGURE 12. Transverse section of the trypanoplasm Cryptobia keysselitzi from tench blood. The attached flagellum (f ) and pellicular micro tubules resemble those of trypanosomes but the large kinetoplast is composed of bundles of DNA fibrils (small arrows) running in all directions (compare anisotropic arrangement in trypanosomes). Cytoplasmic

inclusions shown are lipid (lip), dense bodies, and smooth membranes. Note the plate-like cristae inside the kinetoplast capsule and (at large arrow) the fuzzy coat on the surface membrane. X 60,000 (Electron micrograph kindly provided by T. M. Preston)

74

eventually to be related to the different nature of the surface coat in this flagellate. ASPECTS OF RESPIRATORY METABOLISM

Salivarían trypanosomes

KEITH VICKERMAN

demands on oxygen and glucose by culture forms are one-tenth those of bloodstream forms and phosphorylation accompanies electron transfer along the cytochrome chain. Respiration in culture forms is therefore more conventionally aerobic.

Biochemical studies Old monomorphic laboratory strains of T. brucei syringe-passaged in the bloodstream phase display an active aerobic glycolysis. They break down glucose to pyruvic acid, but no further, for the Krebs tricarboxylic acid cycle does not appear to operate and this product is excreted into the blood: in the absence of oxidative decarboxylation negligible carbon dioxide is produced (summarized by von Brand 1951 ; Ryley 1956, 1962). The energy for the trypanosome's restless movement and all other endergonic processes seems to be supplied solely by the small amount of ATP synthesized during glycolysis. The pyridine nucleotide (nicotinamide adenine dinucleotide, NAD) reduced during glycolysis is oxidized via a non-phosphorylating glycerophosphate oxidase system and not via the more usual phosphorylating cytochrome chain (Fulton and Spooner 1959; Grant and Sargent 1960, 1961 ; Grant, Sargent, and Riley 1961) so that oxygen uptake is insensitive to cyanide. The glycerophosphate oxidase system accounts for the high oxygen demands of these bloodstream trypanosomes. This pattern of respiration is wasteful of both substrate and oxygen and we might expect that on finding itself in a semi-stagnant blood clot in the vector's mid-gut the trypanosome would be forced to implement economy measures. If we again fall back on trypanosomes cultured in vitro as analogues of those in the fly mid-gut, then our expectations are fulfilled. The culture forms in fact respire glucose completely to carbon dioxide and water: the Krebs cycle operates in oxidative decarboxylation, cytochrome pigments can be detected by spectrophotometry, and oxygen uptake is sensitive to cyanide (von Brand and Johnson 1947; Ryley 1956, 1962; Fulton and Spooner 1959). The

Cytological studies The inability of bloodstream T. brucei to oxidize pyruvate and the absence of detectable cytochrome pigments point to the absence of functional mitochondria in these forms, while the presence of an operative Krebs cycle and cytochrome chain in culture forms indicate active mitochondria. These predictions are borne out by electron microscope studies, for quite striking differences in the ultrastructure of the mitochondrion have been observed in these two derivatives of T. brucei (Vickerman 1962). The mitochondrial system of trypanosomes and other

FIGURE 13. Diagram to show changes in form and structure of the mitochondrion of T. brucei throughout its life cycle. The slender bloodstream form lacks a functional Krebs cycle and cytochrome chain. Stumpy forms have a partially functional Krebs cycle but still lack cytochromes. The glycerophosphate oxidase system functions in terminal respiration of bloodstream forms. The fly gut forms have a fully functional mitochondrion with active Krebs cycle and cytochrome chain. Cytochrome oxidase may be associated with the distinctive platelike cristae of these forms. Reversion to tubular cristae in the salivary gland stages may therefore indicate loss of this electron transfer system. All forms other than slender bloodstream forms gave a positive reaction for NADH-tetrazolium reducíase activity.

EXTRACELLULAR BLOOD PROTOZOA

75

IN SLENDER

MAMMALS

TRYPOMASTIGOTE Sparse, short tubular cristae INTERMEDIATE

TRYPOMASTIGOTE

Cristae lengthen

METACYCLIC TRYPOMASTIGOTE STUMPY TRYPOMASTIGOTE Many tubular

Closely-packed

cristae

tubular cristae

Numerous plate-like cristae

EPI MASTICÓTE

MID-GUT & CARDIA TRYPOMASTIGOTES (^CULTURE FORMS)

IN TSETSE FLIES

76

members of the flagellate order Kinetoplastida is of singular interest for each flagellate has just the one mitochondrion which extends the entire length of the flagellate's body. This mitochondrion is remarkable in that it contains a massive amount of DNA, the kinetoplast of light microscopists (reviewed Muhlpfordt 1964). This DNA is usually housed in a capsular expansion of the mitochondrion (Clark and Wallace 1960; Steinert 1960), and this capsule is apposed to the basal body of the locomotory flagellum. During division of the flagellate the kinetoplast-mitochondrion replicates just before the nucleus. This mitochondrion provides one of the most convincing examples of genetic continuity of a cytoplasmic organelle. Electron micrographs of monomorphic (slender) bloodstream T. brucei show that the mitochondrion is a narrow tubular structure extending fore and aft from the kinetoplast capsule (cf. Fig. 14). Internally it has few (if any) of the cristae which are characteristic of an active mitochondrion. Micrographs of cultured trypanosomes (Fig. 16) or of fly mid-gut forms, however, show that the mitochondrion is an elaborate network of kinetoplast-connected canals well furnished with plate-like cristae (Vickerman 1962). It appears that on entering the fly the trypanosomes produce mitochondrial enzymes and a more elaborate mitochondrial system to enable them to switch their pattern of respiration to a more economical one in keeping with their new surroundings. The change in the post-kinetoplastic mitochondrion is particularly noticeable, and I have suggested (Vickerman 1962) that adaptive proliferation of this part of the mitochondrion might be responsible for the change in gross morphology observed in the flagellate at this time, the new mitochondrial system growing out from the kinetoplast region pushing out the posterior end of the trypanosome. Adaptive mitochondriogenesis might in fact help to explain movement of the kinetoplast and the changes in over-all size which are features of trypanosome morphogenesis. But it is quite

KEITH VICKERMAN

likely that the longitudinally arranged microtubules which form a corselet underlying the plasma membrane of the flagellate also play an important part in orienting changes of shape as they do in other cells. To return to the physiological aspect of the transformation of bloodstream forms to fly gut forms, it is worth recalling here that monomorphic bloodstream trypanosomes, whose fine structure and respiration we have just considered, rarely if ever infect a tsetse fly whereas the stumpy forms of pleomorphic strains do : monomorphic bloodstream trypanosomes correspond morphologically to the slender forms of pleomorphic infections. Is there then a metabolic difference between slender (or monomorphic)

FIGURE 14. Longitudinal section of slender bloodstream form of T. brucei rhodesiense in region of flagellar pocket. The kinetoplast is composed of anisotropic DNA fibrils and is contained within a distinct capsular region of the single mitochondrial canal. Profiles of anterior and posterior regions of this canal (m) show its uniform narrow diameter and lack of cristae. X 34,000 FIGURE 15. Transverse (left) and longitudinal (right) sections of stumpy bloodstream forms of T. brucei rhodesiense. The proximity of the nucleus to the flagellar pocket and the unusual prenuclear location of the Golgi apparatus indicate that the trypanosome on the right is a posteronuclear form. Note that the kinetoplast of this trypanosome is not enclosed in a distinct capsular region of the mitochondrial canal, which in this stage is very broad and contains finely tubular cristae. The mitochondrial canal is also visible in the transverse section to the left. X 30,000

EXTRACELLULAR BLOOD PROTOZOA

77

78

and stumpy trypanosomes which confers a metabolic advantage upon the latter when the trypanosomes find themselves in the fly's gut? There is - and it concerns differences in the activity of the mitochondrion in these two forms (Vickerman 1965). I first became aware of this difference in studying electron micrographs of pleomorphic as compared to monomorphic trypanosomes. In populations of the former consisting largely of stumpy forms the mitochondrion is much broader (especially in the pre-kinetoplastic portion) and projecting into its lumen are more and longer tubular cristae (Fig. 15), similar to those found in the mitochondria of many free-living protozoa. A suggestion that this morphological change implied a change in mitochondrial activity came from comparative cytochemical studies on salivarían trypanosomes. NADH-tetrazolium reducíase activity ("diaphorase") can be used to demonstrate cytochemically the presence of active mitochondria in cells (Nachlas, Walker, and Seligman 1958 ), and the single mitochondrion of boodstream T. vivax or T. congolense can be demonstrated satisfactorily by this technique (Vickerman 1965). Monomorphic T. brucei or T. evansi, in contrast, do not stain for activity of this enzyme. In pleomorphic T. brucei, however, although slender forms are as negative as monomorphic strains (Fig. 19), intermediate and stumpy forms resemble T. vivax and T. congolense in giving a positive reaction in the mitochondrion (Figs. 19, 20), suggesting that NADH might be oxidized there. The dependence of motility on a supply of exogenous respirable substrate (Ryley 1956) may also be used to distinguish between trypanosomes with and those without an active mitochondrion. Ryley (1966) noted that T. brucei culture forms could maintain their motility if supplied with the Krebs cycle intermediate a-ketoglutaric acid («-KGA) while monomorphic bloodstream trypanosomes could not. If the experiment is repeated using a pleomorphic strain, the intermediate and stumpy forms can retain their motility in a-KGA long after movement has ceased in the control lacking

KEITH VICKERMAN

substrate (Vickerman 1965), presumably because they have a functional Krebs cycle. Further support for mitochondrial activity in pleomorphic (as opposed to monomorphic) T. brucei is now coming from the more detailed studies of Bowman and his colleagues (Bowman and Flynn 1968; Flynn and Bowman 1970). In a pleomorphic T. b. rhodesiense, they found significantly lower pyruvate production and significantly higher carbon dioxide and succinate production than in monomorphic strains, besides utilization of a-KGA. They conclude that oxidative carboxylation mechanisms are active in the pleomorphic strain but that the Krebs cycle operates at a rate which is quantitatively insignificant in vivo. As yet no cytochromes have been detected in pleomorphic strains, and some other route of electron transfer must be sought within the mitochondrion. Flynn and Bowman (1970) suggest that the L-a-GP oxidase system of bloodstream forms might be augmented by an autooxidizable mitochondrial flavoprotein in the

FIGURE 16. Transverse section of dividing T. brucei rhodesiense culture form. Each daughter flagellate shows a ring-like profile of the mitochondrial network: the cristae (arrowed) are plate-like as in tsetse gut stages of the trypanosome, X 40,000 FIGURE 17. Part of longitudinal section of epimastigote T. b. rhodesiense from salivary glands of Glossina pallidipes. The mitochondrial cristae show tubular profiles. Osmium tetroxide fixation. X 3 3,000 FIGURE 18. Part of longitudinal section of T. evansi equinum, a naturally dyskinetoplastic trypanosome. The dyskinetoplast (dk) is a dense spherical structure which in this case is not apposed to the basal body of the flagellum. X 30,000

80

terminal respiration of those trypanosomes that have developed a functional Krebs cycle, i.e. presumably the intermediate and stumpy forms. The L-a-GP oxidase system was localized by Ryley (1966) using a tetrazolium technique in single-membrane-bound particles (probably the dense bodies of Figs. 5, 15, 18, etc.). These bodies most likely fall within the category of organelles known as microbodies or peroxisomes which are associated with non-phosphorylating electron transfer, and they have now been isolated by Bayne, Muse, and Roberts (1969). So, activation of the T. brucei mitochondrion starts in the mammalian host and seems to occur as the slender trypanosomes transform into stumpy ones. This activation is, as we have seen, only a partial one, and this might be reflected in the tubular form of the mitochondrial cristae in stumpy forms (Fig. 15) as compared to the plate-like cristae of culture (Fig. 16) or fly midgut trypanosomes. Nevertheless, the partial mitochondrial activity of stumpy trypanosomes might explain why they are at an advantage when it comes to infecting the vector: they are one step along the road toward complete mitochondrial activity which seems to be necessary for continuing life in the new environment. Much less can be said of the mitochondrial changes that occur in the trypanosomes once they have left the fly gut for the salivary glands, for extreme scarcity of material has precluded detailed study. Electron microscopy of salivary gland forms (Vickerman 1970) has shown that the epigmastigote stage still has the mitochondrial reticulum but the cristae, though densely packed, have returned to the tubular condition (Fig. 17) : NADH-tetrazolium reducíase activity is found in the mitochondrial network. The metacyclics are also positive for this enzyme, and electron microscopy shows that packed tubular cristae are still present in the mitochondrion but both ultrastructure and cytochemistry indicate that this organelle has now reverted to the single canal found in bloodstream forms. From these cytological studies it looks as though the salivary gland stages show progressive reduc-

KEITH VICKERMAN

tion of mitochondrial activity in preparing to become blood parasites again. The lower salivarian trypanosomes T. vivax and T. con^olense and related species do not appear to have the complete repression of the mitochondrion found in bloodstream T. brucei, if we can judge from what information is available on their metabolism (Agosin and von Brand 1954; Fulton and Spooner 1959; von Brand and Tobie 1959; Ryley 1956), cytochemistry (see above), and ultrastructure (Rudzinska and Vickerman 1968; Vickerman 1969a); they do, however, show similar changes in the mitochondrion on entering the vector (Vickerman 1969a, 1970). Dyskinetoplasty and its effect on the life cycle Bloodstream populations of all salivarian trypanosomes contain individuals apparently lacking the kinetoplast as seen by light microscopy of stained smears. This condition may arise spontaneously, but its incidence can be increased to 100 per cent by treatment with certain polycyclic dyes (Werbitzki 1910). Such kinetoplastless trypanosomes are usually non-viable, the notable exceptions being derivatives of T. brucei and T. evansi. In these species the condition is inherited as a mutation: once lost the kinetoplast cannot be regained. Electron microscopy, however, shows that the kinetoplast is not lost altogether : the fibrous mass of DNA is replaced by a compact spherical mass lacking fibrous organization. For this reason the term dyskinetoplastic (Trager and Rudzinska 1964) is preferable to the older designation akinetoplastic in describing these forms. Dyskinetoplastic trypanosomes retain the inactive mitochondrion of their kinetoplast-bearing forbears. T. brucei rendered dyskinetoplastic by drug treatment can no longer undergo transformation in vitro (Reichenow 1940). It seems probable that the ability to transform in culture has been impaired along with the kinetoplast, and the obvious explanation of this is that the integrity of the kinetoplast is essential for complete activation of the mitochondrion which takes place at this time.

EXTRACELLULAR BLOOD PROTOZOA

FIGURES 19 and 20. Smears of trypanosomes incubated with NADH and Nitro blue tetrazolium salt for NADH-tetrazolium reducíase activity. Black formazan deposits mark sites of enzyme activity. 19 Slender (left) and stumpy (right) forms of bloodstream T. brucei rhodesiense. The slender form shows a deposit in scattered bodies (small arrows) and these may contain the L-a-glycerophosphate oxidase system. The stumpy form also shows these bodies but here the mitochondrial canal

81

(large arrow) is conspicuously filled with formazan. X2,900 20 Intermediate forms of bloodstream T. brucei brucei showing mitochondria filled with formazan. Some trypanosomes show stages in longitudinal splitting of the mitochondrion, which precedes division of the flagellate. In some strains the non-dividing stumpy forms show a split canal and this may be regarded as preparative to formation of the mitochondrial network found in the fly mid-gut stage. X 2,900

82

How far the kinetoplast is necessary for mitochondrial functioning in the vertebrate host is uncertain at present. The ability of dyskinetoplastic T. brucei to survive and multiply is well known, but we do not know whether they can transform from slender to stumpy forms with accompanying mitochondrial changes since dyskinetoplastic forms are usually monomorphic, though strangely enough stumpy forms have been described for dyskinetoplastic T. evansi (Hoare 1952). In T. vivax and T. congolense, which are believed to have some mitochondrial activity throughout the bloodstream phase, dyskinetoplastic forms are rarer and believed to be non-viable (Muhlpfordt 1964). Bloodstream T. brucei and T. evansi can dispense with the mitochondrion and so survive in this phase of their life cycle: indeed, having lost the kinetoplast, they are condemned to spend eternity there. Stercorarian trypanosomes

KEITH VICKERMAN

extracts (Fulton and Spooner 1959). The glycerophosphate oxidase system is of no importance in these trypanosomes. Both multiplicative (Molyneux 1970) and adult trypanosomes (Anderson and Ellis 1965) have a mitochondrial network with plate-like cristae. The culture forms of T. lewisi (unlike T. brucei) are epimastigotes, and as yet relatively little is known about their respiration. Preliminary studies by Hibbard and Dusanic ( 1969) indicate that with this trypanosome a rather different situation to that obtaining in the bloodstream-to-culture form transformation of T. brucei may be found. Unlike the latter species, endogenous respiration is to be reckoned with in both blood and culture forms of stercorarian trypanosomes (Ryley 1956). More glucose and less oxygen are consumed by culture forms of T. lewisi, and the glucose is less completely oxidized to carbon dioxide than in bloodstream forms. Epimastigote forms from the rectum of the flea have a mitochondrial net with plate-like cristae very similar to that of bloodstream stages (Molyneux 1969). It looks, then, as though the adult bloodstream form may represent the peak of mitochondrial activity during cyclical development of T. lewisi, but a more complete study of both the metabolism and the ultrastructure of this trypanosome is needed before we can compare it in any detail with the salivarian trypanosomes.

The inhibition of reproduction of T. lewisi by ablastin is accompanied by alterations in the parasite's respiratory metabolism. Sanchez and Dusanic (1968) have summarized the literature on respiration of bloodstream forms. They conclude that the trypanosomes have a primarily glycolytic respiration in the multiplicative bloodstream form, but there is a shift to dependence on the Krebs cycle with increased oxygen uptake at 8 days post-inoculation when ablastin is beginning to prevent reproduction. They suspected PATHOLOGICAL EFFECTS a swing back to glycolysis at 12 days post-inoculation, but Entner ( 1968a, b) found the specific We do not know why infections with salivarian activity of Krebs cycle isocitrate dehydrogenase trypanosomes are fatal to man and his domestic in 13-day trypanosomes to be four times that of animals, whereas in some game animals no illmultiplicative forms in his strain. The lactic deeffect is apparent.* Von Brand (1966) has rehydrogenase activity of reproducing T. lewisi is over three times greater than that of adult forms (D'Alesandro and Sherman 1964). *Wild game animals show varying degrees of resistance to these trypanosomes (Ashcroft, Burtt, and FairBoth reproducing and adult forms of T. lewisi have a complete cytochrome chain (Ryley 1951 ; bairn 1959). Reedbuck (Redunca redunca) resist T. vivax and harbour T. brucei with no outward sympFulton and Spooner, 1959) and, in both, NADH toms: Korin gazelle (Gazella rufifrons) develop mild oxidation is sensitive to cyanide, though very low infections with T. vivax but succumb to T. brucei (Decytochrome oxidase activity has been found in sowitz 1960).

83

EXTRACELLULAR BLOOD PROTOZOA

viewed the older theories of trypanosome pathogenesis, chief among which was exhaustion of the host's carbohydrate reserves (Schern 1926). We have already noted the extravagant consumption of glucose by multiplicative T. brucei (50100 per cent of their dry weight in one hour according to Christophers and Fulton 1938), and death is always preceded by hypoglycaemia. But the idea that all the pathological effects of sleeping-sickness or nagana stem from this extravagance is a misguided one for two obvious reasons. First, although glucose consumption is high, by far the greater part of this molecule's potential energy is returned by the trypanosome to its host as excreted pyruvic acid which can be utilized by host mitochondria. Secondly, the number of trypanosomes present in the blood is not necessarily high; the flagellates may be undetectable in the blood of sick animals. This last observation has been overlooked by most other theorizers on pathological mechanisms in trypanosomiases, so I shall deal only with a theory which takes this into account. This postulates that the release of pharmacologically active substances, especially kinins, might bring about vascular changes which lead to a shock syndrome (Boreham 1968a, b, 1970; Goodwin 1968; Goodwin and Hook 1968), and that antigen-antibody reactions instigate the liberation of these substances. The pathology of chronic T. brucei trypanosomiasis in man, cattle, and experimental rabbits seems to follow on primary lesions in the walls of blood vessels. Antigen-antibody reactions are known to be accompanied by the release of kinins (Brocklehurst 1960), though the exact causal relationship of these events is not known. Richards (1965) found that in mice acutely infected with T. brucei the concentrations of kinins rises progressively until death. In the blood of chronically infected rabbits and cattle (Boreham 1968a) and patients with sleeping-sickness (Boreham 1970), kinin activity increases two to three days after a peak parasitaemia. The increased kinin levels appear to be related to the production of antibodies to the common antigens rather than

to the variant antigens (Boreham 1968a) : common antigens are released after each crisis when trypanosomes are being cleared from the blood. Thus, in chronic infections with T. brucei the host is subjected to a series of immunological shocks, which might explain why the gross pathological response of the host resembles that of the Arthus reaction. Arthus (1903) sensitized rabbits with subcutaneous injections of horse serum and then induced anaphylaxis by giving a massive intravenous dose: those rabbits that survived subsequently suffered from wasting, anaemia, oedema, and scabby patches on the skin followed by death in a few weeks. Similar symptoms are familiar to all workers with rabbits infected with T. brucei. Goodwin (1968) suggests that the kinins increase vascular permeability, which results in venous congestion, a fall in blood pressure, reflex arteriole constriction, and, eventually, cardiovascular collapse.* It is perhaps significant that even a heavy T. lewisi infection does not cause the release of kinins (Boreham 1966) because this trypanosome is not pathogenic in its rat host. CONCLUDING REMARKS

The considerations of mammalian trypanosome life cycles which I have presented were intended to illustrate a thesis which may be summarized as follows. In both T. brucei and T. lewisi the physiology of the bloodstream trypanosomes undergoes a change as the multiplicative phase comes to an end. Respiration which was primarily glycolytic is then accompanied by increased mitochondrial activity and, in the case of T. brucei at any rate, this change helps to prepare the non-multiplicative forms for enforced respiratory economies in the vector's gut. The host's immune response might induce this switch in *Seed (1969) has suggested that the trypanosomes themselves liberate a vascular permeability increasing factor, and that it is this factor which provokes the inflammatory reaction when T. brucei is injected intradermally. T. lewisi is also said to contain such a factor.

84

respiratory pathways by, for example, inhibiting substrate uptake. Some trypanosomes in the population, however, appear to be able to adapt to host antibodies by changing their surface antigens, and the mechanism of this antigenic change might lie in the synthesis of different surface coat components. The shedding of surface coat material as exoantigen seems to provoke the host into responding primarily to the variable antigens rather than to antigens which are common to all relapses. The pathological effects of T. brucei might be the result of a kinin-mediated form of Arthus reaction induced by sensitization to successive doses of the common antigens which are released at each crisis. The surface coat of the trypanosome is an adaptation to bloodstream living as it is dispensed with in the vector or in culture, but the metacyclics of T. brucei reacquire the coat as they prepare to infect a new mammalian host. The trend to mitochondrial activation is continued in the vector trypomastigote forms but may be reversed in epimastigote and metacyclic forms. The cycle of mitochondrial activity is accompanied by morphogenetic changes in the trypanosomes and is at least partially under the direction of the kinetoplast. The question of how all these cyclical events are controlled is the most fascinating but the least understood aspect of them. I have presented evidence that antigenic variation is the result of induction rather than spontaneous mutation and selection, but how the presence of antibody in the blood would induce the expression of a gene for a new variant antigen I would not care even to speculate upon at the moment. On the control of the mitochondrial changes we have at least got some sort of lead, for as I have outlined we know that the kinetoplast is necessary for mitochondriogenesis, even if the nature of the kinetoplast continues to mystify us. The discovery by electron microscopists that the kinetoplast's DNA (K-DNA) is lodged in the flagellate's mitochondrial system became less puzzling once the universal occurrence of DNA in mitochondria (M-DNA) became recognized. The kinetoplastid flagellates, so it seemed, simply

KEITH VICKERMAN

had more M-DNA than other cells: this amplification of the mitochondrial genome could be associated with the need to synthesize the proteins of the active mitochondrion at short notice when called upon to adapt to a new environment, more DNA providing more templates for protein synthesis. But is the kinetoplast directly comparable to M-DNA? In some ways apparently it is not. First, the efficiency of renaturation of trypanosome K-DNA after thermal dissociation of the strands is much higher than that of M-DNA, indicating greater homogeneity and therefore coding ability for fewer proteins. Secondly, although electron micrographs of extracted and spread KDNA molecules show that they are circular like those of M-DNA, their circumference is only about one-tenth that of most M-DNA s, and so comparable to many viral DNA s - again suggesting that K-DNA encodes less information (Riou and Paoletti 1967). Perhaps the kinetoplast mini-circles represent amplification of only a small part of the usual mitochondrial genome. Experiments with induced dyskinetoplasty are being conducted by many in the hope of clarifying the role of the kinetoplast in controlling cyclical development. The discovery that culture epimastigotes of T. cruzi rendered dyskinetoplastic by acriflavin can differentiate into metacyclics and then complete the whole vertebrate cycle in tissue culture (Deane and Kloetzel 1969) is indeed surprising and reminds us that even when we have solved the problem of the nature of the kinetoplast we are left with the problem of the nature of dyskinetoplasty. The last ten years have seen considerable advances in our understanding of the changes in the lives of these blood-dwelling protozoa, but they still retain most of their secrets. We could reveal many of these if only we could persuade them to desert the bloodstream for our culture tubes - but not to change in doing so!

EXTRACELLULAR BLOOD PROTOZOA

REFERENCES

Agosin, M., and von Brand, T. 1954. Studies on the carbohydrate metabolism of Trypanosoma congolense. Exptl. Parasitol. 3:517—24 Anderson, W. A., and Ellis, R. A. 1965. Ultrastructure of Trypanosoma lewisi: Flagellum, microtubules and the kinetoplast. J. Protozool. 2: 483-9 Arthus, M. 1903. Injections répétées ds serum de cheval chez le lapin. Compt. rend. soc. biol. 55 : 817-20 Ashcroft, M. T. 1957. The polymorphism of Trypanosoma brucei and T. rhodesiense, its relation to relapses and remissions of infections in white rats, and the effect of cortisone. Ann. Trop. Med. Parasitol. 51:301-12 - 1960. A comparison between a syringe-passaged and a tsetse fly-transmitted line of a strain of Trypanosoma rhodesiense. Ann. Trop. Med. Parasitol. 54:44-53 Ashcroft, M. T., Burtt, E., and Fairbairn, H. 1959. The experimental infection of some African wild animals with Trypanosoma rhodesiense, T. brucei and T. congolense. Ann. Trop. Med. Parasitol 53:147-61 Augustine, D. L. 1943. Some factors in the defense mechanism against reinfection with Trypanosoma lewisi. Proc. Am. Acad. Arts S ci. 75:85—93 Bayne, R. A., Muse, K. E., and Roberts, J. F. 1969. Isolation of bodies containing the cyanide-insensitive glycerophosphate oxidase system of Trypanosoma equiperdum. Comp. Biochem. Physiol. 30:1049-55 Beale, G. H. 1954. The Genetics of Paramecium aurelia. London and New York: Cambridge University Press Bevan, L. E. W., and MacGregor, M. E. 1910. Notes on trypanosomes of the dimorphon group. Vet. J. 66:386-90 Boreham, P. L. F. 1966. Pharmacologically active peptldes produced in the tissue of the host during chronic trypanosome infections. Nature 212: 190-1 - 1968a. Immune reactions and kinin formation in chronic trypanosomiasis. Brit. J. Pharmacol. Chemother. 32:493-504

85

- 1968b. In vitro studies on the mechanism of kinin formation by trypanosomes. Brit. J. Pharmacol. 34:598-603 - 1970. Kinin release and the immune reaction in human trypanosomiasis caused by Trypanosoma rhodesiense. Trans. Roy. Soc. Trop. Med. 64: 394-400 Bowman, I. B. R., and Flynn, I. W. 1968. Metabolic properties and arsenical sensitivity of polymorphic brucei subgroup trypanosomes. Vlllth Inter. Congr. Trop. Med. and Malaria, Teheran, Abstr., pp. 315-16 von Brand, T. 1951. Metabolism of Trypanosomidae and Bodonidae. In Biochemistry and Physiology of the Protozoa, ed. A. Lwoff, i: 178-234. New York: Academic Press - 1966. Biochemistry of Parasites. London and New York: Academic Press von Brand, T., and Johnson, E. M. 1947'. A comparative study of the effect of cyanide on the respiration of some Trypanosomidae. /. Cell. Comp. Physiol. 29:33-9 von Brand, T., and Tobie, E. J. 1959. Observations on the metabolism of the culture form of Trypanosoma congolense. J. Parasitol. 45:204—8 Brocklehurst, W. E. 1960. The release of histamine and the formation of a slow reacting substance (SRS-A) during anaphylactic shock. /. Physiol. (London) 151:416-35 Broom, J. C., and Brown, H. C. 1937. Studies in trypanosomiasis i. The electric charge of trypanosomes in tsetse flies. Trans. Roy. Soc. Trop. Med. Hyg. 31:81-6 - 1939. Studies in trypanosomiasis. m. The electric charge of trypanosomes in the salivary gland of tsetse flies. Trans. Roy. Soc. Trop. Med. Hyg. 32:545-8 Brown, K. N., and Williamson, J. 1962. Antigens of brucei trypanosomes. Nature 194:1253-5 Christophers, S. R., and Fulton, J. D. 1938. Observations on the respiratory metabolism of malaria parasites and trypanosomes. Ann. Trop. Med. Parasitol. 82:43-75 Clark, T. B., and Wallace, F. G. 1960. A comparative study of kinetoplast, ultrastructure in the Trypanosomatidae. /. Protozool. 7:115-24

86 Clarkson, M. J., and Awan, M. A. Q. 1968. Studies of antigenic variants of Trypanosoma vivax. Trans. Roy. Soc. Trop. Med. Hyg. 62:127 Coventry, F. A. 1930. The trypanocidal actions of specific antiserums on Trypanosoma lewisi in vivo. Am. J. Hyg. 12:366-80 Cunningham, M. P. 1966. The preservation of viable metacyclic forms of Trypanosoma rhodesiense and some studies of the antigenicity of the organisms. Trans. Roy. Soc. Trop. Med. Hyg. 60:126 Cunningham, M. P., van Hoeve, K., and Lumsden, W. H. R. 1963. Variable infectivity of organisms of T. brucei subgroup during acute relapsing infections in rats, related to parasitaemia, morphology and antibody response. Repî. E. African Try pan. Research Org., p. 21 Cunningham, M. P., and Vickerman, K. 1962. Antigenic analysis in the Trypanosoma brucei group using the agglutination reaction. Trans. Roy. Soc. Trop. Med. Hyg. 56:48-59 D'Alesandro, P. A. 1959. Electrophoretic and ultracentrifugal studies of antibodies to Trypanosoma lewisi. J. Infect. Dis. 105:76-95 - 1966. Immunological and biochemical studies of ablastin, the reproduction-inhibiting antibody to Trypanosoma lewisi. Ann. N.Y. Acad. Sci. 129: 834-52 D'Alesandro, P. A., and Sherman, I. W. 1964. Changes in lactic dehydrogenase levels of Trypanosoma lewisi associated with the appearance of ablastic immunity. Exptl. Parasitol. 15:430-8 Deane, M. P., and Kloetzel, J. K. 1969. Differentiation and multiplication of dyskinetoplastic Trypanosoma cruzi in tissue culture and in the mammalian host. /. ProtozooL 16:121-6 Desowitz, R. S. 1954. Studies on Trypanosoma vivax. x. The activity of some blood fractions in facilitating infection in the white rat. Ann. Trop. Med. Parasitol. 48:142-51 — 1960. Studies on immunity and host-parasite relationships, ii. The immune response of antelope to trypanosome challenge. Ann. Trop. Med. Parasitol. 54:281-92 Entner, N. 1968a. Comparison of mitochondrial enzymes between young and adult forms of Try-

KEITH VICKERMAN

panosoma lewisi. J. ProtozooL 15:636—8 - 1968b. Further studies on antigenic changes in Trypanosoma lewisi. J. ProtozooL 15:638-40 Fairbairn, H., and Culwick, A. T. 1947. The modification of Trypanosoma rhodesiense on prolonged syringe passage. Ann. Trop. Med. Parasitol. 41:26-9 Flynn, I. W., and Bowman, I. R. B. 1970. Comparative biochemistry of monomorphic and pleomorphic strains of Trypanosoma rhodesiense. Trans. Roy. Soc. Trop. Med. Hyg. 64: 175-6 Fulton, J. D., and Spooner, D. F. 1959. Terminal respiration in certain mammalian trypanosomes. Exptl. Parasitol. 8:137-62 Goedbloed, E. 1968. The apparent absence of agglutinating antibodies in chick embryos infected with T. rhodesiense, In Ann. Repi. Med. Research Centre, Nairobi, Kenya Goodwin, L. G. 1968. The effects of trypanosome infections. Carworth Europe Collected Papers 2:61-5 Goodwin, L. G., and Hook, S. V. M. 1968. Vascular lesions in rabbits infected with Trypanosoma (Trypanozoon) brucei. Brit. J. Pharmacol. Chemother. 32:505-13 Grant, P. T., and Sargent, J. R. 1960. Properties of L-a-glycerophosphate oxidase and its role in the terminal respiration of Trypanosoma rhodesiense. Biochem. J. 76:229—37 - 1961. L-ce-glycerophosphate dehydrogenase, a component of an oxidase system in Trypanosoma rhodesiense. Biochem. J. 81:206-14 Grant, P. T., Sargent, J. R., and Ryley, J. F. 1961. Respiratory systems in the Trypanosomidae. Biochem. J. 81:200-6 Gray, A. R. 1962. The influence of antibody on serological variation in Trypanosoma brucei. Ann. Trop. Med. Parasitol. 56:4-13 — 1965a. Antigenic variation in clones of Trypanosoma brucei. i. Immunological relationships of the clones. Ann. Trop. Med. Parasitol. 59:27-36 - 1965b. Antigenic variation in a strain of Trypanosoma brucei transmitted by Glossina morsitans and G. palpalis. J. Gen. Microbiol. 41:195214

EXTRACELLULAR BLOOD PROTOZOA

- 1966. The antigenic relationship of strains of Trypanosoma brucei isolated in Nigeria. J. Gen. Microbiol 44:263-71 Hibbard, J. S., and Dusanic, D. G. 1969. Comparative studies on bloodstream and culture forms of Trypanosoma lewisi. I. Oxygen utilisation, carbon dioxide production and glucose uptake. Com/?. Biochem. Physiol 29:621-30 Hoare, C. A. 1940. Studies on the behaviour of Trypanosoma evansi in tsetse flies with special reference to its phylogeny. Parasitology 32:10521 - 1952. The polymorphism of Trypanosoma evansi. Trans. Roy. Soc. Trop. M éd. Hyg. 46:367-8 - 1964. Morphological and taxonomic studies on mammalian trypanosomes. x. Revision of the systematics. J. Protozool. 11:200-7 - 1966. The classification of mammalian trypanosomes. Ergeb. Mikrobiol. Immunoforsch. Exptl. Therap. 39:43-57 Hoare, C. A., and Wallace, F. G. 1966. Developmental stages of trypanosomatid flagellates : A new terminology. Nature 212:1385-6 Hollingshead, S., Pethica, B. A., and Ryley, J. F. 1963. The electrophoretic behaviour of some trypanosomes. Biochem. J. 89:123-7 Inoki, S., Kitaura, T., Makabayosi, T., and Kurogochi, N. 1952. Studies on the immunological variation in Trypanosoma gambiense. Med. J. Osaka Univ. 3:357-71 Inoki, S., Osaki, H., and Nakabayashi, T. 1956. Studies on the immunological variation in Trypanosoma gambiense. Verification of the new variation system by Ehrlich's and in vitro methods. Med. J. Osaka Univ. 7:165-73 Ketteridge, D. S. 1970. The presence of host serum components on the surface of rodent adapted Trypanosoma vivax. J. Protozool. 17(Suppl) :24 Lanham, S. 1968. Separation of trypanosomes from the blood of infected rats and mice by anión exchangers. Nature 218:1273-4 Leblond, C. P., and Warren, K. B. 1965. The use of radioautography in investigating protein synthesis. Symp. Intern. Soc. Cell Biol. 4 Le Page, R. W. F. 1968. Further studies on the variable antigens of Trypanosoma brucei. Trans.

87

Roy. Soc. Trop. Med. Hyg. 62:131 Levaditi, C., and Mutermilch, S. 1909. Le mécanisme de la création des variétés de trypanosomes résistantes aux anticorps. Compt. rend. soc. biol. 67:49-51 Luckins, A. G. 1969. Pleomorphism in trypanosomes of the Brucei complex. Ph.D. Thesis, London University MacAdam, R. F., and Herbert, W. J. 1970. Fine structural demonstration of cytoplasmic protrusions (filopodia) in trypanosomes. Exptl. Parasitol. 27:1-8 Massaglia, M. A. 1907. Des causes de crises trypanolytiques et des rechutes qui suivent. Compt. rend. 145:687-9 Miller, J. K. 1965. Variation of the soluble antigens of Trypanosoma brucei. Immunology 9:521—8 Molyneux, D. H. 1969. The fine structure of the epimastigote forms of Trypanosoma lewisi in the rectum of the flea, Nosopsyllus fasciatus. Parasitology 59:55-66 - 1970. Personal communication Mühlpfordt, H. 1964. Über den Kinetoplasten der Flagellaten. Z. Tropenmed. u. Parasitol. 15:289323 Nachlas, M. M., Walker, D. C., and Seligman, A. M. 1958. A histochemical method for the demonstration of diphosphopyridine nucleotide diaphorase. /. Biophys. Biochem. Cytol. 4:29— 38 Njogu, A. R. 1966. The nature of variable antigens of, and antigenic variation in, brucei sub-group trypanosomes. Intern. Council for Trypanosomiasis Research (Lagos), Publ. 100:75 Oehler, R. 1914. Untersuchungen iiber den Dimorphismus von Trypanosoma brucei. Z. Hyg. Infektionskrankh. 77:356-70 Ormerod, W. E. 1963. The initial stages of infection with Trypanosoma lewisi: Control of parasitaemia by the host. In Immunity to Protozoa, ed. P. C. C. Garnham, A. E. Pierce, and I. Roitt. Oxford: Blackwell Scientific Publications Petana, W. B. 1964. Effects of cortisone upon the course of infection of Trypanosoma gambiense, T. rhodesiense, T. brucei and 7\ congolense in albino rats. Ann. Trop. Med. Parasitol. 58:192-8

88 Reichenow, E. 1921. Untersuchungen ueber das Verhalten von Trypanosoma gambiense in menschlichen Korper. Z. Hyg. Infektionskrankh. 94:266-85 — 1932. Das Verhalten von Trypanosoma gambiense in der Kultur. Z. Parasitenk. 4:784—93 - 1940. Zur Frage der Bedeutung des Blepharoplasts der Trypanosomen. Arch. Inst. biol., Sâo Paulo 11:433-66 Richards, W. H. G. 1965. Pharmacologically active substances in the blood, tissues and urine of mice infected with Trypanosoma brucei. Brit. J. Pharmacol. Chemother. 24:124-31 Riou, G., and Paoletti, C. 1967. Preparation and properties of nuclear and satellite deoxyribonucleic acid of Trypanosoma cruzi. J. Mol. Biol. 28:377-82 Robertson, M. 1912. Notes on the polymorphism of Trypanosoma gambiense in the blood and its relation to the exogenous cycle in Glossina palpalis. Proc. Roy. Soc. (London), B 85:527-39 Rudzinska, M. A., and Vickerman, K. 1968. The fine structure. In Infectious Blood Diseases of Man and Animals, éd. D. Weinman and M. Ristic, pp. 217-306. New York: Academic Press Ryley, J. F. 1951. Studies on the metabolism of the protozoa. 1. Metabolism of the parasitic flagellate Trypanosoma lewisi. Biochem. J. 49:577-85 - 1956. Studies on the metabolism of Protozoa. 7. Comparative carbohydrate metabolism of eleven species of trypanosomes. Biochem. J. 62:215-22 - 1962. Studies on the metabolism of the Protozoa. 9. Comparative metabolism of bloodstream and culture forms of Trypanosoma rhodesiense. Biochem. J. 85:211-23 — 1966. Histochemical studies on blood and culture forms of Trypanosoma rhodesiense. In Proc. 1st Intern. Congr. Parasitol. 1:41—2 Sanchez, G., and Dusanic, D. G. 1968. Respiratory activity of Trypanosoma lewisi during several phases of infection in the rat. Exptl. Parasitol. 23:361-70 Schern, K. 1926. Über die auf Grund und Zuckerphànomens der Trypanosomen geklarte Pathogenese der Trypanosomiasen. Berlin, tierarztl. Wochschr. 42:665-6

KEITH VICKERMAN

Seed, J. R. 1963. The characterisation of antigens isolated from Trypanosoma rhodesiense. J. Protozool 10:380^-9 - 1964. Antigenic similarity among culture forms of the "brucei" group of trypanosomes. Parasitology 54:593-6 - 1969. Trypanosoma gambiense and T. lewisi: Increased vascular permeability and skin lesions in rabbits. Exptl. Parasitol. 26:214-23 Seed, J. R., Byram, J., and Gam, A. A. 1967. Characterisation and localisation of acid phosphatase activity in Trypanosoma gambiense. J. Protozool. 14:117-25 Seed, J. R., Cornille, R. L., Risby, E. L., and Gam, A. A. 1969. The presence of agglutinating antibody in the Ig M immunoglobulin fraction of rabbit anti-serum during experimental African trypanosomiasis. Parasitology 59:283-92 Seed, J. R., and Gam, A. A. 1966a. The properties of antigens from Trypanosoma rhodesiense. J. Parasitol. 52: 395-8 - 1966b. Passive immunity to experimental trypanosomiasis. /. Parasitol. 52:1134-40 Seed, J. R., and Weinman, D. 1963. Characterisation of antigens isolated from Trypanosoma rhodesiense. Nature 198:197 Sherman, I. E., and Ruble, J. A. 1967. Virulent Trypanosoma lewisi infections in cortisone treated rats. /. Parasitol. 53:258-62 Soltys, M. A. 1957a. Immunity in trypanosomiasis. I. Neutralization reaction. Parasitology 47:37589 - 1957b. Immunity in trypanosomiasis. n. Agglutination reaction with African trypanosomes. Parasitology 47:390-5 Steinert, M. 1960. Mitochondria associated with the kinetonucleus of Trypanosoma mega. J. Biophys. Biochem. Cytol. 8:542 Taliaferro, W. H. 1932. Trypanocidal and reproduction inhibiting antibodies to Trypanosoma lewisi in rats and rabbits. Am. J. Hyg. 16:32-84 Taliaferro, W. H., and Pizzi, T. 1960. The inhibition of nucleic acid and protein synthesis in Trypanosoma lewisi by the antibody ablastin. Proc. Nati. Acad. Sci. U.S. 46:733-45 Tempelis, C. H., and Lysenko, M. G. 1965. Effects

EXTRACELLULAR BLOOD PROTOZOA

of x-irradiation on Trypanosoma lewisi infection in the rat. Exptl. Parasitai. 16:174-81 Thillet, C. J., and Chandler, A. C. 1957. Immunisation against Trypanosoma lewisi in rats by injections of metabolic products. Science 125: 346-7 Thomson, J. G., and Sinton, J. A. 1912. The morphology of Trypanosoma rhodesiense in cultures: A comparison with the developmental forms described in Glossina palpalis. Ann. Trop. Med. Parasitol. 6:331-56 Trager, W., and Rudzinska, M. A. 1964. The riboflavin requirement and the effects of acriflavin on the fine structure of the kinetoplast in Leishmania tarentolae. J. Protozool. 11:133-45 Vickerman, K. 1962. The mechanism of cyclical development in trypanosomes of the Trypanosoma brucei sub-group: An hypothesis based on ultrastructural observations. Trans. Roy. Soc. Trop. Med. Hyg. 56:487-95 — 1965. Polymorphism and mitochondrial activity in sleeping sickness trypanosomes. Nature 208: 762-6 - 1968. The surface coat of bloodstream trypanosomes. Trans. Roy. Soc. Trop. Med. Hyg. 62: 463 - 1969a. Thefinestructure of Trypanosoma congolense in its bloodstream phase. /. Protozool. 16:54-69 — 1969b. On the surface coat and flagellar adhesion in trypanosomes. /. Cell Sci. 5:163-93 - 1970. Cyclical changes in the mitochondrion of Trypanosoma brucei and related flagellates. In preparation Vickerman, K., and Luckins, A. G. 1969. Localization of variable antigens in the surface coat of Trypanosoma brucei using ferritin-conjugated antibody. Nature 224:1125-7 Watkins, J. F. 1964. Observations on antigenic variation in a strain of Trypanosoma brucei growing in mice. /. Hyg. 62:69—80 Weitz, B. G. F. 1960a. A soluble protective antigen of Trypanosoma brucei. Nature 185:788—9 - 1960b. The properties of some antigens of Trypanosoma brucei. J. Gen. Microbiol. 589—600 Werbitzki, F. W. 1910. Über blepharoplastlose

89

Trypanosomen. Zentr. Bakteriol. Parasitenk., Abt. 753:303-15 Wijers, D. J. B. 1960. Studies on the behaviour of trypanosomes, belonging to the brucei subgroup, in the mammalian host. Doctoral Thesis, University of Amsterdam Wijers, D. J. B., and Goedbloed, E. 1967. The development of African pathogenic trypanosomes in chick embryos. WHO/FAO African Trypanosomiasis Information Service. Tryp./Inf. 67.18 Wijers, D. J. B., and Willett, K. C. 1960. Factors that may influence the infection rate of Glossina palpalis with Trypanosoma gambiense. n. The number and morphology of the trypanosomes present in the blood of the host at the infected feed. Ann. Trop. Med. Parasitol. 54:341-50 Williams, J. S., Duxbury, R. E., Anderson, R. I., and Sadun, E. H. 1963. Fluorescent antibody reactions in Trypanosoma rhodesiense and T. gambiense in experimental animals. /. Parasitol. 49:380^4 Williamson, J., and Brown, K. N. 1964. The chemical composition of trypanosomes. m. Antigenic constituents of Brucei trypanosomes. Exptl. Parasitol. 15:44-68 Wilson, A. J. 1968. Studies on African trypanosomes. Ph.D. Thesis, Edinburgh University Wright, K. A., Lumsden, W. H. R., and Hales, H. 1970. The formation of filopodium-like processes by Trypanosoma (Trypanozoon) brucei. J. Cell Sci. 6:285-97

Discussion F. G. WALLACE

Department of Zoology, University of Minnesota, Minneapolis, Minn., USA There can be few organisms that exemplify the evolution of the physiology of a parasite as a response to the ecological conditions imposed by its hosts as well as the trypanosome. Our modern understanding of these relationships is due in large part to the researches of Dr Vickerman.

90 The succession of antigenic types that arise in the Salivaría is unique among infectious organisms and its explanation will have far-reaching effects. The suggestion that an antigen is secreted into the flagellar pocket can be supported by additional observations. The belief that the principal antigens are metabolites of living organisms is borne out by A. Sanders and F. G. Wallace (Exptl Parasitai. 18: 301-4 [1966]) and R. E. Duxbury and E. H. Sadun (/. Parásito!. 55:859-65 [1969]), who show that irradiated trypanosomes, which are alive and active but no longer reproduce, evoke a stronger immunity than any vaccine made from killed organisms. The immunological importance of the flagellar pocket was shown by S. Adler (Advances in Parasitol. 2:35-96 [1964] ). When he grew Leishmania in a medium that incorporated immune serum, one of the manifestations of a reaction with homologous serum was an extraordinary distention of the flagellar pocket. The emergence of an antigen from the flagellar pocket and its adherence to the surface membrane as a visible layer which differs in the Salivada and the Stercoraria are most important observations. If only we could claim that the dense, uniform surface coat of the salivarían trypanosome in the blood of its vertebrate host was produced by the union of antigen and antibody we would have the basis for a truly elegant theory. But our speaker, with his characteristic thoroughness, has studied the trypanosome in various parts of its insect host and found the same coat on the metacyclic trypanosomes in the salivary gland of the tsetse fly. Is, however, the idea that the surface coat is a complex of substances from both host and parasite entirely ruled out? Is it possible that molecular mimicry, that ultimate step in the adaptation of parasite to host, has made the salivary gland of the fly approach the composition of the host serum? H. Fairbairn and J. Williamson (Ann. Trop. Med. Parasitol 50:322-33 [1956]) found that the saliva of the tsetse fly, far from being a mere anticoagulant, is a complex fluid containing, among other things, protein, which appears to be consumed by the trypanosomes in the saliva of the infected fly. If this protein should have some similarity to the verte-

KEITH VICKERMAN

brate host protein, the fly could stimulate the protozoon to a preadaptation to its warm-blooded host, a refinement of the vector service that is denied to the Stercoraria. The other preadaptation, that of the expansion of the mitochondrion in the "stumpy" bloodstream form of Trypanosoma brucei, which are infective to the insect, will be explained only when we know how much of the life history of the trypanosome is merely a response to environmental stimuli and how much, if any, is determined genetically. M. Steinert (Exptl. Cell Research 15:560-9 [1958]) found that urea stimulated cultured Trypanosoma mega to transform into the trypomastigote blood forms, but only those individuals which were competent accepted this stimulus. L. H. P. da Silva and E. P. Camargo (Ciclo evolutivo do Trypanosoma cruzi. Doenca de Chagas. J. Romeu Cançado, éd. Belo Horizonte, pp. 87-99 [Imprensa oficial do Estado de Minas Gérais, 1968]) found that cultured T. cruzi transform to the trypomastigote when the growth has passed the exponential phase and has become stationary. They regard depletion of the medium as a probable stimulus which elicits the appearance of "regulating mutations." R. Geigy and J. Amrein (Abstr. 2nd Intern. Conf. ProtozooL, London, no. 145, pp. 137-8 [1965]) found that trypanosomes of the brucei group in culture became infective for mice after 18 days in culture, regardless of a number of experimental modifications. These observations suggest that under a veneer of environmentally determined changes there is an inherent sequence of stages which follow one another in an appointed course. But when with the needle or the loop we suspend this cycle in an indeterminate number of more or less identical generations, we are not observing the true life cycle of the organism. Only two weeks ago ( J. K. Frenkel, J. P. Dubey, and N. L. Miller, Science 167:893-6 [1970]) we became aware that the life cycle of Toxoplasma, an organism which has been the subject of countless investigations in hosts which it reached through prédation or through experimental inoculation, is normally that of a coccidian in the cat. Mention of inheritance brings us to DNA and to

EXTRACELLULAR BLOOD PROTOZOA

the kinetoplast, that largest of all extranuclear masses of DNA, characteristic of the Trypanosomatidae and their close relatives the Bodonidae, and master of the quick transformations that are required by these versatile protozoa in changing from one ecological niche to another. Most of our knowledge of the function of the kinetoplast is due to the investigations of Dr Vickerman. I wish only to make the point that the kinetoplast is not always the same and there is material for research in its comparative study. In monoxenous parasites of insects, such as Crithidia, the DNA of the kinetoplast is in fine strands. The appearance of this material after different fixatives suggests that it is not associated with basic protein. In Leishmania the DNA is in a dense band which is composed of tubules looped back and forth. In the genus Trypanosoma both arrangements of DNA are seen. In T. cruzi (C. Brack, Acta Trop. 25:289-356 [1968]) the contrast between the dense band of tubular DNA in the cultured organisms and in the intracellular amastigotes and the loose net of finer fibrils in the trypomastigote is very striking.

91

Physiological, morphological, and ecological considerations of some microsporidia and gregarines JIRÍ VAVRA

The two groups of protozoan parasites to be discussed in this paper differ considerably. Gregarines belong to the subclass Gregarinia of the subphylum Sporozoa, and microsporidia represent the class Microsporidea of the subphylum Cnidospora (Honigberg et al 1964). The groups are distinct morphologically and biologically, their only common characteristics being: (1) they are protozoans, (2) they share a large spectrum of hosts, and (3 ) they are highly adapted parasites. The results of the adaptation to parasitism are, however, not the same in the two groups. A comparative study of how such taxonomically different organisms have adapted to similar environments should be of some interest. PHYSIOLOGICAL AND ECOLOGICAL CONSIDERATIONS OF SOME MICROSPORIDIA

Some of the data presented are results obtained during a period of study spent by the author at the College of Veterinary Medicine, University of Illinois, Urbana, Illinois, with the support of National Science Foundation Research Grant GB-5667x to Dr. N. D. Levine. The author gratefully acknowledges the kindness of Dr Norman D. Levine which made this possible.

All microsporidia are intracellular parasites; except for their resistant spores they seem to be totally unable to live outside cells. The relationship of the microsporidian parasite to its host cell is intimate, both morphologically and physiologically. The microsporidian cell is in direct contact with its host's cytoplasm, the only boundary between the parasite and cytoplasm being the microsporidian cell membrane (Vavra 1965 ). It has even been suggested that the membrane system of the parasite is a continuation of that of the host cell (Sprague and Vernick 1968). In some microsporidia the surface of contact with the host cytoplasm is increased by the presence of vesicular formations or bristle-like tubular expansions formed from the parasite's cell membrane (Sprague, Vernick and Lloyd 1968; Vavra 1968; Weidner 1970). At an early stage of infection the host cell is not damaged by the presence of the parasite, but later considerable lysis occurs in its cytoplasm, which is almost completely destroyed when the spores of the parasite are formed. Some cells are profoundly

MICROSPORIDIA AND GREGARINES

affected by the microsporidian infection; for example, "cell hypertrophy tumours" may be formed in sticklebacks infected by Glugea weissenbergi (Sprague and Vernick 1968) or the nucleus itself may develop polytene chromosomes as in the fat body of black-flies parasitized by Plistophora debaisieuxi (Vavra, unpublished). Also the existing chromosome patterns may be changed by microsporidian infection (Pavan, Perondini, and Picard 1969). The intracellular location of microsporidia implies that they obtain nutrients from their host's cells. The absence of any defined opening in the microsporidian cell suggests further that this must be done by diffusion or micropinocytosis. Some of the developmental stages of microsporidia have an extremely thin membrane (single membrane of the unit type) which shows signs of pinocytotic activity, Vavra 1968). No metabolic study has yet been made of the trophic stages of microsporidia. It would be of interest to determine whether the microsporidia possess oxidative respiratory enzymes. All electron microscope studies performed thus far on microsporidia have indicated the complete absence of mitochondria or mitochondria-like structures at all stages of the life cycle. However, the cytoplasm is rich in membranous and vesicular formations to which the respiratory enzymes may be attached if present. Movement in microsporidia Whether an active movement exists in microsporidia is not clear. Older, light-microscope observations described amoeboid movement and the active penetration of early vegetative stages ("sporoplasm") into the host's tissues. More recent studies have endorsed the view that the sporoplasm is introduced passively into the host's tissue (Lorn and Vavra 1963b; Ishihara 1968a). This seems to imply no need for an active movement of the sporoplasm. The question remains open as to how the later (schizogonic and sporogonic) stages of the microsporidia are able to spread from one cell or tissue to another. In

93

some cases phagocytes may be responsible for the migration of the infection; for example, in Nosema tracheophila the schizogonic stages are located inside the phagocytes which are later incorporated into the host's connective tissue where the sporogony occurs (Cali and Briggs 1967). Spore structure and junction The apparent lack of any conspicuous sort of movement in the trophic stages is compensated to some extent by the particular germ-release mechanism which enables the sporoplasm to travel several hundred microns in a fraction of a second and to reach the host tissue. The mechanism of spore dehiscence in microsporidia is unique among protozoa and other spore-forming organisms. It is one of the most elaborate adaptations to parasitism found and a unique morphological adaptation of microsporidia to their environment (for a detailed description of the spore germination process and a history of its theories see Lorn and Vavra 1963b). The spores usually germinate when swallowed into the digestive tract of a host. Digestive enzymes presumably act upon the polysaccharide material ("polar cap") occurring at one of the two poles of the spore, allowing water to be imbibed by the swelling structures of the spore (polaroplast and the vacuole). Swelling causes a sudden increase in pressure within the spore, which soon exceeds the mechanical resistance of the thinner portion of the spore shell at the apical pole of the spore. This portion then breaks and allows the polar filament to be evaginated from the spore. The filament is turned upside down like a finger of a glove during eversión, and when it is fully everted from the spore, the sporoplasm is forced through its lumen and discharged at its tip (Ohshima 1966). During the spore extrusion the pressure inside the spore, as well as in the everting filament, is high, the filament having considerable rigidity and penetrative power so that it can easily pierce not only a cell membrane (Lorn and Vavra 1963b) but the whole gut of a Tribo-

94

Hum larva. In Tenebrio the filaments penetrate only through the gut walls of young, second- and third-instar larvae which have moulted 8 hours or less before feeding by spores (Fisher and Sanborn 1962). Thus, the sporoplasm is injected directly by means of the filament into the body or a tissue of the host, and the action of the host's proteolytic enzymes in the gut is avoided. The mechanism of spore germination is conditioned by mechanical and structural properties of individual spore components: the shell, the filament, swelling structures, and the sporoplasm. The shell is a firm single piece of impervious material, which serves the purpose of making the spore resistant to the external environment as well as to the pressure generated within it during germination. Electron microscope and chemical examinations show that the spore has two layers. The outer layer ("exospore," Vavra 1964b) is proteinaceous and sometimes forms projections and ornamentations (Vavra 1963, 1968) or bears mucous layers (Lorn and Vavra 1963a), which are supposed to give spores the buoyancy needed for dispersal. The inner layer ("endospore," Vavra 1964b) is thicker and consists of a-chitin (Vavra 1967; Vavra, unpublished) . This is an extremely non-reactive substance which can be hydrolysed in nature only by its specific enzyme, the chitinase. The presence of this substance helps to explain the unusual resistance of microsporidian spores, which is claimed to be in the range of weeks under dry conditions and years under wet conditions (Ohshima 1964e; Weiser 1961b). It has been shown by observations using the electron microscope (Vavra 1965; Vavra, Joyon, and de Puytorac 1966; Vavra, unpublished) that the polar filament consists of tightly bound and spirally wound fibres. This structural arrangement gives both the firmness and elasticity required for the passage of the sporoplasm through the lumen. The physicochemical nature of the swelling structures in the spore is unknown. Structurally one of them, the polaroplast, in an unswelled state consists of tightly packed lamellae. During

JIRI VAVRA

swelling, the interlamellar spaces increase and the whole organelle augments in volume. The polaroplast is presumed to have some structural relationship with the large vacuole situated near the posterior pole of the spore. This assumption is supported by evidence that both the polaroplast and the posterior vacuole originate by differentiation of the Golgi apparatus of the microsporidian sporoblast (Sprague and Vernick 1969). The vacuole functions during the last phase of spore germination, when it expands and pushes the contents of the spore into the evaginated filament (Lorn and Vavra 1963b). Little information is available on the structure of the sporoplasm while inside the spore, but it is known that it has a single membrane, two nuclei surrounded by a double membrane, ribosomes, and some membranous structures when it is discharged from the spore (Ishihara 1968a). Ribosomes isolated from sporoplasm have the physical characteristics of monoribosomes with a sedimentation coefficient 705-, resembling those of lower protists (bacteria and blue-green algae) (Ishihara and Hayashi 1967). Structurally the sporoplasm is a simple cell. This simplicity is evidently a prerequisite for the elasticity required during the passage through the filament. The cellular nature of the sporoplasm as revealed by the observations of Ishihara (1968a) is opposed to the view of Sprague and Vernick (1968), who suggest that the microsporidian germ is a simple subcellular structure consisting mostly of DNA which is injected into the host tissue in a manner analogous to certain viruses. The exact mechanism of spore activation is poorly understood. The digestive fluids of specific hosts as well as a number of physical and chemical stimuli are effective in causing the spore extrusion (Kudo 1918; Ohshima 1964a, b, c, d; 1965a, b). Ishihara (1967) postulated that there is a negatively charged receptor on the spore with which different univalent ions can associate to cause the activation of the spore. Passage through the digestive tract of a host which is taxonomically distant from the original one usually does not stimulate spore germina-

95

MICROSPORIDIA AND GREGARINES

tion. In unpublished experiments we have proved that spores of Thelohania cladocera pass without any change through the digestive tract of fish, although they germinate promptly when ingested by their original host Daphnia pulicaria. Similarly, the hornet Vespa crabro cannot be infected with Nosema apis although the spores which have passed through the hornet's intestine remain infective for the bee (Reng and Dorntlein 1965). Animals in which spores cannot germinate may thus still function as factors spreading the infective agent. Nosema curvidens from the bark beetle is not distributed in nature through the faeces of its host but by arthropods such as mites or the staphylinid beetle which feed on the dead beetle containing spores (Weiser 1961a). A particular situation occurs in microsporidia parasitizing cestodes. Tapeworms lack a digestive tract, and spore germination may be induced by stimuli within the lumen of the intestine of the host and outside the cestode. The germs of the microsporidia must then be injected by the filaments directly through the tegument of the tapeworm, as can be corroborated by the successful experiments of Dissanaike (1958), who transmitted Nosema helminthorum from Moniezia expansa to Hymenolepis nana and Taeniorhynchus saginatus by feeding the spores of the parasite to the host of the cestode. The ability of the spores to release the sporoplasm in the body of a potential host is undoubtedly the first requirement for infection, provided that the microsporidia are not transmitted transovarially, which does occur frequently. Other factors can determine whether the pathogen becomes established and whether the infection is manifested as patent. The physiological state of the host and the activity of its immune systems play a role. The sex of the host may be of some importance. For example in Culex tarsalis only male larvae may acquire patent infection by Thelohania (Kellen and Wills 1962; Kellen et al 1965). Further, the genetic constitution of the host may interfere with the infection as shown by the example of the hybrids of Culex tarsalis X C. erythrothorax in which the infection by The-

lohania californica is prevented by the presence of an "erythrothorax factor" (Kellen et al. 1966). Other factors are not understood. Nosema lymantriae, a parasite of Lymaniría dispar, is not infective for Euproctis chrysorrhoea, but when its spores are fed simultaneously with Thelohania similis (which alone is pathogenic for E. chrysorrhoea) a mixed infection occurs and this complex of two microsporidia can be further transmitted to E. chrysorrhoea (Weiser 1961b). Experiments with the cultivation of microsporidia in tissue cultures have shown a considerable lack of specificity of these parasites at the cellular level. Nosema bombycis can grow in vitro not only in silkworm tissue (Ishihara and Sohi 1966) but also in tissues of mammalian and chicken embryos (Ishihara 1968b). This may indicate that the specificity of infection is perhaps determined not so much by the suitability of the cell for infection but by factors determining whether the microsporidia can reach the cell. This assumption is further corroborated by the fact that surgically introduced Nosema sp. develops in hosts in which it cannot be introduced per os (Fisher and Sanborn 1962). The influence of ecological factors on microsporidian infections Little is known about the influence of external conditions on microsporidian infections. Sudden fluctuations in the rate of infection are known to occur in nature. For example, the peak of infection of honey-bees by Nosema apis may be expected each spring, the level of infection is low during the summer and a rise may occur in the autumn (Doull and Cellier 1961 ). It is supposed that changes in the metabolism of the bees due to stress allow the parasite to develop quickly, which results in an epizootic (Doull 1961 ). In other instances the reasons for fluctuations of microsporidian infections are less clear. Unpublished observations indicate marked seasonal fluctuations in infection by Gurleya sp, a parasite of the ovarian tissue of Cyclops vicinus and C. strenuus. Heavy infection is common during

96 the winter and early spring, but later the infection completely disappears although the host crustaceans are abundant throughout the year. Trnkova (1968) observed for several years the frequency of occurrence of two microsporidian pathogens in Daphnia pulicaria in a carp pond in Czechoslovakia. This cladoceran is the dominant zooplanctonic species of the pond and occurs there throughout the year. In five years of observations the infection by Thelohania dadocera appeared regularly in May, reaching its peak at the end of the month and the begining of June. The rate of infection decreased sharply thereafter and completely disappeared in the summer period. Important variations were observed in the rate of infection in different years. In 1961 20 per cent of adult Daphnia were infected, but in 1962-5 the rate of infection was about 1 per cent. In 1966 the infection persisted in the pond until August. The second pathogen (Gurleya sp. ) from the same pond and the same host has different seasonal abundance. Sparse infections can be found throughout the year, but frequent infections occur from June to October with a peak in late August and early September. As in the previous species, the maximum rate of infection, 39 per cent, was found in 1961 ; in 1962 the rate was only 6 per cent and between 1963 and 1965 less than 1 per cent. So far there is no sound explanation for these seasonal fluctuations. The habitat of the host seems to be highly important in laboratory studies of infections by microsporidia. It seems feasible to divide the microsporidia into two groups : one in which experimental infections are readily obtained and the other in which they are not. Microsporidia from terrestrial hosts belong to the first group: if a sensitive host is used, the infection can be readily established by feeding spores of the parasite, even though some forms (e.g. different insect instars) may be more sensitive than others. In the second group the feeding of spores usually does not lead to infection. Microsporidia from waterdwelling animals belong mostly to this group, although in some of them the infection can be

JIRI VAVRA

transmitted experimentally, such as in Bacillidium cyclopis from Acanthocyclops (Vavra 1962). Some crustaceans are completely resistant to experimental infections (Daphnia spp., Vavra 1964a), as are black-fly larvae (Debaisieux 1919; Weiser, personal communication), mosquito larvae (Chapman 1967;Kellen 1962), and Chaoborus larvae (Sikorowski and Madison 1968). In the midge Chironomus plumosus the fourth-instar larvae are resistant to infection by Thelohania while the first-instar larvae are sensitive (Hilsenhoff and Lovett 1966). The inability to produce infection in some water-dwelling hosts suggests either that the spores in the aquatic environment become infective only in a particular physiological state (Sikorowski and Madison 1968) or that the water-dwelling hosts have developed mechanisms blocking the development of the parasite under certain conditions. Our unpublished observations show that a single Daphnia pulicaria infected with Thelohania cladocera contained about three million spores. These are released after the death of the host and are distributed easily through the aquatic environment. If there were no barrier between the host and the parasite, soon most of the daphnia in the locality would be infected. The effect of microsporidia upon their hosts Although the microsporidia are rather well adapted parasites, some of them are highly pathogenic. Others are less pathogenic or even seemingly harmless to their hosts. Sometimes the microsporidian infection may increase the sensitivity of the host to adverse conditions or to an insecticide (Rosicky and Weiser 1951 ). A classical example is the increased sensitivity of the weevil Otiorrhynchus ligustici to DDT when infected with microsporidia (Rosicky and Weiser 1951). In some microsporidian infections the host may be affected by substances with juvenile hormone activity, the presence of which was demonstrated experimentally by Fisher and Sanborn (1964). The profound effect of a microsporidian infection on the host is demonstrated

MICROSPORIDIA AND GREGARINES

by the fact that even its sex may be affected by the presence of microsporidia in its body. Octosporea effeminans is a benign parasite of the ovarian tissues of the amphipod crustacean Gammarus duebeni. It has been demonstrated that the oocytes of the amphipod which are latently infected with this organism differentiate only into females. So-called thelygenic strains of the crustacean, i.e. strains which give rise exclusively to female progeny, were shown to be latently infected. The infection is completely harmless for vital functions of the host and seemingly favours the host by inducing the production of more females, although in a closed habitat the gradual replacement of males in the population would make the fertilization of eggs difficult (Bulnheim and Vavra 1968). MORPHOLOGICAL AND PHYSIOLOGICAL CONSIDERATIONS OF SOME GREGARINES

The morphological adaptations to parasitism which have evolved in different developmental stages of gregarines are of various kinds. Some of them are not unique for gregarines but are common to all Sporozoa. For example, the apical part of the sporozoite in most sporozoa contains the conoid and associated organelles. In gregarines, however, this structure evolved into the fixation apparatus (muerons and epimerites), which is sometimes of amazing complexity. This apparatus has the function of attaching securely large gregarine trophozoites to the cells of their hosts; in addition, a physiological function of feeding has developed in it (Schrevel and Vivier 1965). Other morphological adaptations in gregarines are unique to them, such as the epicytic folds which enlarge the surface of the organism considerably and which participate in its movement. Structures related to the dehiscence of gametocysts also exist exclusively in gregarines. Organelles of attachment and nutrition In the trophic phase of the life cycle some gregarines are intracellular, some are partially intra-

97

cellular, and others simply adhere to the surface of the cells. The fixation of the sporozoite is performed by means of its apical part (containing conoid and associate structures) which later changes into mucron or epimerite (Desportes 1967; Vavra and McLaughlin 1969). In some presumably primitive archigregarines such as Selenidium, the conoid is preserved in the adult trophozoite (Schrevel 1968). There is an amazing variety of forms of epimerites or muerons, many of the approximately 150 gregarine genera having different forms. This wide variety of forms would seem to result from the rather narrow host-specificity of the gregarine species and the wide range of microecological factors within their hosts. In intracellular gregarines the fixation organelle may be partially regressed, evidently through disuse (Vavra 1969). In extracellular gregarines the epimerite may be either intracellular or extracellular. Grebnickiella has the epimerite differentiated into a system of rhizoids, which pierce the host cell (Tuzet and Galanghan 1968). In other extracellular gregarines the epimerite is deeply insinuated into the host cell, but never penetrates into its interior and remains extracellular (Devauchelle 1968). As has been mentioned, the second function of the mucron or epimerite is physiological: the feeding of the organism. In Selenidium hollandei, for example, the mucron performs the function of a ultracytostóme (Schrevel 1969). Large food vacuoles are formed at the base of the conoid situated at the apex of the mucron. These vacuoles migrate into the cytoplasm, where micropinocytosis occurs at their borders (Schrevel 1968). Similar vacuoles originate in the mucron of Lankesteria culicis and L. barretti and penetrate through almost the entire body of the trophozoite (Vavra 1969). In Lecudina pellucida one can observe in the mucron strands of material of the same appearance as that outside the gregarine (Schrevel and Vivier 1966; Schrevel 1969). Electron-dense, club-shaped organelles in the anterior part of Selenidium are homologous to the dense anterior structures - "toxo-

98

nemes," "paired organelles" etc. - of other Sporozoa. They contain acid phosphatase and are supposed to participate in the lysis of host epithelium and to contribute to the digestion of the materials ingested into the cell (Schrevel 1969). An important question of the physiology of gregarines is how much the rest of the body besides the epimerite or mucron participates in the uptake of food. In young gregarines the epimerite is excessively large, its final size being reached before that of the rest of the body. At this time also the formation of the paraglycogen takes place in the epimerite. These facts suggest that in the early stage the epimerite is the most active metabolic site. Later in the development, however, ( 1 ) the relative size of the epimerite decreases until (2) at the moment of detachment from the host cell it may even be lost, and (3) the paraglycogen is formed in the protomerite (Devauchelle 1968). In later stages of development the gregarine must therefore feed through other segments of the body. The way in which this is done is not clear: either the epicyte is permeable to the nutrients or the numerous micropores occurring in the pellicle may be the sites of a pinocytosis. The extremely large surface of gregarine cells (made possible by folding) and the permeability of the cell membrane for small molecules such as neutral red (Schrevel 1969) seem to favour the first hypothesis. However, it is difficult to imagine how larger molecules can get through the thick epicytic membrane, which is known to consist of three superimposed unit membranes (Vivier 1969). The interpretation of pores as pinocytosis sites (Vivier and Schrevel 1964) is not corroborated by Devauchelle (1968), who interprets them rather as sites of extracellular excretion. Because of the failure to incorporate either colloidal thorium dioxide ("Thorotrast") (Schrevel, personal communication) or ferritin (Vavra, unpublished) into the gregarines, the question remains open. Movement in gregarines The movement of relatively large (of the order

JIRI VAVRA

of several hundred microns) gregarine gamonts in narrow spaces within the host, filled sometimes with rather viscous fluids, has necessitated the evolution of a special mechanism of movement. Gregarines move by bending and metaboly as do other organisms, but such types of movement cannot be considered as a special adaptation to parasitism. However, in common with other parasitic and non-parasitic organisms, they have a third mode of movement, gliding, and for this, gregarines have developed a unique mechanism, one that has been a puzzle since it was first observed in the 19th century. Several hypotheses were proposed to explain it. Two of the most popular sought the principle either in local contractions of myonemes or in the excretion of mucus. The mucus was supposed to act by piling behind the cell or by pushing the organism by reactive force during its excretion or to move the gregarine by unidirectional swelling (for more details on theories of gregarine gliding see Vavra and Small 1969). In 1967 Vivier observed, in the transmission electron microscope, that the epicytic folds exhibit undulating patterns and concluded that the undulations might be the actual cause of the gliding. This conclusion has been confirmed by observations of several species and genera of gregarines in the scanning electron microscope (Vavra and Small 1969). The pellicular folds not only undulate but do so in an organized way in which two neighbouring folds come into apposition periodically and then separate. The liquid surrounding the gregarine and filling the grooves between the folds is pushed backwards and by the reactive force the gregarine moves forward. This type of movement is somewhat reminiscent of the vertebrate gastric pump. Such a mechanism can also explain the role of the mucus secreted by gregarines, for, even if mucus is not a prerequisite for movement, the undulations of the folds will be more effective in a more viscous environment. An indirect confirmation of this mechanism of gliding is suggested by the fact that the epicytic folds in nongliding gregarines do not exhibit undulating patterns.

99

MICROSPORIDIA AND GREGARINES

Dissemination mechanisms for oocysts In some types of gregarines the gametocysts open and liberate the oocysts by simple rupture of the gametocyst wall, and in others the gametocyst develops structures which help to disseminate oocysts a considerable distance. In a more primitive type represented by Dactylophoridae, the oocysts gather inside the gametocyst in a saclike structure which is later evaginated outside the gametocyst. This sac is pressurized by gas, and on the slightest stimulus it ruptures, throwing the oocysts a considerable distance (Grasse 1953). Gregarinidae possess a more sophisticated way of gametocyst dehiscence. During maturation, the gametocyst develops long interior tubes called sporoducts. These are later evaginated by the pressure within the gametocyst, and the spores within them are then discharged one by one, being glued into long chains. In Gregarina blattarum one gametocyst can develop 5 to 30 sporoducts and the chains of oocysts may reach up to 87 mm, representing approximately 10,000 spores (Sprague 1941). Undoubtedly this dissemination increases the chances that an oocyst will be swallowed accidentally by a new host. The mode of gametocystic dehiscence is an adaptation to the ecology of the host. In monocystid gregarines from soil-inhabiting oligochetes where the gametocysts and oocysts are liberated after the death and decay of the host (Miles 1962), the gametocysts do not possess any special mechanism of dehiscence. The same is true in Diplocystidae, where the infection is usually transmitted by cannibalism. The gregarines which have developed sporoducts are parasites of terrestrial insects (Orthoptera, Coleóptera). The Dactylophoridae with the sac type of dehiscence of gametocysts are all parasites of myriapods. In gregarines of Diptera the gametocysts open while still inside the host and the oocysts contaminate the eggs during oviposition and defaecation - as in Lankesteria culicis and L. barretti (Vavra 1969) - or are glued to the surface of eggs by the secretion of accessory glands

- as in Monocystis chagasi ( Adler and May rink 1961). CONCLUSIONS

This brief and necessarily incomplete account of some morphological and physiological adaptations in microsporidia and gregarines proves the introductory statement that both groups are welladapted parasites, and shows that the type and results of adaptation differ in the two groups. Except for their spores, microsporidia are simple cells, spatially, and most probably also physiologically, fully dependent on the host cell. It seems that the main evolutionary principle for these organisms is the structural simplification which is conditioned by their exclusively intracellular mode of life, the uniformity of the intracellular environment being unfavourable to morphological diversification. This is demonstrated by the fact that all microsporidian types are simple variations of the same structural principle. The relative structural complexity of the spore reflects the fact that it is the only stage coming into direct contact with the outer environment. The spore is the product of simple cytoplasmic differentiation of the sporoblast, however, and the germ it carries is again a very simple cell. In contrast, it seems that the evolutionary principle for gregarines, as can be judged from their morphology and biology, is diversification. This process is evidently supported by their mostly extracellular way of life.

REFERENCES Adler, S., and Mayrink, W. 1961. A gregarine, Monocystis chagasi n.sp., of Phlebotomus longipalpis. Remarks on the accessory glands of P. longipalpis. Rev. Inst. med. trop., Sâo Paulo 3: 230-8 Bulnheim, H. P., and Vavra, J. 1968. Infection by the microsporidian Octosporea effeminans sp.n., and its sex determining influence in the amphipod Gammarus duebeni. J. Parásito]. 54:241-8

100 Cali, A., and Briggs, J. 1967. The biology and life history of Nosema tracheophila sp.n. (Protozoa: Cnidospora: Microsporidea) found in Coccinella septempunctata Linnaeus (Coleóptera: Coccinellidae). /. Invert. Pathol. 9:515-22 Chapman, H. C. 1967. Plistophora caecorum sp.n., a microsporidian of Culiseta inornata (Díptera: Cullicidae) from Louisiana. /. Invert. Pathol. 9:500-2 Debaisieux, P. 1919. Microsporidies parasites des larves de Simulium. Thelohania varians. La Cellule 30:47-79 Desportes, I. 1967. Ultrastructure et évolution du sporozoite de Stylocephalus africanas, Théodoridès, Desportes and Jolivet, Eugrégarine Stylocephalidae. Compt. rend. 265:423-6 Devauchelle, G. 1968. Etude ultrastructurale du développement des gregarines du Tenebrio molitor L. Protistologica 4:313-30 Dissanaike, A. S. 1958. Expérimental infection of tapeworms and oribatid mites with Nosema helminthorum. Exptl Parasitol. 7:306-18 Doull, K. M. 1961. A theory of the causes of development of epizootics of Nosema disease of the honey bee. J. Insect Pathol. 3:297-309 Doull, K. M., and Cellier, K. M. 1961. A survey of the incidence of Nosema disease (Nosema apis Zander) of the honey bee in South Australia. /. Insect Pathol. 3:280-8 Fisher, F. M., Jr., and Sanborn, R. C. 1962. Observations on the susceptibility of some insects to Nosema (Microsporidia, Sporozoa). /. Parasitol. 48:926-32 - 1964. Nosema as a source of juvenile hormone in parasitized insects. Biol. Bull. 126:235-52 Grasse, P. P. 1953. Traité de Zoologie, i, Fase. IL Protozoaires: Rhizopodes, Actinopodes, Sporozoaires, Cnidosporidies. Paris: Masson Hilsenhoff, W., and Lovett, O. L. 1966. Infection of Chironomus plumosus (Díptera: Chironomidae) by a microsporidian (Thelohania sp.) in Lake Winnebago, Wisconsin. /. Invert. Pathol. 8:51219 Honigberg, B. M., et al. 1964. A revised classification of the phylum Protozoa. /. Protozool. 11:7-20

JIRÍ VAVRA

Ishihara, R. 1967. Stimuli causing extrusion of polar filaments of Glugea fumiferanae spores. Can. J. Microbiol. 13:1321-32 - 1968a. Some observations on thefinestructure of sporoplasm discharged from spores of a microsporidian, Nosema bombycis. J. Invert. Pathol. 12:245-58 — 1968b. Growth of Nosema bombycis in primary cell cultures of mammalian and chicken embryos. /. Invert. Pathol. 11:328-9 Ishihara, R., and Hayashi, Y. 1967. Some properties of ribosomes from the sporoplasm of Nosema bombycis. J. Invert. Pathol. 11:377-85 Ishihara, R., and Sohi, S. S. 1966. Infection of ovarian tissue culture of Bombyx mor i by Nosema bombycis spores. /. Invert. Pathol. 8: 538-40 Kellen, W. R. 1962. Microsporidia and larval control. Mosquito News 22:87-95 Kellen, W. R., Chapman, H. C., Clark, T. B., and Lindegren, J. E. 1965. Host-parasite relationships of some Thelohania from mosquitoes (Nosematidae: Microsporidia). /. Invert. Pathol. 7: 161-6 Kellen, W. R., Clark, T. B., Lindegren, J. E., and Sanders, R. D. 1966. Development of Thelohania californica in two hybrid mosquitoes. Exptl. Parasitol. 18:251-4 Kellen, W. R., and Wills, W. 1962. The transovarial transmission of Thelohania californica Kellen and Lipa in Culex tarsalis Coquillett. /. Insect. Pathol. 4:321-6 Kudo, R. 1918. Experiments on the extrusion of polar filaments of cnidosporidian spores. /. Parasitol. 4:141-7 Lorn, J., and Vavra, J. 1963a. Mucous envelopes of spores of the subphylum Cnidospora (Doflein 1901). Acta Soc. Zool. Bohemoslov. 27:4-6 - 1963b. The mode of sporoplasm extrusion in microsporidian spores. Acta Protozoologica 1 : 81-9 Miles, H. B. 1962. The mode of transmission of the acephaline gregarine parasites of the earthworms. /. Protozool. 9:303-6 Ohshima, K. 1964a. Difference of filament evagination of Nosema bombycis Nâgeli in the diges-

MICROSPORIDIA AND GREGARINES

tive juice of starved and non-starved larvae of silkworm. Japan. J. Zool. 14:175-87 - 1964b. Influence of digestive juice of silkworm larvae starved for different periods of time, its viscosity and surface tension, on filament evagination of Nosema bombycis Nâgeli. Japan. J. Zool. 14:189-208 - 1964c. Stimulative or inhibitive substance to evaginate the filaments Nosema bombycis Nàgeli. i. The case of artificial buffer solution. Japan J. Zool. 14:209-29 - 1964d. Effect of potassium ion on filament evagination of spores of Nosema bombycis as studied by neutralization method. Annot. Zool. Japon. 37:102-3 - 1964e. Method of gathering and purifying active spores of Nosema bombycis and preserving them in good condition. Annot. Zool. Japon. 37:94— 101 - 1965a. Substance stimulating or inhibiting evagination of the filament of Nosema bombycis Nágeli. ii. The case of digestive juice and its electrodialyzed juice of starved silkworm larvae. Annot. Zool. Japon. 38:134-9 — 1965b. Substance stimulating or inhibiting the evagination of the filament of Nosema bombycis Nâgeli. m. Action of HCO3 in the digestive juice of starved silkworm larvae and the method effective in evaginating the filament in vitro. Annot. Zool. Japon. 38:198-206 - 1966. On morphological and physical nature of the filament and spore membrane of Nosema bombycis. Japan. J. Zool. 15:183-202 Pavan, C., Perondini, A. L. P., and Picard, T. 1969. Changes in chromosomes and in development of cells of Sciara ocellaris induced by microsporidian infections. Chromosoma 28:328—45 Reng, G., and Dorntlein, D. 1965. Infektionsversuche an Vespa crabro L. mit Nosema apis Z. Berlin tieràrztl. Wochschr. 78:394-7 Rosicky, B., and Weiser, J. 1951. Nosematosis of Otiorrhynchus ligustici. n. The influence of the parasitation by Nosema otiorrhynchí W. on the susceptibility of the beetles to insecticides. Acta. Soc. Zool. Bohemoslov. 15:219-34 Schrevel, J. 1968. L'ultrastructure de la région an-

101

térieure de la gregarine Selenidium et son intérêt pour l'étude de la nutrition chez les Sporozoaires. /. Microscopic 7:391-410 - 1969. Etude de la nutrition chez les Sporozoaires: cas des gregarines. Progr. in Protozool., IHrd Intern. Congr. Protozool., Leningrad, pp. 69-70 Schrevel, J., and Vivier, E. 1965. Ultrastructural and cytochemical aspects of the mucron and the epimerite of some gregarines, parasites of polychaetous annelids. Progr. in Protozool. llnd Intern. Congr. Protozool., London, p. 161 a - 1966. Etude de l'ultrastructure et du rôle de la région antérieure (mucron et epimerite) de gregarines parasites d'annelides polychaetes. Protistologica 2:17-26 Sikorowski, P. P., and Madison, C. H. 1968. Hostparasite relationship of Thelohania corethrae (Nosematidae: Microsporidia) from Chaoborus astictopus (Díptera: Chaoboridae). /. Invert. Pathol. 11:390-7 Sprague, V. 1941. Studies on Gregarina blattarum with particular reference to the chromosome cycle. Illinois Biol. Monographs 18:7—44 Sprague, V., and Vernick, S. H. 1968. Light and electron microscope study of a new species of Glugea (Microsporidia, Nosematidae) in the 4spined stickleback Apeltes quadracus. J. Protozool. 15:547-71 — 1969. Light and electron microscope observations on Nosema nelsoni Sprague 1950 (Microsporidia, Nosematidae) with particular reference to its Golgi complex. /. Protozool. 16:26471 Sprague, V., Vernick, S. H., and Lloyd, B. C., Jr. 1968. The fine structure of Nosema sp. Sprague 1965 (Microsporidia, Nosematidae) with particular reference to stages in sporogony. /. Invert. Pathol. 12:105-17 Trnkova, M. 1968. The production of Daphnia pulicaria as related to microsporidian infections in two carp ponds. Master of Biology Dissertation, Charles University, Prague (In Czech) Tuzet, O., and Galanghan, V. 1968. Ultrastructure des rhizoides de la gregarine Grebnickiella gracilis Bhatia (Petrocephalus nobilis Schneider,

102 Nina gracilis Grebnicki). Compt. rend. 266: 1401-2 Vavra, J. 1962. Bacillidium cyclopis n.sp. (Cnidospora, Microsporidia), a new parasite of copepods. Acta Soc. Zool. Bohemoslov. 26:295—9 - 1963. Spore projections in microsporidia. Acta Protozoologica 1:154—5 - 1964a. A failure to produce an artificial infection in cladoceran microsporidia. /. Protozool. 11 (Suppl.):35-6 - 1964b. Some recent advances in the study of microsporidian spores. Proc. 1st Intern. Congr. Parasitol., Rome i:443—4 Milan: Pergamon Press and Tamburini Ed., 1966 - 1965. Etude au microscope électronique de la morphologie et du développement de quelques microsporidies. Compt. rend. 261:3467-70 - 1967. Hydrolyse enzymatique des spores de microsporidies. /. Protozool. 14(Suppl.) :49 - 1968. Ultrastructural features of Caudospora simulii Weiser (Protozoa, Microsporidia). Folia Parásito!. 15:1-9 - 1969. Lankesteria barretti, n.sp. (Eugregarinida, Diplocystidae), a parasite of the mosquito Aedes triseriatus (Say) and a review of the genus Lankesteria Mingazzini. /. Protozool. 16:546—70 Vavra, J., Joyón, L., and de Puytorac, P. 1966. Observation sur Fultrastructure du filament polaire de microsporidies. Protistologica 2:109-11 Vavra, J., and McLaughlin, R. E. 1969. The fine structure of some of the developmental stages of Mattesia grandis McLaughlin (Sporozoa, Neogregarinida) parasite of the boll-weevil Anthonomus grandis Boheman. /. Protozool., in press Vavra, J., and Small, E. B. 1969. Scanning electron microscopy of gregarines (Protozoa, Sporozoa) and its contribution to the theory of gregarine movement. /. Protozool 16:745-57 Vivier, E. 1969. L'ultrastructure des formes végétatives des sporozoaires. Progr. in Protozool., lllrd Intern. Congr. Protozool., Leningrad, pp. 47-9 Vivier, E., and Schrevel, J. 1964. Etude, au microscope électronique, d'une gregarine du genre Selenidium, parasite de Sabellaria alveolata L.

JIRI VAVRA

/. Microscopie 3:651-70 Weidner, E. H. 1970. Ultrastructural study of microsporidian development, i. Nosema sp. Sprague, 1965 in Callinectes sapidus Rathbun. Z. Zellforsch. 105:33-54 Weiser, J. 196la. A new microsporidian from the bark beetle Pityokteines curvidens Germar (Coleóptera, Scolytidae) in Czechoslovakia, J. Insect Pathol. 3:324-9 - 1961b. Die Mikrosporidien aïs Parasiten des Insekten. Monogr. angew. Entomol. 17:1—14

Discussion MARSHALL LAIRD

Department of Biology, Memorial University, St. John's, Nfld. Dr Vavra's thought-provoking paper recalls to me that he and another Prague friend, Jiri Lorn, once questioned some old observations of mine from Singapore. These concerned the dehiscence of spores of Plistophora collessi, a parasite of mosquitoes; at a time when a wave of antipesticide hysteria is sanctifying, so to speak, everything that has to do with biological control candidacy, this must certainly be regarded as an organism that may well prove to have very real "mission relevance" — to use the current jargon. For female mosquitoes parasitized by Plistophora collessi lay not eggs but, instead, egg-sized cysts replete with huge numbers of infective microsporidan spores. Lorn and Vavra demonstrated beautifully, in the early sixties, the eversión of the elongate polar filament of certain microsporidans and the discharge of the elastic sporoplasm through its incredibly narrow-bore lumen. Nevertheless, I have distinct recollections of phase-contrast observations of spore-masses of Plistophora collessi fresh from the Singapore mosquito host, compressed under a coverslip and showing a general and explosive release of polar filaments, which continued to lash about briefly, the sporoplasms still attached to their

MICROSPORIDIA AND GREGARINES

proximal ends among the now-empty spores. These observations parallel an earlier description by Dr Elizabeth Canning. I should hasten to add that while, years ago, I was first trying to correlate these observations with those made by Drs Lom and Vavra, John Kramer sent me some dried Octosporea muscae domesticae spores, with instructions to flood the slide with distilled water, add a coverslip, and watch. I did. Most aggravatingly, the polar filaments inexorably everted and in due course a tiny globule appeared at the distal end just as described by Lom and Vavra. What, then, of polar filament discharge and sporoplasm emergence in the still more diversified and bizarre myxosporidans? Setting aside questions of the closeness or otherwise of the relationship of these to Microsporida, they undoubtedly have from one (as in species ofMyxobolus) to several polar capsules. Where there is more than one, through which does the sporoplasm exit, and why? And in some cases at least, doesn't the sporoplasm simply leave via the dehisced spore wall? And, if so, mightn't this happen in some microsporidans too? In any event, the issue is pertinent to any consideration of the ecology and physiology of these parasites. Conceivably, knowing the diversity of microsporidans, the manner of sporoplasm discharge is not necessarily the same throughout the class. Possibly the mechanism of spore dehiscence in the Microsporida is not unique but may occur in some Myxosporida also. That leaves me with the gregarines, and remarkably little of my allotted time. Having just read with the greatest of admiration the superbly illustrated paper by Vavra and Small in the November 1969 issue of the Journal of Protozoology, on the scanning electron microscopy of gregarines and its contribution to the theory of gregarine movement, may I simply open the discussion to the floor by observing that morphologically minded protozoologists without access to a scanning electron microscope are currently rather in the position of pre-Copernican astronomers; and by warmly congratulating our colleague and visitor for his distinguished contributions to our discipline.

103

Helminths as vectors of micro-organisms D. L. LEE

It is common knowledge that arthropods are important vectors of micro-organisms and as such are responsible for disseminating many diseases of man, other animals, and plants. Few people seem to realize, however, that some parasitic worms are also important as vectors of micro-organisms. This interesting aspect of the host-parasite relationship and of parasite ecology is one that is comparatively neglected. The helminths which have been implicated in the transmission of micro-organisms are the nematodes, the trematodes, and the cestodes, the most important of these being the nematodes. (By micro-organisms I mean viruses, rickettsia, bacteria, and protozoa.) The literature up to 1960 on helminths as vectors of micro-organisms was reviewed in two papers in a symposium on the biological transmission of disease agents held by the Entomological Society of America. One paper by Hewitt and Raski (1962) was on nematode vectors of plant viruses and concentrated on the transmission of grape-vine fan-leaf virus by the plantparasitic nematode Xiphinema index, and the other, by Philip (1962), was on helminths as carriers of microbial disease agents of man and animals. Here I shall review only the more thoroughly established examples in the literature up to 1960, leaving the reader to look up the more speculative relationships between helminths and microbial agents in these earlier reviews. Since digenetic trematodes and cestodes usually have complex life cycles involving one or more intermediate hosts, they should, theoretically, be unsuitable as vectors of micro-organisms. There is, however, one well-documented example of a trematode, and a less well documented one of a cestode, acting as a vector.

105

HELMINTHS AS VECTORS OF MICRO-ORGANISMS

TREMATODES

Nanophyetus salmonicola is a digenetic trematode, less than 1 mm in length, which lives in the small intestine of fish-eating mammals, including the dog. It is found only in the northwest Pacific area of the United States, its distribution being limited by the distribution of its first intermediate host, the snail Goniobasis silicula. This trematode is important as a vector of a rickettsia, Neorickettsia helminthoeca, which causes a fatal disease called salmon-poisoning in dogs, foxes, and other Canidae, but interestingly does not cause disease in Felidae. Work on this disease has been reviewed by Philip (1955, 1962), and the relationship between this trematode and the rickettsia has been well established. The trematode uses the salmon as a secondary intermediate host and the final host becomes infected when it eats fish infected with metacercaria. The stages of the trematode in the snail, in the fish, and in the dog have all been shown to carry the infection. Apparently the rickettsia is passed from one generation of trematode to another through the eggs and the larval stages in the snail and in the fish. Philip considers it probable that the proliferating stages of the trematode in the snail could act as a reservoir for the rickettsia. It is remarkable that a trematode with such a complex life cycle involving two intermediate hosts should act as the vector of a rickettsia, especially when one remembers that most rickettsial diseases are transmitted by arthropod vectors which spread the micro-organisms when feeding upon the body fluids of the host. Those interested in this fascinating relationship should read the review by Philip (1955). CESTODES

There is only one example of a cestode being implicated in the transmission of micro-organisms. Rift Valley fever virus has been shown to be present in larval stages of the cat tapeworm, Taenia crassicollis, removed from the livers of

experimentally infected mice (Findlay and Howard 1951 ). The invaginated scoleces were removed from the cysts, washed, and diluted from 10~2 to 10~6 before being injected intraperitoneally into mice which then died in 2 to 7 days. As the scoleces were within a cyst, the virus presumably was able to enter the cyst and either became adsorbed onto the surface of the scoleces or became intracellular. It would be interesting to clarify this point. The cysts were not fed to cats to see if the virus could be transmitted to the final host. This is very interesting work, especially as it has wider implications in the possible transmission of viruses by the larval stages of other tapeworms. NEMATODES

Nematodes are much more important in the transmission of micro-organisms than are the trematodes and cestodes and, although only a few examples are known, possibly they are more important as vectors than has been realized to date. Many plant parasitic nematodes would seem to be ideally adapted to act as vectors because many species feed as ectoparasites upon the roots and aerial parts of their plant hosts and they can move from plant to plant. They also have a stylet which is used to pierce host cells and sometimes to withdraw the fluid contents; in this they resemble aphids, which are well-known vectors of virus diseases of plants. The first demonstration that plant parasitic nematodes can act as vectors of viruses was made by Hewitt, Raski, and Goheen (1958) who showed that Xiphinema index was the vector of the virus which causes fan-leaf disease of grapevines in California. Extensive work by this group of workers has shown that uninfected nematodes, after feeding upon roots of infected plants, transmit the disease to uninfected plants when put to feed upon their roots and that even early larval stages of the nematode can transmit the virus. The nematodes became infected after only

106

one day spent feeding upon infected roots and could retain the virus for at least 30 days in the absence of any plant roots. Preliminary work indicated that there was no transovarian passage of the virus. Das and Raski (1968) later showed that X. index acquired and transmitted grape-vine fanleaf virus within an access time of 15 minutes and that the virus was maintained for at least 12 weeks when the nematodes were maintained on fig but not in fallow soil. Various workers have now shown that species of Xiphinema can transmit several virus diseases. Breece and Hart (1959) implicated X. americanum in the transmission of yellow bud mosaic of peach. Jha and Posnette (1959) and Harrison and Cadman (1959) showed that species of Xiphinema were responsible for the transmission of arabis mosaic virus to strawberry, raspberry, and white clover. Harrison (1967) later showed that all stages of X. diversicaudatum could transfer strawberry latent ringspot virus to cucumber seedlings and that when infected nematodes were transferred to fresh plants every two to four days for three weeks, single nematodes transmitted the virus on up to three successive occasions. Species of Trichodorus have been shown by various workers to transmit tobacco rattle virus (TRY). Bj0rnstad and St0en (1967) showed that T. pachydermus could transmit TRV but that Longidoms spp. extracted from the same soil samples did not do so. Van Hoof (1968) has shown that nine species of Trichodorus in the Netherlands are all potential vectors of TRV. Raspberry ringspot virus and tomato black ring virus are transmitted by Longidoms elongatus. There are now many records of these nematode/plant virus relationships but the nematodes involved are mostly species of Xiphinema, Trichodorus, and Longidoms. Apparently nematodes transmitting plant viruses fall into two groups: Longidoms and Xiphinema spp. transmit polyhedron-shaped viruses, while Trichodorus spp. transmit rod-shaped rattle viruses. Taylor and Raski (1964) suggested that viruses

D. L. LEE

transmitted by Longidorus are "non-persistent" and are mechanically transmitted whereas those in Xiphinema are more closely associated with the tissues of the nematode. Until recently no one knew the whereabouts of the virus particles in the nematode but Taylor and Robertson (1969, 1970) have now found virus particles in Longidorus elongatus and Trichodorus pachydermus by studying the ultrastructure of these nematodes. This is a very important advance in this field. Longidorus elongatus is the natural vector of raspberry ringspot virus and tomato black ring virus and Taylor and Robertson (1969) have found virus-like particles closely associated with the cuticle of the stylet guiding sheath and in the lumen of the buccal capsule of infected nematodes (Figs. 1,2). Similarly, they have found rod-shaped virus particles, presumably of tobacco rattle virus, in the lumen of the oesophagus of Trichodorus pachydermus taken from a population known to transmit the virus (1970). The virus particles were mostly orientated lengthwise in close association with the cuticle which lines the oesophagus (Fig. 3 ). They found more particles associated with the cuticle of the glandular part of the oesophagus than elsewhere, and suggested that this may be because of the presence

FIGURES 1 and 2 1 Electron micrograph of a longitudinal section through the guiding sheath of Longidoms elongatus to show particles of raspberry ringspot virus in association with the cuticle of the guiding sheath. (Courtesy of C. E. Taylor) 2 Electron micrograph of a transverse section through the top part of the stylet of L. elongatus to show raspberry ringspot virus in association with the guiding sheath. (Courtesy of C. E. Taylor) g, guiding sheath; 1, lumen of buccal capsule; s, stylet; v, virus particles.

HELMINTHS AS VECTORS OF MICRO-ORGANISMS

107

108

of a layer of mucus-like material associated with the cuticle in this part of the oesophagus (Fig. 4). The size of the rod-shaped particles agrees closely with the dimensions of TRV particles. They envisage the transmission process in these plant parasitic nematodes as a mechanical contamination of the stylet. The nematodes pierce the epidermal cells of root tips of infected plants by thrusting their solid stylet into the cell. They then suck out the contents by the pumping action of the oesophagus. TRV particles taken in with the cytoplasm of the host cell are then apparently adsorbed onto the cuticle of the oesophagus. Subsequent transmission of the virus particles presumably takes place when the secretions from the oesophageal glands flow out of the mouth of the nematode when it thrusts out its stylet to pierce another cell. There is a specific association between a species of nematode and a particular virus because it has been shown that other species of nematode will ingest the virus but do not retain it on the cuticle of the oesophagus and do not transmit it. It has been suggested by Taylor and Robertson (1970) that surface charges of the virus particle and the nematode cuticle may be important in the adsorption of the particle on to the cuticle and that differences in surface charge may account for the specificity of the nematode/virus relationship. Many plant parasitic nematodes are able to move from host to host during feeding and one individual can, therefore, transmit the virus from one host to another. In animal parasitic nematodes, however, the nematode is confined to one host at a time and new hosts are infected by ingesting embryonated eggs or by the acquisition of active larvae. It is obvious, therefore, that if animal parasitic nematodes are to act as vectors of micro-organisms they must be capable of transferring the micro-organisms to their offspring through the egg. The first authentic record of an animal parasitic nematode acting as a vector for a virus was in 1941 when Shope implicated the swine lung-

D. L. LEE

worms Metastrongylus elongatus and M. pudendotectus in the transmission of swine influenza virus. Apparently the virus remains in an undetectable, or masked, phase while the first three larval stages of the nematode develop outside of the definitive host. The eggs of the nematode normally hatch in earthworms; the larvae penetrate the tissues of the earthworm and develop to the third-stage infective larva. When infected earthworms are eaten by pigs the nematode larvae penetrate the intestinal wall and are carried to the lungs via the circulatory system. The virus is not activated in the pig until some extra stimulus, such as an adverse change in climate or stress due to some other infection, triggers it off (Shope 1943, 1955). The work by Shope has been confirmed by Sen et al. ( 1961 ) who used pathogen-free, antibody-free pigs and migrating Ascaris suum larvae as the stimulus needed to "unmask" the virus which had been transmitted to the pigs by Metastrongylus. Shotts et al. (1968) set up a model system which can be investigated in the laboratory, using Strongyloides ratti and swine influenza virus in rats and mice. They showed that S. ratti can act as a carrier of swine influenza virus and infected mice this way. They were also able to recover the virus from homogenates of S. ratti. It will be interesting to find out where the virus is in the

FIGURES 3 and 4 3 Electron micrograph of a transverse section through the oesophagus of Trichodorus pachydermus with particles of tobacco rattle virus associated with the cuticular lining of the oesophagus. (Courtesy of C. E. Taylor) 4 Electron micrograph of a transverse section through the oesophagus of T. pachydermus to show the particles of tobacco rattle virus within the mucus layer on the surface of the cuticle. (Courtesy of C.E.Taylor) c, cuticle; 1, lumen of oesophagus; m, mucus layer; v, virus particles.

HELMINTHS AS VECTORS OF MICRO-ORGANISMS

109

110

various stages of the nematode, especially in the adult female, as the virus presumably must pass from the alimentary tract of the nematode to the ovary if it is to infect the eggs. Syverton, McCoy, and Koomen (1947) have implicated Trichinella spiralis in the transmission of lymphocytic choriomeningitis as they found that encysted larvae taken from muscle of guinea pigs infected with this virus were able to transmit the virus to other guinea pigs when the larvae were fed to, or injected into, them. The authors felt confident that they had avoided all possibility of surface contaminations of the larvae in their experiment. Possibly this is not an important relationship in nature as animals may become infected from infected flesh of the prey as well as by the nematode larvae because this virus is resistant to digestive juices. Stefanski and Zebrowski (1958) found that Newcastle disease virus became incorporated into Ascaridia galli when chickens were concurrently infected with the nematode and the virus. They stated that although the nematode is not usually associated with outbreaks of Newcastle disease in poultry, it may act as a reservoir of the infection in a given area as the embryonated eggs of the nematode can remain viable for many months. Various workers have implicated nematodes in the mechanical transmission of bacteria into other hosts but these seem not to be specific relationships between a nematode and a bacterium but due to the introduction of bacteria from the environment or from the alimentary tract into the tissues of the host. There is, however, a specific association between a species of Neoplectana (DD 136) which is parasitic in insects and the bacterium Achromobacter nematophilus. This association was first discovered by Dutky and Hough (1955) who recovered the nematode and its associated bacterium from diseased codlingmoth larvae. This nematode apparently parasitizes several species of insect and attempts have been made to use it, together with its associated bacterium, to control insect pests. Welch and

D. L. LEE

Bronskill (1962) found that the nematode is quickly encapsulated by larvae of Aedes aegypti after it has penetrated the gut wall but encapsulation does not prevent the host from dying of the associated bacterium. Poinar (1966) found that infective nematode larvae contained Achromobacter in the ventricular portion of the intestinal lumen and in two instances he found bacterial cells, presumably Achromobacter, inside intestinal cells of the nematode. When the nematode penetrated into the body cavity of a suitable host insect the bacteria were released through the anus and multiplied rapidly in the host's body causing rapid death of the insect. The nematodes continue to feed upon the tissues of the dead insect and on the bacteria. Finally, I wish to spend some time on associations between protozoa and nematodes and especially upon the relationship between Histomonas meleagridis and Heterakis gallinarum on which I am currently working. H. meleagridis is a protozoon which causes an enterohepatitis disease of chickens and turkeys. The disease is variously referred to as blackhead, infectious enterohepatitis, or histomoniasis. Histomonas meleagridis has affinities with both the amoebae and with the flagellates, the stage which lives in the caecal lumen of the bird and the stage which grows in vitro possessing flagella and also putting out pseudopodia. None of the stages in the bird, nor the cultured stage, possess mitochondria (Lee etal 1969; Schuster 1968). It does not possess a cystic stage but is transmitted from bird to bird by the embryonated eggs of the nematode Heterakis gallinarum, which is itself a parasite in the caecum of the bird. This surprising relationship was first discovered by Graybill and Smith (1920) and has been verified on numerous occasions. Indeed, the normal laboratory method of infecting birds with Histomonas is to give them embryonated eggs of Heterakis. Several workers have described what they interpreted as Histomonas in various tissues of the nematode, notably Tyzzer (1934) and Kendall (1959) in the larva, and Niimi ( 1937), Deso-

HELMINTHS AS VECTORS OF MICRO-ORGANISMS

witz (1950), and Gibbs (1962) in the adult nematode. The most thorough studies were those of Niimi, whose paper was unnoticed for many years, and Gibbs, and it is obvious that both had found the protozoon inside the nematode. Niimi suggested that Histomonas is ingested by Heterakis in the caecal lumen of the bird and passes through the wall of the intestine into the pseudocoelom, where it disappears in the male but invades the ovaries and eggs of the female. He stated that while in the intestine of the nematode the protozoon is similar to the stages found in the caecum of the bird, but the stages in the ovary and in the eggs are much smaller ( 1 to 1.4 jam ) in diameter. Gibbs ( 1962 ), who used both infected and uninfected Heterakis, found the protozoon in the reproductive system of both sexes and also in the intestine. He described penetration of the ovary wall by the protozoon and found it in the oocytes and eggs of the female and between the spermatozoa of the male. This was unquestionably Histomonas, but all of the work was done with the light microscope and so little knowledge was gained about the structure of the protozoon and its relationship with the cytoplasm of the nematode's cells. I have recently shown (Lee 1969) that there is a definite development of Histomonas in the reproductive tract of female Heterakis and that the nematode is not only a vector of the protozoon but is also a true intermediate host as the protozoon undergoes development in the nematode. I will briefly describe these results together with other, unpublished, results, on Histomonas in the intestine of the nematode and in male Heterakis. The aims of my work were to find out ( 1 ) if the protozoon differs structurally from the stages in the avian host (Lee et al 1969), (2) if it multiplies within the nematode, (3 ) if there is a "resting" stage in the nematode, (4) if it feeds upon the nematode, and (5 ) how it gains entry into the egg of the nematode. Apparently the adult nematode, which feeds upon the caecal contents of the bird, becomes infected when it ingests a histomonad. This invades

111

the intestinal epithelium of the nematode and, once inside a cell (Fig. 5 ), multiplies to such an extent that the protozoa almost completely fill the cell (Fig, 6). Although there is extensive enlargement of the cell there is little evidence of cytolytic activity by the protozoa and no obvious reduction in the glycogen stores of the intestinal cell. This appears to be a transient stage in the nematode and also of relatively infrequent occurence; I have seen it rarely in infected nematodes and only a small percentage of the nematodes in an infected bird become infected with Histomonas. After multiplication in the intestinal cell the protozoa apparently break out into the body cavity of the nematode and penetrate the reproductive system. In the female they cross the ovary wall and lie between the oogonia (Fig. 7). Here they feed and divide to give many individuals which gradually pass down the ovary into the growth zone. The protozoa are easily recognized at this stage as they are much less dense than the oogonia, have no mitochondria, have a much smaller nucleus, and have a row of microtubules associated with the Golgi apparatus and with the two centrioles. The oogonia develop into oocytes in the growth zone of the ovary and when they begin to develop into larger cells the protozoon breaks through the oolemma (Figs. 9, 10) and becomes intracellular (Fig. 8). The damage to the oolemma is repaired after the protozoon has entered. Inside the oocyte the Histomonas continues to feed and divide, although at a much slower rate than before, and it remains in the cytoplasm during formation of the egg-shell (Figs. 11,12, 13 ). There is no membrane of host origin between the cell membrane of the Histomonas and the cytoplasm of the oocyte, and the protozoon appears to be feeding upon the oocyte, there being usually a small space between the protozoon and the cytoplasm of the oocyte or egg (Figs. 11, 12). The amount of damage is, however, very slight. This lack of pathogenicity in the nematode contrasts sharply with the course of the infection in the bird, where the protozoon is very pathogenic, but this has ob-

112

vious survival value for the nematode. If Histomonas were pathogenic to the nematode, then it would curtail or stop egg production by the worm, which would not be to the advantage of the protozoon since it is transmitted to new avian hosts inside the egg of the nematode. There is a gradual reduction in size of the protozoon from the intestine of the nematode to the ovary and along the length of the ovary to the fertilized egg so that the relatively large protozoon (about 10 /¿m) in the intestine becomes 4 or 5 fim in size in the egg of the nematode (Fig. 12 ). Again, this is of obvious survival value since a large parasite inside the nematode egg would certainly interfere with embryonation of the egg whereas a small one will be less likely to prevent embryonation. Histomonas is obviously parasitic in Heterakis and is not a resting stage, but the nematode is apparently able to repair the slight damage caused by the protozoon and the two are well adapted to each other. In male Heterakis Gibbs (1962) found histomonads in the intestine; they ranged in size from 5 to 9 ¡mm and some appeared to be dividing. Some showed pseudopod-like extensions, and a blepharoplast could be identified. He also found histomonads in the reproductive system. In the testis the protozoa occurred among the developing spermatozoa but were most numerous in the cells lining the rest of the reproductive tract.

FIGURES 5 and 6 5 Histomonas meleagridis in intestinal cells of Heterakis gallinarum. l^cm section of Aralditeembedded material, stained with toluidine blue. 6 Histomonas meleagridis in a single cell of Heterakis gallinarum. The protozoon has multiplied within the intestinal cell. 1 ¡mm section of Aralditeembedded material, stained with the periodic acid/ Schiif method for polysaccharides. e, egg; h, Histomonas; hn, nucleus of Histomonas; 1, lumen of intestine; n, nucleus of intestinal cell; p, pseudocoelom; u, uterus.

D. L. LEE

FIGURES 7 and 8 7 Electron micrograph of a section through the germinal zone of the ovary of Heterakis gallinarum to show Histomonas meleagridis lying between the oogonia and the rachis. 8 Electron micrograph of a section through the growth zone of the ovary of Heterakis gallinarum to show Histomonas meleagridis inside an oocyte. Note the difference in size of the nuclei of the oocyte and of the protozoon. h, Histomonas; hn, nucleus of Histomonas; m, mitochondrion; n, nucleus of oogonium or oocyte; 0, oogonium; sp, shell precursor. FIGURES 9 and 10. Electron micrographs of sections through the growth zone of the ovary of Heterakis gallinarum to show Histomonas meleagridis in the process of entering an oocyte. The large arrow indicates the direction of entry of the Histomonas into the oocyte. b, break in oolemma; c, centriole; f, parabasal filament; g, Golgi complex; hn, nucleus of Histomonas; 1, lipid; o, oolemma. FIGURES 11-13 11-12 Electron micrographs of sections through the egg of Heterakis gallinarum to show Histomonas meleagridis dividing (Fig. 11) and the structure of the protozoon (Fig. 12). 13 Light micrograph of a 1 /¿m section through the egg of Heterakis gallinarum to show three histomonads in the egg. c, centriole; es, egg-shell; f, parabasal filament; h, Histomonas; hn, nucleus of Histomonas. FIGURES 14 and 15. Electron micrographs of sections through the testis of Heterakis gallinarum to show Histomonas meleagridis lying between spermatids (Fig. 14) and between the wall of the testis and the spermatids (Fig. 15). Note that the protozoon is much larger here than in the female reproductive tract (see Fig. 8). fv, food vacuole; h, Histomonas; hn, nucleus of Histomonas; sp, spermatid; w, wall of testis.

HELMINTHS AS VECTORS OF MICRO-ORGANISMS

113

114

D. L. LEE

HELMINTHS AS VECTORS OF MICRO-ORGANISMS

117

118

Many were found in the lumen of the seminal vesicle and in the vas deferens and showed tubelike extensions. These individuals ranged in size from 4 to 17 /¿m. My studies with the electron microscope on the male reproductive tract of Heterakis infected with Histomonas are not yet complete. I have, however, seen the protozoon in the lower part of the testis between developing spermatocytes and spermatids (Figs. 14,15) and in the wall of the vas deferens. It appears to have the same structure in these various locations. In the testis it lies between the spermatocytes and the wall of the testis and pushes out pseudopodia between the germ cells (Figs. 14,15). It is about 10 /¿m in diameter but the pseudopodia extend out from this central mass of the protozoon. The nucleus is spherical and about 2.5 ¡jum in diameter. There are numerous electron-dense whorls of membranes enclosing vacuoles, which probably represent food vacuoles. A row of microtubules, which resembles a striated fibre, is closely associated with the nucleus and with a Golgi complex, but centrioles have not yet been observed. There are numerous small vesicles in the cytoplasm but no mitochondria. This stage of Histomonas appears to be much more active than those in the female nematode. It is interesting that Springer, Johnson, and Reid ( 1969) were able to transmit Histomonas to turkey poults by feeding them whole male Heterakis but not whole female Heterakis. This, at first, appears odd but in the light of my electron microscope evidence it would appear that Histomonas in the male nematode is in an active feeding stage which resembles the invasive stage in the bird (Leee/fl/. 1969) and could be physiologically and structurally able to infect birds, whereas the stages in the female are much smaller and less active and possibly unable to infect birds because of some physiological or structural inability. If this is so, then infectivity must be regained during development in the embryonating egg and the larva of the nematode. Although we have studied embryonating eggs and artificially hatched larvae of Heterakis with the electron microscope we

D. L. LEE

have so far not observed Histomonas. Heterakis gallinarum has also been shown to transmit the non-pathogenic Histomonas w enrichi (Lund 1968) and my studies on this protozoon indicate that it has the same relationship with the nematode as does H. meleagridis (Lee 1969). These are the only well-established examples of transmission of a protozoon by a nematode. There are, however, indications that there may be others. Burrows and Swerdlow (1956) suspected that they had found Dientamoeba fragilis, an amoeboid parasite of the alimentary canal of man, in eggs of Enterobius vermicularis. Both the protozoon and the nematode are parasites of the human colon and, as Dientamoeba has no cystic stage, it may possibly use Enterobius as a vector in the same way as Histomonas uses Heterakis. If this is so, it will be interesting to see if there is also development of Dientamoeba in the reproductive tract of Enterobius. Hutchison (1965) thought that Toxoplasma gondii may be transmitted by the embryonated eggs of Toxocara cali. He fed mice infected with Toxoplasma to a cat which had been infected with T. cati two months previously. Two weeks later the faeces of the cat were subjected to zinc sulphate flotation, and nematode ova, together with other particulate matter, were collected. After three months six mice were each given some of this suspension, containing ova and other particles, by stomach tube. After tests it was shown that these mice had been infected with T. gondii. Larval forms of Toxocara were found in the brains of some of the mice. Hutchison found only bacteria, oocysts of Isospora, and ova of T. cati in the cat's faeces and stated: "On the basis of the facts presented, it appears that some form of Toxoplasma is capable of passing into the external environment in the faeces of infected cats." He also stated that it was impossible to say whether the infection was transmitted in eggs of Toxocara or by some other means. This paper by Hutchison stimulated many investigations into the possible role of nematodes in the transmission of Toxoplasma. Dubey (1968) found that as few as two artificially

119

HELMINTHS AS VECTORS OF MICRO-ORGANISMS

hatched larvae of T. cati taken from faeces of a cat infected with both Toxoplasma and Toxocara would infect mice with Toxoplasma when injected intraperitoneally. Recently, however, Hutchison, Dunachie, and Work (1968) have shown that faeces produced by cats artificially infected with Toxoplasma but not infected with Toxocara can transmit the protozoon to susceptible mice. The infectivity of the faecal extracts was retained at room temperature, and they concluded that there was evidence for the existence of unknown forms of Toxoplasma which are passed independently in faeces and which can survice in moist conditions for at least three months. Work and Hutchison (1969) later found a cystic organism, capable of producing Toxoplasma antibodies and typical Toxoplasma infections in mice, in the faeces of infected cats. This, they claim, is a new cystic form of T. gondii. Also, Frankel, Dubey, and Miller (1969) found that washing faeces from cats infected with Toxoplasma gondii and Toxocara cati through a 44 /xm sieve retained the nematode eggs but allowed the protozoon through. The protozoa were retained by a 10-15 ¡mm filter. These stages of T. gondii in the faeces developed infectivity within a week and could be stored at room temperature for at least three months. They also obtained the faecal forms of T. gondii from apparently Toxocara-îree cats which were infected with T. gondii. These recent findings would seem to disprove the theory that Toxoplasma gondii is transmitted by nematodes, although Dubey's (1968) work on isolated larvae suggests otherwise. We are left, therefore, with the relationship between Histomonas meleagridis and H. ^enrichi and Heterakis gallinarum as the only proven examples of transmission of a protozoon by a nematode. I wish to thank Dr C. E. Taylor for giving me access to unpublished work on the location of plant viruses in plant parasitic nematodes and for allowing me to use some of his electron micrographs.

REFERENCES

Bj0rnstad, A., and St0en, M. 1967. Rattelvirus med Trichodorus pachydermus som vektor. Norsk Landbruksakad. 8:12-14 Breece, J. R., and Hart, W. H. 1959. A possible association of nematodes with the spread of peach yellow bud mosaic virus. Plant Dis. Rept. 43:989-90 Burrows, R. B., and Swerdlow, M.A. 1956. Enterobius vermicularis as a probable vector of Dientamoeba fragilis. Am. J. Trop. Med. Hyg. 5: 258-65 Das, S., and Raski, D. J. 1968. Vector-efficiency of Xiphinema index in the transmission of grapevine fanleaf virus. Nematologica 14:55-62 Desowitz, R. S. 1950. Protozoon hyperparasitism of Heterakis gallinae. Nature 165:1023 Dubey, J. P. 1968. Feline toxoplasmosis and its nematode transmission. Vet. Bull. 38:495-9 Dutky, S. R., and Hough, W. S. 1955. Note on a parasitic nematode from codling moth larvae, Carpocapsa pomonella (Lepidoptera, Olethreutidae). Proc. Entomol. Soc., Wash. 57:244 Findlay, G. M., and Howard, E. M. 1951. Notes on Rift Valley fever. Arch. ges. Virusforsch. 4: 411-23 Frenkel, J. K., Dubey, J. P., and Miller, N. L. 1969. Toxoplasma gondii: fecal forms separated from eggs of the nematode Toxocara cati. Science 164:432-3 Gibbs, B. J. 1962. The occurrence of the protozoon parasite Histomonas meleagridis in the adults and eggs of the cecal worm Heterakis gallinae. J. Protozool. 9:288-93 Graybill, H. W., and Smith, T. 1920. Production of fatal blackhead in turkeys by feeding embryonated eggs of Heterakis papulosa. J. Exptl. Med. 21:647-62 Harrison, B. D. 1967. The transmission of strawberry latent ringspot virus by Xiphinema diversecaudatum (Nematoda). Ann. Appl. Biol. 60: 405-9 Harrison, B. D., and Cadman, C. H. 1959. Role of a dagger nematode (Xiphinema sp.) in outbreaks

120 of plant diseases caused by arabis mosaic virus. Nature 184:1624-6 Hewitt, W. B., and Raski, D. J. 1962. Nematode vectors of plant viruses. In Biological Transmission of Disease Agents, éd. K. Maramorosch, pp. 63-72. Symposium of the Entomological Society of America, Atlantic City, 1960. New York and London: Academic Press Hewitt, W. B., Raski, D. J., and Goheen, A. C. 1958. Nematode vector of soil-borne fanleaf virus of grapevines. Phytopathology 48:586-95 Hoof, H. A. van. 1968. Transmission of tobacco rattle virus by Trichodorus species. Nematologica 14:20-4 Hutchison, W. M. 1965. Experimental transmission of Toxoplasma gondii. Nature 206:961-2 Hutchison, W. M., Dunachie, J. F., and Work, K. 1968. The faecal transmission of Toxoplasma gondii. Acta Pathol. Microbio]. Scand. 74:462-4 Jha, A., and Posnette, A. F. 1959. Transmission of a virus to strawberry plants by a nematode (Xiphinema sp.). Nature 184:962 Kendall, S. B. 1959. The occurrence of Histomonas meleagridis in Heterakis gallinae. Parasitology 49:169-72 Lee, D. L. 1969. The structure and development of Histomonas meleagridis (Mastigamoebidae: Protozoa) in the female reproductive tract of its intermediate host Heterakis gallinarum (Nematoda). Parasitology 59:877-84 Lee, D.L., Long, P. L.,Millard,B.J. 5 and Bradley,!. 1969. The fine structure and method of feeding of the tissue parasitizing stages of Histomonas meleagridis. Parasitology 59:171—84 Lund, E. E. 1968. Acquisition and liberation of Histomonas wenrichi by Heterakis gallinarum. Exptl. Parasitol. 22:62-7 Niimi, D. 1937. Studies on blackhead, n. Mode of infection. /. Japan. Soc. Vet. Sci. 16:23-6 Philip, C. B. 1955. There's always something new under the "parasitological" sun (the unique story of helminth-borne salmon poisoning disease). J. Parasitol. 41:125-48 - 1962. Helminths as carriers of microbial disease agents of man and animals. In Biological Transmission of Disease Agents, éd. K. Maramorosch,

D. L. LEE

pp. 159-69. Symposium of the Entomological Society of America, Atlantic City, 1960. New York and London: Academic Press Poinar, G. O. 1966. The presence of Achromobacter nematophilus in the infective stage of a Neoaplectana sp. (Steinernematidae: Nematoda). Nematologica 12:105-8 Schuster, F. L. 1968. Ultrastructure of Histomonas meleagridis (Smith) Tyzzer, a parasitic ameboflagellate. /. Parasitol. 54:725-37 Sen, H. G., Kelley, A. W., Underdahl, N. R., and Young, G. A. 1961. Transmission of swine influenza virus by lungworm migration. J. Exptl. Med. 113:517-20 Shope, R. E. 1941. Swine lungworm as reservoir and intermediate host for swine influenza virus; transmission of swine influenza virus by swine lungworm. J. Exptl. Med. 74:49-68 - 1943. The swine lungworm as a reservoir and intermediate host for swine influenza virus, m. Factors influencing transmission of the virus and provocation of influenza. /. Exptl. Med. 77:11138 - 1955. The swine lungworm as a reservoir and intermediate host for swine influenza virus, v. Provocation of swine influenza by exposure of prepared swine to adverse weather. /. Exptl. Med. 102:567-72 Shotts, E. B., Foster, J. W., Brugh, M., Jordan, H. E., and McQueen, J. L. 1968. An intestinal threadworm as a reservoir and intermediate host for swine influenza virus. /. Exptl. Med. 127:35969 Springer, W. T., Johnson, J., and Reid, W. M. 1969. Transmission of histomoniasis with male Heterakis gallinarum (Nematoda). Parasitology 59: 401-5 Stefanski, W., and Zebrowski, L. 1958. Investigations on the transmission of Newcastle disease virus by Ascaridia galli and the pathogenic synergism of both agents. Bull. acad. polon. sci. Classell 6:67-72 Syverton, J. T., McCoy, O. R., and Koomen, J. 1947. The transmission of the virus of lymphocytic choriomeningitis by Trichinella spiralis. J. Exptl. Med. 85:159-69

HELMINTHS AS VECTORS OF MICRO-ORGANISMS

Taylor, C. E., and Raski, D. J. 1964. On the transmission of grape fanleaf by Xiphinema index. Nematologica 10:489-95 Taylor, C. E., and Robertson, W. M. 1969. The location of raspberry ringspot and tomato black ring viruses in the nematode vector, Longidorus elongatus (de Man). Ann. Appl. Biol. 64:233-7 - 1970. Location of tobacco rattle virus in the nematode vector, Trichodorus pachydermus Seinhorst. J. Gen. Virol. 6:179-82 Tyzzer, E. E. 1934. Studies on histomoniasis, or "blackhead" infection, in the chicken and the turkey. Proc. Am. A cad. Art s S ci. 69:189-264 Welch, H. E., and Bronskill, J. F. 1962. Parasitism of mosquito larvae by the nematode, DD 136 (Nematoda: Neoaplectanidae). Can. J. Zool. 40:1263-8 Work, K., and Hutchison, W. M. 1969. A new cystic form of Toxoplasma gondii. Acta Pathol. Microbiol. Scand. 75:191-2

Discussion K. A. WRIGHT

Department of Parasitology, School of Hygiene, University of Toronto Dr Lee has given a most interesting review of vector/micro-organism associations in the helminths. If we expand this group of helminths to include leeches, we have yet another group of vectors transmitting trypanosomes and haemogregarines to amphibians and reptiles. We could site Trypanosoma diemyctyli of salamanders transmitted by Batrachobdella picta (cf. J. H. Barrow, Trans. Am. Microscop. Soc. 72:197-216 [1953]), T. chrysemydis of turtles transmitted by Placobdella spp. (cf. P. T. K. Woo, Can. J. Zool. 47:1139-51 [1969]), Haemogregarina sp. of turtles transmitted by Emys orbicular is (cf. E. Reichenow, Arch. f. Protistol. 20:251-350 [1910]), and Crytobia salmositica transmitted to fish by a rhynchobdellid (cf. C. D. Becker and M. Katz, J. Parasitol. 51:95-99 [1965]).

121

Clear understanding of the mechanisms of transmission will require detailed knowledge of the microanatomy of the vector. Consider the feeding mechanisms of nematodes that transmit plant viruses. A summary of electron microscope examinations of Trichodorus by H. Hirumi, T. A. Chen, K. J. Lee, and K. Maramorosch (J. Ultrastruct. Research 24:434-53 [1968]) and of Xiphinema by K. A. Wright (Can. J. Zool 43:689-700 [1965]) and D. R. Roggen, D. J. Raski, and N. O. Jones (Nematologica 13:1-16 [1967]) allows the interpretation of their stylet apparatus. In Trichodorus the stylet is a solid tooth-like structure developed in the dorsal wall of the oesophagus, while in Xiphinema (representative of the family Longidoridae) it is a hollow rolled structure inserted into the anterior cuticle of the oesophagus. In both nematodes muscular extrusion of the stylet presumably results in rupture of the plant cell. In Trichodorus food materials enter through the simple channel of the buccal capsule into the oesophagus, while in Xiphinema and Longidorus food is taken in through the stylet itself while the stylet is inserted into the plant. In reference to the mechanism of virus transfer suggested by Dr Taylor and reported here by Dr Lee, I would suggest that viruses adsorbed onto any region of the buccal capsule or oesophagus of Trichodorus may be carried into host tissues with secretions of the oesophageal glands. In contrast, in the Longidoridae, viruses adsorbed onto the anterior oesophageal cuticle may be subjected to simple transport into the host, while viruses adsorbed onto the stoma cuticle, or the cuticle forming the guide ring, as illustrated by Taylor, may not be so readily injected. May these viruses be at a dead end for transmission, or may they serve as a more secluded reservoir of virus that would prolong the period of virus retention by the vector? Furthermore, Roggen's analysis of muscle action in the posterior oesophageal bulb of X. index suggests that only secretions of the dorsal gland may be extruded through the stylet, while subventral gland secretions are passed posteriorly. Thus, the chemical nature of the dorsal gland exudates become of special interest. It may be suggested that viruses located posteriorly to the dorsal

122

gland orifice are as valueless as those passed into the intestine, where Taylor considers them to be of no significance in transmisión. In the Longidoridae, we are then limited to the area between the stylet and the anterior tip of the posterior oesophageal bulb as the significant portion of the feeding apparatus from which virus may be washed back into host plants. Considering the Histomonas-Heterakis transmission discussed by Dr Lee, again the microanatomy of the nematode and protozoan may be important in relation to the movement of histomonads through the vector. The protozoan passes through three major compartments of the nematode during its development. After passive entry to the intestinal lumen, it enters intestinal cells, presumably through phagocytic activity of the host, or active penetration of the protozoan. If penetration occurs, it would be of interest to know its mechanism, as penetration organelles apparently do not occur. The histomonads may be prevented from migrating between cells by the junctional complexes that bind the nematode's intestinal epithelium together. Further information is required on the structure of the nematode's reproductive tract in order to determine if migration into it resembles the protozoan's entry into the intestine. The musculoepithelium of the female tract of the trichuroids is discontinuous, leaving sizable gaps between cells, although its basal lamina is continuous. If this occurs in Heterakis, it could facilitate passage of the histomonads intercellularly into the female tract, although penetration of the basal lamina would still be required.

D. L. LEE

Site-finding behaviour in helminths in intermediate and definitive hosts MARTIN J. ULMER

Unusually complex and subtle factors characterize most host-parasite adaptations, and despite continuous advances in physiology, electron microscopy, in vitro culture techniques, and immunology, numerous aspects of the relationships between parasite and host remain enigmatic. The analysis of helminth behaviour has recently begun to attract the attention of many parasitologists, and correlative morphological studies on sensory structures attest to a growing interest in this neglected area. Until recently, most behavioural studies on helminths have been limited to incidental observations made during the course of life cycle studies. Experimental approaches to behavioural problems, particularly with reference to freeliving stages of helminths, now appear with increasing frequency, however. Salt (1935) and Laing (1937) observed that host-finding by parasitic insects involves a twofold process involving selection of host environment (ecological selection) followed by finding of the host within the host's environment (psychological selection). Referring to trematode miracidia, Wright (1959a) suggested that the establishment of contact between the host and parasite involved three stages : ( 1 ) attraction of the parasite to the environment in which the host is found; (2) random movements until the parasite comes within chemical range of its host; and (3 ) chemotactic responses of the parasite to the host ("host-selection"). This sequential pattern proposed by Wright may be typical for other helminths as well. The present paper concerns itself in part with host-selection, but principally with another puzzling and intriguing aspect, namely, the site-finding behaviour of helminths after they have reached their intermediate and definitive hosts. How helminths recognize and respond to those essential sites which are prerequisite to their establishment and survival constitutes an almost virgin field for parasitologists. Monoge-

124

netic flukes and phytoparasitic nematodes hâve not been included in this discussion, for they are considered elsewhere in this symposium. TREMATODES

Miracidia The occurrence of free-living stages in trematode and nematode life cycles makes these helminths more accessible subjects than cestodes and acanthocephalans for studies on adaptive behaviour. This is reflected in the more extensive data based on studies of the former, and the paucity of available information on the latter groups. Considerable data have appeared on trematode behaviour associated with host-selection by miracidia, and significant review articles on this topic have been published in recent years (Wright 1959a; Cheng 1967; Schwabe and Kilejian 1968). With few exceptions, these studies have been based on species having medical or economic significance. For more than half a century, reports of chemoattraction of free-living mir acidia to their molluscan intermediate hosts have appeared, but an almost equal number of papers refute the concept. Despite conflicting reports, it is generally assumed that such attraction is characteristic of many digenetic flukes. Controversy still exists, however, as to whether or not such adaptive behaviour is related to host-specificity (Cheng 1968). The first published account of host-attractants for miracidia appears to be that of Leiper and Atkinson (1915), who stated that the snail Oncomelania nosophora "showed an extraordinarily marked attraction" for the miracidia of Schistosoma japónica, "as contrasted to other species." Later workers who favoured the concept of chemoattraction include Faust (1924), Faust and Meleney ( 1924 ), Faust and Hoffman (1934), Brumpt (1940), Plempel (1964), Etges and Decker (1963), Kloetzel (1958, 1960), Lutz( 1919), Neuhaus( 1952), Davenport, Wright, and Causley (1962), Maclnnis

MARTIN J. ULMER

(1963, 1965), Plempel, Gonnert, and Federmann (1966), Shiff (1968, 1969), and Shiff and Kriel (1970), all of whom studied miracidia of schistosomes;Neuhaus (1941, 1953), Kendall (1965), Wilson (1969), and Wilson and Denison (1970b), who dealt with Fasciola\ Barlow (1925), who suggested attraction of Fasciolopsis buski miracidia to molluscan hosts; Campbell ( 1961 ), who reported on Fascioloides magna\ Kawashima, Tada, and Miyazaki (1961), on Paragonimus; Tubangui and Pasco (1933), on Echinostoma\ and Mathias (1925 ), on Cotylurus. Some investigators report little or no evidence of attraction to normal snail hosts by specific chemotactic stimuli, and consequently consider contact between miracidia and molluscs to be only a chance phenomenon. These include Mattes (1926, 1936, 1949a, 1949b) and Griffiths ( 1939), who reported on Fasciola\ Swales (1935) and Campbell and Todd (1955), on Fascioloides\ Abdel-Malek (1950), Stirewalt (1951), Chemin and Dunavan ( 1962), Sudds ( 1960), and Barbosa (1960), on Schistosoma mansoni; Najim (1956), on Gigantobilharzia; Chu and Cutress (1954), on Austrobilharzia\ Sudds (1960), on Trichobilharzia and Schistosomatium\ Stunkard (1943 ), on Zoogonoides\ Wajdi (1964) and Cowper (1947), on Schistosoma:, Crandall (1960) and Ulmer and Sommer (1957), onHeronimus. LaRue (1951) in a review article on host-parasite relationships among digenetic trematodes concluded that under laboratory conditions "trial and error are much more important than chemotactic response in bringing the miracidium to the proper snail" but added "nevertheless, a chemical stimulus operating over very short distances probably explains the selection of the mollusk as one suitable for penetration." Etges and Decker (1963 ), in studying Schistosoma mansoni, concluded that light and gravity were far more powerful stimuli in orienting miracidia than chemical ones produced by the molluscan host. Although Chemin and Dunavan (1962) reported that snails do not attract miracidia, they were able to show that miracidia

SITE-FINDING BEHAVIOUR IN HELMINTHS

125

of S. mansoni are capable of locating snails at considerable distances. In a subsequent study of behavioural responses of S. mansoni miracidia to snail-emitted substances, Chemin (1970) concluded that "snail elaborated stimulants probably play a role in the natural host-finding process." Experimental devices for testing the host-finding capabilities of miracidia are of several types and include the slide-drop method used by Campbell and Todd (1955), various rectangular, tubular, or circular chambers (Neuhaus 1953;Kloetzel 1958;Etges and Decker 1963; Plempel 1964; Wilson and Denison 1970a), Ytube choice apparatus (Kawashima et al. 1961 ; Shiff 1968; Shiff and Kriel 1970), or the "flyingspot microscope" used by Davenport et al. ( 1962) for visualizing and photographing a televised image of moving miracidia on a screen. Single and multiple exposures of miracidia have been reported by investigators, such exposures having been made ( 1 ) to suitable and unsuitable molluscan hosts, (2) to pieces of filter paper impregnated with snail tissue extracts (Campbell 1961), (3) to gastropod mucus, or (4) to agar blocks or pyramids containing snail exudates or selected chemicals (Maclnnis 1965;

Cheng 1968). The studies of Maclnnis involving the use of agar test pyramids impregnated with aqueous solutions of amino acids, short chain fatty acids, sugars, and various salts indicate that specific attractants are involved in host-finding by S. mansoni and Schistosomatium douthitti miracidia. Movements exhibited by schistosome miracidia near intermediate hosts were categorized by Maclnnis into six recognizable types and, in addition, six types of behaviour following contact with a suitable host were described. Chemin ( 1968 ) described a linear channel involving the introduction of "decoy" snails between miracidia and susceptible snails, and reported a markedly reduced infection rate. The first record of the probable significance of secretions from snail body surfaces in host-finding by miracidia was that of Thomas (1883) in his work on Fasciola hepática. That mucus of molluscs may provide a clue to the understanding of chemotactic responses on the part of miracidia was suggested by Wright ( 1959b), who demonstrated that body-surface mucus of lymnaeid snails contains species-specific substances. He further showed that paper chromatography might serve as a useful tool in demonstrating chemotactic means employed by miracidia in

TABLE I Summary of experiments on the snail-finding ability of Megalodiscus temper at us miracidia

Diffusion time (hours)

Age of miracidia

Per cent positive

0

1 min to 1 h 2h 7h 1-15 min 3-5 h 8-8ih 1-10 min 2-4 h 8h 1-30 min 4-5 h 10-12 h

0 30 30 50 100 80 25 100 80 76 100 78

2-5 5-8

9-12

Average time (minutes); range in parentheses H (4-34) 21 (1-44) 24 (1-34)

U (4-24) if (4-34)

34 (24-4)

l (4-24) 2 (4-34) 14 (4-14) l (4-14) 2 (1-24)

Number of miracidia

10 10 10 10 22 10 12 30 10 17 10 18

126

MARTIN J. ULMER

DIFFUSION

TIME

AGE

OF

MIRACIDIA

(H)

AGE

OF

MIRACIDIA

(H)

FIGURE 1. Summary of experiments showing the effect of miracidial age and diffusion time of the

snail on the ability of Megalodiscus temperatus miracidia to locate Helisoma trivolvis.

selecting appropriate intermediate hosts. The probable relationship between mucus secretion and behavioural patterns in helminths deserves more attention; its significance in invertebrates has been repeatedly emphasized in a series of papers by Jakowska (1963, 1965, 1966). Few investigators appear to have considered the age of miracidia in connection with their ability to locate suitable hosts. Barlow (1925) and Campbell and Todd (1955) have shown, however, that miracidia of Fascioloides magna are more infective at certain ages than at others.

Experimental evidence providing additional support for the belief that factors such as age of miracidia, the presence of snail exudates, and incubation time of unreleased miracidia may be decisive in host-finding is given in a series of studies completed recently in our laboratory. The frog amphistome, Megalodiscus temperatus, was used as a source for miracidia. Shortly after gravid flukes of this species are transferred from the colon of their definitive host (Rana pipiens) to lake water, eggs are released, and hatching occurs at room temperature within a few minutes

SITE-FINDING BEHAVIOUR IN HELMINTHS

127

FIGURE 2. Apparatus for observing site-finding behaviour of Megalodiscus miracidia.

to an hour. The relative ease with which one may obtain adequate numbers of Megalodiscus miracidia makes this species particularly well suited for experimental studies, especially since miracidia are large and live for slightly more than 12 hours at room temperature. These experiments (summarized in Table i and Fig. 1 ) involved the individual exposure of numerous miracidia of varying ages to snails (Helisoma trivolvis) whose shell diameters ranged from 15 to 20 mm. Each exposure was made by releasing a single miracidium in the centre of a standard 9-cm-diameter petri dish containing 25 ml of filtered lake water (pH 8.0— 8.4). Each miracidium was observed constantly for 5 minutes and at each 15-second interval its position was recorded on a sheet of paper alongside the apparatus. A single snail was placed at the periphery of the petri dish. Movement of the snail was restricted by a piece of fibre adhesive tape extending above the water, from the shell to the edge of the container. This arrangement (Fig. 2 ) permitted the mollusc to extend its foot

at the bottom of the container, but prevented movement of the animal. Chernin and Dunavan (1962) concluded that the physical properties of the container in dispersion studies of S. mansoni miracidia influenced the distribution of miracidia - i.e. they tended to concentrate more at the outer margins than at the centre of the petri dishes used in their experiments. However, in 10 runs of individual miracidia of M. temperatus in petri dishes containing only water, random distribution of the organisms resulted. In our experiments, miracidia of various ages ( 1 minute to 12 hours after hatching) were used. When a snail was placed at the periphery of a petri dish, the time elapsing between its introduction and the introduction of a single miracidium in the centre of the dish was recorded as "mucus-diffusion time," which varied from 0 to 12 hours. Observations of miracidia were facilitated by use of a circular fluorescent 22-watt light placed approximately 4 inches above the water surface, thereby providing diffuse lighting. A black glass plate under the petri dish aided in observing miracidial

128

movements. Readings were taken from the moment of miracidial release until contact with the snail had been established and penetration had been initiated, or until 5 minutes (20 readings) after release in instances where miracidia failed to reach the mollusc. Results of this series of experiments involving more than 150 miracidia may be summarized as follows : ( 1 ) the ability of M. temperatus miracidia to locate a suitable snail host successfully within a given time varies greatly in accordance with their age; (2) recently hatched miracidia (1 minute to 1 hour old) as well as older miracidia (7 to 12 hours) are less apt to locate the mollusc than are miracidia 2 to 5 hours old, and when successful require more time to do so; (3 ) diffusion of exudates from the snail host enhances the host-finding ability of miracidia, as evidenced by the greater percentage of successful contacts made when snail exudates were present in the surrounding medium; and (4) under favourable conditions, miracidia may reach the snail in a remarkably short time (in these experiments, sometimes as little as a quarter of a minute; see Fig. 3). These results tend to confirm the findings of Campbell and Todd (1955) based on work with Fascioloides magna. Despite their conclusion that under experimental conditions "chemotaxis plays little or no part in establishing contact between this miracidium and the snail," they reported that miracidia are more infective "when between 1.5 and 2 hours old than when very young (less than 1 hour) or very old." The divergent opinions of previous workers on miracidial host-finding may have resulted from their failure to consider the age of miracidia in their experiments. Whereas Chemin and Perlstein (1969) observed that snail-derived substances (excretions, haemolymph, or tissue extract) "failed to interfere with host-finding" of S. mansoni miracidia, our studies demonstrate that such substances definitely aid in host-finding by miracidia of M. temperatus.

MARTIN J. ULMER

All miracidia used in our experiments on Megalodiscus noted above were obtained from eggs recently shed from gravid worms at room temperature. To determine if miracidia were equally effective when hatching was delayed by lower temperatures, freshly shed eggs were refrigerated (8-10° c) for 5 hours and were then allowed to hatch at room temperature. Of these miracidia, all less than an hour old, 90 per cent established contact with Helisoma in an average time of II minutes with no diffusion time. This represents a significantly higher percentage of successful runs in less time than obtained with the miracidia that were not refrigerated before hatching (Table i). Also, freshly hatched miracidia of M. temperatus were refrigerated for 5 hours and then introduced into the test chamber containing Helisoma with a 5-hour diffusion time. One hundred per cent contact resulted from 10 runs, with an average time of 1/4 minutes, results which approximate those of similar runs of unrefrigerated miracidia. These results provide experimental support for the suggestion by Campbell and Todd (1955) that freshly hatched miracidia have not yet fully developed a sensitivity to stimuli provided by snail hosts. Van

A

B

FIGURE 3. Path of miracidium of Megalodisciis temperatus in the presence of a snail host (Helisoma trivolvis). Position of miracidium indicated at 15second intervals. A Miracidium 7 hours old, no mucus-diffusion time (time: 4& minutes). B Miracidium 3 hours old, 5-hour mucus-diffusion time (time: 15 seconds).

SITE-FINDING BEHAVIOUR IN HELMINTHS

Gundy ( 1965 ) has also referred to this factor of physiological aging in unhatched juvenile nematodes. Additional experiments were conducted to determine if miracidia of Megalodiscus exhibited preference for a suitable intermediate host when exposed to snails of two different species. In these experiments, Stagnicola reftexa, an unsuitable host for Megalodiscus, was placed at one side of a petri dish and Helisoma trivolvis directly opposite. Mucus-diffusion time in these experiments varied from 2 to 5 hours. In 38 tests, 34 (89 per cent) of the miracidia were attracted to H. trivolvis in an average time of 1 minute (% to 2^2 minutes). Snail mucus has been emphasized repeatedly as the substance involved in host attraction of miracidia (Faust 1924; Faust and Meleney 1924; Wright 1959b, 1960; Campbell 1961; Kendall 1965 ), yet relatively little experimental evidence confirming this appears in the literature, although Maclnnis's studies (1963, 1965) on schistosomes suggests it is a very important factor. Wright (1964) conducted tests on various response-stimulating substances in snails and classified them into three types in accordance with the degree of change in behaviour of miracidia. In decreasing order of significance these were: ( 1 ) whole ground extracts of head and foot, digestive gland and faeces, and filtered lake water in which snails had remained overnight; (2) mantle; and (3) blood and mucus. Wilson (1968) showed that mucus contains a wide range of potential attractants for Fasciola hepática miracidia, and Wilson and Denison ( 1970a, b) presented experimental data indicating that the behaviour of miracidia of this species is greatly modified by the presence of snail mucus. It should be emphasized that mucus is a complex heterogeneous mixture, and that other secretory products of various types are constantly being brought to the surface of the gastropod foot. Wright (1960) has noted that differences in chemical composition of surface substances (including mucus) of snails may modify chemotac-

129

tic phases of the host-finding behaviour of miracidia. Among gastropods, however, other possible attractant-incorporating exudates may be associated with mucus. Opening into or near the mantle chamber of pulmonate gastropods are the digestive and reproductive systems, as well as the ureter. The latter leads to the renal organ, which is directly connected to the pericardial chamber via a renopericardial canal. Cheng (1963 ) has suggested that haemocoelomic fluid may diffuse from the viscera to the intestine of gastropods and may thus appear in faecal deposits. Haemolymph, faecal matter, and fluids from the excretory system have been subjected to very few experimental procedures to determine their possible attractant qualities for miracidia (or cercariae). Studies by Shiff and Kriel ( 1970 ) and by Chemin (1970), however, have demonstrated that some unidentified thermostable, water-soluble substance or substances produced by snail metabolism are especially attractive to certain species of miracidia. The source and mode of action of such substances remain to be determined. Studies on Fasciola miracidia by Wilson and Denison (1970b) suggest that short chain fatty acids may be involved as part of the mechanism of host-location by miracidia. The significance of haemolymph as an attractant is suggested in a recent report by Muftic ( 1969 ), who showed that it serves as a culture medium for development of post-miracidial larval stages. Combined molluscan secretions may have an even greater significance than has been realized in modifying responses of miracidia and subsequent developmental stages. Intramolluscan stages Localization of sporocysts and rediae of digenetic flukes may vary considerably. Mother sporocysts generally do not occur in nutritionally rich areas such as hepatopancreas and ovotestis of molluscan hosts, sites commonly employed by daughter sporocysts and rediae. Tissues of the mantle, tentacles, head, and foot are commonly

130

invaded by mother sporocysts or rediae, but less frequently utilized areas have also been reported for digeneans. Thus, certain echinostome and philophthalmid rediae may move to the cavity or wall of the ventricle of the heart (Alicata 1962; Heyneman 1966; Lie and Umathevy 1965a) or develop in the buccal mass (Lie and Umathevy 1965b), salivary glands (F. G. Rees 1934), haemolymph and kidney (W. J. Rees 1936), rectum (Dobrovolny 1939), and male reproductive system (F. G. Rees 1934). Additional developmental sites include the albumen gland, hermaphroditic duct, uterus, and vagina (Cheng and Cooperman 1964), the blood vascular system of gastropods (Uzmann 1953), and gills of pelecypod molluscs (Goodchild 1948). Reasons for such organ or tissue specificity are still unknown, but the regularity of its occurrence suggests that factors other than random behaviour are operative. The unconfirmed report by Muftic (1969) on in vitro culture of post-miracidial stages of S. mansoni in molluscan haemolymph is particularly germane. A crystalline substance isolated from snail tissue, whose chemical and biological properties were similar to those of ecdysone, was added to the haemolymph and provided a medium for successful development of cercariae. Muftic concluded that miracidia probably respond to the presence of morphogenetic substances produced by the ovotestis and that these are probably steroids. His study should generate considerable interest and strongly suggests that the predilection shown by certain trematode larvae for hepatopancreas and ovotestis is based on a response to such substances. Migration of larvae within the molluscan host appears to be principally via the blood system (Agersborg 1924; Alicata 1962; Probert and Erasmus 1965 ). Some larval stages are aided in their journey within the host by active movement of the parent generation as in Postharmostomum, where a combination of swaying, undulating movements of the long branches of the mother sporocyst, together with muscular constrictions of the sporocyst walls, results in forc-

MARTIN J. ULMER

ible expulsion of small daughter sporocyst embryos (Ulmer 1951b). Larvae usually emerge from molluscan hosts as cercariae, although examples of normal release of rediae and sporocysts are well known. Release of daughter sporocysts via the pneumostome of land molluscs is characteristic of several species of dicrocoeliid flukes (Timon-David 1960; Patten 1952). Such sporocysts are believed to migrate actively to the respiratory chamber of the snail, although actual release is passive. According to Dobrovolny ( 1939), cercariae of Plagioporus sinitsini encyst within sporocysts localized in the rectum of the snail. Sporocysts emerge from the snail only when metacercariae are fully developed (Fig. 5). Fish, the definitive hosts, are attracted to the bright red or yellow sporocysts and ingest them. Palombi (1942) reported on the highly unusual sporocyst of Ptychogonimus which leaves its scaphopod host (Dentalium) and actively migrates on the ocean floor where it is later ingested by crustacean second intermediate hosts. The emergence of cercariae may be affected by various environmental factors including temperature, light, humidity, pH of water, carbon dioxide concentration, oxygen depletion, and such emergence may be either active or passive (Kendall 1965). But activities of the host and of the parasite itself may be involved as well. Thus, Kendall and McCullough (1951) considered the unusual activity of the mantle of snails infected with Fasciola hepática cercariae to result in increased pressure on the terminal part of the perivisceral haemocoel with resultant passive but forceful extrusion of large numbers of cercariae. Passive emergence of cercariae via the intestinal canal of molluscs with the appearance of larvae in faeces was reported by Campbell and Todd (1956) in Fascioloides magna. Another unusual method of escape noted by Brumpt (1941) and confirmed by Duke (1952) as well as by Etges and Gresso (1965) involves the use of female genital ducts resulting in the release of cercariae enclosed within egg clutches of the host.

SITE-FINDING BEHAVIOUR IN HELMINTHS

Active emergence of cercariae and their subsequent escape was demonstrated by Pearson (1956, 1959, 1961), who observed that strigeoid cercariae of three genera actively migrate via blood vessels to the edge of the mantle and then rupture the vessel and adjacent integument. In so doing, they produce a temporary "escape pore" at a fixed point in the tissue walls. Schistosome cercariae may escape through intact integument by use of cercarial escape glands as shown by Duke (1952). Richards (1961) reported active emergence of nine species of cercariae through vessels of the mantle region. Probert and Erasmus ( 1965 ), in a detailed study of cercarial migration within a gastropod, also reported active boring of larvae through the body wall, particularly at the inner surface of the mantle, but concluded it was unlikely that directional mechanisms on the part of cercariae were involved in such migrations. Emergence of cercariae may also occur through the pneumostome, as shown by Najarían (1954) for cercariae of Echinoparyphium, by Robinson (1949) and Ulmer ( 1951b) for cercariae of Postharmostomum, and by Carney (1967) for slime-balls containing cercariae of Brachylecithum. Cercariae Varying responses of free-living cercariae to environmental conditions such as temperature, light and shadow, gravity, touch, water turbulence and currents have been reviewed by Smyth ( 1966). Such responses vary greatly and are of importance in environment-finding as are their sometimes bizarre swimming activities and peculiar morphological modifications. Whether or not there exists a predilection of cercariae for their hosts, or if chemoattraction occurs, is not agreed upon by investigators. Faust and Meleney (1924), who reported the occurrence of chemical attraction in miracidiummollusc relationships, could find no evidence for it in their study of contact behaviour of schistosome cercariae and mammalian skin. Numerous other investigators who reached similar conclu-

131

sions are noted by Wesenberg-Lund (1934). McCoy (1935) and Smyth (1966) stated that no evidence is at hand that chemotaxis is involved in host-location by cercariae. Cercariae of Clonorchis sinensis "never actively seek a fish host" according to Komiya (1966) in a review article dealing with that species, and are stimulated to penetrate as a result of water disturbances caused by a fish. McCoy (1935) concluded that cercariae do not exhibit chemotactic responses but their phototactic responses have "practical significance in their life history." More recently, however, increasing evidence points to the likelihood of chemoattraction. The normally quiescent cercariae of Gorgodera amplicava, according to Cheng (1963 ), are stimulated by molluscan "sera" following ingestion of cercariae by their second intermediate hosts, thus enabling cercariae to escape from their enclosing tails. Cercariae then migrate to appropriate areas and encyst. The source of the stimulating factor is said by Cheng to be the haemocoelomic fluid which seeps into the alimentary tract of the snail, but no experimental evidence has appeared to give support to this supposition. Chemoattraction of cercariae of Trichobilharzia was reported by Neuhaus (1952), who demonstrated cercarial stimulation by placing various substances on the tip of a needle near papillae at the anterior end of the cercarial body, and concluded that they serve as sensory receptors. Studies by R. A. Campbell currently in progress at our laboratory indicate that cercariae of Cotylurus flabellijormis are able to locate molluscan hosts within a few seconds, apparently in response to some substance diffusing from the snail (Campbell, unpublished). The nature and functions of secretions from the glandular complex of schistosome cercariae have been studied intensively and reviewed by Stirewalt ( 1963,1966). In addition to their lytic, lubricative, adhesive, and protective activities, they also appear to function in localization of suitable sites for cercarial penetration (Stirewalt and Kruidenier 1961; Stirewalt 1959). Stirewalt ( 1966), in reviewing the literature of

132

schistosome penetration, reported that schistosomules inside the skin of a host may first wander at random through the skin and subcutaneous tissues "along lines of least resistance" rather than in response to specific stimuli. But in the deeper tissues, they move into sebaceous glands, an activity apparently typical of several species of mammalian and avian schistosomes, and Stirewalt (1959) demonstrated that this was in direct response to a certain specific characteristic of these glands. Stirewalt and Uy (1969) have recently referred to the "cercarial awareness" of and posture response to selected conditions which thus serve as "penetration stimuli" for cercariae of S. mansoni. They suggest that diffusion of some stimulus from the skin of the definitive host initiates penetration behaviour. Detailed information on the role of sebaceous glands of the skin in effecting entry or in establishing migratory routes of schistosomes is not known. Stirewalt and Kruidenier ( 1961 ) concluded that the film of sebum covering the skin surface serves to initiate secretion of preacetabular glands used in penetration of definitive hosts. That glandular secretions of skin may play some role in entry or migration pathways of avian schistosomes was observed this past summer in our laboratory at Lake Okoboji, Iowa. Cercariae of the avian schistosome Gigantobilharzia huronensis swim to the surface after emerging from physid snails (including^/?lexahypnorum) and form clumps of numerous individuals situated in a thin film of apparently mucoid material. If a clean glass slide is placed in the water directly under them, no cercarial activity is discernible. But when a clean slide is rubbed over the surface of human skin (nose or forehead) and then placed under the water surface near them, the cercariae almost immediately begin to move very actively and continue to do so until the slide is removed. This suggests a definite chemotactic response. However, whether it is due to sebum (the product of sebaceous glands) or other lipids from the skin is not known. Wagner (1960) presented experimental evidence that cercariae of Schistosomatium douthitîi respond to various

MARTIN J. ULMER

soluble extracts present in the skin of mice and that cercarial invasion was very low when such materials were ether-extracted from the skin. He concluded that acids derived from the skin are responsible in part for cercarial stimulation. Clegg (1969) has published experimental data demonstrating that free sterols (particularly cholesterol) in the surface lipid of bird skin stimulates cercarial penetration by schistosomes of the genus Austrobilharzia. How this stimulating effect is produced remains unknown. In Clegg's experiments cercariae were not attracted by cholesterol, but required contact with it before penetration was initiated. Nor does cholesterol cause secretion of cercarial penetration glands. The intriguing possibility that cholesterol triggers a hormonal response within the cercarial body was suggested by Clegg in view of the recent findings by Thorson et al. (1968) ofhormonal-like lipids in Fasciola miracidia, protoscolices of Echinococcus, and Trichinella juveniles.

FIGURES 4-9. Examples of migratory routes and developmental sites of selected trematodes in their intermediate hosts. 4 Leucochloridium sporocyst (broodsac) in tentacle of amphibious snail. 5 Sporocysts of Plagioporus sinitsini containing encysted metacercariae in rectum of snail (Goniobasis) (after Dobrovolny 1939). 6 Metacercariae of Echinoparyphium flexum in kidney of Lymnaea palustris (after Najarían 1954). 7 Anguispira altérnala, showing migratory route of Postharmostomum helicis cercariae to pericardial chamber (Ulmer 195la). 8 Cross-section through Anguispira alternata, showing metacercariae of Postharmostomum helicis in pericardial chamber, kidney, and renopericardial canal (Ulmer 1951a). 9 Sagittal section through kidney region of Mesodon thyroidus, showing metacercariae of Brachylaima virginianum within the renal chamber (Ulmer 1952).

SITE-FINDING BEHAVIOUR IN HELMINTHS

133

134

MARTIN J. ULMER

Metacercariae Unusual localization of many species of metacercariae reflects a high degree of site-finding behaviour and organ specificity (Figs. 6-15). Only a limited number of instances are discussed below to indicate the variety of situations known to exist. Certain strigeoid cercariae migrate to the cranial region of fishes where they become diplostomula-type metacercariae. Hoffman and Hoyme (1958) studied migration of Diplostomum baeri cercariae in stickleback and found all metacercariae to localize in the brain region, particularly in the choroid plexus and optic lobes, apparently reaching these sites via the bloodstream. Ferguson (1943 ) demonstrated that cercariae of Diplostomum flexicaudum reach the lens of fish (rainbow trout, blackhead minnow) in a remarkably short time (Fig. 13) and that if eyes are removed surgically prior to exposure of the host to cercariae, the parasites do not localize in the orbital region. Although the eye (intact or lensless) was believed to provide a cercariaattracting stimulus, Ferguson could not demonstrate that the lens itself contained factors influencing cercarial migrations. Erasmus (1958, 1959 ) showed that cercariae of a related species ("cercaría x") migrate in a similar manner in stickleback and also invade the lens. No specific chemotactic influence of the eye was shown, but Erasmus concluded that chance migration alone could scarcely explain the phenomenon and that migration must be directive. Szidat (1969) stressed the probable significance of hormonal influences of the host on localization of such metacercariae in the eyes. Szidat (1966, 1969) has introduced an additional factor that he considers of possible significance in localization of metacercariae. This relates to his hypothesis of the supposed presence of "memory" in helminths. Apparently influenced by the conclusions of Best (1963) that memory probably exists in lower invertebrates (particularly planarians), Szidat (1969) sug-

FIGURES 10-15. Examples of migratory routes and developmental sites of selected trematodes in their intermediate hosts (continued). 10 (a) Adult dragonfly with encysted metacercariae of Haematoloechus in vestige of nymphal branchial basket (after Krull 1930). (b) Encysted metacercariae in gill lamellae of naiad, (c) Dragonfly naiad showing location of metacercariae in gill lamellae of branchial basket. 11 (a)Rana pipiens, showing location of Fibricola crátera diplostomula in hind limb musculature, (b) Diplostomula of F. crátera in gastrocnemius muscle of frog. 12 (a) Perch, showing location of Psilostomum ondatrae metacercariae in lateral line canal, (b) Metacercariae of P. ondatrae in lateral line canal of fish scale (after Beaver 1939). 13 (a) Eye of fish, showing diplostomula of Diplostomum flexicaudum in lens, (b) Minnow, showing location of D. flexicaudum metacercariae (diplostomula) . (c) Section of fish eye, showing diplostomula in cortex of lens. 14 Cross-section of vertebral column and spinal cord of Rana pipiéns, showing diplostomula. 15 (a) Position of ant (Formica) infected with metacercariae of Dicrocoelium, on blade of grass (after Anokhin 1966). (b) Diagram of ant head, showing position of suboesophageal ganglion, (c) Metacercaria of Dicrocoelium in suboesophageal ganglion of ant (after Anokhin 1966).

SITE-FINDING BEHAVIOUR IN HELMINTHS

135

136

gested that strigeoid metacercariae in fish eyes "are guided to their selected infection site by an 'engram,' or a specific memory, developed and fixed in the course of many generations." Thus, metacercariae within the nervous system "are able to obtain a plan (engram) of the total properties of the environment. Where successful localization occurs, metacercariae retain the engram and it becomes fixed." This interesting avenue of speculation as a proposed explanation of localization based in part on "memory" appears somewhat premature in my estimation, particularly in view of our very limited knowledge on behaviour patterns and sensory mechanisms of parasitic helminths. Metacercariae of Fibricola crátera (Fig. 11) develop within the coelomic cavities of tadpoles, but upon their metamorphosis move into the hind-limb musculature of frogs and encapsulate (Cuckler 1940; Hoffman 1955; Turner 1957, 1958). Similar migratory behaviour was reported for metacercariae of Neodiplostomum intermedium by Pearson (1961), for mesocercariae of Strigea elegans by Pearson (1959), and for two species oí Alaria by Pearson (1956). Such migratory movements of helminths at the time of metamorphosis of the host strongly suggests the influence of host endocrines. Cercariae of several echinostome species appear to have a predisposition for renal organs of gastropod molluscs (Fig. 6) and migrate and encyst there to the exclusion of almost all other tissues (Heyneman 1966). Najarían (1954), in reporting on the life cycle of Echinoparyphium flexum, could not explain this, nor was he able to cause encystation of cercariae when they were placed in water containing macerated kidney tissue. Metacercariae of Psilostomum ondatrae, according to Beaver ( 1939), encyst principally in the lateral line canal of perch (Fig. 12). Cercariae of numerous species of the family Brachylaimatidae enter land snails where they become metacercariae. In Postharmostomum helicis, metacercariae develop only within the pericardial chamber of the land snail Anguispira alternata, but require for this a snail other than

MARTIN J. ULMER

one harbouring sporocysts and cercariae (Fig. 8 ). Hence, a migration of the parasite from one snail to another is mandatory. Using the respiratory orifice (pneumostome) as a point of entry, cercariae pass through the ureter and move to the kidney, thence through the renopericardial canal to ultimately reach the pericardial chamber where metacercariae develop and may live for more than a year (Ulmer 195la, b). Cercariae of a related species, Brachylaima virginianum, follow a similar pathway (Fig. 9) in another terrestrial gastropod (Mesodon thyroidus), but develop only in the renal organ (Ulmer 1952). Cercariae of both species, as shown by Ulmer (1951b) andbyKrull (1935) respectively, possess numerous papillae (presumably sensory) which may function in site-location, although no confirmatory experimental evidence can be cited. Elwell (1967) has found that free-living cercariae of P. helicis respond to substances diffusing over a distance of only a few millimetres or less from their snail host. Experiments on the encystment of cercariae of the amphistome Megalodiscus temperatus on frog skin were conducted by Krull and Price (1932), who found that metacercariae were limited to dark pigmented spots and that encystment appeared to be in response to a complex of chemotactic-phototactic stimuli. Cercariae of the frog lung fluke Haematoloechus medioplexus encyst in the unique respiratory structure (branchial basket) of dragonfly naiads (Krull 1931) and are never found in other regions of the host (Fig. 10). Metacercariae remain in this limited area even after metamorphosis of the dragonfly to the adult stage, at which time the parasites become restricted to a small vestige of the branchial organ in the posterior end of the digestive tract. Bizarre changes in host behaviour may sometimes occur as a result of the presence of larval flukes in intermediate hosts. Such alterations of behaviour may increase the likelihood that worms will reach appropriate definitive hosts. The movement of amphibious snails parasitized by the peculiar broodsacs of Leucochloridium

SITE-FINDING BEHAVIOUR IN HELMINTHS

and Neoleucochloridium provide striking evidence of this (Fig. 4). The highly modified sporocysts (broodsacs) of these genera develop within the snail and some make their way to tentacles (Wesenberg-Lund 1931; Kagan 1952; Halik 1931 ; Hecker and Thomas 1965 ). Here they pulsate in response to light, their rhythmic movement possibly continuing throughout the day. At dusk and at night, broodsacs are withdrawn from the tentacles into the body of the snail. Actively pulsating broodsacs resemble insect larvae and hence may attract birds to ingest infected snails or broodsacs that have erupted spontaneously from tentacles of the host. Expelled broodsacs may continue to pulsate even after complete separation from the sporocyst producing them (Kagan 1952). Unlike uninfected snails, infected individuals are characteristically found on marshy vegetation, their antennae exposed to light, thus making the parasite very conspicuous. To my knowledge, no one has yet provided an explanation for the unusual pulsation of broodsacs. Neither Monnig (1922) nor Kagan (1952) could demonstrate evidence of innervation in broodsacs and Kagan suggested that their movement may be myogenic. Another example of aberrations in host behaviour resulting from site-finding by helminths is that OÍDicrocoelium dendriticum. Recent investigations (Hohorst and Graefe 1961 ; Anokhin 1966) on certain aspects of its life cycle indicate that in second intermediate hosts (ants) a single cercaria (occasionally two) migrates from the abdominal cavity to the suboesophageal ganglion where it develops to the metacercarial stage ("Hirnwurm" or "brainworm" of Hohorst and Graefe ( 1961 ) ). This results in a characteristic type of behaviour of infected ants, for they become torpid at the tips of grass, thus increasing the opportunity of their being ingested by definitive hosts (Fig. 15). Carney (1967, 1969) has shown that in another dicrocoeliid (Brachylecithum) a single metacercaria localizes in the supraoesophageal ganglion of an ant host, resulting in the latter's modified behaviour and a pronounced obesity. The sluggish circling

137

behaviour of infected ants, together with their greatly increased size, makes them more conspicuous to their definitive bird hosts. It is of interest that in all instances cited above metacercariae localize in the sites indicated to the almost complete exclusion of other areas of the host. To ascribe such behavioural patterns to chance appears unwarranted. Adults Helminthologists have long been aware that almost every species of adult trematode appears to have a preferred site in its host, but factors responsible for this localization remain speculative or wholly unknown. The examples cited below are only a few of many that might be mentioned. The peculiar restriction of adult Dendritobilharzia pulverulenta to arterial rather than venous vessels (Ulmer and VandeVusse 1970) and their almost exclusive localization in the lower dorsal aorta (VandeVusse, unpublished) contrast strikingly with the preference of other schistosomes for veins. Within a single genus, considerable variation in site-selection may occur. Thus, adults of Alaria canis may be scattered throughout the duodenum, but adults of A. arisaemoides are generally in a single clump in the jejunal wall (Pearson 1956). Certain flukes appear to migrate along very definite pathways in order to become established. Maldonado (1945) showed that Tamerlania, a eucotylid fluke, is able to reach the kidney of its pigeon host within five hours and that in its migration it consistently passes through the ureters. Adults of the gorgoderid fluke, Phyllodistomum staff ordi, mature in the urinary bladder of bullheads, but require a sojourn in the ureter before migrating to the bladder (Waffle 1968). The lengthy migration of Alaria adults, involving passage through the stomach wall, diaphragm, lungs, trachea, and oesophagus to the duodenum is unidirectional, according to Savinov (1953). Some flukes, once established in their host,

138

characteristically undertake a second migration later in the life cycle. Adults of Heronimus chelydrae live within lung tissue of turtles, but occasionally are found in the larger bronchi. When gravid, they migrate through the larger bronchi to the pharyngeal and mouth regions, in order to release miracidia (Ulmer and Sommer 1957; Crandall 1960). That migration may be associated with host metamorphosis has been shown in the frog amphistome, Megalodiscus îemperatus. In young tadpoles and mature frogs, this fluke is almost always in the colon, even in heavy infections (Herber 1939). Upon metamorphosis of the tadpole, young worms migrate to the stomach, apparently in response to the shortening of the intestine. With concomitant changes in diet following metamorphosis, worms migrate back to the colon. Herber considered this initial migration to the stomach as a response to a change in amount and type of food available. The later movement to the colon he attributed to the high acid content of the frog stomach. Although Krull and Price (1932) had suggested that crowding causes such movements, Herber demonstrated that even in heavily infected young tadpoles no movement of flukes occurred until metamorphosis. Efford andTsumura (1969) agreed with Herber but could provide no reason for a similar migration in a related species of Megalodiscus. Increasing evidence suggests that adult flukes move in response to chemical stimuli (McCoy 1935). Faust and Khaw (1927) indicated that bile serves as a chemical stimulus attracting young Clonorchis sinensis to and through the bile ducts and thence to the liver. Yoshida (1931) stated that adults of this species respond chemotactically to pancreatic secretions as well. Wykoff and Lepes (1957 ), however, showed that in rabbits whose bile ducts had been ligatured, young C. sinensis were nonetheless able to migrate to the liver, probably by intestinal penetration and transfer to the liver via the portal system. Dawes and Hughes (1964) attribute the wandering of young Fasciola adults to a lowered nutritional state and subsequent food-searching that results

MARTIN J. ULMER

in chance migration to the liver, such movement being "neither predestined nor inevitable." In a review article on the pathogenesis of Fasciola, Sinclair (1967) observed that, in all studies relating to migratory routes of the adult and the means whereby young flukes find the liver, the more generally accepted explanation appears to be some type of chemotaxis. West ( 1961 ) observed that adult Philophthalmus gralli, which mature in the orbit of the eye of birds near the exit of the lachrymal duct, reach that site with great rapidity and suggested that secretions of the lachrymal gland "may elicit a chemotactic response on the part of the parasite." In a recent review of Paragonimus, Yokogawa ( 1965 ) indicated that migration of adults differs in accordance with the type of definitive host and with the species of Paragonimus, but suggested no reasons to explain such behavioural variations. The experimental studies on Paragonimus kellicotti by Sogandares-Bernal (1966), however, provide convincing evidence that adults respond to attractants of some sort. In his experiments, a single metacercaria was fed to a domestic cat and approximately eight weeks later a second metacercaria was ingested. In four of six such experiments, worms were able to find one another and to encyst in the lungs. Among unisexual digeneans, the possibility of sex attractants has been mentioned as a factor enabling adult flukes to locate one another, even if very few worms are present. Particularly as a result of recent attention directed to the biologically active pheromones (originally termed ectohormones), considerable interest is now being directed to the possible effects of such substances on site-finding and reproductive behaviour of helminths. As suggested by Karlson and Liischer (1959), minute amounts of pheromones secreted to the outside by one individual may cause a second individual of the same species to modify either its behaviour or development. Attempts were made by Armstrong (1965 ) to determine if pheromones are operative in sex-attraction and in initiation and maintenance of sexual maturity of mated female schistosomes. His results indi-

SITE-FINDING BEHAVIOUR IN HELMINTHS

cated that in three genera of schistosomes (Schistosoma, Heterobilharzia, and Schistosomatiuni), pairing of worms resulted principally from trial and error and from thigmotaxis rather than from chemotaxis. However, the presence of a pheromone produced by male worms was suggested by Armstrong, who believed it might stimulate sexual maturation of females. CESTODES The restricted number of sites occupied by larval and adult cestodes in intermediate and definitive hosts is in marked contrast to the diversity of sites occupied by digenetic trematodes. The paucity of data on site-finding among cestodes is related in part to the virtual absence of free-living stages in their life cycles. No evidence of hostattraction appears in the literature relative to oncospheres, and in those groups where free-living coracidia occur, ingestion by suitable crustacean intermediates appears entirely fortuitous (Mueller 1959). Following ingestion of eggs or coracidia by a suitable host, oncospheres are released, presumably by action of the host's digestive enzymes. In pseudophyllideans, movement to the haemocoel of the invertebrate host is aided by action of oncospheral hooks; in cyclophyllideans, movement through the gut wall of arthropod or vertebrate intermediaries is facilitated by the activity of oncospheral (epidermal) glands (Sil verm an and Maneely 1955). Wardle and McLeod (1952) stated that localization of cestode larvae may result from mechanical factors such as size of blood vessels, as in Hydatigera taeniaejormis, whose oncospheres are probably filtered out in the portal circulation of the liver, so that distribution of strobilocerci is limited to that organ. Brumpt (1936) suggested that a definite oncospheral tropic response accounts for the distribution of Taenia solium cysticerci in muscles of swine, whereas they appear predominantly in the nervous system of man. Slais (1967) quotes a study by Dixon and Lipscomb ( 1961 ) in which the authors confirm a

139

predilective migration of T. solium cysticerci to the human central nervous system. Reports of several investigators indicate that metacestodes demonstrate choice in their localization within hosts. Migration to a very specific site is shown by procercoids of Archigetes iowensis, which migrate to the posterior seminal vesicles of tubificid annelids (Calentine 1964). In the related caryophyllaeid genus Biacetabulum, procercoids show a preference for the same organ, but as their size increases, they generally break through its wall and enter the coelom (Calentine 1965). Kennedy (1965) in his study of A. limnodrili also reported that procercoids move to the anterior end of the coelomic cavity of tubificids and ultimately localize in the "testes sac" (seminal vesicle?). Larvae of those cyclophyllideans and pseudophyllideans utilizing invertebrate intermediates usually localize in the haemocoel, but an exception is shown by Hymenolepis compressa, whose cysticercoids occur in the stomach of a lymnaeid snail (Supperer 1959). In some instances, preference of larvae of a cestode species for a particular site varies with the hosts utilized. Plerocercoids of Triaenophorus, according to Newton ( 1932), are almost exclusively limited to epaxial musculature between the skull and dorsal fin of ciscoes (Leucichthyes), but in burbots (Lota) they are restricted almost entirely to the liver (Miller 1945). McCaig and Hopkins (1963) showed that Schistocephalus plerocercoids localize in specific areas of the intestine, depending upon the species of mammalian or avian hosts parasitized. Encysted plerocercoids of the proteocephalan tapeworm Ophiotaenia filaroides, according to Wood (1965 ), have the ability to migrate from cysts attached to stomach and intestinal walls of the tiger salamander and make their way to the intestine. Wood indicated that the apical (end) organ of the scolex of this species may function as an exocrine gland whose secretion lyses host tissue. Further evidence of the secretory function of the apical gland was presented by Fischer and

140

Freeman (1969), who described stages in the penetration of Proteocephalus ambloplitis plerocercoids. They followed the migration of plerocercoids through the intestinal wall and into the intestinal lumen of the fish host, and noted a subsequent decrease in size of the apical organ. The exact stimulus for such migration is unknown, but apparently it is related to a spring rise in temperature. Plerocercoids migrated in all bass which attained a certain size. The authors further suggest that the hormonal condition of the host may enhance the effect of temperature in the migratory behaviour of this species. Physicochemical and morphological features of the alimentary tract influence greatly the establishment of adult cestodes at particular sites; these have been reviewed extensively by Read (1950, 1955, 1963), Rogers (1962), Smyth (1963, 1969),Voge ( 1967),SchwabeandKilejian (1968), and Hopkins (1970). It is generally agreed that initial establishment of adults in the intestine may depend upon responses to such stimuli as bile (reviewed by Smyth and Haslewood 1963), hydrogen-ion concentration, digestive enzymes, etc., all of which may serve as trigger mechanisms in localization. Localization of three genera of tetraphyllidean tapeworms in the spiral valve of Raja was discussed by Williams ( 1968). A sequential distribution occurs as one proceeds posteriad, each species apparently situated within its preferred area within this organ. Preference of various caryophyllidean tapeworms for particular regions of fish intestines was noted by Lawrence ( 1969), who reported the association of pairs of worms of different genera occupying the same niche in white suckers. The frequency of occurrence of such groupings of worms in specific locations within the intestines was so great that it could not be attributed to chance alone. Recent experimental studies have shed additional light on localization and migratory phenomena of adult tapeworms. That migration of adults occurs following their initial establishment in the intestines has been well documented for a

MARTIN J. ULMER

variety of cestode species. As early as 1939, Chandler showed that Hymenolepis diminuta adults usually attach well behind the stomach, but that most of them move forward 7 to 10 days after infection. Recent studies by Holmes (1961, 1962) and by Brâten and Hopkins (1969) confirm this. Brâten and Hopkins provide convincing evidence that worms recognize specific regions of the host's intestine. Goodchild ( 1958), in a series of surgical implantation studies, demonstrated that Hymenolepis diminuta when introduced into the small intestine of recipient rats migrated from implantation sites into regions of "optimal character." Immature Proteocephalus filicollis occur in the rectum of stickleback, but larger mature and gravid individuals migrate anteriad and attach near the pyloric valve (Hopkins 1959). Adult Diphyllobothrium sp., according to Archer and Hopkins ( 1958 ), first become established in the posterior part of the small intestine, but after two days begin to migrate toward the stomach. Although no reason could be given for such migration, the authors believed it to be due to an extrinsic stimulus rather than the result of an innate behaviour pattern. Schistocephalus adults are capable of developing in a variety of avian and mammalian hosts (McCaig and Hopkins 1963 ). In chickens, they migrate anteriad after their initial attachment, but such migration is not typical in other hosts. Hymenolepis microstoma was shown by Dvorak, Jones, and Kuhlman (1961) and by DeRycke ( 1966) to move upward from the intestine and to localize in the bile ducts. DeRycke assumed that some growth-stimulating substance in bile caused such migration. In a later study involving in vitro culture from the cysticercoid to the young adult, DeRycke and Berntzen ( 1967 ) provided experimental evidence for the importance of bile in the growth and development of this species. Movement of adult worms posteriorly is also known to occur among cestodes. Foster and Daugherty (1959) showed experimentally that distribution of the chicken tapeworm Raillietina varied considerably in accordance with age of

SITE-FINDING BEHAVIOUR IN HELMINTHS

infection. Cestodes first localized in the duodenum anterior to the opening of the bile duct, but later migrated posteriad to the jejunal area, a region of high enzyme activity. Hopkins ( 1969 ), Read and Kilejian ( 1969 ), and Chappell, Arai, Dike, and Read (1970) have provided experimental evidence demonstrating the existence of circadian migratory behaviour of H. diminuta adults in the small intestine of rats, such migrations occurring in accordance with feeding patterns of the host, but no specific stimulus for migration was determined. An attempt to explain migration of adult cestodes in the definitive host was made by Holmes (1961,1962), who believed thatHymenolepis selected optimum sites along the length of the intestine. How such selectivity occurred was not stated, however. Crompton and Whitfield (1968) used published data of previous workers on Hymenolepis and concluded that migration occurred so that worms might "maintain their absorptive areas in the most favorable region for their development." Although this suggests possible benefits for the parasite, it does not explain the mechanisms underlying such migration. More meaningful reasons were presented by Brâten and Hopkins ( 1969 ), whose studies on H. diminuta indicated that worms can overcome peristaltic activity of the host's intestine, and that they possess sensory abilities enabling them to recognize direction as well as normal and abnormal locations in the host's intestine. Their convincing evidence of the probable existence of complex sensory systems for location-specificity points to a need for increased study in the neglected aspect of site-finding by cestodes. NEMATODES

Parasitic nematodes have become extraordinarily well adapted to diverse habitats and a considerable body of information has accumulated relative to site-finding. Their extremely varied life cycles include those with no free-living stages, others with free-living juvenile phases,

141

and those with either one or two intermediate hosts. Both external and internal environmental factors may influence site-finding by these helminths. Chemical attraction and oriented migration of phytoparasitic nematodes to root diffusâtes have been demonstrated by numerous investigators and were reviewed by Jones (1959, 1960), Blake (1962),Klinger ( 1965),Dropkin(1965), and Webster ( 1969). Little positive evidence of specificity of the attractants, however, appears to have been shown. Stimulation of moulting of preadults by such diffusâtes was shown by Rhoades and Linford ( 1959 ). Selection of appropriate hosts and of preferred sites by juveniles depends upon chemoreception, according to Steiner (1925), and Poinar and Leutenegger (1968) showed that amphids and tactile-sensitive papillae play an important role in site-finding. Specific substances causing orientation of phytoparasitic nematodes have been investigated by Thorpe^ al (1947), and Bird (1959), and include carbon dioxide, certain amino acids, and dissolved sugars at very low concentrations. Literature on site-finding by zooparasitic nematodes is scattered. It is well known that adults may localize in surprisingly unusual areas. Those invading the digestive tract are by far the most common and may be scattered widely throughout its length, or may be limited to very restricted sites, such as the proventriculi of avian hosts, where the highly modified females of Tetrameres and Microtetrameres occur. Many other unusual sites occupied by nematodes might be cited. Skrjabingylus parasitizes the frontal sinuses of mustelids; Angiostrongylus, according to Sodeman, Sodeman, and Richards ( 1969), is limited to the pulmonary arteries of rats, and Spirocerca lupi invades the walls of various arteries of dogs. Among filarial worms, each species appears to have its own preferred habitat. Litomosoides, the cotton rat filarial worm, occurs in the pleural cavities (Bertram 1966) ; Splendidofilaria quiscali adults are found in the lateral ventricles of the grackle (Odetoyinbo 1960) ; Onchocerca armillata occurs in the

142

aortae of cattle. Microfilariae also show a definite organotropism and each species within its insect vector has a specific site for development (thoracic muscles, Malpighian tubules, or fat bodies). Nematodes employ various methods in establishing contact with their hosts, particularly in species where free-living juvenile stages exist. Wallace (1961) and Rogers and Sommerville (1963 ) reviewed data showing how movement of juveniles after hatching could significantly influence their dispersal to new hosts. Infective strongyloid juveniles (Haemonchus, Trichostrongylus, etc.) migrate from faeces to vegetation and are thus available for ingestion by their definitive hosts. Crofton (1954) in studying strongyloid juveniles concluded that migration from host faeces to grass is a result of random movements unrelated to geotropisms or sensory reception. Juveniles of Necator increase their opportunities for host contact if they remain in the soil where they respond to carbon dioxide concentrations by extending themselves and moving from side to side (Rogers and Sommerville 1963). Parker and Haley (1960) provided experimental evidence that movement of filariform juveniles of Nipposirongylus is greatly influenced by even minute temperature differences, and that these may be of vital significance in enabling the nematodes to make contact with their hosts. Dispersal of nematodes as a result of their association with organisms other than their hosts has been demonstrated rather frequently. Perhaps the most unusual is the sequence of events enhancing the likelihood of transfer of bovine lungworm juveniles (Dictyocaulus) to definitive hosts as a result of the association of juveniles with the dung-inhabiting fungus Pilobolus. Robinson (1962) has shown that juvenile nematodes migrate to sporangiophores of the fungus and, when these ripen, spores are discharged violently, sending the airborne juveniles for a distance of several yards. Reasons for the establishment of nematodes at specific sites within the host are not known,

MARTIN J. ULMER

but evidently site-finding involves a multiplicity of factors relating both to the host as well as to the parasite. In general, host factors provide a broad array of stimuli to which nematodes are somehow able to respond. Some of these factors are indicated in the selected examples cited below. Among host factors, age and sex influence the distribution of adult Trichinella spiralis in mice, according to Larsh and Hendricks (1949 ), for in experimental infections most adult worms occur in the posterior half of the intestines of younger hosts, but significantly more develop in the anterior half of the intestines of older mice. These authors showed that intestinal emptying time was probably responsible for this varying distribution of adults in the intestine. Larsh, Gilchrist, and Greenburg ( 1952) showed that host immune responses also affect distribution of Trichinella adults, for in immunized mice worms remained in the anterior half of the intestine for a much shorter time than in controls. Migratory behaviour of Toxocara canis juveniles in dogs is related to host age, as shown by Webster (1958), who found that in hosts under 3 months old, trachéal migration was common, but that migration via the heart and circulatory system is typical in those hosts older than 6 months. Schad ( 1963 ) studied niche diversification and localization of congeneric species of oxyuroid nematodes of turtles and concluded that age of the host, as well as sex, was important in the distribution of adults in the intestine. Additional evidence that sex is significant in localization of nematodes has been shown by several studies. During pregnancy and lactation, migratory behaviour and development of Toxocara canis in the dog is greatly modified, for fewer juveniles are found in tissue phases at such times (Oshima 1961 ). Prenatal infection of dogs by T. canis has been recorded frequently, but the stimulatory mechanism initiating juvenile migration to the foetus during pregnancy is unknown (Soulsby 1965 ). Effects of gonadectomy on migration of Nippostrongylus were studied by Solomon (1966), who found that fewer juveniles

SITE-FINDING BEHAVIOUR IN HELMINTHS

reached the lung following gonadectomy, indicating either that they were detained in the skin or that their rate of migration was markedly slowed. Solomon (1969) reviewed the effects of host hormones on parasitic infections. Tissue specificity of juveniles of filarial worms is well documented (some limited exclusively to skin, others found only in the bloodstream), but reasons for this are not entirely known. It is of interest, however, that such localization apparently favours the likelihood of transmission to suitable vectors. Nelson (1964) has summarized behavioural data concerning filarial nematodes in intermediate and definitive hosts. Fiilleborn's (1912) experimental studies suggested that chemotactic responses are involved in the movement of microfilariae from canine blood to fluid from Malpighian tubules. The migratory behaviour of some microfilariae may be related to the activities of their hosts, as in Loa loa. Feeding activities of Chrysops, its intermediate host, results in the rupture of the insect's hypopharyngeal membrane, thereby releasing microfilariae from the head of the fly for transmission to the definitive host (Lavoipierre 1958). Diurnal and nocturnal periodicity of a number of species has been studied intensively by Hawking, but despite more than a decade of detailed studies on mechanisms resulting in the circadian cycles of Wuchereria bancrofti microfilariae in the peripheral blood of definitive hosts Hawking (1964) concluded that a "fixative force" retains microfilariae in pulmonary capillaries, and that a "switch mechanism" occurs which "switches this force on by day and off by night." Both forces, however, are associated with the parasite and not the host. In reviewing his studies in 1967P however, Hawking concluded that accumulation of microfilariae in lungs is due to increased oxygen tension, to which microfilariae react by some type of reflex, but that rhythmic changes in 24hour cycles of the host dominate rhythms of microfilariae, according to the species involved. Virtually nothing is known of causal factors resulting in the sometimes elaborate migrations of nematodes within their hosts. One of the most

143

unusual migratory cycles involved in localization is that of Spirocerca lupi, whose juveniles develop in beetles. When infected beetles are swallowed by dogs, released juveniles migrate within the walls of the gastric, coeliac, abdominal, and thoracic arteries until they reach the heart region. Worms then migrate from the wall of the aorta to the wall of the oesophagus, finally burrowing into its lumen (Faust 1927; Hu and Hoeppli 1935, 1936, 1937). A remarkably similar migratory route has recently been described by Stockdale and Anderson (1970) for juveniles of Filaroides martis in mink. Many other examples of nematode migration could be cited, yet few investigators are inclined to go beyond a statement similar to that of Twohy (1956) who, in observing the mass migrations of Nippostrongylus from skin to lungs of mice, concluded that "certain conditions must attract or force larvae to undertake the migration." Infective stages of nematodes (eggs or juveniles) within the host are affected by a complex assemblage of stimuli provided by the host. These have been reviewed by Rogers (1960, 1966), by Rogers and Sommerville (1963), and by Thorson ( 1969 ), and include carbon dioxide, various salts, hydrogen-ion concentration, and temperature. Such stimuli serve as important physiological triggers in controlling such activities as hatching and moulting in ensuing stages of the life cycle. Tetley (1937) studied the distribution of intestinal nematodes of sheep and found the greatest numbers to occur "in a regular order" in the jejunum, the site of infection being determined by responses of juveniles to stimuli present in the intestinal contents. Such stimuli, more abundant in the duodenum, were believed to be associated with bile and/or pancreatic secretions. Increasing evidence suggests that both moulting and exsheathment are controlled in part by some intrinsic mechanism and that neurosecretory cells are involved (Davey 1964,1966). Indirectly or directly, site-finding very probably is also under the influence of such factors. Rogers ( 1957) indicated that factors causing exsheath-

144

ment in turn govern distribution of nematodes along the alimentary tract. Whitlock (1966) considered the sheaths of juveniles to be involved in sensory functions resulting in a high degree of host and organ specificity of trichostrongyloid and strongyloid nematodes. Chemically mediated sexual attraction, long known to exist in insects, has recently been documented among nematodes in several papers based on studies of mate attraction. Doerr and Menzi ( 1933 ) found that feeding rats only two or three trichinae juveniles was usually sufficient to establish an infection, with resultant development of gravid females and subsequent muscular invasion by juveniles. Beaver (1955) reported that Necator infections could be established from exposure of a host to only three juveniles. Beaver and Little (1964) reported that female ascarids are attracted to males and that their characteristic movement in attempting to seek out males could result in their becoming localized in the appendix. Roche ( 1966) reported that female Ancylostoma caninum produce a "messenger substance" which moves in the direction of flow within the intestinal lumen and attracts male worms. Bonner and Etges (1967) demonstrated by means of a linear test channel that adult male and female Trichinella spiralis respond to one another's presence, but the nature and source of the stimulus were not determined. These studies strongly suggest the existence of pheromones, noted earlier with reference to schistosomes, and indicate that their significance in site-finding has yet to be appraised. ACANTHOCEPHALA

The complete absence of any free-living stages in acanthocephalan life cycles limits discussion of site-finding behaviour to stages within their intermediate and definitive hosts. Unfortunately, data are very scanty and little experimental work has been done to provide an adequate explanation for site-finding behaviour by these helminths. Ingestion of embryonated acanthocephalan

MARTIN J. ULMER

eggs by suitable arthropod hosts appears to be entirely a matter of chance. Following hatching, the acanthor may wander about for some time before penetrating the intestinal epithelium in order to reach the host's haemocoel. Chemical factors in the host appear to influence triggering mechanisms for hatching. Such factors also influence movement of the acanthor through the intestinal wall, but the wandering of larvae prior to penetration of the intestinal epithelium is not understood (Bird and Wallace 1969). Where second intermediate hosts are requisite to completion of the life cycle, acanthocephalans may invade specific areas and encyst. In Neoechinorhynchus cylindratus, liver of the bluegill is the site of cystacanth accumulation (Ward 1940). Juveniles of Neoechinorhynchus emydis, according to Hopp ( 1954 ), encyst principally in the foot of viviparous snails (Campeloma). In the pyloric region of the stomach of definitive hosts, recently encysted juveniles become active. Their subsequent distributional patterns are dependent upon physicochemical conditions whose nature is apparently unknown, although bile and carbon dioxide have been shown to be important (Graff and Kitzman 1965; Awachie 1966). Generally, adult acanthocephalans are severely limited in their distribution within definitive hosts. Adults of Echinorhynchus truttae may occur throughout the entire length of the digestive tract (Awachie 1965), but in other species such as Macracanthorhynchus hirudinaceus, the jejunum is the preferred site (Kates 1944). Adults of Corynosoma constrictum are found in the lower third of the small intestine of teal and mallards (Keithly 1969) ; Paulisentis fractus localizes very close to the first flexure of the intestine below the stomach in creek chubs (Cable and Dill 1967 ) ; and Polymorphus minutus becomes established just beyond the yolk stalk of the intestine of ducks (Nicholas and Hynes 1958; Crompton and Harrison 1965). Other acanthocephalans demonstrate a decided preference for pyloric caeca, as shown by DeGiusti (1949) and by Venard and Warfel (1953) for Leptorhynchoides thecatus.

SITE-FINDING BEHAVIOUR IN HELMINTHS

Detachment and subsequent reattachment of adults farther posteriad has been reported for several species. Another type of migration has been reported for adults of Moniliformis moniliformis (— M. dubius)by Burlingame and Chandler ( 1941 ). A concentration of adults in experimental infections develops in a preferred zone ("zone of viability"), worms reaching the area by active migration. Females localize somewhat more anteriad than males, and as age of the infection increases, the population tends to move forward, localizing near the anterior end of the zone of viability. When a second infection is superimposed upon the first, competition for suitable attachment sites seems to occur. After a certain age, older primary worms may prevent younger secondary ones from becoming established in the zone of viability, so that younger worms attach more posteriorly and never intermingle with adults of the initial infection. Somewhat similar results from superimposed infections were reported by Nicholas and Hynes (1958) in Polymorphus minutas, for adults developed from a second infection were more widely scattered than usual, and fewer developed. Renewed interest in life cycles of acanthocephalans in recent years, and the successful in vitro hatching of Moniliformis eggs (Edmonds 1966), should stimulate additional studies on site-finding in this unusual group of helminths. SENSORY MECHANISMS IN SITE-FINDING BEHAVIOUR

Recent researches centring on sensory mechanisms in helminths suggest that site-finding behaviour may be related to nervous, sensory, and possibly neurosecretory structures. Trematodes within their hosts probably are aided in finding their preferred sites by some type of chemical sensitivity, but few if any experimental data have appeared to confirm this. Although it is generally assumed that flukes possess far fewer sensory mechanisms than their freeliving turbellarian relatives, the recent studies by

145

Rohde (1966, 1968b) indicate that in some groups rather complex sensory systems have developed. Papillae on miracidia were reported as early as 1896 by Coe, who observed conical protuberances on Fasciola hepática miracidia, but considered them as openings of glands. Bettendorf (1897) reviewed sensory papillae of trematodes and cestodes and concluded that they were "taste papillae." Donges (1964) considered sensory papillae situated in various regions of Posthodiplostomum miracidia to serve as taste receptors. Ginetsinskaya and Dobrovolskii ( 1963 ), who used the term sensilla (sensory papillae together with a sensory hair) for such a structure, observed that sensillae appear only on those miracidia which are not ingested by their intermediate hosts. No experimental evidence, however, has been provided as to the significance of the presumed chemoreceptive function of sensory papillae. In sporocysts and rediae, fewer sensory structures exist. Pearson (1956, 1961) illustrated sensory hairs on Alaria and Neodiplostomum sporocysts, and similar structures were reported at the anterior end of daughter sporocysts of Diplostomum by Ginetsinskaya and Dobrovolskii (1963). Daughter rediae of a species of Catatropis possess sensory hairs (Sinitsin 1904). It would be of interest to determine if these structures exist among sporocysts and rediae of other groups of digenetic trematodes and if they are related to localization of such intramolluscan stages. Sensory structures in the form of individual receptors are extremely abundant in cercariae, occurring in diverse families (Wagner 1961 ; Lie 1966). Sensory hairs are known to occur in bodies and tails of cercariae (Donges 1964), and surface papillae may be concentrated in particular areas. Although the function of these structures has not been determined experimentally, their morphology suggests a sensory role. Their distribution is so constant and regular that they have been suggested as a useful criterion for taxonomic differentiation of species (Vercammen-

146

Granjean 1951; Wagner 1961; Ginetsinskaya andDobrovolskii 1963).Neuhaus (1952) described a number of sensillae in Trichobilharzia cercariae and considered them as chemosensory receptors. Dixon and Mercer (1965) described the ultrastructure of a presumed tangoreceptor at the edge of the oral sucker in Fasciola hepática cercariae. Very limited information on sensory papillae in metacercariae appears to have been published. The fate of sensory papillae of cercariae at the time of their encystment is not known with certainty, according to Dixon and Mercer (1965 ). Sensillae are said to degenerate in the metacercarial stage according to Ginetsinskaya and Dobrovolskii (1963), but Dobrovolny (1939) found more papillae with sensory hairs in Plagioporus metacercariae than in cercariae of that species. Sensory receptors in adult digeneans have been studied under light and electron microscopy, but opinions vary as to the functions of such receptors. Morris and Threadgold (1967) considered sensillae of Fasciola adults to be rheoreceptors, used in detecting direction of flow of a fluid medium. Erasmus ( 1967 ) concluded that concentrations of sensillae on lobes of the holdfast of the strigeoid, Cyathocotyle, may facilitate attachment of the adult to host mucosa, and that their structure suggests a tangoreceptive function. Many prominent sensory papillae in an adult allocreadiid fluke (Macrolecithus) were described by G. Rees (1968 ), who noted their greater abundance in the preacetabular zone and considered them to be either tactile or chemoreceptive. Rohde ( 1968a) has shown that in at least one aspidogastrid fluke (Multicotyle), the nervous system and sensory receptors are extraordinarily specialized, suggesting that such complexity is related to the fluke's ability to locate favourable niches within its host. Rohde (1968a), in comparing sensory structures of various groups of trematodes, considered them to be best developed among the aspidogastrids. Hair-like structures have been described on the body surface of Proteocephalus plerocercoids

MARTIN J. ULMER

(Freeman 1964) and "cuticular hairs" are known to occur in plerocercoids of various other orders (Voge 1967) ; their function, however, remains enigmatic. Bulb-like tegumental sensory structures, each with a cilium-like terminal extension, have recently been reported from adult cestodes, including Echinococcus, by Morseth ( 1967 ). Evidence of neurosecretion in Hymenolepis diminuta was provided by Davey and Breckenridge (1967). Neuroanatomical studies on H. diminuta suggest that the apical organ of that species may function as a sensory chemoreceptor (Wilson and Schiller 1969). Nematodes are known to possess an abundance of sensory organs of diverse types, including chemoreceptors (Lee 1965). Little is known of the relationships of these to site-finding, but in Neoaplectana, according to Poinar and Leutenegger (1968), size of amphids in the pharyngeal region of juveniles is greatly increased when juveniles become infective and hence the authors suggest that in host-location, chemoreception by amphids plays a more important role than do tactile-sensitive papillae. Other nematode structures may also be involved in site-finding. Nelson (1964) has suggested that the commonly encountered caudal papillae of microfilariae may be involved in orientation of juvenile worms, and Davydov and Rusak (1968) have shown that in Ascaris the cerebral ganglia are responsible for regulation of forward movement. Little evidence of the existence of sensory structures in acanthocephalans has been reported, according to Pratt ( 1969), and in those instances where presumed tactile-sensory organs have been found, they are reduced to a few occurring in the proboscis and several associated with the male bursa. Whether such structures are related to site-finding has apparently not been determined. Continuing evidence of neurosecretory glands and neurohormones in trematodes (Ude 1962; Grasso 1967a, b; Matskasi 1970), cestodes (Davey and Breckenridge 1967), and nematodes (Davey 1964, 1966) suggests that these, too, may play some role in localization of hel-

SITE-FINDING BEHAVIOUR IN HELMINTHS

minths, but this avenue of investigation remains unexplored. Experimental proof of neurophysiological functions of the various sensory structures thought to be involved in site-finding behaviour remains to be demonstrated. Among larger invertebrates including insects, electroantennograms have been used in gaining insights into sensory mechanisms. Dropkin (1965) suggested that in the larger nematodes sensory structures might be investigated by microsurgical methods employing microbeams. Extirpation experiments to determine if presumed sensory structures of helminth larvae and adults are indeed chemoreceptive have yet to be undertaken, although such studies among turbellarians are well known (Mueller 1936).

147

tions, are generally based on the elimination of one or more factors associated with normal behaviour; analysis of behavioural patterns is hence exceedingly difficult, and progress understandably slow. The study of behaviour has not yet developed as a recognized area in helminthology, yet accumulating evidence suggests more and more that in every stage, highly specialized, complex behaviour patterns occur in response to exacting requirements of each species. The careful and critical analysis of adaptive behaviour for each life cycle stage, and the elucidation of trigger mechanisms including chemical, hormonal, sensory, and neurosensory stimuli, undoubtedly will provide challenging areas of inquiry for the intellectually curious helminthologist. ACKNOWLEDGMENTS

Fifteen years ago, in a comprehensive general paper concerning relationships of behaviour to specificity of parasites, Davenport (1955) called attention to the paucity of information then available on that subject. The significance of specific behavioural responses in determining distribution of helminths within their hosts was also stressed by Read (1958) and by Thorson (1969). Despite conspicuous advances in helminthology, however, there appear to be very few host-parasite associations in which the stimuli determining behaviour have been clearly identified and in which the factors resulting in sitefinding have been unequivocally demonstrated. The complexity of helminth life cycles necessitates control mechanisms at various points. Berntzen's (1966) analogy that a parasite is computer-like in its being governed by various physiological clocks and keys, and that the host's environment (both external and internal) serves as a switching station for a key-punch system, is very apropos relative to site-finding behaviour. Our knowledge of the mechanisms involved is very superficial and the need for intensified studies is abundantly apparent. The small size of most parasitic helminths and the nature of their habitat make precise observations difficult. Laboratory studies, made under artificial condi-

Grateful acknowledgment is expressed to Dr A. S. Elwell for critical reading of the manuscript, to Mr Edward A. Dykstra for assistance in the miracidial host-finding experiments noted in the text, and to Mr Harvey Blankespoor for aid in photography. Experimental studies on Megalodiscus miracidia were supported in part by NSF Grant GB-5465X and by the NSF Undergraduate Research Participation Program. REFERENCES

Abdel-Malek, E. T. 1950. Susceptibility of the snail Biomphalaria boissyi to infection with certain strains of Schistosoma mansoni. Am. J. Trop. Med. 30:887-94 Agersborg, H. P. L. 1924. Studies on the effect of parasitism upon the tissues, i. With special reference to certain gastropod molluscs. Quart. J. Microscop. ScL 68:361-403 Alicata, J. E. 1962. Life cycle and developmental stages of Philophthalmus gralli in the intermediate and final hosts. /. Parasitai. 48:47-54 Anokhin, I. 1966. Diurnal cycle of the activity and behaviour of ants (Formica pratensis Retz.) invaded by metacercaria of Dicrocoelium lanceaturn during the grazing period. Zool. Zhur. 45: 687-92

148 Archer, D. M., and Hopkins, C. A. 1958. Studies on cestode metabolism, m. Growth patterns of Diphyllobothrium sp. in a definitive host. Exptl. Parasitai. 7:125-44 Armstrong, J. C. 1965. Mating behavior and development of schistosomes in the mouse. /. Parasitol 51:605-16 Awachie, J. B. E. 1965. The ecology of Echinorhynchus truttae Schrank, 1788 (Acanthocephala) in a trout stream in North Wales. Parasitólogy 55:747-62 - 1966. The development and life history of Echinorhynchus truttae Schrank, 1788 (Acanthocephala). J. Helminthol 40:11-32 Barbosa, F. A. S. 1960. Alguns aspectos das relacóes hopedeiroparasito entre as fases larvarias do trematódeo Schistosoma mansoni e o molusco Australorbis glabratus. Bol. fundaçâo Gonçalo Moniz 5:1-44 Barlow, C. H. 1925. The Life Cycle of the Human Intestinal Fluke Fasciolopsis buski (Lankester). Am. J. Hyg. Monogr. Ser. 4 Beaver, P. C. 1939. The morphology and life history of Psilostomum ondatrae Price, 1931 (Trematoda: Psilostomidae). /. Parasitai. 25:383-93 - 1955. Observations on Necator infections resulting from exposures to three larvae. Rev. ibérica parasitai. Tomo Extr. 713:721 Beaver, P. C., and Little, M. D. 1964. The genital girdle in relation to estrus and mating in Ascaris lumbricoides. J. Parásito!. 50:128-31 Berntzen, A. K. 1966. A controlled culture environment for axenic growth of parasites. Ann. N.Y. Acad. Sci. 139, part 1:176-89 Bertram, D. S. 1966. Dynamics of parasitic equilibrium in cotton rat filariasis. Advances in Parasitol 4:255-319 Best, J. B. 1963. Protopsychology. Sci. American 208:54-62 Bettendorf, A. 1897. Ueber Musculatur und Sinneszellen der Trematoden. Zoo/. Jahrb. Abt. Anat. 10:307-58 Bird, A. F. 1959. The attractiveness of roots to the plant parasitic nematodes Meloidogyne javanica and M. hapla. Nematologica 4:322—35 Bird, A. F., and Wallace, H. R. 1969. Chemical

MARTIN J. ULMER

ecology of Acanthocephala and Nematoda. Chem. Zoo/. 3:561-92 Blake, C. D. 1962. Some observations on the orientation of Ditylenchus dipsaci and invasion of oat seedlings. Nematologica 8:177-92 Bonner, T. P., and Etges, F. J. 1967. Chemically mediated sexual attraction in Trichinella spiralis. Exptl. Parasitai. 21:53-60 Brâten, T., and Hopkins, C. A. 1969. The migration of Hymenolepis diminuta in the rat's intestine during normal development and following surgical transplantation. Parasitology 59:891-905 Brumpt, E. 1936. Précis de Parasitologie. 5th ed. Paris: Masson - 1940. Confirmation des observations de A. Lutz sur les lesions tentaculaires de Planorbis glabratus (P. guadeloupensis) déterminées par l'évolution des miracidies de S. mansoni. Compt. rend, soc. biol 133:625-8 - 1941. Observations biologiques diverses concernant Planorbis (Australorbis) glabratus, hôte intermédiare de Schistosoma mansoni. Ann. Parasitol. 18:9-45 Burlingame, P. L., and Chandler, A. C. 1941. Host parasite relations of Moniliformis dubius (Acanthocephala) in albino rats, and the environmental nature of resistance to single and superimposed infections with this parasite. Am. J. Hyg. 33 (D) : 1-21 Cable, R. M., and Dill, W. T. 1967. The morphology and life history of Paulisentis frac tus VanCleave and Bangham, 1949 (Acanthocephala: Neoechinorhynchidae). /. Parasitol. 53:810-17 Calentine, R. L. 1964. The life cycle oiArchigetes iowensis (Cestoda: Caryophyllaeidae). J. Parasitol. 50:454-8 - 1965. The biology and taxonomy of Biacetabulum (Cestoda: Caryophyllaeidae). /. Parasitol. 51:243-8 Campbell, W. C. 1961. Notes on the egg and miracidium of Eascioloides magna (Trematoda). Trans. Am. Microscop. Soc. 80:308-19 Campbell, W. C., and Todd, A. C. 1955. Behavior of the miracidium of Fascioloides magna (Bassi 1875) Ward 1917 in the presence of a snail host. Trans. Am. Microscop. Soc. 74:342-6

SITE-FINDING BEHAVIOUR IN HELMINTHS

- 1956. Emission of cercariae and metacercariae in snail feces. Trans. Am. Microscop. Soc. 75: 241-3 Carney, W. P. 1967. Notes on the life cycle of Brachylecithum mosquensis (Skrjabin and Isaitschikoff, 1927) from the bile ducts of the robin, Turdus migratorius, in Montana. Can. J. Zool. 45:131-4 — 1969. Behavioral and morphological changes in carpenter ants harboring dicrocoeliid metacercariae. Am. Midland Naturalist 82:605-11 Chandler, A. C. 1939. The effects of number and age of worms on development of primary and secondary infections with Hymenolepis diminuta in rats, and an investigation into the true nature of "premunition" in tapeworm infections. Am.J.Hyg. 29:105-14 Chappell, L. H., Arai, H. P., Dike, S. C., and Read, C. P. 1970. Circadian migration of Hymenolepis (Cestoda) in the intestine, i. Observations on H. diminuta in the rat. Com p. Biochem. PhysioL 34:31-46 Cheng, T. C. 1963. Activation of G orgoder a amplicava cercariae by molluscan sera. Exptl. Parasitol. 13:342-7 - 1967. Marine molluscs as hosts for symbioses; with a review of known parasites of commercially important species. In Advances in Marine Biology, éd. F. S. Russell, vol. 5. New York: Academic Press - 1968. The compatability and incompatability concept as related to trematodes and molluscs. Pacific Sci. 22:141-60 Cheng, T. C., and Cooperman, J. S. 1964. Studies on host-parasite relationships between larval trematodes and their hosts, v. The invasion of the reproductive system of Helisoma trivolvis by the sporocysts and cercariae of Glypthelmins pennsylvaniensis. Trans. Am. Microscop. Soc. 83:12-23 Chemin, E. 1968. Interference with the capacity of Schistosoma mansoni miracidia to infect the molluscan host. J. Parasitol. 54:509-16 - 1970. Behavioral responses of miracidia of Schistosoma mansoni and other trematodes to substances emitted by snails. /. Parasitol. 56:

149

287-96 Chemin, E., and Dunavan, C. A. 1962. The influence of host-parasite dispersion upon the capacity of Schistosoma mansoni miracidia to infect Australorbis glabratus. Am. J. Trop. Med. Hyg. 11:455-71 Chemin, E., and Perlstein, J. M. 1969. Further studies on interference with the host-finding capacity of Schistosoma mansoni miracidia. /. Parasitol. 55:500-8 Chu, G. W. T. C., and Cutress, C. E. 1954. Austrobilharziavariglandis (Miller and Northup, 1926) Penner, 1953 (Trematoda: Schistosomatidae) in Hawaii with notes on its biology. J. Parasitol. 40:515-23 Clegg, J. A. 1969. Skin penetration by cercariae of the bird schistosome Austrobilharzia terrigalensis: the stimulatory effect of cholesterol. Parasitology 59:973-89 Coe, W. R. 1896. Notizen über den Bau des Embryos von Distomum hepaticum. Zool. Jahrb. Abt.Anat. 9:561-70 Cowper, S. G. 1947. Observations on the life cycle of Schistosoma mansoni in the laboratory, with a discussion on the snail vectors of S. mansoni and S. haematobium. Ann. Trop. Med. Parasitol. 41:173-7 Crandall, R. B. 1960. The life history and affinities of the turtle lung fluke, Heronimus chelydrae MacCallum, 1902. J. Parasitol. 46:289-307 Crofton, H. D. 1954. The vertical migration of infective larvae of strongyloid nematodes. /. Helminthol. 28:35-52 Crompton, D. W. T., and Harrison, J. G. 1965. Observations on Poly morphus minutus (Goeze, 1782) (Acanthocephala) from a wildfowl reserve in Kent. Parasitology 55:345—55 Crompton, D. W. T., and Whitfield, P. J. 1968. A hypothesis to account for the anterior migrations of adult Hymenolepis diminuta (Cestoda) and Moniliformis dubius (Acanthocephala) in the intestine of rats. Parasitology 58:227-9 Cuckler, A. C. 1940. The life cycle of Fibricola crátera (Barker and Noll, 1915) Dubois, 1932 (Trematoda: Strigeata). /. Parasitol. 26(Suppl.): 32-3

150 Davenport, D. 1955. Specificity and behaviour in symbioses. Quart. Rev. Biol. 30:29-46 Davenport, D., Wright, C. A., and Causley, D. 1962. Technique for the study of the behavior of motile micro-organisms. Science 135:1059-60 Davey, K. G. 1964. Neurosecretory cells in a nematode, Ascaris lumbricoides. Can. J. Zool. 42: 731-4 — 1966. Neurosecretion and molting in some parasitic nematodes. Am. Zool. 6:243-9 Davey, K. G., and Breckenridge, W, R. 1967. Neurosecretory cells in a cestode, Hymenolepis diminuta. Science 158:931-2 Davydov, O. N., and Rusak, L. V. 1968. The role of cerebral ganglia in the regulation of locomotor activity of asear ids. Folia Parasitai. 15:351-8 Dawes, B., and Hughes, D. L. 1964. Fascioliasis: the invasive stages of Fasciola hepática in mammalian hosts. Advances in Parasitai. 2:97-168 DeGiusti, D. L. 1949. The life cycle ofLeptorhynchoides thecatus (Linton), an acanthocephalan offish. /. Parasitai. 35:437-60 DeRycke, P. H. 1966. Development of the cestode Hymenolepis microstoma in Mus musculus. Z. Parasitenk. 27:350-4 DeRycke, P. H., and Berntzen, A. K. 1967. Maintenance and growth of Hymenolepis microstoma (Cestoda: Cyclophyllidea) in vitro. J. Parasitol. 53:352-4 Dixon, H. B. F., and Lipscomb, F. M. 1961. Cysticercosis: An analysis and follow-up of 450 cases. Brit. Med. J. (5235) : 1320 Dixon, K. E., and Mercer, E. H. 1965. Fine structure of the cercarial nervous system. /. Parasitol. 51:967-76 Dobrovolny, C. G. 1939. Life history oiPlagioporus sinitsini Mueller and embryology of new cotylocercous cercariae (Trematoda). Trans. Am. Microscop. Soc. 58:121-55 Doerr, R., and Menzi, E. 1933. Studien über der Trichinellen Infektion. vin. Vergleichende Untersuchungen über die Empfanglichkeit der Ratte und des Meerschweinchens fur die Infektion per os. Zentr. Bakteriol. Parasitenk. Abt. I, Orig. 128:177-88 Dônges, J. 1964. Der Lebenszyklus von Posthodi-

MARTIN J. ULMER

plostomum cuticola (v. Nordmann 1832) Dubois 1936 (Trematoda: Diplostomatidae). Z. Parasitenk. 24:169-248 Dropkin, V. H. 1965. Behaviour of nematodes in relation to that of other invertebrates. Nematologica 11:1—3 Duke, B. O. L. 1952. On the rate of emergence of the cercariae of Schistosoma mansoni from Australorbis glabratus. J. Helminthol. 26:133—46 Dvorak, J. A., Jones, A. W., and Kuhlman, H. H. 1961. Studies on the biology of Hymenolepis microstoma (Dujardin, 1845). /. Parasitol. 47: 833-8 Edmonds, J. S. 1966. Hatching of the eggs of Moniliformis dubius. Exptl. Parasitol. 19:216-26 Efford, I. E., and Tsumura, K. 1969. Observations on the biology of the trematode Megalodiscus microphagus in amphibians from Marion Lake, British Columbia. Am. Midland Naturalist 82: 197-203 Elwell, A. S. 1967. Biology of Anguispira altérnala (Gastropoda: Endodontidae) with special reference to the digenetic trematode Postharmostomum helicis. Ph.D. Dissertation, Iowa State Univ. (Univ. Microfilm No. 67-12,953) Erasmus, D. A. 1958. Studies on the morphology, biology and development of a strigeoid cercaría (CercanaX Baylis 1930). Parasitology 48:31235 - 1959. The migration of Cercaría X Baylis (Strigeida) within the fish intermediate host. Parasitology 49:173-90 — 1967. The host-parasite interface of Cyathocotyle bushiensisKhan, 1962 (Trematoda: Strigeoidea). ii. Electron microscope studies of the tegument. J. Parasitol. 53:703-14 Etges, F. J., and Decker, C. L. 1963. Chemosensitivity of the miracidium of Schistosoma mansoni to Australorbis glabratus and other snails. /. Parasitol. 49:114-16 Etges, F. J., and Gresso, W. 1965. Effect of Schistosoma mansoni infection upon fecundity in Australorbis glabratus. J. Parasitol. 51:757—60 Faust, E. C. 1924. The reactions of the miracidia of Schistosoma japonicum and S. haematobium in the presence of their intermediate hosts. /.

SITE-FINDING BEHAVIOUR IN HELMINTHS

Parasitai. 10:199-204 - 1927. Migration route of Spirocerca sanguinolenta in its definitive host. Proc. Soc. Exptl. Biol. Med. 25:192-5 Faust, E. C., and Hoffman, W. A. 1934. Studies on schistosomiasis mansoni in Puerto Rico. in. Biological studies, i. The extra-mammalian phases of the life cycle. Puerto Rico J. Publ. Health Trop. Med. 10:1-49 Faust, E. C., and Khaw, O. K. 1927. Studies on Clonorchis sinensis (Cobbold). Am. J. Hyg. Monogr. Ser. 8 Faust, E. C., and Meleney, H. E. 1924. Studies on Schistosomiasis Japónica. Am. J. Hyg. Monogr. Ser. 3 Ferguson, M. S. 1943. Migration and localization of an animal parasite within the host. /. Exptl. Zool. 93:375-401 Fischer, H., and Freeman, R. S. 1969. Penetration of parenteral plerocercoids of Proteocephalus ambloplitis (Leidy) into the gut of smallmouth bass. /. Parasitai. 55:766-74 Foster, W. B., and Daugherty, J. W. 1959. Establishment and distribution of Raillietma cesticillus in the fowl and comparative studies on amino-acid metabolism of R. cesiicillus and Hymenolepis diminuta. Exptl. Parásito!. 8:413-26 Freeman, R. S. 1964. On the biology of Proteocephalus parallacticus MacLulich (Cestoda) in Algonquin Park, Canada. Can. J. Zool. 42:387408 Fiilleborn, F. 1912. Untersuchungen über die chemotaktische Wirkung der Malpighischen Gefasse von Stechmucken auf Hundemikrofilarien. Zentr. Bakteriol. Parasitenk. Abt. I, Orig. 65: 349-52 Ginetsinskaya, T. A., and Dobrovolskii, A. A. 1963. A new method for observing sensillae of trematode larvae and the systematic importance of these structures. Doklady Akad. Nauk BioL Sci. 151:1087-9 (Eng. trans.) Goodchild, C. G. 1948. Additional observations on the bionomics and life history of Gorgodera amplicavaLooss 1899 (Trematoda: Gorgoderidae). /. Parasitai. 34:407-27 - 1958. Transfaunation and repair of damage in

151

the rat tapeworm, Hymenolepis diminuta. J. Parasitai. 44:345-51 Graff, D. J., and Kitzman, W. B. 1965. Factors influencing the activation of acanthocephalan cystacanths. J. Parasitai. 51:424-9 Grasso, M. 1961 a. Prime indagini sulla presenza di cellule neurosecretrici in Fasciola hepática. Atti accad. nazi. Lincei Rend. Classe sci. fis., mat. e nat.42:S5-7 - 1967b. Distribuzione e áttivitá délie cellule neurosecretrici in Fasciola hepática. Atti accad. nazi. Lincei Rend. Classe sci. fis., mat. e nat. 42:903-5 Griffiths, H. J .1939. Observations on the bionomics of ova and miracidia of Fasciola hepática Linn, in eastern Canada. Can. J. Research, Sec. D 17:205-11 Halik, L. 1931. Über die rhythmischen Bewegungen der in Bernsteinschnecken parasitierenden Sporozystenschlauche von Leucochloridium macrostomum Rud. ( = paradoxum Carus). Z. vergleich. Physiol. 14:462-78 Hawking, F. 1964. The periodicity of microfilariae. vin. Further observations on Wuchereria bancrofti. Trans. Roy. Soc. Trop. Med. Hyg. 58 : 212-27 - 1967. The 24-hour periodicity of microfilariae; biological mechanisms responsible for its production and control. Proc. Roy. Soc. (London), B 169:59-76 Hecker, U., and Thomas, E. 1965. Über Sporozystenschlauche von Leucochloridium macrostomum Rud. Verhandl. deut. zool. Ges. 1964: 444-56 Herber, E. C. 1939. Studies on the biology of the frog amphistome, Diplodiscus temperatus Stafford. /. Parasitai. 25:189-95 Heyneman, D. 1966. Successful infection with larval echinostomes surgically implanted into the body cavity of the normal snail host. Exptl. Parasitai. 18:220-3 Hoffman, G. L. 1955. Notes on the life cycle of Fibricola crátera (Trematoda: Strigeida). /. Parasitai. 41:327 Hoffman, G. L., and Hoyme, J. B. 1958. The experimental histopathology of the "tumor" on the brain of the stickleback caused by Diplostomum

152 baeri eucaliae Hoffman and Hundley, 1957 (Trematoda: Strigeoidea). /. Parasitai 44:374-8 Hohorst, W., and Graefe, G. 1961. Ameisen-obligatorische Zwischenwirte des Lanzettegels (Dicrocoelium dendriticum). Naturwissenschaften 48:229-30 Holmes, J. C. 1961. Effects of concurrent infections on Hymenolepis diminuta (Cestoda) and Moniliformis dubius ( Acanthocephala). i. General effects and comparison with crowding. /. Parasitai 47:209-16 - 1962. Effects of concurrent infections on Hymenolepis diminuta (Cestoda) and Moniliformis dubius (Acanthocephala). n. Effects on growth. /. Parasitai 48:87-96 Hopkins, C. A. 1959. Seasonal variations in the incidence and development of the cestode Proteocephalus filicollis (Rud. 1810) in Gasterosteus aculeatus (L. 1766). Parasitology 49:529-42 - 1969. The influence of dietary methionine on the amino-acid pool of Hymenolepis diminuta in the rat's intestine. Parasitology 59:402-27 - 1970. Location-specificity in adult tapeworms with special reference to Hymenolepis diminuta. J. Parásito!. 56, no. 4, sec. n, pt. 3:561-4 Hopp, W. B. 1954. Studies on the morphology and life cycle of Neoechinorhynchus emydis (Leidy), an acanthocephalan parasite of the map turtle, Graptemys geographica (LeSueur). /. Parasitai 40:284-99 Hu, C. H., and Hoeppli, R. J. C. 1935. Route of migration of Spirocerca sanguinolenta in experimentally infected dogs. Proc. Sac. Exptl Biol Med. 32:1393-4 - 1936. The migration route of Spirocerca sanguinolenta in experimentally infected dogs. Chinese Med. J. (Suppl. 1) 50:293-311 - 1937. Further study on the migration route of Spirocerca sanguinolenta in experimentally infected dogs. Chinese Med. J. 51:489-95 Jakowska, S. 1963. Mucus in invertebrates - introductory remarks. Proc. XVI Intern. Congr. Zool 1:81 - 1965. The role of zoologists in modern research in mucous secretions. In "Mucus in Invertebrates - Collected Papers." Ann. N.Y. Acad.

MARTIN J. ULMER

Sci. 118:903-11 - 1966. The challenge of research on mucus. In "Interdisciplinary Investigation of Mucus Production and Transport." Ann. N.Y. Acad. Sci. 130(Art. 3): 871-973 Jones, F. G. W. 1959. Ecological relationships of nematodes. In Plant Pathology: Problems and Progress: 1908-58, éd. A. S. Holton et al Madison: University of Wisconsin Press - 1960. Some observations and reflections on host finding by plant nematodes. RijkslandbouwhogeschoolMed. Landbouw. Gent. 25:1009-24 Kagan, I. G. 1952. Further contributions to the life history of Neoleucochloridium problematicum (Magath, 19201) new comb. (Trematoda: Brachylaematidae). Trans. Am. Microscop. Soc. 71:20-44 Karlson, P., and Luscher, M. 1959. "Pheromones" - a new term for a class of biologically active substances. Nature 183:55—6 Kates, K. C. 1944. Some observations on experimental infections of pigs with the thorn-headed worm, Macracanthorhynchus hirudinaceus. Am. J. Vet. Research 5:166-72 Kawashima, K., Tada, L, and Miyazaki, I. 1961. Host preference of miracidia of Paragonimus ohirai Miyazaki 1939 among three species of snails of the genus Assiminea. Kyushu J. Med. Sci. 12:99-106 Keithly, J. S. 1969. Life history of Corynosoma constrictum VanCleave (Acanthocephala: Polymorphidae). Dissertation Abstr. 29:3544-6 Kendall, S. B. 1965. Relationships between the species of Fasciola and their molluscan hosts. Advances in Parasitai 3:59-98 Kendall, S. B., and McCullough, F. S. 1951. The emergence of the cercariae of Fasciola hepática from the snail Limnaea truncatula. J. Helminthol 25:77-92 Kennedy, C. R. 1965. The life history oiArchigetes limnodrili (Yamaguti) (Cestoda: Caryophyllaeidae) and its development in the invertebrate host. Parasitology 55:427-37 Klinger, J. 1965. On the orientation of plant nematodes and of some other soil animals. Nematologica 11:4-18

SITE-FINDING BEHAVIOUR IN HELMINTHS

Kloetzel, K. 1958. Observacóes sobre o tropismo do miracidio de S. mansoni pelo molusco A. glabratus. Rev. brasil biol. 18:223-32 — 1960. Novas observacoes sobre o tropismo do miracidio de Schistosoma mansoni pelo molusco Australorbis glabratus. Rev. Inst. med. trop., Sao Paulo 2:341-6 Komiya, Y. 1966. Clonorchis and clonorchiasis. Advances in Parasitol. 4:53—106 Krull, W. H. 1930. The life history of two North American frog lung flukes. /. Parasitol. 16:207— 12 - 1931. Life history studies on two frog lung flukes, Pneumonoeces medioplexus and Pneumobites parviplexus. Trans. Am. Microscop. Soc. 50:215-77 - 1935. Some observations on the life history of Brachylaemus virginiana (Dickerson) Krull, N. 1934. Trans. Am. Microscop. Soc. 54:118-34 Krull, W. H., and Price, H. F. 1932. Studies on the life history of Diplodiscus temperatus Stafford from the frog. Occasional Papers Museum Zool Univ. Mich. 237:1-38 Laing, J. 1937. Host finding by insect parasites, i. Observations on the finding of hosts by Alysia manducator, Mormoniella vitripennis and Trichogramma evanescens. J. Animal Ecol. 6:298— 317 Larsh, J. E., Jr., Gilchrist, H. B., and Greenburg, B. G. 1952. A study of the distribution and longevity of adult Trichinella spiralis in immunized and non-immunized mice. /. Elisha Mitchell Sci.Soc.68 (no. 1):1-11 Larsh, J. E., Jr., and Hendricks, J. R. 1949. The probable explanation for the difference in the localization of adult Trichinella spiralis in young and old mice. /. Parasitol. 35:101-6 LaRue, G. R. 1951. Host-parasite relations among the digenetic trematodes. /. Parasitol. 37:333-42 Lavoipierre, M. M. J. 1958. Studies on the hostparasite relationships of filarial nematodes and their arthropod hosts, i. The sites of development and the migration of Loa loa in Chrysops silacea, the escape of the infective forms from the head of the fly, and the effect of the worm on the insect host. Ann. Trop. Med. Parasitol.

153

52:103-21 Lawrence, J. L. 1969. Host-parasite relationships in Catostomus commersoni with emphasis on the caryophyllidean tapeworms. Dissertation Abstr. 29:4895-6 Lee, D. L. 1965. The Physiology of Nematodes. San Francisco: W. H. Freeman and Co. Leiper, R. T., and Atkinson, E. L. 1915. Observations on the spread of Asiatic schistosomiasis. Brit. Med. J. 201-2 Lie, K. J. 1966. Studies on Echinostomatidae (Trematoda) in Malaya, xm. Integumentary papillae on six species of echinostome cercariae. /. Parasitol 52:1041-8 Lie, K. J., and Umathevy, T. 1965a. Studies on Echinostomatidae (Trematoda) in Malaya, vin. The life history of Echinostoma audyi sp.n. /. Parasitol 51:781-8 - 1965b. Studies on Echinostomatidae (Trematoda) in Malaya, x. The life history of Echinoparyphium dunni sp.n. /. Parasitol 51:793-9 Lutz, A. 1919. O Schistosomum mansoni e a schistosomatose segundo observacoes, feitas no Brasil. Mem. inst. Oswaldo Cruz 11:121-55 Maclnnis, A. J. 1963. Chemical attraction for Schistosoma mansoni miracidia. Dissertation Abstr. 24:1299 - 1965. Responses of Schistosoma mansoni miracidia to chemical attractants. J. Parasitol 51:73146 Maldonado, J. F. 1945. The life cycle of Tamerlania bragai Santos 1934 (Eucotylidae), a kidney fluke of domestic pigeons. /. Parasitol 31:30614 Mathias, P. 1925. Recherches expérimentales sur le cycle évolutif de quelques trématodes. Bull Biol 59:1-123 Matskasi, I. 1970. On the neurosecretory cells of Opisthodiscus diplodiscoides Cohn (Trematoda), and their structural changes during the day. Folia Parasitol 17:25-30 Mattes, O. 1926. Zur Biologie der Larvenentwicklung von Fasciola hepática besonders über den Einfluss der Wasserstoffionenkonzentration auf das Ausschlupfen der Miracidien. Zool Anz. 69: 138-56

154 - 1936. Zur Frage der Wirtsauffindung der Parasiten auf Grund experimenteller Untersuchungen an Leberegelmiracidien. Verhandl. deut. zool. Ces. 38:183-6 - 1949a. Wirtsfindung und Wirtsspezifitàt bel Parasiten. Verhandl. deut. zool. Ces. 1948:165-72 - 1949b. Wirtsfindung, Invasionsvorgang und Wirtsspezifitàt beim Fasciola Miracidium. Z. Parasitenk. 14:320-63 McCaig, M. L. O., and Hopkins, C. A. 1963. Studies on Schistocephalus solidus. IL Establishment and longevity in the definitive host. Exptl. Parasitai. 13:273-83 McCoy, O. R. 1935. The physiology of helminth parasites. Physiol. Revs. 15:221-40 Miller, R. B. 1945. Studies on cestodes of the genus Triaenophorus from fish of Lesser Slave Lake, Alberta, m. Notes on Triaenophorus nodulosus (Pallas) in the second intermediate host. Can. J. Research, D 23:1-5 Monnig,H.O. 1922. Über Leucochloridium macrostomum (Leucochloridium paradoxum Carus), ein Beitrag zur Histologie der Trematoden. Inaugural Dissertation, Jena Morris, G. P., and Threadgold, L. T. 1967. A presumed sensory structure associated with the tegument of Schistosoma mansoni. J. Parasitai. 53:537-9 Morseth, D. J. 1967. Observations on the fine structure of the nervous system of Echinococcus granulosas. J. Parasitai. 53:492-500 Mueller, H. G. 1936. Untersuchungen über spezifische Organe niedere Sinne bei rhabdocoelen Turbellarien. Z. vergleich. Physiol. 23:253-92 Mueller, J. F. 1959. The laboratory propagation of Spirometra mansonoides (Mueller, 1935) as an experimental tool. n. Culture and infection of the copepod host, and harvesting the procercoid. Trans. Am. Microscop. Sac. 78:245-55 Muftic, M. 1969. Metamorphosis of miracidia into cercariae of Schistosoma mansoni in vitro. Parasitology 59:365-71 Najarían, H. 1954. Developmental stages in the life cycle of Echinoparyphium flexum (Linton, 1892) Dietz, 1910 (Trematoda: Echinostomatidae). J. Morphol 94:165-97

MARTIN J. ULMER

Najim, A. T. 1956. Life history of Gigantobilharzia huronensis Najim, 1950, a dermatitis-producing bird blood-fluke (Trematoda-Schistosomatidae). Parasitology 46:443-69 Nelson, G. S. 1964. Factors influencing the development and behaviour of filarial nematodes in their arthropodan hosts. In Host-Parasite Relationships in Invertebrate Hosts, 2nd Symposium, British Society for Parasitology, pp. 75-119 Neuhaus, W. 1941. Beitráge zur Frage der Wirtsfindung des Miracidiums von Fasciola hepática. Z. vergleich. Physiol. 28:563-79 — 1952. Biologie und Entwicklung von Trichobilharzia szidati n.sp. (Trematoda, Schistosomatidae), einem Erreger von Dermatitis beim Menschen. Z. Parasitenk. 15:203-66 - 1953. Über den chemischen Sinn der Miracidien von Fasciola hepática. Z. Parasitenk. 15:476-90 Newton, M. V. B. 1932. The biology of Triaenophorus tricuspidatus (Bloch, 1779) in Western Canada. Contribs. Can. Biol. and Fisheries 7: 341-60 Nicholas, W. L., and Hynes, H. B. N. 1958. Studies on Polymorphus minutus (Goeze, 1782) (Acanthocephala) as a parasite of the domestic duck. Ann. Trop. Med. Parasitai. 52:36-47 Odetoyinbo, J. A. 1960. Biology of Spie ndido filaría quiscali (von Linstow 1904) n. comb. (Nematoda: Onchocercidae). Dissertation Abstr. 21: 2836-7 Oshima, T. 1961. Influence of pregnancy and lactation on migration of the larvae of Toxocara canis in mice. /. Parasitai. 47:657-60 Palombi, A. 1942. II ciclo biológico áiPtychogonimus megastoma (Rud.). Osservazioni sulla morfologia e fisiología délie forme larvali e considerazioni filogenetiche. Riv. parassitol. 6:117-72 Parker, J. C., and Haley, A. J. 1960. Phototactic and thermotactic responses of the filariform larvae of the rat nematodeNippostrongylus mûris. Exptl. Parasitol. 9:92-7 Patten, J. A. 1952. The life cycle of Conspicuum icteridorum Dentón and Byrd, 1951 (Trematoda: Dicrocoeliidae). /. Parasitol. 38:165-82 Pearson, J. C. 1956. Studies on the life cycles and morphology of the larval stages of Alaria arisae-

SITE-FINDING BEHAVIOUR IN HELMINTHS

moldes Augustine and Uribe, 1927 and Alaria cam's LaRue andFallis, 1936 (Trematoda: Diplostomidae). Can. J. Zool. 34:295-387 - 1959. Observations on the morphology and life cycle of Strigea elegans Chandler and Rausch, 1947 (Trematoda: Strigeidae). /. Parasitai. 45: 155-74 - 1961. Observations on the morphology and life cycle of Neodiplostomum intermedium (Trematoda: Diplostomatidae). Parasitology 51:133-72 Plempel, M. von 1964. Chemotaktische Anlockung der Miracidien von Schistosoma mansoni durch Australorbis glabratus. Z. Naturforsch. 196: 268-9 Plempel, M., Gônnert, R., and Federmann, M. 1966. Ver suche zur Wirtsfindung von Miracidien. Proc. 1st Intern. Congr. Parasitai. 1:32 Poinar, G. O., Jr., and Leutenegger, R. 1968. Anatomy of the infective and normal third stage juveniles of Neoaplectana carpocapsae Weiser (Steinernematidae: Nematoda). /. Parasitai. 54: 340-50 Pratt, I. 1969. The biology of the Acanthocephala. Chem.Zool. 3:245-52 Probert, A. J., and Erasmus, D. A. 1965. The migration of Cercaría X Baylis (Strigeida) within the molluscan intermediate host Lymnaea stagnalis. Parasitology 55:77-92 Read, C. P. 1950. The vertebrate small intestine as an environment for parasitic helminths. Rice Inst. Pam. 37, no. 2 - 1955. Intestinal physiology and the host-parasite relationship. Rutgers Univ. Bur. Biol. Research, Ann. Conf. Protein Metabolism, pp. 27-43 - 1958. Status of behavioral and physiological "resistance." Rice Inst. Pam. 45:36-54 - 1963. Biochemistry and physiology of tapeworms. Physiol. Revs. 43:263-305 Read, C. P., and Kilejian, A. Z. 1969. Circadian migratory behavior of a cestode symbiote in the rat host. /. Parasitai. 55:574-8 Rees, F. G. 1934. Cercaría patellae Lebour, 1911, and its effect on the digestive gland and gonads of Patella vulgata. Proc. Zool. Sac. London 1 : 45-53 Rees, G. 1968. Macrolecithus papilliger sp. nov.

155

(Digenea: Allocreadiidae, Stossich, 1904) from Phoxinus phoxinus (L.). Morphology, histochemistry and egg capsule formation. Parasitology 58:855-78 Rees, W. J. 1936. The effect of parasitism by larval trematodes on the tissues of Littorina littorea (Linné). Proc. Zool. Soc. London 2:357—68 Rhoades, H. L., and Linford, M. B. 1959. Molting of preadult nematodes of the genus Paratylenchus stimulated by root diffusâtes. Science 130: 1476-7 Richards, C. S. 1961. Emergence of cercariae of Schistosoma mansoni from Australorbis glabratus. J. Parasitai. 47:428 Robinson, E. J. 1949. The life history of Postharmostomum helicis (Leidy, 1847) n. comb. (Trematoda: Brachylaemidae). J. Parasitai. 35:51333 Robinson, J. 1962. Pilobolus spp. and the translation of the infective larvae of Dictyocaulus viviparous from faeces to pastures. Nature 193: 353-4 Roche, M. 1966. Influence of male and female Ancylostoma caninum on each other's distribution in the intestine of the dog. Exptl. Parasitai. 19 : 327-31 Rogers, W. P. 1957. An alternative approach to the study of host-parasite specificity. Intern. Union Biol. Sci. Ser. B, no. 32:309-11 - 1960. The physiology of infective processes of nematode parasites : the stimulus from the animal host. Proc. Roy. Sac. (London), B 152:36786 — 1962. The Nature of Parasitism. New York and London : Academic Press - 1966. Reversible inhibition of a receptor governing infection with some nematodes. Exptl. Parasitai. 19:15-20 Rogers, W. P., and Sommerville, R. I. 1963. The infective stage of nematode parasites and its significance in parasitism. Advances in Parasitai. I : 109-77 Rohde, K. 1966. Sense receptors of Multicotyle purvisi Dawes 1941 (Trematoda: Aspidobothria). Nature 211:820-2 - 1968a. The nervous systems of Multicotyle pur-

156 vw/Dawes 1941 (Aspidogastrea) and Diaschistorchis multitesticularis Rohde 1962 (Digenea). Z. Parasitenk. 30:78-94 - 1968b. Lichtmikroskopische Untersuchungen an den Sinnesrezeptoren der Trematoden. Z. Parasitenk. 30:252-77 Sait, G. 1935. Expérimental studies in insect parasitism, m. Host selection. Proc. Roy. Soc. (London), B 117:413-35 Savinov, V. A. 1953. Developmental peculiarities of Alaria alata (Goeze 1782) in definitive and reservoir hosts. In Contributions to Helminthology Published to Commemorate the 75th Birthday of K. I. Skrjabin (Eng. trans. 1966), pp. 612-16 Schad, G. A. 1963. Niche diversification in a parasitic species flock. Nature 198:404 Schwabe, C. W., and Kilejian, A. 1968. Chemical aspects of the ecology of Platyhelminthes. Chem. Zool. 2:467-549 Shift, C. J. 1968. Location QÍ Bulinus (Physopsis) globosus by miracidia of Schistosoma haematobium. J. Parasitai. 54:1133-40 - 1969. Influence of light and depth on location of Bulinus (Physopsis) globosus by miracidia of Schistosoma haematobium. J. Parasitol. 55:10810 Shiff, C. J., and Kriel, R. L. 1970. A water soluble product of Bulinus (Physopsis) globosus attractive to Schistosoma haematobium miracidia. /. Parasitol. 56:281-6 Silverman, P. H., and Maneely, R. B. 1955. Studies on the biology of some tapeworms of the genus Taenia, m. The role of the secreting gland of the hexacanth embryo in the penetration of the intestinal mucosa of the intermediate host, and some of its histochemical actions. Ann. Trop. Med. Parasitol. 49:326-30 Sinclair, K. B. 1967. Pathogenesis of Fasciola and other liver flukes. Helminthol. Abstr. 36:115-34 Sinitsin, D. 1904. Über einige neue und wenig bekannte Organe der digenetischen Trematoden. Zool.Anz. 27:767-70 Slais, J. 1967. The location of the parasites in muscle cysticercosis. Folia Parasitol. 14:217—24 Smyth, J. D. 1963. Biology of cestode life cycles.

MARTIN J. ULMER

Tech. Commun. Commonwealth Bur. Helminthol 34:1-38 — 1966. The Physiology of Trematodes. Edinburgh and London: Oliver and Boyd — 1969. The Physiology of Cestodes. San Francisco: W. H. Freeman & Co. Smyth, J. D., and Haslewood, G. A. D. 1963. The biochemistry of bile as a factor in determining host specificity in intestinal parasites, with particular reference to Echinococcus granulosus. Ann. N.Y. Acad. Sci. 113:234-60 Sodeman, T .M., Sodeman, W. A., and Richards, C. S. 1969. The intrapulmonary localization of Angiostrongylus cantonensis in the rat. Proc. Helminthol. Soc. Wash. 36:143-6 Sogandares-Bernal, F. 1966. Studies on American paragonimiasis. iv. Observations on the pairing of adult worms in laboratory infections of domestic cats. /. Parasitol. 52:701-3 Solomon, G. B. 1966. Development oiNippostrongylus brasiliensis in gonadectomized and hormone-treated hamsters. Exptl. Parasitol. 18:37496 - 1969. Host hormones and parasitic infection. Intern. Rev. Trop. Med. 3:101-58 Soulsby, E. J. L. 1965. Textbook of Veterinary Clinical Pathology. Vol. 1. Helminths. Philadelphia: F. A. Davis Co. Steiner, G. 1925. The problem of host-selection and host specialization of certain plant-infesting nemas and its application in the study of nemic pests. Phytopathology 15:499-534 Stirewalt, M. A. 1951. The frequency of bisexual infections of Schistosoma mansoni in snails of the species Australorbis glabratus (Say). /. Parasitol. 37:42-7 - 1959. Chronological analysis, pattern and rate of migration of cercariae of Schistosoma mansoni in body, ear and tail skin of mice. Ann. Trop. Med. Parasitol. 53:400-13 - 1963. Chemical biology of secretions of larval helminths. Ann. N.Y. Acad. Sci. 113:36-53 - 1966. Skin penetration mechanisms of helminths. In Biology of Parasites, ed. E. J. L. Soulsby, pp. 41-57. New York: Academic Press Stirewalt, M. A., and Kruidenier, F. J. 1961. Acti-

SITE-FINDING BEHAVIOUR IN HELMINTHS

vity of the acetabular secretory apparatus of cercariae of Schistosoma mansoni under experimental conditions. Exptl. Parasitol. 11:191-211 Stirewalt, M. A., and Uy, A. 1969. Schistosoma mansoni: cercarial penetration and schistosomule collection in an in vitro system. Exptl. Parasitol 26:17-28 Stockdale, P. H. G., and Anderson, R. C. 1970. The development, route of migration, and pathogenesis of Filaroides mart is in mink. /. Parasitol. 56:550-8 Stunkard, H. 1943. The morphology and life history of the digenetic trematode, Zoogonoides laevis Linton, 1940. Biol Bull. 85:227-37 Sudds, R. H., Jr. 1960. Observations of schistosome miracidial behavior in the presence of normal and abnormal snail hosts and subsequent tissue studies of these hosts. J. Elisha Mitchell Soc. 76: 121-33 Supperer, R. 1959. Untersuchungen iiber Parasiten der Hausente, Anas platyrhynchos dom. Z. Parasitenk. 19:259-77 Swales, W. E. 1935. The life cycle of Fascioloides magna (Bassi 1875) the large liver fluke of ruminants in Canada. Can. J. Research, Sec. D 12:177-215 Szidat, L. 1966. Fundamentos para estudios sobre compartamiento en helmintos. Proc. 1st Intern. Congr. Parasitol. 1:485 - 1969. Structure, development and behaviour of new strigeatoid metacercariae from subtropical fishes of South America. /. Fisheries Research Board Can. 26:753-86 Tetley, J. H. 1937. The distribution of nematodes in the small intestine of the sheep. New Zealand J. Sci. Technol. 18:805-17 Thomas, A. P. 1883. The life history of the liver fluke (Fasciola hepática). Quart. J. Microscop. Sci. 23:99-133 Thorpe, W. H., Crombie, A. C., Hill, R., and Darrah, J. H. 1947. The behaviour of wireworms in response to chemical stimulation. J. Exptl. Biol. 23:234-66 Thorson, R. E. 1969. Environmental stimuli and the responses of parasitic helminths. Bioscience 19:126-30

157

Thorson, R. E., Digenis, G. A., Berntzen, A., and Konyalian, A. 1968. Biological activities of various lipid fractions from Echinococcus granulosus scolices on in vitro cultures of Hymenolepis diminuta. J. Parasitol. 54:970-3 Timon-David, J. 1960. Recherches expérimentales sur le cycle de Dicrocoelioides petiolatum (A. Railliet 1900) (Trematoda, Dicrocoeliidae). Ann. Parasitol. 35:251-67 Tubangui, M. A., and Pasco, A. M. 1933. The life history of the human intestinal fluke, Euparyphiurn ilocanum (Garrison 1908). Philippine J. Sci. 51:581-603 Turner, H. F. 1957. Preliminary notes on the life cycle of Fibricola crátera (Barker and Noll, 1915) Dubois 1932 (Trematoda: Diplostomatidae). /. Alabama Acad. Sci. 29:43-4 - 1958. The life history of Fibricola crátera (Barker and Noll, 1915) Dubois 1932 (Trematoda: Diplostomatidae). Dissertation Abstr. 19:609 Twohy, D. W. 1956. The early migration and growth of Nippostrongylus mûris in the rat. Am. J.Hyg. 63:165-85 Ude, J. 1962. Neurosekretorische Zellen im Cerebralganglion von Dicrocoelium lanceolatum St. u. H. (Trematoda-Digenea) Zool. Anz. 169: 455-7 Ulmer, M. J. 195la. Postharmostomum helicis (Leidy, 1847) Robinson 1949 (Trematoda), its life history and a revision of the subfamily Brachylaeminae. I. Trans. Am. Microscop. Soc. 70: 189-238 — 1951b. Postharmostomum helicis (Leidy, 1847) Robinson 1949 (Trematoda), its life history and a revision of the subfamily Brachylaeminae. n. Trans. Am. Microscop. Soc. 70:319—47 - 1952. Morphological features of Brachylaima virginianum metacercariae (Trematoda: Brachylaimatidae), and migration route of cercariae in the 2nd intermediate host. Iowa State Coll. J. Sci. 27:91-103 Ulmer, M. J., and Sommer, S. C. 1957. Development of sporocysts of the turtle lung fluke, Heronimus chelydrae MacCallum (Trematoda: Heronimidae). Proc. Iowa Acad. Sci. 64:601-13 Ulmer, M. J., and Vande Vusse, F. 1970. Morpho-

158 logy of Dendritobilharzia pulverulenta (Braun, 1901) Skrjabin, 1924 (Trematoda: Schistosomatidae) with notes on secondary hermaphroditism in mâles. /. Parasitai. 56:67-74 Uzmann, J. R. 1953. Cercaría milfordensisnov. sp., a microcercous trematode larva from the marine bivalve Mytilus edulis L. with special reference to its effect on the host. /. Parasitai 39:445-51 Van Gundy, S. D. 1965. Nematode behaviour. Nematologica 11:19-32 Venard, C. E., and Warfel, J. H. 1953. Some effects of two species of Acanthocephala on the alimentary canal of the largemouth bass. J. Parasitol 39:187-90 Vercammen-Granjean, P. H. 1951. Sur la chaetotaxie de la larve infestante de Schistosoma mansoni. Ann. Parasitai. 26:412-14 Voge, M. 1967. The post-embryonic developmental stages of cestodes. Advances in Parasitai. 5:24797 Waffle, E. L. 1968. The life history and host-parasite relationships of Phyllodistomum staff or di (Trematoda: Gorgoderidae). Dissertation Abstr. 28:4808-B-4809-B Wagner, A. 1960. Stimulation of Schistosomatium douthitti cercariae to penetrate their host. Dissertation Abstr. 20:3449 - 1961. Papillae on three species of schistosome cercariae. /. Parasitai. 47:614-18 Wajdi, N. 1964. Some observations on the relationship of Schistosoma haematobium with its intermediate host. /. Helminthol. 38:383-90 Wallace, H. R. 1961. The bionomics of freeliving stages of zooparasitic and phytoparasitic nematodes - a critical survey. Helminthol. Abstr. 30: 1-22 Ward, H. L. 1940. Studies on the life history of Neoechinorhynchus cylindratus ( VanCleave, 1913) (Acanthocephala). Trans. Am. Microscop. Soc. 59:327-47 Wardle, R. A., and McLeod, J. A. 1952. The Zoology of Tapeworms. Minneapolis: University of Minnesota Press Webster, G. A. 1958. On prenatal infection and the migration of Toxocara canis Werner, 1782 in dogs. Can. J. Zool 36:435-40

MARTIN J. ULMER

Webster, J. M. 1969. The host-parasite relationships of plant-parasitic nematodes. Advances in Parasitai. 7:1-40 Wesenberg-Lund, C. 1931. Contributions to the development of the Trematoda Digenea. Part I. The biology of Leucochloridium paradoxum. Mem. acad. ray. sci. Danemark, Sect. sci. 9:90-142 - 1934. Contributions to the development of the Trematoda Digenea. Part n. The biology of the freshwater cercariae in Danish freshwaters. Mem. acad. roy. sci. Danemark, Sect. sci. 9:1— 223 West, A. F. 1961. Studies on the biology of Philaphthalmus gralli Mathis and Léger 1910 (Trematoda: Digenea). Am. Midland Naturalist 66: 363-83 Whitlock, J. H. 1966. The environmental biology of a nematode. In Biology of Parasites, ed. E. J. L. Soulsby, pp. 185-96. New York: Academic Press Williams, H. H. 1968. Phyllobothrium piriei sp. nov. (Cestoda: Tetraphyllidea) from Raja naevus with a comment on its habitat and mode of attachment. Parasitology 58:929-37 Wilson, R. A. 1968. An investigation into the mucus produced by Lymnaea truncatula, the snail host of Fasciola hepática. Camp. Blochem. Physiol. 24:629-33 - 1969. Chemoreception in the miracidium of Fasciola hepática. Parasitology 59:4P (Abstr.) Wilson, R. A., and Denison, J. 1970a. Studies on the activity of the miracidium of the common liver fluke, Fasciola hepática. Camp. Biochem. Physiol. 32:301-13 - 197Ob. Short chain fatty acids as stimulants of turning activity by the miracidium of Fasciola hepática. Camp. Biochem. Physiol. 32:511-17 Wilson, V. C. L. C., and Schiller, E. L. 1969. The neuroanatomy of Hymenolepis diminuta and H. nana. J. Parasitai. 55:261-70 Wood, D. E. 1965. Nature of the end organ in Ophiotaenia filar aides (LaRue). /. Parasitai. 51:541-4 Wright, C. A. 1959a. Host location by trematode miracidia. Ann. Trop. Med. Parasitai. 53:28892

SITE-FINDING BEHAVIOUR IN HELMINTHS

- 1959b. The application of paper chromatography to a taxonomic study in the molluscan genus Lymnaea. J. Linnean Soc. London (Zool.) 44: 222-37 - 1960. Relationships between trematodes and molluscs. Ann. Trop. Med. Parasitol. 54:1-7 - 1964. Miracidial responses to molluscan stimuli. Proc. 1st Intern. Congr. Parásito!. 11:1058-9 Wykoff, D. E., and Lepes, T. J. 1957. Studies on Clonorchis sinensis. I. Observation on the route of migration in the definitive host. Am. J. Trop. Med. Hyg. 6:1061-5 Yokogawa, M. 1965. Paragonimus and paragonimiasis. Advances in Parásito!. 3:99-158 Yoshida, T. 1931. Experimentelle Untersuchungen iiber die Bedingungen fur die Einwanderung von Clonorchis sinensis in den Ductus pancreaticus. Okay ama Igakkwai Zasshi 43:920-36 (Japanese text; German summary)

Discussion R. S. FREEMAN

Department of Parasitology, School of Hygiene, University of Toronto That some free-living stages of parasites respond to stimuli to increase their chances of transferring to the next host came as no great surprise. To discover that researchers reached diametrically opposing conclusions while supposedly evaluating the effect of identical factors on the same species of worms was more of a surprise. Presumably some of this indicates both a lack of rigorous experimental procedures and uncritical analysis of the data obtained. Sensilla-like organs and other presumably sensory structures have been seen on most parasites studied with both light and electron microscopy. But what do such structures sense? What are required, it would appear, are suitable isolated systems which lend themselves to rigorous in vitro experiment. As Schwabe and Kilejian (in: Florkin and Scheer, Chem. Zool. 2:467 [1968]) pointed out, in vivo experiments with flatworms are difficult to

159

perform and control, and more difficult to interpret as effects of isolated variables, and in vitro cultivation until quite recently has lacked sufficient sophistication to yield important results. A major advancement is the relatively recent awareness and acceptance that once parasites reach their hosts, and definitive sites within these hosts, they do not necessarily settle down to life as passive receivers - the truly degenerate organism many textbooks describe - but rather they continue responding to stimuli in various ways including directional movement. Of course complicated migratory patterns by juvenile helminths have been recognized for some time, e.g., for Ascaris lumbricoides. However, even such migrations were, and still are, considered more host-dominated than parasiteselected. Consider, for example, what is meant when one reads that the larvae "are carried to the liver via the mesenteric lymphatics or venules" (Cheng, The Biology of Animal Parasites [Philadelphia: W. B. Saunders, 1964]). Possibly this attitude arose because few host-parasite relationships were recognized where the parasite showed a clearcut series of on-off responses to stimuli, as does for example the nematode Spirocerca lupi (see Thorson, Bioscience 19:126 [1969]). Equally intriguing is the suggestion by Michel (Parasitology 53:63 [1963]) that newly arrived juveniles of the nematode Ostertagia somehow detect that adults of their kind are already in the cow and do not leave the mucosa and enter the gut lumen until the adult nematodes pass out. And what about the recent report by Lie, Basch, and Hoffman (/. Parásito!. 53:1205 [1967]) that rediae of one species of fluke preyed on other flukes in the viscera of snails? Is this a veritable "seek and destroy" mission, or a predatorprey relationship, or perhaps a defence of territory? Also striking are several recent reports of parenteral plerocercoids of cestodes leaving the viscera, penetrating the gut wall, and differentiating to adults in the gut lumen of the same host (Wood, J. Parásito!. 51:541 [1965]; Eckert, von Brand, and Voge, /. Parásito!. 55:241 [1969]; Fischer and Freeman, /. Parásito!. 55:766 [1969]). Plerocercoids of Proteocephalus ambloplitis, for example, migrated when the temperature surrounding the fish host was raised

160

from 4 to 7°c. Presumably the stimulus is more than a simple temperature trigger, however, since some plerocercoids migrated in all large fish, but in only few small fish. This system should lend itself to various types of simple experiments since penetration can be stimulated in vitro (Fischer and Freeman, 1969, op. cit.}. Equally exciting is the recently demonstrated circadian and other migrations of adult cestodes along the gut (Brâten and Hopkins, Parasitology 59:891 [1969]; Read and Kilejian, /. Parasitol. 55:574 [1969]). In short, helminths are now known from all major groups which show specific reactions to a variety of stimuli. Responses of parasitic helminths to pheromones, although such substances were not necessarily called that until recently, have been postulated for some time. This includes, for example, the way small numbers of dioecious and even monoecious helminths locate each other within the host (see Thorson, 1969, op. cit.). Such movements suggest a behavioural or developmental response by one member of a species to substances, the pheromones, produced by another member of the same species. Recently Brown, Eisner, and Whittaker (Bioscience 20:21 [1970]) carried this concept even further, and discussed the need for the term "allomone" to designate substances produced by one species which evoke behavioural or physiological reactions in another species which are adaptively favourable to the emitter. Ablastin produced by the rat and inhibiting sexual reproduction by Trypanosoma lewisi probably qualifies. These same workers suggest the term "kairomone" for a transspecific chemical messenger which elicits an adaptive benefit from the recipient rather than the emitter. Such substances might be sex hormone produced by a frog triggering sexual behaviour by polystomid flukes parasitizing it. Certainly the true appreciation of the role of pheromones, allomones, and kairomones in the life of parasites has only begun, although they probably are important keys to understanding parasitic behaviour in the host.

MARTIN J. ULMER

The physiology and behaviour of the monogenean skin parasite Entobdella soleae in relation to its host (Solea solea) G. C. KEARN

In the past Polystoma integerrimum and Gyrodactylus sp. have invariably been employed in textbooks to illustrate the main features of the anatomy and biology of monogenean (platyhelminth) parasites. Unfortunately these parasites are in many ways highly specialized monogeneans, Polystoma coping with the special problems of life first on the gills of an aquatic tadpole and then in the bladder of an air-breathing amphibian, and Gyrodactylus having suppressed the free-swimming larval stage and invading new hosts presumably by direct transfer of adult parasites. In contrast, the monogenean skin parasite Entobdella soleae (van Beneden and Hesse 1864; Johnston 1929) is less specialized; in the following account of the biology of this parasite an attempt is made to demonstrate its suitability both for teaching purposes and for research in parasitology. Adult specimens of E. soleae are about 3-5 mm in length and are found attached to the "ventral" (lower) surf ace of the host. Although Solea solea has been observed during diving operations in close proximity to pleuronectid and elasmobranch flatfishes on the sea bottom (Dr N. S. Jones, personal communication) E. soleae has been found on only Solea solea (apart from an isolated record by Sproston 1946, who found the parasite on Solea lascar is at Roscoff, France). During my own work I have examined the skin of many hundreds of freshly caught teleost and elasmobranch flatfishes and I have found many infected specimens of S. solea in England both at the Plymouth Laboratory and at the Lowestoft Laboratory, and in France at the Station biologique at Arcachon. Apart from two parasites which most probably transferred themselves to the skin of a ray in a tightly packed trawl, I found E. soleae on only S. solea. The organs which are most conspicuous in living or preserved specimens of E. soleae are the posteriorly situated adhesive organ, the anter-

162

G. C. KEARN

iorly situated pharynx, and the organs of the reproductive system which fill most of the available space in the rest of the body (Fig. 2). But before going on to consider the function of these organs it will be necessary to know something about the behaviour of the host and about the structure of the fish's skin. THE BIOLOGY OF THE COMMON SOLE

Solea solea is a teleost fish adapted for life on the sea-bed. The fish is flattened laterally and lies on the sea bottom on its left (lower) side. The skin of the right (upper) surface contains chromatophores which enable the fish to change colour

and match its surroundings but the lower side contains no chromatophores and is white in colour. Soles are found at depths of from 10 to 70 metres on sandy bottoms (Poll 1947). Kruuk ( 1963 ) studied these fishes in the field and in the laboratory and found that they spend most of the day-time buried in the sand with only the top of the head and the eyes uncovered. During the night they leave the sand and cruise slowly close to the sea bottom. Long excursions away from the bottom, which are a feature of the behaviour of other flatfishes such as plaice (Pleuronectes platessa), are rarely made. This difference in behaviour appears to be related to a difference in

A

BV

C

FIGURE 1. The structure of the skin of the common sole (S. solea). A, longitudinal section through the skin; B, c, a freshly removed upper scale viewed from above (B) and in sagittal section (c) ; D, E, a scale which has been kept in sea-water for about 3 hours after removal from the fish, viewed from

E D above (D) and in sagittal section (E). c., dermal chromatophore; d., dermis; e., fish epidermis; 1., weak line where skin fractures; la., bony lamina of the scale; m., epidermal mucus cell; s., spine. (From Kearn 1967a)

PHYSIOLOGY AND BEHAVIOUR OF ENTOBDELLA SOLEAE

feeding habits, soles feeding mainly at night and by touch, using sensory filaments on the head, and plaice relying more heavily on sight for finding food (Bateson 1890). The skin of the sole consists of two layers, the outer epidermis and the inner dermis, which differ from each other in their component parts and in their properties (Fig. 1 ). The epidermis is a delicate layer easily damaged if the fish is handled roughly. It consists of six or seven layers of epidermal cells together with superficial mucus cells which produce the layer of slime covering the body of the fish. The dermis is thicker than the epidermis and less easily damaged since it contains a good deal of fibrous tissue as well as nerves and blood vessels. Furthermore the dermis produces the overlapping bony scales which cover the upper and lower surfaces of the fish. Each scale consists of a proximal lamina enclosed in a dermal pocket and a distal spiny region which projects above the general surface of the skin but is nevertheless covered by a tightly adhering thin layer of dermis and epidermis. The bony lamina of the scale is not attached to the surrounding dermal tissue so that by grasping the free spiny end of a scale with forceps and exerting a slight pull the skin fractures and the scale can be withdrawn from its pocket (Fig. IB, c). Damage to the skin of a fish induces increased cell division in the epidermis and a migration of some of the epidermal cells across the wound; the dermis, however, has poor powers of regeneration after injury. This can be clearly demonstrated in S. solea by withdrawing a scale from its dermal pocket, immersing the scale in sea-water, and then observing the skin attached to its distal end at intervals (Kearn 1967a). After only 3 hours at 17° c the small patch of epidermis at the spiny end of the scale has spread down the lamina almost to its proximal border while the dermis has not migrated (Fig. ID, E; Fig. 17). The area of skin covering the eye of the sole forms the transparent cornea. The epidermal component (the conjunctiva) consists of uniform flattened cells and contains no mucus cells. The

163

dermal component of the cornea contains no blood vessels. THE BIOLOGY OF Eutobdella soleae Movement and the attachment mechanism* The adult parasite (Fig. 2) has a flattened leaflike body and a posterior cup-shaped adhesive organ (haptor) which the parasite uses to attach itself to the skin of its host. By placing a fish in a glass-bottomed tank and plotting the positions of the parasites on the lower surface of the fish every few hours, it has been demonstrated that E. soleae is able to move about on the skin of the fish. Parasites removed from the fish will attach themselves by means of the haptor to glass and this permits more detailed observation of movement. Movement is aided by two adhesive pads, one on each side of the head, which are used for temporary attachment while the position of the haptor is being changed. It is not unusual for parasites in a glass dish to take two or three successive "steps" by using the haptor and adhesive pads alternately. The effectiveness of the adhesive secretion can be readily demonstrated by attempting to dislodge, with a strong jet of water from a pipette, a parasite attached only by its adhesive pads. There is little doubt that the adhesive properties of the secretion are more than adequate to cope with the weaker water currents flowing over the surface of the host in nature. Lyons (1970) has found that the region of the epidermisf of the adult parasite penetrated by the ducts of the adhesive gland is, *A detailed account of attachment has been published by Kearn (1964). tLyons has shown in the same paper that the outer covering of monogeneans, like that of cestodes and digeneans, is a living layer containing mitochondria and inclusions (probably mucoproteinaceous). The nuclei of this syncytial epidermal covering lie in cell bodies which are situated in the parenchyma and communicate with the outer epidermal layer by means of cytoplasmic connections.

164

in sections prepared for the electron microscope, more dense than the epidermis covering the rest of the animal. Furthermore, the epidermis covering the anterior adhesive pads is separated from the syncytial epidermal covering of the rest of the body by cell boundaries, and bears numerous short microvilli. Lyons suggests that these microvilli may serve to spread the adhesive secretions over the skin of the host into a thin "tacky" film and may help to mix the products of different gland cells which may have to interact with each other or with water before adhesive properties develop. The adhesive pads on the head are used for temporary attachment during locomotion and the parasite spends most of its time attached to the host's skin by means of the haptor. This cupshaped adhesive organ bears three pairs of median hooks and seven pairs of relatively small marginal booklets (Fig. 2). Lyons (1966) has shown that these sclerites consist of a scleroprotein resembling keratin. There is evidence from studies of the embryonic development of the median hooks (seeKearn 1963a) that the members of the anterior pair (accessory sclerites) develop from an extra, centrally placed, eighth pair of marginal booklets, which unlike those of the other seven pairs continue to grow and change their shape. The other two pairs of median hooks are hamuli, i.e. extra sclerites which develop later than the marginal booklets and take over the main job of attachment as the parasite grows (see Llewellyn 1963 ). The posterior hamuli are relatively small in the adult and probably play only a small part in attachment. The accessory sclerites and the anterior hamuli become functionally associated in the adult to bring about more effective attachment. The orientation of these large median sclerites can be studied by viewing the inside of the unflattened, cup-shaped, adhesive organ with a stereomicroscope. The stout, slightly curved shafts of the anterior hamuli run in a longitudinal direction across the roof of the cup. Posteriorly these sclerites taper and form hooks which protrude from the tissues of the haptor. In contrast,

G. C. KEARN

the shafts of the accessory sclerites are orientated almost perpendicularly to the plane of the roof of the attachment organ. Their distal halves protrude from the haptor and are also hook-like in shape. Further information on the mode of action of the adhesive organ can be obtained by studying the living parasite attached to the skin of its host. If the edge of the haptor is lifted with a sharp needle, the cup, except for the posterior border of the disk which is effectively pinned to the skin by the hooked distal regions of the two anterior hamuli, separates immediately from the skin. This indicates that suction plays a part in attachment and shows that although the accessory sclerites are hook-like in shape they do not penetrate the skin but act as props and simply push against the skin of the fish. This interpretation is further supported by the appearance of the "footprint" left on the skin of the sole at the site of attachment of the haptor. The region of the fish's skin previously enclosed by the haptor is raised to form a "blister" which clearly bears centrally the indentations made by the "props" (Fig. 18). The way in which the "props" and the large hooks are used to generate suction may best be appreciated by considering a longitudinal section through the attached adhesive organ (Fig. 3). Lying at the posterior end of the body is a pair of prominent muscles. Each of these gives rise to a long tendon which runs into the haptor towards the ventral surface of the cup. Here the tendons pass beneath a bridge of transverse fibres which are attached at each end to the ventral integument; each tendon then passes through a notch at the proximal end of the prop and becomes attached to the proximal end of the long anterior hamulus. When the extrinsic muscles contract, the anterior ends of the shafts of the anterior hamuli and the roof of the cup in which they are embedded are lifted, thereby reducing the pressure inside the sea-water-filled cup. The suction force generated will be proportional to the pull exerted by the extrinsic muscles. The transverse fibres

FIGURE 2. The anatomy of an adult specimen of E. soleae (ventral view), ac.s., accessory sclerite; ad. a., adhesive area; a.ha., anterior hamulus; b., bladder; e., egg; e.d., ejaculatory duct; e.m., extrinsic muscle; e.r., external reservoir; ex.p., excretory pore; ey., eye; f., transverse fibres; g.G., glands of Goto; h., haptor; i., intestine; i.r., internal reservoir; lu., lumen of pharynx; lp., lip tucked inside the

pharyngeal lumen; m., mouth; m.h., marginal hooklet; m.v., marginal valve; oot., ootype; ov., ovary; p., "penis"; p.ha., posterior hamulus; ph.g., pharyngeal gland cell; sp.d., efferent duct from spermatophore gland; s.r., seminal receptacle; t., testis; te., tendon; v., vagina; v.f., vitelline follicle; v.o., vaginal opening; v.r., vitelline reservoir; v.s., vesícula seminalis.

166

G. C. KEARN

which bind the tendons to the ventral surface of the anterior region of the haptor will also be lifted and so contribute towards spreading the load more evenly over the roof of the haptor. The lifting of the roof of the adhesive organ is made possible by the prop-like accessory sclerites. Clearly it is important that these sclerites penetrate as little as possible into the skin of the fish. Deep penetration is not possible because of the bony scales in the skin and penetration of the superficial layers of the skin may be prevented by the blunt distal ends of the sclerites. There are two additional problems associated with a cup-shaped suction device of a flexible na-

ture. If the roof of the cup were lifted the suction force would be reduced by any flow of water under the edge of the cup into the central cavity and by inward movement of the edges of the disk. E. soleae has overcome the problem of influx of water by means of a flexible flap or valve around the edge of the disk. Furthermore, the marginal booklets, although they do not grow during postlarval life, are retained in the adult around the edge of the cup and the points of these booklets protrude in a ventral direction from the tissue of the haptor. It seems possible that they may serve to pin down the edge of the disk and prevent inward movement of the edges of the haptor.

FIGURE 3. A diagrammatic parasagittal section through the adhesive organ of E. soleae. ep., epidermis of fish; pe., peduncle. The arrows show the direction of movement of the tendon when the ex-

trinsic muscle contracts. The posterior hamulus has been omitted. Other lettering as in Figure 2. (Redrawn from Kearn 1964)

PHYSIOLOGY AND BEHAVIOUR OF ENTOBDELLA SOLEAE

167

FIGURE 4. A diagrammatic sagittal section through a feeding parasite, ep., epidermis of fish; i., intestine; lp., pharyngeal lip; ph.g., pharyngeal gland cell.

Feeding* The mouth of E. soleae lies on the ventral side of the head region and opens into a spacious cavity containing the pharynx (Fig. 2). The pharyngeal lumen opens posteriorly into a short oesophagus which divides into two gut caeca each of which runs longitudinally down one side of the body and gives off short side branches. There is no anus. The anterior opening of the pharyngeal lumen is surrounded by a fibrous lip which is tucked inside the lumen when the parasite is not feeding. The pharynx wall contains large elongated gland cells running longitudinally. Anteriorly each gland cell opens at the tip of a papilla which projects into the pharyngeal lumen, and radial and circular fibres are associated with the opening at the tip of each papilla. The glands have a strong proteolytic activity. This can be demonstrated by exposing the pharyngeal glands by cutting across a freshly frozen animal with a razor blade and placing the cut surface in contact with the gelatin layer of a piece of photographic film. Pharyngeal gland tissue, in contrast with the tissues of the head surrounding the pharynx, readily dissolves the gelatin in contact with it. Occasionally a preserved specimen is found in *A detailed account of feeding has been published byKearn (1963b).

which the mouth is wide open and the fibrous lips surrounding the anterior opening of the pharynx are everted and protrude through the mouth;the application of pressure to the anterior region of the living parasite sometimes brings about a similar eversión of the pharynx. Living parasites which have been detached from the host and kept for some time in a glass culture vessel will often push the pharynx out through the mouth and apply it to the bottom of the vessel, suggesting that eversión of the pharynx may play an important part in feeding. This has been confirmed by starving parasites by separation from the host for up to 24 hours and replacing them on the upper skin of a freshly killed sole. Soon after replacement on the skin the pharynx is everted and applied to the skin of the fish so as to enclose a circular area of skin (Fig. 4). The protruding bell-like pharynx is held in contact with the skin for about 5 minutes (at about 19° c) and the proximal region of the pharynx undergoes peristaltic contraction. As a result of this a colourless fluid, containing no resolvable particulate matter except for an occasional black particle, is pumped from the pharynx through the oesophagus into the intestine. During feeding the anterior adhesive pads on the head are not used for attachment. Then the pharynx is detached and retracted into its cavity, the liquid in the gut caeca sometimes continuing to oscillate back and forth for a few minutes after the cessation of

168

G. C. KEARN

A

B

FIGURE 5. The main paths of water currents produced by undulating body movements in E. soleae. A, ventral view; B, lateral view. (Simplified and redrawn from Kearn 1962)

feeding. Since the dermal chromatophores in the circular area of skin recently enclosed within the pharynx are now more clearly visible (Fig. 19), it is evident that superficial host tissue has been eroded; this may be confirmed by histological sectioning of the feeding wound. The parasite erodes only the superficial epidermis and the dermis beneath is not attacked (Fig. 20). In a search of a Bouin-preserved fish carrying more than 100 parasites,* 21 circular epidermal wounds were found. Their size and appearance and the close proximity of all but two of these * Infested soles caught in the English Channel near Plymouth carry three (one to nine) adult parasites per fish but when infected soles are kept in a small aquarium tank in the laboratory the population of parasites builds up steadily through repeated reinfection so that eventually they may carry as many as two hundred parasites.

wounds to attached parasites give every indication that these are feeding wounds. The depth of these superficial skin wounds varies : some are deep wounds in which all the epidermis has been eroded, and others are shallower with a thin covering of epidermis over the floor of the crater. The shallow wounds may have been freshly made by parasites which stopped feeding before they reached the dermis, or they may be old wounds in which epidermal migration and regeneration have begun. Evidence of the speed at which sole epidermis migrates has been presented above. Berlin (1951) reported that in the loach (Misgurnus fossilis) deep skin wounds 25 to 100 mm2 in area and involving both epidermis and dermis were completely covered by migrating epidermis in as little as 24 to 36 hours, and the epidermis had returned to its original thickness after 3 to 4 days. The dermis of the loach, however, regen-

PHYSIOLOGY AND BEHAVIOUR OF ENTOBDELLA SOLEAE

crates slowly and Berlin found that it was still not completely reconstituted after 4 months. Thus any damage inflicted on a host by an epidermal skin feeder would be repaired rapidly by the host and any danger of host infection by micro-organisms would be minimal. Breathing movements Rhythmical undulating movements of the body are a prominent feature of the behaviour of living specimens of E. soleae attached either to the host or to the glass bottom of a dish (Kearn 1962). The addition of small amounts of finely powdered carmine to the sea-water surrounding an undulating parasite shows that these movements draw water from the region behind the parasite and propel it forwards beneath the body and out in the head region (Fig. 5 ). Furthermore, at constant temperature the rate of undulation and the amplitude of the body waves change in relation to the oxygen content of

Oxygen (ml/1)

FIGURE 6. The relationship between the rate of body undulation and the oxygen content of the medium in three separate adult individuals of E. soleae. (FromKearn 1962)

169

the ambient sea-water. In sea-water deoxygenated by passing nitrogen through it, the rate of undulation is very rapid and the amplitude of the waves is large (Fig. 6). In contrast, in welloxygenated sea-water the rate of undulation and wave amplitude are much reduced and indeed the undulating movements are much less noticeable when the parasite is viewed with a stereomicroscope. In normally oxygenated sea-water (i.e. water containing 6 ml of oxygen per litre) at 10° c the rate of undulation in an adult specimen is approximately thirty-five waves per minute. A rise in temperature of 10° c approximately doubles the rate of undulation, and carbon dioxide dissolved in the ambient sea-water has a narcotic effect so that the rate of undulation decreases rapidly and the parasites become detached from the substrate. The responses of the parasite to water containing different amounts of dissolved oxygen indicate that the undulating movements of E. soleae have a breathing function and that conditions of abnormally low oxygen concentration may prevail at the sea bottom where the host lives. Indeed present knowledge indicates that the oxygen content of marine sediment and of the sea-water just above it is abnormally low (provided that there are no disturbing bottom currents) over large areas of ocean bottom (Perkins 1957; Richards 1957). Consequently a parasite such as E. soleae living on the lower surface of a somewhat sedentary, bottom-dwelling, flatfish is likely to require some kind of breathing movement to obtain sufficient oxygen for its needs, whereas gill-parasitic monogeneans such as Gastrocotyle trachuri have no problem since they are continually washed by the host's gill ventilating current (Llewellyn 1964). There is also evidence that E. soleae can adjust itself to low ambient oxygen concentrations in another way. When the oxygen content of the surrounding water is high the surface of the parasite is wrinkled and internal organs cannot be seen clearly. On transfer to water containing little dissolved oxygen the length and breadth of the parasite increase (Fig. 7) and the wrinkles

170

G. C. KEARN

disappear so that the internal organs can be seen clearly. Extension of the body in the facial plane will permit the parasite to move a greater volume of sea-water and bring more oxygen into contact with the body, and if there is in addition a stretching of the integument this will increase the surface area of the parasite and also decrease the diffusion distance to deeper-seated tissues.

Reproduction and host-finding Sperm exchange (seeKearn 1970) The parasite has a protrusible penis-like organ and a vagina (Fig. 2 ) but the distal region of the latter is so narrow (little more than 10 /¿m in diameter in the living parasite) that without considerable distension it would be unable to accommodate the much larger "penis." Furthermore, the "penis" does not serve as an outlet for

Oxygen (ml/l)

FIGURE 7. The relationship between "surface area" (see below) and the oxygen content of the medium in three individuals of E. soleae. An index of surface area was computed by multiplying the maximum body length and the breadth of the body. This index is represented on the graph as a proportion of the "surface area" in normally oxygenated seawater (x on graph). (From Kearn 1962)

sperms only; inside the muscular wall enclosing the "penis" sac is a conspicuous reservoir containing a jelly-like material. Continuous observation of living parasites reveals that at intervals of 4 hours at 16° c the "penis" is protruded and extrudes a spermatophore consisting of a centrally situated mass of sperms enclosed by the jelly-like material. These spermatophores, sometimes cylindrical in shape but frequently rolled up to form a spherical mass, are found frequently lying freely among the eggs laid by adult specimens of E. soleae. The spermatophores have no rigid outer wall and stick readily to glass or to a living parasite. The sperms of E. soleae are very long. Estimates of the lengths of relatively straight sperms in squashed spermatophores vary from 180 to 220 ¡mm. Dr K. M. Lyons (personal communication) has found that each sperm contains two axial units, each consisting of nine peripheral pairs of microtubules and a single central microtubule (the so-called 9+1 fibre arrangement). In this respect the sperms of E. soleae resemble those of other platyhelminths (Burton 1967). Surprisingly these freely deposited spermatophores are not picked up by other parasites. This has been investigated by allowing about twentyfive parasites, freshly removed from the fish, to reattach themselves to a circular coverglass. This coverglass was placed with the parasites attached to its upper surface on a smaller circular Perspex boss so that the coverglass overlapped the Perspex. The attached parasites were not able to negotiate the overlapping edge of the coverglass and so the overcrowded parasites were unable to disperse. Because of their close proximity to one another the chances of their picking up spermatophores or exchanging sperms was increased. In spite of this none of the spermatophores shed into the surrounding water was picked up; in fact it was found that spermatophores are passed directly from one individual to another. Mutual exchange of spermatophores takes place during a lengthy mating ceremony (lasting for about 30 minutes) which begins when the head regions of neighbouring parasites come into contact. The

171

PHYSIOLOGY AND BEHAVIOUR OF ENTOBDELLA SOLEAE

A

B

C

FIGURE 8. Successive stages in the mating behaviour of E. soleae (ventral view). p.s., "penis" sac; sp., spermatophore; v.o., vaginal opening. (From Kearn 1970) region of the body containing the vagina in each parasite then becomes stretched so that this region of the animal (the left side of the body) appears more transparent than the rest of the body in transmitted light (Fig. SA) . After the heads of the two animals have touched many times, the head regions become loosely interlocked in such a way that the opening of the "penis" of each individual lies next to the vaginal opening of the other. This is achieved by tucking the head below the left edge of the body of the other parasite (Fig. SB; Fig. 15). Mating parasites may separate and after a few seconds reassociate, but when the parasites separate for the last time each parasite carries a spermatophore

attached to the outside of the body in the region of the vaginal opening (Fig. 8c, 9A). These spermatophores are not difficult to dislodge and there seems to be no insertion of the "penis" in the vagina during mating. Soon after mating the vaginal region of each individual contracts strongly at intervals of 5 to 10 minutes (Fig. 9fi), and these movements serve to force the spermatophore, or at least the central core of sperms, into the narrow distal vaginal canal. The proximal part of the vagina is a convoluted tube in which inactive sperms are stored (Fig. 2). This convoluted storage chamber opens into the vitelline reservoir. Spermatophore exchange may take place be-

172

G. C. KEARN

A

B

FIGURE 9. Assimilation of a spermatophore by E. soleae. A, animal seen in lateral view with attached spermatophore. B, attitude adopted by parasite while sucking spermatophore into the vagina, g., common genital opening; sp., spermatophore. (From Kearn 1970)

tween young parasites which have a fully functional spermatophore-making apparatus but no vitellaria, and also between fully mature, egglaying adult parasites. Unions between a young and an old parasite have not been observed. Parasites which have insufficient spermatophore material to make a spermatophore themselves can still take part in mating and receive a spermatophore from another individual. Although it seems mechanically possible for a parasite to deposit a spermatophore on its own ventral surface, no attached spermatophores were found on parasites which had been isolated from their fellows for some time, and the inference is that self-fertilization is rare or never takes place. The muscular "penis" sac in which the spermatophores are assembled encloses the reservoir containing the jelly-like spermatophore matrix

and the vesicula seminalis containing sperms from the testes (Fig. 2). The reservoir opens into an ejaculatory tube running the length of the "penis" and opening just short of its distal rounded tip. A sphincter muscle controls the efflux of matrix from the reservoir into the ejaculatory duct and the vesicula seminalis has a narrow opening into the ejaculatory tube. Some clues about the way in which spermatophores are assembled can be obtained by applying pressure to the anterior region of a living parasite. This treatment causes protrusion of the "penis" and the ejection of a spermatophore containing the same relative amounts of matrix and sperm as in freely laid spermatophores. In parasites preserved while undergoing this treatment and stained subsequently in Ehrlich's Haematoxylin it can be seen that sperms from the vesicula

PHYSIOLOGY AND BEHAVIOUR OF ENTOBDELLA SOLEAE

173

FIGURE 10. Embryonated eggs of E. soleae attached to sand grains, ap., appendage; e., egg; o., oncomiracidium; op., operculum; s.g., sticky globules; sa.p., sand particle. (FromKearn 1963a)

seminalis are injected into the matrix as the latter is pushed along the ejaculatory tube. This indicates that the spermatophore is assembled immediately prior to ejection simply by a contraction of the muscular wall surrounding the "penis" sac, bringing about an injection of sperms into the jelly as the jelly itself is extruded. Outside the muscular "penis" sac is a second reservoir communicating by a narrow duct with the internal reservoir (Fig. 2). Ducts running longitudinally and carrying spermatophore material from each side of the body converge on the external reservoir and open into it. The spermatophore gland itself has not yet been positively identified but it seems likely to be a diffuse follicular gland interspersed with the vitellaria. Spermatophores of a different kind have been described in Entobdella diadema from the skin of

the sting ray Dasyatis pastinaca by Llewellyn and Euzet (1964). These spermatophores are spindle-shaped and have a rigid outer wall. The terminal male organ which makes these spermatophores also differs from that of E. soleae. The protrusible distal region works in the manner of a cirrus, there are two reservoirs of spermatophore-forming material inside the cirrus sac (no external reservoir), and the sperm itself is stored inside the sac in a spindle-shaped region of the vesícula seminalis which communicates with the more distal of the two internal reservoirs. The way in which the spermatophores of E. diadema are exchanged is not known but the vagina of E. diadema has a spacious distal region (cf. E. soleae, Fig. 2) which could well accommodate the spermatophore and perhaps digest its wall to liberate the sperms.

174

FIGURE 11. The oncomiracidium of E. soleae (ventral view), a.e.c., anterior longitudinal excretory canal; c., collar surrounding mouth; ci., cilia; d., domus; d.g.c., ducts of gland cells; f.c., flame cell; g.c., gland cells; 1. lens of eye; p.e.c., posterior longitudinal excretory canal; ph., pharynx; pi.c., pigment cup. Other lettering as in Figure 2. (From Kearn 1963a)

G. C. KEARN

PHYSIOLOGY AND BEHAVIOUR OF ENTOBDELLA SOLEAE

The egg

It seems probable that sperms are transported or perhaps make their own way from the storage area in the vagina via the vitelline reservoir to one to four (usually two) spherical seminal receptacles, each of which opens into the oviduct. The sperms stored in the seminal receptacles usually are active. Careful examination of unflattened specimens reveals the presence of active sperms in the anterior part of the ovary among the fully developed oocytes, and therefore fertilization may take place before the egg cell leaves the ovary. On reaching the ootype the oocyte accompanied by vitelline cells is churned back and forth for one or two minutes. Then the churning process ceases and the egg assumes a tetrahedral shape imparted by four pads on the inner wall of the ootype (Kearn 1963a). The vitellaria have all the components of a quinone-tanning system, and the freshly laid egg (Fig. 10) is pale yellow in colour, rapidly darkening after laying until the shell is brown. Each side of the tetrahedron measures about 165 /¿m and the egg-shell material is prolonged at one of the corners to form a slender appendage (about 880 jam long) which at intervals along its length bears sticky droplets (Fig. 10). It can be demonstrated that the eggs laid by parasites attached to the lower surface of the host do not adhere to the surface of the host. Eggs laid by parasites on an infected sole kept in a shallow glass tank all become attached to the surface of the glass below the fish. (This is a useful way of collecting large numbers of eggs for experimental work. ) Adult parasites lay eggs at the rate of two per hour at 15° c but often complete detachment of the eggs from the parasite is delayed because the appendage, which trails behind the egg as it passes up the uterus, remains lodged in the uterus. Specimens are frequently found with two or three tetrahedral eggs hanging freely in the surrounding water but still attached to the parasite by their appendages. Parasites with eggs protruding from the uterus frequently lift the head region above the substrate in such a

175

way that the eggs are uppermost. Such movements may assist the separation of the eggs from the parasite when the parasite is attached beneath the fish and may serve to implant the eggs in the sand. When the parasite is attached to glass with no suitable surface above on which the eggs may be implanted, the egg appendages are eventually released and the eggs fall to the bottom and become attached to the glass by their sticky droplets. The oncomiracidium When incubated at 14° c the eggs of E. soleae develop into a ciliated larva (oncomiracidium). In just less than four weeks one of the corners of the tetrahedral egg becomes detached, permitting the larva to swim out into the surrounding water (Fig. 10). At temperatures from 9 to 14° c (annual range of bottom temperatures at depths of 60 metres in the English Channel) the oncomiracidium continues to swim for up to 24 hours. The larva dies if it fails to find a host within that time. A detailed description of the oncomiracidium has been published elsewhere (Kearn 1963a), but several features of this larva are of special interest since they may contribute to the hostfinding mechanism and to early life on the host. The oncomiracidium bears a large number of ciliated epidermal cells which are arranged in three areas (Figs. 11, 16). There is a ciliated area on the anterior region of the body, another on the posterior region of the body, and a third on the haptor. Recently Dr K. M. Lyons (personal communication) has had success with a silver-staining technique for outlining the ciliated epidermal cells of marine larvae and this has revealed that my early estimates of the numbers of ciliated cells, based on counts of cells which became detached from a living unstained larva (Kearn 1963a), were too low. Dr Lyons counted a total of 53 epidermal cells on the body, 30 cells forming the anterior ciliated band and 23 cells forming the posterior body band. She was unable to obtain an accurate estimate of the cells in the

176

G. C. KEARN

Solea solea (3)

Buglossidium /ufeum (5)

¿/manda ¡imanda (3) P/euronecfes p/afessa (1) Rai a sp. (3)

Length of ant. hamulus [um] Body length [mm]

FIGURE 12. The numbers of parasites of various sizes collected from the skin of flatfishes kept in a crowded tank with infected specimens of S. solea. The number of individuals of each species of fish examined is given in parentheses. The length of the anterior hamulus is used as a measure of the age of each parasite, and for convenience in interpretation a scale showing approximate body length is provided. The arrow indicates the size at which parasites begin to lay eggs. (Redrawn from Kearn 1967a)

PHYSIOLOGY AND BEHAVIOUR OF ENTOBDELLA SOLEAE

ciliated band on the haptor, and my earlier estimate of nine cells seems likely to be too low. The eyes, each consisting of a transparent lens partly enclosed by a pigment cup, are very conspicuous and are fully developed in the larva (Fig. 16), indicating that they may play an important part in the behaviour of the oncomiracidium. Although they undergo no further growth they are retained by the adult. In the free-swimming larva the haptor is folded up so that the ventral edges of the cup are in contact with each other and the various sclerites differ in proportional sizes and in shape from those of the adult. The marginal booklets are relatively large in the oncomiracidium and make active gaffing movements. None of the median sclerites reach their definitive size in the larva but the posterior hamuli, which are relatively small in the adult, have well-developed hooked regions and are more prominent than the anterior hamuli and accessory sclerites. This suggests that the marginal booklets and the posterior hamuli play an important part in the attachment of the larva, later to be superseded by the anterior hamuli and by the accessory sclerites. Although the extrinsic muscles and tendons associated with the accessory sclerites are visible in the larva, it is most unlikely that the adult's attachment apparatus is operative at this stage since the hooked region of the anterior hamulus is not yet present. The oncomiracidium has three kinds of gland cells giving rise to ducts opening on the anterolateral borders of the head region. The most prominent of these glands have long lateral ducts running forward from two groups of five cells, one such group lying on each side of the body just behind the eyes. The other two kinds of gland cells lie close together in the mid-line anterior to the eyes and their ducts follow different courses to the anterolateral borders of the head region. The alimentary system is represented by the pharynx and the intestine, which at this stage is ring-shaped with no lateral diverticula, and there are nine pairs of flame cells with ducts connecting them with the excretory bladders.

177

Host-finding (for a detailed account see Kearn 1967a) Although E. soleae is strictly specific to its host, S. solea (see Introduction), the freshly hatched larva of E. soleae is likely to encounter a variety of both teleost and elasmobranch fishes on the sea bottom. Flatfishes caught in the trawl together with S. solea at Plymouth include other soleid fishes (the thickback sole S. variegata and the solenette Buglossidium luteum), pleuronectid fishes (the plaice Pleuronectes platessa, the dab Limanda limanda, and the flounder Platichthys flesus), and various elasmobranch fishes of the genus Raia. Therefore the larvae of E. soleae either attach themselves selectively to S. solea, or are less selective but are unable to survive on alien hosts. That the larvae show a strong preference for S. solea can be demonstrated by maintaining a variety of flatfishes in a small laboratory tank including 5. solea infected with adult specimens of E. soleae. Over a period of several weeks all the fishes in the tank are exposed to large numbers of free-swimming larvae. The numbers of parasites of different sizes can then be studied by scraping the skin of each fish and picking out the parasites with a stereomicroscope. In such an experiment scrapings from three common soles yielded a total of 194 attached parasites, including large numbers of freshly attached larvae (differing from the free-swimming larvae only in the absence of ciliated cells) as well as adults (Fig. 12). The soleid fish B. luteum was far less attractive to the larvae than S. solea but nevertheless acquired small numbers of young parasites (33 individuals from five fishes, three of which were oncomiracidia; no adults were found, see following section). The pleuronectid fishes carried few parasites (eight parasites from four fishes) and all of these were well-developed individuals or adults. Only one parasite was found in skin scrapings from three rays. The free-swimming larvae of E. soleae attach themselves readily to small pieces of detached common sole skin (Fig. 17), which can be ob-

178

tained most conveniently by extracting scales from their dermal sockets with a pair of forceps (see Fig. 1 and associated text). This not only shows that it is the skin and not the whole fish to which the larvae respond but permits detailed observations of the invading larvae. The larvae first attach themselves to the skin by means of the head region, presumably using the secretions of one or perhaps more than one of the groups of gland cells situated in the anterior region of the body (Fig. 11). The larva then unfolds the haptor and attaches itself to the skin by means of the hooks, and the head region becomes detached. About 20 seconds after attachment by the haptor the ciliated epidermal cells begin to fall away from the body and after a further 30 seconds all the ciliated cells have been lost. Scales or small pieces of excised skin from two or three different species of flatfishes were offered to large numbers of free-swimming larvae and this confirmed that the larvae have a strong preference for the skin of S. solea. Out of 156 larvae 140 preferentially attached themselves to the scales of S. solea in a total of 21 separate two-choice and three-choice experiments. A much weaker response to the skin of related soles was found (nine larvae on the skin of thickback soles and four on the skin of the solenette), and the skin of pleuronectids was unattractive (plaice, one attached larva; dab, two attached larvae). The larvae of E. soleae are able to make the correct choice in total darkness when the scales of two or three different fishes are offered. Thus even though the eyes of the larva are well developed they play no part in the close-range identification of the host fish, although they may well influence the orientation of the larva prior to the host selection and so have an indirect role in host-finding. The selective behaviour of the invading larvae of E. soleae indicates that the larva may respond to specific chemical substances produced by the fish's skin. This has been investigated by absorbing the secretions from the skin of an immobilized living sole with a sheet of agar jelly held in contact with the skin, and offering the impreg-

G. C. KEARN

nated jelly together with unimpregnated controls to free-swimming larvae. In such experiments the impregnated jelly was arranged so that the larvae had access only to substances diffusing through the jelly and not to epidermal cells adhering to the jelly. In 16 experiments a total of 212 larvae attached themselves to the impregnated jelly and cast off their ciliated cells and only 19 larvae attached themselves to the unimpregnated controls. In a separate experiment larvae showed no response to jelly impregnated with mucus from Platichthys flesus whereas 82 larvae attached themselves to jelly impregnated with S. solea mucus offered in the same dish. It is possible to isolate sole epidermal tissue by allowing the epidermis to migrate down the lamina of a detached scale (Fig. ID, E) . Since the dermis does not migrate, the distal region of the scale carrying the dermis can be cut off (along the line xy in Fig. 1 ) and discarded and the proximal lamina bearing only epidermis can then be offered to free-swimming larvae. Using this technique it has been found that the larvae attach themselves to isolated epidermis just as readily as they do to epidermis and dermis together. In three such exepriments 33 larvae attached themselves to isolated epidermis and only two larvae to intact scales boiled to destroy their chemical attraction. Thus it seems that the attractive substance or substances are produced by the epidermis of the sole and not by the underlying dermis. Furthermore, the larvae show no response to sole cornea (modified epidermis containing no mucus cells), indicating that the attractive substance is a component of the mucus secreted by mucus cells in the epidermis. Host-specificity It might be supposed that adult skin parasites transferred experimentally to the skins of alien hosts would be able to attach, feed, and survive on their new hosts, but experiments have shown that this is not so (Kearn 1961 a). Adult specimens of E. soleae survive transfer from one specimen of S. solea to another and remain attached

PHYSIOLOGY AND BEHAVIOUR OF ENTOBDELLA SOLEAE

for several weeks. The parasites attach readily to glass and survive without food for 2 to 6 days at 14 to 17° c. In contrast, parasites placed on the skin of the solenette and on plaice and dabs invariably become detached after only 24 to 30 hours. Perhaps the adhesive organ is unable to function properly on the skin of these alien hosts or the foreign skin may produce substances which are repellent to the parasite. The solenette is of special interest because the skin shows some attraction for oncomiracidia (see p. 178) and yet is inhospitable to the adults (see also the results of long-term infection experiments with solenettes, Fig. 12 and p. 177). Surprisingly parasites which were transferred to rays survive for longer periods (2 to 8 days) even though the skin is unattractive to oncomiracidia (see p. 177). The similarity between the survival time on ray skin and on glass indicates that the parasite is probably unable to feed on ray skin even though the attachment organ can operate successfully. It seems therefore that physiological rather than morphological adaptations to S. solea prevent the survival of adult parasites on alien hosts. These adaptations may have been brought about by continued isolation on S. solea as a result of the response of the invading larvae only to specific substances secreted by the host's skin, but it is also possible that specificity in host-finding evolved because of prior physiological adaptation to a single host species. Post-oncomiracidial migration Although adult specimens of E. soleae are almost invariably found on the lower white surfaces of soles, it has been found by scraping the skin both of freshly caught and of laboratorykept heavily infected fishes that immature parasites (below about 1.5 mm in length) occur on the upper pigmented surfaces of the fish (Fig. 13 and Kearn 1963c). Moreover, more young parasites were found on the head region of the upper surface than elsewhere. It has been shown that the free-swimming

179

larvae show no preference for dorsal skin (Kearn 1967a) ; in three experiments they attached in approximately equal numbers to dorsal skin (43 attached larvae) and ventral skin (31 attached larvae). It seems therefore that larvae invade the upper surface of the sole not because it is chemically more attractive but because this is the only part of the fish exposed to the free-swimming larvae when the sole is lying on the bottom or in the sand with just the eye region of the head exposed (Fig. 14). The migration from the upper to the lower surface of the host (Fig. 14) corresponds with the onset of sperm production (at a length of 1.5 mm, Kearn 1963a) and egg production begins at a length of about 2 mm. It is possible that a pair of organs called the "glands of Goto" (see Fig. 2) may play an important part in early development. These organs are situated in the adult in the posterior angle between the testes, but in small post-oncomiracidia from the upper surface of soles they are fully developed before the testes appear (Kearn 1963a). Nervous system Very little is known about the nervous system and sense organs of E. soleae or indeed about those of any other monogeneans. Peripheral sense organs must play a considerable part in attachment, movement, feeding, and particularly in host-location and mating, and recent investigations by Lyons (1969a, b) have thrown some light on the nature of some of these organs. Like many other invertebrates, monogeneans make extensive use of cilia in their peripheral sense organs and each receptor unit or sensillum consists of a terminal cilium embedded in a sensory neurone. These sensilla may be isolated or grouped together to form compound sensilla and the sense organs are sealed into the epidermis by means of septate desmosomes. Single isolated sensilla have been found in the adult epidermis of E. soleae and in the oncomiracidium between the bands of ciliated cells on the

180

G. C. KEARN

Upper surface

Lower surface

Lower surface of the head

Number of parasites Upper surface

Lower surface

Lower surface of the head

Number of parasites

FIGURE 13. The number of parasites of various sizes on the surfaces of (a) 28 freshly caught soles and (b) a single heavily infected sole. (From Kearn 1963c)

PHYSIOLOGY AND BEHAVIOUR OF ENTOBDELLA SOLEAE

body. The ultrastructure of the projecting cilia in those isolated receptors so far studied closely resembles that of typical motile cilia, that is each cilium contains a single central pair of longitudinally orientated microtubules surrounded by nine peripheral pairs of microtubules (the socalled 9 + 2 fibre arrangement). Sensilla of a similar kind are grouped together on the anterior border of the head of adult specimens of E. soleae to form compound receptors but two sorts of compound receptors, each with cilia of a different kind, have been located on the head region of the oncomiracidium. These larval compound organs possess projecting cilia which contain large numbers of longitudinal fibres (30 to 100 in the so-called cone sensillum). In the cone sensillum each modified cilium opens separately through the presumptive adult epidermis of the larva, but in the second kind of compound organ at least five modified cilia are located in a pit situated on the side of the head region of the larva. Because of the small size of monogenean peripheral sense organs studies of the physiological responses of these organs to different stimuli are as yet difficult. However, by comparison with similar organs in other invertebrates Lyons (1969a) tentatively ascribed a tangoreceptive or rheoreceptive function to the single isolated sensilla bearing cilia with a typical 9 + 2 fibre arrangement and found on the adult and on the oncomiracidium. There is a little evidence (Lyons 1969b) which points to a chemoreceptive function for the pit organs containing modified cilia which are found on the head of the oncomiracidium, and indeed the presence of a chemoreceptor in the larva would be expected in view of the specific response of the larva to substances secreted by the sole's epidermis. DISCUSSION This is a convenient point at which to summarize the main features of the biology of Entobdella soleae and to look briefly at our knowledge of other skin-parasitic monogeneans. The adult

181

parasite attaches itself to the skin of the lower surface of a common sole by means of a cupshaped haptor, and because the sand-dwelling sole lives in a situation where the oxygen content of the ambient water is likely to fall to a low level, the parasite is obliged to perform undulating breathing movements to obtain sufficient oxygen for its requirements. The parasite feeds on the fish's epidermis (which the sole can replace readily) by means of a glandular protrusible pharynx, and lays tanned tetrahedral eggs which become attached by an egg stalk bearing sticky droplets to sand grains at the sea bottom. A ciliated oncomiracidium develops within the egg and this free-swimming larva is attracted to its host, the common sole, by specific chemical substances secreted by the epidermis of the fish. Because of the sole's habit of lying on or burying itself in the sand at the sea bottom, invasion of the lower surface of the fish is rarely possible. After settling on the upper surface of the fish the ciliated epidermal cells are shed and later (when the larva has grown from about 0.2 mm to about 1.5 mm in length and shortly before the spermatophore-making apparatus becomes functional) the larvae migrate, using the anterior sticky pads and the haptor, to the lower surface. Mating, involving mutual exchange of spermatophores, takes place on the lower surface of the fish. The spermatophores are deposited externally in the region of the vaginal opening and are sucked into the vagina soon after the mating parasites separate. Maturation of the vitellaria and production of the first egg follows spermatophore exchange but growth continues after sexual maturity is reached and further spermatophore exchange takes place between adult parasites. The biology of Acanthocotyle lobianchi, a monogenean parasite found on the lower surface of rays, is in many ways similar to that of E. soleae. Acanthocotyle has a glandular protrusible pharynx which is used to erode the host's epidermis (Kearn 1963b) and adult parasites make vigorous breathing movements (Kearn 1962). In Acanthocotyle efficient attachment

182

G. C. KEARN

B

A

C

E D

FIGURE 14. The life cycle of E. soleae. A, two eggs attached to sand particles at the sea bottom; B, freeswimming oncomiracidia invade the upper surfaces of the heads of soles buried in the sand; c, the postoncomiracidium attached to the skin of the upper surface of the fish; D, parasites migrate to the lower surface; and E, become mature, c.s., ctenoid scale; e., egg; o., oncomiracidium; p.c., pigment cell; p.o., post-oncomiracidium; s.a.d., stalk and adhesive globules of the egg; s.p., sand particle. (From Kearn 1963c)

FIGURES 15-20. The biology of E. soleae. (The approximate lengths of the larvae and adult parasites are 0.25 mm and 2.0 mm respectively.) 15 Two specimens of E. soleae exchanging spermatophores. Living specimens photographed in ventral view by transmitted light at I/125 second.

16 A living free-swimming oncomiracidium of E. soleae photographed with an electronic flash, b., bladder; ci., cilia. 17 A detached dorsal scale of S. solea which has been exposed to free-swimming larvae of E. soleae. Eight larvae have attached themselves to the epidermis which has already migrated almost to the posterior edge of the lamina. Scale preserved in Bouin's fluid. 18 "Footprint" in the skin of a sole produced by the attached haptor of E. soleae. The skin was preserved in Bouin's fluid before the parasite was removed from the fish. 19 A feeding wound (AB) inflicted by the adjacent attached specimen of E. soleae on the skin of the upper surface of a sole. The skin was preserved in Bouin's fluid immediately after feeding had ceased in order to render the epidermis of the fish opaque. This treatment caused the parasite to contract and separated the head region of the animal from the wound, h., haptor. 20 A section (AB) through the feeding wound AB in Figure 19 (ch., chromatophore; d., dermis; ep., epidermis; m.c., mucus cell).

184

has been achieved not as in Entobdella by the acquisition of extra large sclerites (hamuli) and the modification of the centrally placed larval booklets but by the development of a completely new attachment organ derived from the posterior region of the body and armed with radial rows of sclerites (Kearn 1967b). The larval haptor, which carries one central and seven peripheral pairs of larval booklets, is retained by the adult but is relatively small compared with the new attachment organ or pseudohaptor. The larvae of Acanthocotyle have no ciliated epidermis and so must wait passively on the sea bottom until a ray settles on top of them (Kearn 1967b). Consequently newly attached specimens of A. lobianchi as well as adults are found on the lower surface of the host (cf. Entobdella, p. 179). Acanthocotyle elegans also has a larva which lacks cilia but the adult parasites are found on the upper surface of their host Raía clávala. The inference is that young specimens of A. elegans invading the lower surface of the fish must migrate to the upper surface before reaching maturity (a migration in the reverse direction to that shown by E. soleae}. Nothing is known about sperm exchange in acanthocotylids. In contrast with E. soleae and Acanthocotyle spp. Leptocotyle minor inhabits the skin of a more active round-bodied fish, the dogfish Scyliorhinus canícula. It may be significant that in this parasite sperm is deposited directly inside the vagina of the co-copulant (Kearn 1965). Spermatophores deposited on the outside of the parasite's body like those of E. soleae would be exposed to strong currents on the skin of an active host and might be dislodged and lost. Similarly it might be expected that breathing movements would be unnecessary for a parasite living on an active fish. Indeed Leptocotyle does not perform the vigorous body undulations observed in Entobdella and Acanthocotyle. Occasionally the rigid body of Leptocotyle is seen oscillating slowly backwards and forwards (Kearn 1962), but it is not known whether these gentle movements contribute in any significant way to the

G. C. KEARN

capture of oxygen. L. minor attaches itself to one of the host's denticles, the projecting parts of which are hard plates completely devoid of a covering of soft skin. Consequently hooks are unsuitable for attachment and are absent in Leptocotyle (although there is some evidence of hooks in the ciliated larva) and the adhesive organ of the parasite is fixed to the denticle by means of a layer of cement (Kearn 1965 ). The parasite can attach the head region temporarily to another denticle and after slowly stripping the posterior adhesive pad away from the denticle it can move in a leech-like manner. The parasite feeds on the epidermis lying between the denticles. It is worth while to speculate briefly on the way in which these monogenean skin parasites have evolved. It seems likely that the early monogeneans were descendants of free-living rhabdocoel-like turbellarians which moved about on the sea or river bottom by means of cilia (Llewellyn 1965). Perhaps, like some modern turbellarians, these animals used chemoreceptors to locate their food and ingested small aquatic organisms or the tissues of larger animals by means of a protrusible pharynx. Since the earliest vertebrates were most probably sluggish bottomliving creatures, their epidermis provided an abundant and readily accessible source of food for the free-living flatworms; it was not until these vertebrates became more active that the problems of hanging on to the host became acute. This problem of attachment was solved by the acquisition of a posterior adhesive organ or haptor, bearing, to begin with, probably 16 marginal booklets. Later, perhaps because of increasing size of the protomonogeneans or increased activity of the host, these marginal booklets were supplemented in acanthocotylids by a new attachment disk derived from the body and in other monogeneans first by one pair and then by a second pair of larger hooks (hamuli). Continued selection pressure for improvement of the attachment organ has led in entobdellids to a modification of the centrally placed pair of mar-

PHYSIOLOGY AND BEHAVIOUR OF ENTOBDELLA SOLEAE

ginal booklets and the association of these modified booklets with one of the pairs of hamuli to provide an efficient suction apparatus. The ancestral mode of feeding on epidermis has survived evidently with little change in modern skin parasites. The chemotactic hostfinding mechanism of the larva of Entobdella is strongly reminiscent of the homing behaviour of free-living flatworms in response to juices emanating from the food. Indeed it seems likely that this specific response to fish odours has played a big part in establishing host-specificity. The ciliated epidermis of the parasitic flatworm still retained the important job of transporting the larva to its new host, but once attached to the fish's skin there was no longer a need for cilia and the ciliated epidermis was shed. Locomotion on the host was still important to ensure cross-fertilization, but a new method was called for which avoided any risk of the parasite being dislodged. This was overcome by the development of sticky glands on the head which could attach the parasite effectively for short periods while the position of the haptor was being changed. The advantages of a mutual exchange of spermatophores are by no means clear. It is conceivable that the free-living bottom-dwelling ancestors of Entobdella used freely deposited spermatophores as a means of distributing their sperms and promoting cross-fertilization and to avoid the problems of sperm storage, thereby permitting an increase in sperm production. It is clear that spermatophores deposited on the skin of a moving fish are likely to be washed away and lost. Although this problem is not as critical on a slow-moving sole as on a faster-moving fish, this danger may still have led to deposition of spermatophores directly on the body of a mating parasite followed almost immediately by assimilation into the vagina in E. soleae. Leptocotyle is washed by more vigorous currents and deposits its sperms directly in the vagina of the cocopulant. Some of the early monogeneans with two pairs of hamuli took to living on the well-

185

ventilated gills of their hosts and although the descendants of some of these parasites (e.g. Dipleclanum aequans) produce spermatophores these are deposited inside the vagina (Paling 1966). These gill-parasitic monogeneans still spend the early part of their life on the outer body skin of their hosts (Kearn 1968 ). The knowledge of the physiology and behaviour of E. soleae described above can be attributed to the high degree of suitability of the parasite for research purposes. The relatively large size of the adult parasite and the transparency of its body are considerable advantages, and large numbers of free-swimming larvae can be obtained from eggs incubated at 14° c in less than four weeks. Furthermore, the common sole is a hardy fish which will accept food in captivity and will survive for many months in a laboratory tank. Because of repeated reinfection in the confined space of a laboratory tank the population of parasites on a freshly caught fish may increase tenfold, providing a readily available source of animals for experimental work. Precisely the same qualities, together with the parasite's lack of specialization, make E. soleae equally suitable for teaching purposes. ACKNOWLEDGMENTS

I would like to record my thanks to the Cambridge University Press and to the Editors of Parasitology and The Journal of the Marine Biological Association of the United Kingdom for permission to republish all of the figures except 2,4, 16, and 17. REFERENCES Bateson, W. 1890. The sense-organs and perceptions of fishes; with remarks on the supply of bait. /. Marine Biol. Assoc. U.K. 1:225-56 Beneden, P. J. van, and Hesse, C. E. 1864. Recherches sur les bdellodes (hirudinées) et les tré matodes marins. Mém. Acad. roy. sci. lett. belg. 34:1-142 Berlin, L. B. 1951. Compensatory regeneration of

186 thé epidermis of the groundling. Dokl. Akad. Naiik, S.S.S.R. 80:245-8 (In Russian) Burton, P. R. 1967. Fine structure of unique central region of the axial unit of lung-fluke spermatozoa. /. Ultrastruct. Research 19:166-72 Johnston, T. H. 1929. Remarks on the synonymy of certain tristomatid trematode genera. Trans. Roy. Soc. S. Australia 53:71-8 Kearn, G. C. 1962. Breathing movements in Entobdella soleae (Trematoda, Monogenea) from the skin of the common sole. /. Marine Biol. Assoc. U.K. 42:93-104 - 1963a. The egg, oncomiracidium and larval development of Entobdella soleae, a monogenean skin parasite of the common sole. Parasitology 53:435-47 - 1963b. Feeding in some monogenean skin parasites: Entobdella soleae on Solea solea and Acanthocotyle sp. on Raía clávala. J. Marine Biol. Assoc. U.K. 43:749-66 - 1963c. The life cycle of the monogenean Entobdella soleae, a skin parasite of the common sole. Parasitology 53:253-63 - 1964. The attachment of the monogenean Entobdella soleae to the skin of the common sole. Parasitology 54:327-35 - 1965. The biology oiLeptocotyle minor, a skin parasite of the dogfish, Scyliorhinus canícula. Parasitology 55:473-80 - 1967a. Experiments on host-finding and hostspecificity in the monogenean skin parasite Entobdella soleae. Parasitology 57:585-605 - 1967b. The life cycles and larval development of some acanthocotylids (Monogenea) from Plymouth rays. Parasitology 57:157-67 - 1968. The development of the adhesive organs of some diplectanid, tetraonchid and dactylogyrid gill parasites (Monogenea). Parasitology 58: 149-63 - 1970. The production, transfer and assimilation of spermatophores by Entobdella soleae, a monogenean skin parasite of the common sole. Parasitology 60:301-11 Kruuk, H. 1963. Diurnal periodicity in the activity of the common sole, So lea vulgaris Quensel. Neth. J. Sea Research 2:1-28

G. C. KEARN

Llewellyn, J. 1963. Larvae and larval development of monogeneans. Advances in Parasitol. 1:287326 - 1964. The effects of the host and its habits on the morphology and life cycle of a monogenean parasite. In Parasitic Worms and Aquatic Conditions, pp. 147-52. Prague: Czechoslovak Academy of Sciences - 1965. The evolution of parasitic platy helminths. In Third Symp. Soc. Parasitol., pp. 47-78. Oxford: Blackwell Llewellyn, J., and Euzet, L. 1964. Spermatophores in the monogenean Entobdella diadema Monticelli from the skin of sting-rays, with a note on the taxonomy of the parasite. Parasitology 54: 337-44 Lyons, K. M. 1966. The chemical nature and evolutionary significance of monogenean attachment sclerites. Parasitology 56:63-100 - 1969a. Sense organs of monogenean skin parasites ending in a typical cilium. Parasitology 59: 611-23 - 1969b. Compound sensilla in monogenean skin parasites. Parasitology 59:625-36 - 1970. The fine structure and function of the adult epidermis of two skin parasitic monogeneans, Entobdella soleae and Acanthocotyle elegans. Parasitology 60:39-52 Paling, J. E. 1966. The functional morphology of the genitalia of the spermatophore-producing monogenean parasite Diplectanum aequans (Wagener) Diesing, with a note on the copulation of the parasite. Parasitology 56:367-83 Perkins, E. J. 1957. The blackened, sulphide containing layer of marine soils with special reference to that found at Whitstable, Kent. Ann. Mag. Nat. Hist., Ser. 12, 10:25-35 Poll, M. 1947. Faune de Belgique. Poissons Marins. Brussels : Musée royale d'histoire naturelle de Belgique Richards, F. A. 1957. Oxygen in the ocean. Mem. Geol. Soc. Am. 67(1) : 185-238 Sproston, N. G. 1946. A synopsis of the monogenetic trematodes. Trans. Zool. Soc. London 25: 185-600

PHYSIOLOGY AND BEHAVIOUR OF ENTOBDELLA SOLEAE

Discussion L. MARGOLIS

Fisheries Research Board of Canada, Biological Station, Nanaimo, B.C. Dr Kearn has presented us with a lucid, condensed account of his researches over the past decade on the biology of a representative of a group of flatworms, the Monogenea, whose affinities with other parasitic platyhelminths are still being debated. For a long time they were regarded as closely related to the Digenea and grouped with them in the class Trematoda, but recent opinion, although not unanimous, is shifting in favour of closer ties with the cestodes. Among these main groups of parasitic platyhelminths, from the point of view of their physiology, least is known about the Monogenea. J. D. Smyth, in his book The Physiology of Trematodes (Edinburgh and London: Oliver and Boyd, 1966), said that "it is clear that knowledge of the physiology of the Monogenea is extremely limited." This situation likely has come about because of the concentration of effort on parasites of medical and veterinary significance and on parasites that can be maintained readily in traditional laboratory animals. The practical importance of monogeneans will undoubtedly increase as our use of the world's fisheries resources progressively moves from one dominated by a hunting economy to one in which farming practices play an important role. Under the unnatural, crowded conditions usually encountered in fish culture establishments, parasites with a direct life cycle, such as monogeneans, flourish and reach population sizes far beyond that usually encountered in nature. At these high population densities they may seriously affect the health of the fish and cause high mortalities, particularly among young fish. This has been the experience in carp farms of central and eastern Europe and Israel. As marine fishes increasingly become the objects of artificial culture, one can anticipate that monogeneans will make their presence known in an unwelcome way. Indeed, there are already cases of this

187

on record (e.g. Benedenia seriolae and Heteraxine heterocera in culture of Serióla quinqueradiata in Japan [T. Hoshina, Proc. 3rd Symp. Comm. Off. Intern Epizoot. Etude Malad. Poissons, Stockholm, 1968]). To aid in the development of rational methods of control of monogeneans on cultured fishes there is need for basic knowledge of the biology of these parasites. Fundamental research on ecology, physiology, and behaviour of monogeneans thus not only contributes to our general understanding of the phenomenon of parasitism but clearly also has a practical side. A search of the literature leads one to conclude that the Monogenea, in relation to their abundance in the aquatic environment, have not received their share of research interest from experimental parasitologists. Hopefully, Dr Kearn's stimulating paper will help to improve this situation. To conclude my brief remarks, and before opening the paper to general discussion from the floor, I would like to address two specific questions to Dr Kearn. First, his description of the feeding activity in Entobdella soleae indicates that this helminth feeds intermittently. In contrast, in a recent paper Dr M. Wiles (Can. J. Zool. 48:69-73 [1970]) reported that Diplozoon paradoxum, a blood-feeding gill parasite of cyprinids, feeds almost continuously. Is the difference in feeding schedule between these two species likely to be a characteristic difference between the two major groups of Monogenea which they represent, namely Monopisthocotylea and Polyopisthocotylea, considering that the former are basically tissue-feeders and the latter blood-feeders? Secondly, regarding the increased respiratory response of E. soleae to reduced oxygen tension, has any work been done on the mechanism which brings about this response?

The microcosm of intestinal helminths CLARK P. READ

In 1950,1 published a monographic examination of the small intestine as a place to live. From the information available, certain general conclusions were drawn: (1) There is an exocrinoenteric circulation of organic compounds, i.e. some organic compounds are regularly secreted into the intestinal lumen and subsequently resorbed. (2) There are linear gradients in the intestine with respect to pH, oxidation-reduction potential, and certain chemical components; these may vary with feeding pattern. (3 ) It was postulated that the region near the mucosa, termed the paramucosal lumen, differs in its chemical character from the central lumen. (4) It was concluded that the lumen of the small gut is a highly regulated space within the body and should not be regarded merely as a tube running through the body or as a space outside the body. Examination of the literature of the past 20 years has shown that the concepts just described were soundly based, although a number of newer findings have shown that intestinal function is even more complex than visualized in 1950. I should like to review some of this more recent literature in terms of its bearing on our concepts of the gut as a habitat. CHEMICAL FLOW INTO THE INTESTINE

Several years ago it was shown that the amount of nitrogen recoverable from the small intestine of the rat could exceed the amount of nitrogen ingested, even in diets containing protein (Rosenthal and Nasset 1958). These and earlier observations (reviewed by Read 1950) suggested that the gut can deliver a significant amount of endogenous nitrogen into the lumen in response to a meal. Further, it was shown that the molar ratios of free amino acids found in the lumen of the small intestine of the dog, the rat, and the dogfish are remarkably stable regardless of the quality of the food ingested (Nasset, Schwartz,

THE MICROCOSM OF INTESTINAL HELMINTHS

189

TABLE I The molar proportions of free amino acids in the small intestine of the rat for 16 animals. Values calculated as micromoles per micromole of valine (from J. E. Simmons, unpublished) 10p.m.

ASP THR SER PRO* GLU GLY ALA VAL MET ILEU LEU TYR PHE LYS TRY HIS ARG

4a.m.

10a.m.

4p.m.

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

1.67

2.03 1.06 1.66 1.47 4.13 2.63 2.34 1.00 0.43 0.81 1.50 0.56 0.56 2.06 0.05 0.52 1.29

1.48 0.88 1.60 1 .00 3.04 2.26 2.74 1.00 0.41 0.76 1.53 0.59 0.55 2.07 0.03 0,45 1.26

1.39 0.75 1.51 1 .00 2.65 2.17 2.41 1.00 0.40 0.77 1.60 0.62 0.60 2.10 0.02 0.44 1.34

1.64 0.95 1.40 0.80 3.12 2.28 2.07 1.00 0.31 0.75 1.29 0.40 0.45 1.64 0.03 0.43 0.95

1.35 0.82 1.30 0.82 2.35 2.14 2.14 1.00 0.31 0.71 1.44 0.46 0.57 1.92 0.01 0.43 1 .08

1.41 0.49 1.01 1 .08 2.56 2.09 2.42 1.00 0.25 0.58 1.13 0.47 0.45 1.57 0.02 0.32 1 .01

1.51 0.71 1.21 0.96 3.05 2.41 2.33 1.00 0.32 0.78 1.46 0.55 0.49 1.97 0.03 0.47 1.09

1.50 0.72 1.56 0.81 2.50 2.17 2.04 1.00 0.35 0.78 1.55 0.65 0.63 2.24 0.02 0.45 1 .40

2.08 0.92 1.88 1.26 3.95 2.87 2.40 1.00 0.36 0.79 1.59 0.70 0.70 2.27 0.03 0.50 1.27

1.77 0.97 1.71 0.91 2.61 2.62 2.29 1.00 0.43 0.73 1.62 0.66 0.64 2.35 0.05 0.44 1.50

1.80 0.69 1.71 1.29 2.81 3.58 2.76 1.00 0.45 0.78 1.61 0.87 0.78 2.64 0.09 0.44 1 .60

1.91 1.19 1.60 0.95 3.58 2.35 1.97 1.00 0.41 0.68 1.66 0.74 0.70 2.49 0.06 0.51 1 .44

1.57 0.80 1.34 0.95 2.72 3.36 2.07 1.00 0.40 0.67 1.44 0.56 0.56 2.05 0.03 0.44 1.05

1.30 0.98 1.48 1 .44 2.23 2.54 2.48 1.00 0.41 0.79 1.43 0.72 0.68 2.14 0.10 0.44 1.09

1.51 1.03 1.61 1.42 2.79 3.75 2.29 1.00 0.40 0.78 1.33 0.64 0.59 2.05 0.05 0.49 1.06

0.91 1.65 1 .10 2.97 2.11 1.85 1.00 0.39 0.73 1.58 0.56 0.58 2.12 0.04 0.47 1 .41

*Single determination. AH other values are averages of two determinations.

TABLE II Correlation of molar proportions of free amino acids in the lumen of the rat small intestine calculated from data of Table I Data set (rat no.) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Correlated with (rat no.)

Mean correlation coefficient*

Srt

2-16

0.953 0.939 0.960 0.962 0.953 0.966 0.947 0.967 0.955 0.958 0.960 0.942 0.940 0.941 0.937 0.920

0.026 0.032 0.022 0.021 0.025 0.018 0.027 0.018 0.024 0.023 0.021 0.030 0.032 0.029 0.032 0.039

3-16 -2, 4-16 -3, 5-16 1-4, 6-16 1-5, 7-16 1-6, 8-16 1-7, 9-16 1-8, 10-16 1-9, 11-16 1-10,12-16 1-11,13-16 1-12, 14-16 1-13,15-16 1-14,16 1

1-15

*Mean values of 15 coefficients. fStandard error of mean correlation coefficient ; from pooled standard errors of individual coefficients.

and Weiss 1955;Nasset and Ju 1961; Read etal. 1960; Simmons, unpublished). In acute feeding experiments in dogs, the protein zein was used as a nitrogenous constituent. Zein is virtually devoid of the amino acids lysine and tryptophan. The jejunal contents contained lysine and tryptophan at relative concentrations which did not differ significantly from the relative concentrations observed in animals receiving egg albumen or no food at all (Nasset 1962). In our laboratory, Simmons (unpublished) found that the molar ratios of amino acids in the small intestine of the rat are remarkably stable at various times of the day and independent of the feeding behaviour of the animal (Tables I, n). These observations are all consistent with the view that the dietary proteins must be markedly diluted by nitrogenous compounds of endogenous origin. Nasset (1965 ) furnished data on the extent of dilution of exogenous protein by endogenous nitrogen compounds and estimated a dilution factor of about sevenfold. We have verified this in experiments in which 14C-labelled casein was fed

190

CLARK P. READ

TABLE III Exogenous and endogenous nitrogen in the small intestine of the rat after feeding 1 g of 14 C-labelled casein. Two rats were used at each time interval Time after feeding (min)

Total N (mg)

Exogenous N* (mg)

Endogenous Nf (mg)

Exogenous N, as per cent of total

60 90 120

34.1 29.4 23.4

5.3 4.1 3.4

28.8 25.3 20.0

15.5 13.9 14.4

*Calculated from radioactivity. •¡•Calculated by difference.

to rats and the contents of the small intestine subsequently analysed for total nitrogen and 14C. Results are shown in Table in. It has been assumed that the carbon/nitrogen ratio of casein and its hydrolysed products remained constant. In the small gut the percentage of exogenous nitrogen remains astonishingly constant and, as may be seen, the results agree well with those of Nasset ( 1965 ). What may be the sources of such remarkable quantities of endogenous nitrogen? First, digestive enzymes and other secreted proteins are found in the saliva, gastric juice, pancreatic juice, intestinal juice, and bile. In man, this may constitute more than 200 g of protein per day. A second important source of endogenous protein is the sloughing off of the intestinal epithelium. This tissue proliferates at the rate of about 1 per cent per hour in man (Lipkin 1965 ), which results in the liberation of up to 90 g of protein per day in the gut lumen. A third and significant source of amino acids entering the gut lumen from endogenous sources is the flux of free amino acids across the mucosa. A bidirectional flux of amino acids between gut tissue and the lumen was reported by Christensen, Feldman, and Hastings (1963) and by Orten and Doppke (1964). Arme and Read ( 1969 ) found that, in the rat, the concentration of non-metabolizable amino acid, 1-aminocyclopentane carboxylic acid, rapidly reached a steady state in the blood, in the intestinal lumen, and in the parasite Hymenolepis diminuta. The efflux of amino acids into the lumen of the intestine has been examined in detail by Jacobs

and his colleagues (Jacobs 1965 ). Perfusion of the gut with saline resulted in the efflux of amino acids into the lumen of the gut. Efflux rates were increased when the perfusing fluid contained one or more amino acids. Arme and Read (1969) showed that when the rat is given a single amino acid by mouth, there is a rapid increase in the concentrations of other free amino acids in the lumen of the small intestine. The efflux of labelled amino acids from the blood into the lumen of the gut occurs very rapidly (Jacobs 1965), the label being detectable in the gut less than a minute after injection of an amino acid into the vena cava (Fig. 1 ). All of the above are consistent with the view that there is indeed a circulation of nitrogen compounds from the body into the digestive tract with a subsequent reabsorption of these substances or the products of their hydrolysis. There is now a large body of evidence that the fats are digested to the level of monoglycerides or fatty acids before being absorbed in the digestive tract (Senior 1964;Porter 1969). Further, there is evidence that a considerable portion of the lipid found in the digestive tract is derived from endogenous sources. Baxter (1966) reported that 50 per cent of the lymphatic fatty acids of rats, maintained on a fat-free diet, was derived from secreted bile. Bile is known by analysis to contain fatty acids and lecithins (Spitzer, Kyriakides, and Balint 1964; Balint et al. 1965) and the biliary lecithins are subjected to rapid hydrolysis in the intestine (Leat 1965). Ginger and Fairbairn (1966) showed that the

191

THE MICROCOSM OF INTESTINAL HELMINTHS

FIGURE 1. The movement of 14C-l-aminocyclopentane carboxylic acid from the blood into the lumen of the small intestine of the rat. A segment corresponding to the jejunum was submitted to a singlepass perfusion with Ringer's solution at the rate of 1 ml per minute. The perfusate was sampled at 20second intervals. Ten microcuries of labelled amino acid was injected into the vena cava at the time indicated by the arrow.

composition of the lipids in the small intestine of the rat differed sharply from that of the food, the gut containing a much higher proportion of unsaturated fatty acids. These workers suggested that the measured differences between the diet and the components of bile in the rat might indicate that the bile is a major factor in the dilution of the dietary fatty acids. In a subsequent study it was shown that biliary fats are indeed important in determining the lipid constituents of the intestinal lumen contents, but that other intestinal sources must also be involved (Kilejian, Ginger, and Fairbairn 1968). When labelled palmitate or linoleate was administered intravenously to the rat, labelled lipid was recoverable from the lumen of the intestine 30 minutes later and was also found in the tissue lipid of the parasite Hymenolepis diminuta. Kilejian et al. (loe. cit.) carefully pointed out that, although the fatty acids of the gut were clearly diluted by lipids from endogenous sources and thus were

homeostatically controlled, homeostasis did not appear to be as precise as that observed with the amino acids. This might be related to the fact that animals in general may tolerate variations in the tissue lipid constituents over a broader range than is true for proteins. The necessity for homeostasis may be less rigorous for lipids than for proteins. Kilejian et al. observed that the presence of the parasite Hymenolepis appeared to induce an increase in intestinal lumen lipids. This parasite does not have a parallel effect on the amino acids in the small intestine (Simmons, unpublished). CHEMICAL GRADIENTS

Linear chemical gradients in the intestine will be determined by several parameters, including the source of the substance under consideration, its site of absorption, and the degree to which it may interact with other substances in the gut. For purposes of this discussion, we shall restrict our consideration to the sites of absorption of some particular compounds. Although there are no obvious differences in mucosal cell structure in different segments of the small intestine, it has been known for many years that at least some substances were preferentially absorbed in certain parts of the small gut. Tappeiner (1887) showed that bile salts were maximally absorbed in the ileum. Later studies have shown that vitamin B12 also has a specific absorption site in the

192

CLARK P. READ

TABLE IV

Amino acids recovered from the lumen of the small intestine of the dog after feeding egg albumin (from data of Nasset 1965) Histidine

Duodenum Jejunum Upper ileum Lower ileum

uM

Relative cone.

53 99 129 116

1.0 1.0 1.0 1.0

Valine

Phenylalanine

uM

Relative cone.

165 398 523 487

3.2 4.0 4.1 4.2

ileum (Booth and Mollin 1959). However, there are a number of substances which may be absorbed by all sections of the small intestine. That is, the intestine does not seem to be markedly differentiated in its capacity to absorb sugars, amino acids, or fatty acids. Functionally, however, differential absorption may occur. There is good evidence that glucose is absorbed in the proximal portion of the small intestine if glucose is fed (Borgstrom et al. 1957 ), whereas the absorption of sucrose or the products of its hydrolysis seems to occur more distally (Gray and Ingelfinger 1965). Such observations suggest that there are gradients in the concentrations of a variety of compounds, although good data on such gradients are not readily available. Nasset et al. ( 1955 ) studied the concentrations of several amino acids in various sections of the gut after feeding dogs a meal of egg albumen. These data (Table iv) again show the constancy of the molar ratios of these amino acids in different sections of the gut and also that there is a concentration gradient in the absolute amounts of the amino acids found in these four segments of the gut. Since, as mentioned above, bile salts seem to be specifically absorbed in the distal portion of the small intestine, it might be expected that a gradient in bile salt concentration would exist in the small intestine. This is indeed the case. However, the form of this gradient is not a simple linear one. Dietschy ( 1967) carried out a careful study of the distribution of conjugated and unconjugated bile salts in the intestine of the rat.

uM

Relative cone.

91 200 262 254

1.7 2.0 2.0 2.2

Methionine

Lysine

uM

Relative cone.

uM

Relative cone.

38 78 99 111

0.7 0.8 0.8 1.0

105 286 396 328

2.0 2.9 3.1 2.8

His data are reproduced in Fig. 2. Hepatic bile with a concentration of bile salts at about 33 mM is delivered to the duodenum, where a rapid dilution to about 8 mM occurs. During passage down the digestive tract the absorption of water and solutes occurs more rapidly than the diffusion of bile salts into the mucosa. This results in an increasing concentration of bile salts until the region of specific absorptive activity is reached (about segment 8). In the distal part of the small intestine there is a rapid decline in the intraluminal concentration of bile salts. The form of this gradient appears to differ from that of the amino acids, but rigorous comparisons of the data available do not appear feasible. In contrast, comparisons of the fats and fatty acids of the intestinal lumen with the amino acids of the lumen in the dog show that there are real similarities (Knoebel 1959; Knoebel and Nasset 1957; Nasset et al. 1955 ). There are increasing concentrations of fatty acids and amino acids from proximal to distal regions of the gut in the dog with a tendency to a drop in concentrations in the ileum. THE PARAMUCOSA

Digestion and absorption in the intestine have been regarded as individual and separate, though sequential, processes. Crane (1968) pointed out that they have been traditionally viewed as separated in space and time. However, they now appear to be different aspects of a single process termed digestive-absorptive function. This may

THE MICROCOSM OF INTESTINAL HELMINTHS

193

Conjugated Unconjugated

MIDNIGHT

HEPATIC SMALL BOWEL SEGMENT NUMBER BILE FIGURE 2. Bile salt concentrations down the length of the small intestine of the rat. Rats (200-225 g) were allowed food and water ad libitum and then killed at midnight when their stomachs and intestines were filled with food. The small bowel was divided into 10 segments of equal length, numbered for purposes of identification from 1 to 10, proximal to distal. The intestinal contents of each seg-

ment were centrifuged, and the concentration of conjugated and unconjugated bile salts in the supernatant fraction was quantitated using thin-layer chromatography. In another group of animals, the concentration of bile salt in hepatic bile was determined in samples of bile obtained within 10 minutes after cannulation of the common hepatic duct. (From Dietschy 1967).

have profound implications in determining the nature of substances available to a lumen-dwelling parasite and merits some examination. The vertebrate mucosa has a structurally elaborate brush border facing the lumen of the gut, and it was long considered that this surface was mainly differentiated for absorptive function. However, it has become apparent that the brush border functions in digestion as well and that conceptual separation of functional digestion and absorption obscures the integrated nature of microvillar function. Several workers showed that alkaline phosphatase was associated with the surface of intestinal epithelium and, in 1950, Granger and Baker showed that it was localized in the brush border. Using histochemical methods, a peptidase was found in the brush border (Nachlasetal. 1960) and, with the development of methods for the isolation of mucosai cell brush borders, Miller and Crane ( 1961 ) found that the digestive disaccharidases are found in this part of the intestinal cell. These disacchari-

dases include maltase, isomaltase, sucrase, lactase, and trehalase (Eicholz 1967; Eicholz and Crane 1965; Johnson 1967; Miller and Crane 1961 ). Study of the brush border membrane has revealed surface particles about 50 A in diameter, attached by stalks to the luminal surface (Overton, Eicholz, and Crane 1965; Johnson 1967). These knobs are removed by papain treatment as are the disaccharidases and are recovered from Sephadex columns in the same fraction as the enzymes. Thus, the knobs are strongly implicated as the site of disaccharidase activity (Johnson 1967, 1969; Oda and Sato 1964). How may these digestive functions of the mucosal cell be related to absorptive function? There is evidence that a spatial arrangement of membrane components confers a kinetic advantage for the absorption of the products of sugar hydrolysis (Crane 1967). By this term is meant that the monosaccharide products of brush border hydrolase activity are better absorbed than

194

free monosaccharide added to the lumen. It appears that the cellular digestive hydrolases and the membrane transport system of the brush border are in close physical proximity and because of this organization of the membrane the cell has an advantage in acting against the diffusional loss of free energy. This kinetic advantage in the utilization of the terminal products of carbohydrate digestion will have important effects in the availability of carbohydrate to lumendwelling parasites, particularly those which appear to be sharply limited in the quality of carbohydrate which can be absorbed and metabolized. Special mention must also be made of another phenomenon associated with the digestive-absorptive function of the mucosa. In the past few years it has come to be appreciated that the surfaces of cells possess a mucopolysaccharide layer external to what has been termed the cell membrane. It is usually referred to as the glycocalyx (Bennett 1963 ; Ito 1969). It is well known that proteins adsorb to the charged surface presented by this layer (Bell 1962). It is thus not surprising that enzymes of pancreatic origin bind to the surface of the brush border and function in that location (De Laey 1966; Ugolev 1965). Further, the kinetic parameters of pancreatic amylase are altered by adsorption on the mucosal surface (De Laey 1967). Ugolev referred to the potentiation of hydrolase activity observed after adsorption as membrane (contact) digestion. It might be expected that membrane (contact) digestion would yield an additional kinetic advantage in the digestive-absorptive functions of the intestinal mucosa, but definitive experiments to explore this point do not seem to have been performed. Interestingly, there is some evidence for membrane (contact) digestion in tapeworms. Taylor and Thomas (1968) reported that the presence of living tapeworms of three species increased the rate of starch hydrolysis by alpha-amylase, although the worms lacked an intrinsic alphaamylase. The author has restudied this phenomenon with Hymenolepis diminuta using a pancreatic amylase (Read, in press). These studies

CLARK P. READ

have shown that membrane (contact) digestion may occur at the tapeworm surface and involve enzymes of host origin. The adsorbed enzyme is readily removed by washing, unlike the intrinsic surface phosphatases of tapeworms (Arme and Read 1970; Dike and Read 1971 ). It will be of great interest to determine whether tapeworms may "share" other digestive enzymes of the host, particularly those of pancreatic origin. A FINAL NOTE

It is important that we recognize that parasitic organisms may alter their environment. The author has recently reviewed the changes which occur in the well-known parasitism trichinosis (Read 1970). Symons and Fairbairn (1962) described changes which occur in nippostrongyliasis. The most obvious ones are those related to digestive-absorptive function of the microvilli described earlier in this paper. Symons and Fairbairn (1962) showed that there was a decrease in the relative activity of the enzymes maltase and leucine aminopeptidase in brush border preparations. They pointed out that the changes which they observed in nippostrongyliasis were relatively non-specific in character, in that they seemed to resemble those seen in non-tropical sprue, hookworm disease, and niacin deficiency. The studies of Symons and Fairbairn and those of Padykula et al. ( 1961 ) leave no doubt that there are significant changes in the villar pattern of the intestine associated with nippostrongyliasis and with sprue. The studies of Trier (1967) leave no further doubt that these changes in nontropical sprue involve modification of the microvilli of the small gut. Further, the impaired digestive-absorptive function observed in these varying conditions of disease indicates that the functions affected are those of the normal microvillar surface. Thus, the changes associated with trichinosis and with nippostrongyliasis appear to constitute a brush border disease. The changes associated with infection by Hymenolepis do not seem to be identical and should be studied further. The above data are not reviewed with a

THE MICROCOSM OF INTESTINAL HELMINTHS

view to developing immediate startling conclusions concerning their significance to organisms living in the digestive tract. However, they illustrate the complexities of the environment of intestinal parasites. Space and time do not permit an analysis of other data available on appropriate aspects of gut physiology. The author intends to deal with this in more detail elsewhere. However, it has become more apparent that the gut is in dynamic relation with the body and that a parasitic organism living in the intestine of a vertebrate has a variety of substances available which may be available in the food of the host at quite different relative concentrations. No attempt has been made to deal with the complications introduced by coprophagy or with the cyclic changes which may be induced by feeding pattern. The latter may be very important in considering the behaviour and feeding habits of parasitic organisms (Read and Kilejian 1969; Chappell et al 1970). Finally, it is hoped that we may lay to rest what I have termed the "vacuum cleaner hypothesis" of Chandler (1943 ). It was postulated by Chandler that parasitic organisms living primarily in the gut lumen may somehow directly remove from the mucosa substances which are required. Recently (November 1969) I heard a parasitologist reiterate this hypothesis in its unregenerated form : I winced but resisted the temptation to comment. Chandler recognized the fallacy of his earlier conclusions and, in co-authorship with the present author, clearly said so (Chandler, Read, and Nicholas 1950). Clearly, the newer knowledge of normal intestinal physiology now allows more precise and meaningful examination of the molecular basis for changes in the intestinal microcosm originating in the interactions between hosts and parasites. REFERENCES Arme, C., and C. P. Read. 1969. Fluxes of amino acids between the rat and a cestode symbiote. Comp. Biochem. Physiol. 29:1135-47

195

- 1970. A surface enzyme in Hymenolepis diminuta (Cestoda). /. Parasitai. 56:514-16 Balint, J. A., Kyriakides, E. C., Spitzer, H. L., and Morrison, E. S. 1965. Lecithin fatty acid composition in bile and plasma of man, dogs, rats and oxen. /. Lipid Research 6:96—9 Baxter, J. J. 1966. Origin and characteristics of endogenous lipid in thoracic duct lymph in rat. /. Lipid Research 7:158-66 Bell, L. G. E. 1962. Polysaccharide and cell membranes. J. Theoret. Biol 3:132-3 Bennett, H. S. 1963. Morphological aspects of extracellular polysaccharidases. /. Histochem. Cytochem. 11:14-23 Booth, C. C., and Mollin, D. L. 1959. The site of absorption of Vitamin B12 in man. Lancet 1 : 18-21 Borgstrom, B., Dahlquist, A., Lundh, G., and Sjovall, J. 1957. Studies of intestinal digestion and absorption in the human. J. Clin. Invest. 36: 1521-36 Chandler, A. C. 1943. Studies on the nutrition of tapeworms. Am. J. Hyg. 37:121-30 Chandler, A. C., Read, C. P., and Nicholas, H. O. 1950. Observations on certain phases of nutrition and host-parasite relations of Hymenolepis diminuta in white rats. /. Parasitai. 36:523-35 Chappell, L. H., Arai, H. P., Dike, S. C., and Read, C. P. 1970. Circadian migration of Hymenolepis (Cestoda) in the rat intestine, i. Observations on H. diminuta in the rat. Comp. Biochem. Physiol. 34:31-46 Christensen, H. N., Feldman, B. H., and Hastings, A. B. 1963. Concentrative and reversible character of intestinal amino acid transport. Am. J. Physiol. 205:255-60 Crane, R. K. 1967. Structural and functional organization of an epithelial cell brush border. In Intracellular Transport, ed. K. B. Warren. New York: Academic Press - 1968. A concept of the digestive absorptive surface of the small intestine. In Handbook of Physiology. Alimentary Canal, ed. C. F. Code, sect. 6, vol. 5. Washington, D.C.: Am. Physiol. Soc. De Laey, P. 1966. Die Membranverdauung der Starke. 2. Mitt der einfluss von mucinen auf die

196 membranverdauung. Die Nahrung 10:649 - 1967. Die Membranverdauung der Starke. 5. Mitt zur zweigestaltigkeit der membranverdauung der Starke. Die Nahrung 11:9 Dietschy, J. M. 1967. Effects of bile salts on the intermediate metabolism of the intestinal mucosa. Federation Proc. 26:1589-98 Dike, S. C., and Read, C. P. 1971. Tegumentary phosphohydrolases of Hymenolepis diminuta. J. Parasitai., in press. Eicholz, A. 1967. Structural and functional organization of the brush border of intestinal epithelial cells, in. Enzymic activities and chemical composition of various fractions of Tris-disrupted brush borders. Biochim. Biophys. Acta 135:47582 Eicholz, A., and Crane, R. K. 1965. Studies of the organization of the brush border in intestinal epithelial cells, i. Tris-disruption of isolated hamster brush borders and density gradient separation of fractions. /. Cell. Biol. 26:687-91 Ginger, C. D., and Fairbairn, D. 1966. Lipid metabolism in helminth parasites, n. The major origins of the lipids of Hymenolepis diminuta (Cestoda). J. Parasitai. 52:1097-107 Granger, B., and Baker, R. F. 1950. Electron microscope investigation of the striated border of intestinal epithelium. Anat. Record 107:423-42 Gray, G. M., and Ingelfinger, F. J. 1965. Intestinal absorption of sucrose in man: The site of hydrolysis and absorption. /. Clin. Invest. 44:390-98 Ito, S. 1969. Structure and function of the glycocalyx. Federation Proc. 28:12-25 Jacobs, F. A. 1965. Bi-directional flux of amino acids across the intestinal mucosa. Federation Proc. 24:946-52 Johnson, C. F. 1967. Disaccharidase localization in hamster intestine brush borders. Science 155: 1670 - 1969. Hamster intestinal brush-border surface particles and their function. Federation Proc. 28:26-9 Kilejian, A., Ginger, C. D., and Fairbairn, D. 1968. Lipid metabolism in helminth parasites, iv. Origins of the intestinal lipids available for absorption by Hymenolepis diminuta (Cestoda). /. Parásito!. 54:63-8

CLARK P. READ

Knoebel, L. K. 1959. The gastrointestinal digestion of fats in dogs fed triglycérides, partial glycerides, and free fatty acids. /. Nutrition 68:393403 Knoebel, L. K., and Nasset, E. S. 1957. The digestion and absorption of fat in dog and man. J. Nutrition 61:405-19 Leat, W. M. F. 1965. Possible function of bile and pancreatic juice in fat absorption in the ruminant. Biochem. J. 94:21-22P Lipkin, M. 1965. Cell proliferation in the gastrointestinal tract of man. Federation Proc. 24:10^15 Miller, D., and Crane, R. K. 1961. The digestive function of the epithelium of the small intestine, ii. Localization of disaccharide hydrolysis in the isolated brush border portion of intestinal epithelial cells. Biochim. Biophys. Acta 52:293-8 Nachlas, M. M., Monis, B., Rosenblatt, D., and Seligman, A. M. 1960. Improvement in the histochemical localization of leucine aminopeptidase with a new substrate, L-leucyl-4-methoxy-2naphthylamide, J. Biophys. Biochem. Cytol. 1: 261-4 Nasset, E. S. 1962. Amino acids in gut contents during digestion in the dog. /. Nutrition 76:1314 - 1965. Role of the digestive system in protein metabolism. Federation Proc. 24:953-8 Nasset, E. S., and Ju, J. S. 1961. Mixture of endogenous and exogenous protein in the alimentary tract. /. Nutrition 74:461-5 Nasset, E. S., Schwartz, P., and Weiss, H. V. 1955. The digestion of proteins in vivo. J. Nutrition 56:83 Oda, T., and Sato, R. 1964. Elementary particles of the microvilli of rabbit intestinal epithelial cells. Symp. Am. Soc. Cell Biol., Cleveland, Ohio Orten, A. U., and Doppke, H. J. 1964. The fine structure of time-curves for the absorption of amino acids from a Thiry loop in man. Federation Proc. 23:339 Overton, J., Eicholz, A., and Crane, R. K. 1965. Studies on the organization of brush border in intestinal epithelial cells, n. Fine structure of fractions of Tris-disrupted hamster brush border. J. Cell Biol. 26:693-706 Padykula, H. A., Strauss, E. W., Ladman, A. J.,

THE MICROCOSM OF INTESTINAL HELMINTHS

and Gardner, F. H. 1961. A morphologic and histochemical analysis of the human jejunal epithelium in nontropical sprue. Gastroenterology 40:735-65 Porter, K. R. 1969. Independence of fat absorption and pinocytosis. Federation Proc. 28:35-40 Read, C. P. 1950. The vertebrate small intestine as an environment for parasitic helminths. Rice Inst.Pam.ll(2):l-94 - 1970. Chemical pathology of trichinosis. In Trichinosis in Man and Animals, éd. S. E. Gould. Springfield, 111. : Charles C. Thomas - 1971. Contact digestion in a tapeworm. J. Parasitol., in press Read, C. P., and Kilejian, A. Z. 1969. Circadian migratory behaviour of a cestode symbiote in the rat host. /. Parasitai. 55:574-8 Read, C. P., Simmons, J. E., Jr., Campbell, J. W., and Rothman, A. H. 1960. Permeation and membrane transport in parasitism : Studies on a tapeworm-elasmobranch symbiosis. Biol. Bull. 119:120-33 Rosenthal, S., and Nasset, E. S. 1958. Gastric emptying and intestinal absorption of carbohydrate and protein as influenced by the nature of the test meal. /. Nutrition 66:91-103 Senior, J. R. 1964. Intestinal absorption of fats. /. Lipid Research 5:495-521 Spitzer, H. L., Kyriakides, E. D., and Balint, I. A. 1964. Biliary phospholipids in various species. Nature 204:288 Symons, L. E. A., and Fairbairn, D. 1962. Pathology, absorption, transport, and activity of digestive enzymes in rat jejunum parasitized by the nematode Nippostrongylus brasiliensis. Federation Proc. 21:913-18 Tappeiner, H. 1887. Cited by Read, 1950. Taylor, E. W., and Thomas, J. N. 1968. Membrane (contact) digestion in the three species of tapeworm Hymenolepis diminuta, Hymenolepis micro stoma, and Moniezia expansa. Parasitology 58:535-46 Trier, J. S. 1967. Structure of the mucosa of the small intestine as it relates to intestinal function. Federation Proc. 26:1391-404 Ugolev, A. M. 1965. Membrane (contact) digestion. Physiol. Revs. 45:555-95

197

Discussion D. F. METTRICK

Department of Zoology, University of Toronto In the short time that I have available to discuss Dr Read's paper, I should like first to comment on the chemical flow into, and chemical gradients within, the intestine, and to suggest that the quantity and molar ratios of amino acids in the small intestine are not so remarkably stable as has previously been thought. In 1963 C. P. Read, A. R. Rothman, and I. E. Simmons (Ann.N.Y. Acad.Sci. 113:154) put forward the concept that a tapeworm parasitized the homeostatic mechanism of its host. This means that any specific concentration and ratio of amino acids in the lumen leads to an equally specific, although not necessarily the same, concentration and ratio in the tapeworm. It follows that in the absence of a homeostatic mechanism a change in the amino acids in the lumen would result in an imbalanced ratio in the amino acid pool of the tapeworm. D. F. Mettrick and H. N. Munro (Parasitology 55:453 [1965]) and Mettrick (Parasitology 58:37 [1968]) showed that feeding diets containing only one or two amino acids resulted in retardation of growth of Hymenolepis diminuta. Their suggestion that this retardation in growth was attributable to an imbalance in the tapeworm's amino acid pool is supported by C. A. Hopkins (Parasitology 59:407 [1969]), who proved that a dietary imbalance of Lmethionine affects the amino acid pool of the tapeworm. Similarly L. H. Chappell and C. P. Read (Abstr. 3, p. 30, 44th Annual Meeting, Am. Soc. Parasitologists, Nov. 1969) reported imbalances in the worms' amino acid pool following feeding of the amino acids lysine, proline, and phenylalanine. If, therefore, the homeostatic mechanism of the host is as efficient as suggested by E. S. Nasset (/. Am. Med. Assoc. 164:172 [1957]; also in The Role of the Gastrointestinal Tract in Protein Metabolism, ed. H. N. Munro [Oxford: Blackwell, 1964]) and E. S. Nasset and J. S. Ju (/. Nutrition 74:461 [1961]), it is difficult to see why an imbalanced amino acid dietary source was not buffered by en-

198 dogenously secreted protein. It is harder still to explain recent results from our laboratory which show (a) that increasing quantities of the amino acid L-methionine in the diet result in increased retardation of worm growth (Table i) and (b) that if a series of neutral amino acids are arranged in a sequence corresponding to their rate of absorption by the intestinal mucosa from the luminal amino acid pool (C. Gitler and D. Martinez-Rojas, in The Role of the Gastrointestinal Tract in Protein Metabolism) the order representing decreased rate of absorption is nearly identical with that for increased retardation effect on the worms (Table n). Apart from this independent evidence from studies on tapeworms, Nasset's (1957, 1964) homeostatic concept has been criticized by E. Geiger, L. E. Human, and M. J. Middleton (Proc. Soc. Exptl. Biol Med. 97:232 [1958]), who concluded that the volume of endogenously secreted amino nitrogen in the intestine was not sufficient to maintain homeostasis, and by C. Gitler (in Mammalian Protein Metabolism, éd. H. N. Munro [New York: Academic Press, 1964]), who suggested that the relatively constant amino acid level in the intestine was due to the slow digestion and accumulation of endogenous protein rather than the rapid production of a large volume of endogenous protein to buffer exogenous recruitment. Although there may be doubt as to the efficiency and mode of operation of homeostasis within the intestine, there is evidence, as we have seen, that the molar proportions of free amino acids in the rat small intestine appear to be surprisingly constant. We can question this sort of data in two ways: first, is there a difference in time, and, secondly, is there a difference for individual amino acids through time? The molar proportions of free amino acids in the rat intestine over a 4-hour period following feeding (Table in) are similar to Simmon's figures covering 24 hours. However, in this case there are highly significant differences in the ratios of any particular amino acid over the 4-hour period examined, and the difference between the time intervals was only just outside the level of statistical significance (Table iv). I would predict from these pooled data that

CLARK P. READ

TABLE I Effect on worm dry weight of adding increasing quantities of the amino acid L-methionine Amount of dietary amino acid (mg)

Per cent decrease in worm weight

9 17 35 70

2 7 12 17

TABLE II Correlation between rate of uptake by the intestinal mucosa of dietary amino acids and the effect of the amino acids on worm dry weight Amino acids*

Per cent decrease in worm weight!

L-cysteine L-methionine L-isoleucine L-tryptophan L-serme L-glycine L-threonine

5 10 10 7 15 20 26

*Listed in decreasing order of rate of absorption from lumen of intestine. f Effect on worm development increases from top to bottom. TABLE III The molar proportions of free amino acids in the small intestine of the rat, over a 4-hour period following feeding* (values calculated as micromoles per micromole of valine)

11 a.m. LYS HIS ARG ASP THR SER GLU PRO GLY ALA VAL MET

ILEU LEU TYR PHE AMM

1.24 0.55 0.55 1.48 0.76 1.55 1.72 0.40 1.07 1.37 1.00 0.17 0.57 0.78 0.13 0.10 10.31

12a.m.

1.64 0.95 1.45 2.07 0.98 1.64 1.70 0.26 1.46 1.17 1.00 0.53 0.81 1.02 0.27 0.21 16.67

1 p.m.

2 p.m.

0.75 0.29 0.31 0.87 0.75 0.80 1.12 1.13 0.62 0.71 1.00 0.54 0.49 0.96 0.54 0.42 2.27

1.21 0.63 0.49 1.76 0.69 1.22 1.54 0.91 1.33 1.46 1.00 0.52 0.58 0.90 0.27 0.24 3.67

*Food presented at 10 a.m. for 15 minutes.

THE MICROCOSM OF INTESTINAL HELMINTHS

199

TABLE IV Analysis of variance showing the significant effect of time on the variation of individual amino acids Degrees of freedom

Mean square

Variance ratio

19

13.31 4.44

5.19* 1.73(n.s.)

48

2.56

Source of variation

Sum of squares

Total Between amino acids 213.03 Between time intervals 13.31

349.41 226.34 lí

67

Residual error

123.07

*P < 0.001.

when specific regions of the intestine are analyzed separately, there will be significant differences between the molar ratios of the free amino acids in the different parts of the gastrointestinal tract. Although I have followed Simmons in expressing the molar ratios of the amino acids in the intestinal amino acid pool as micromoles per micromole of valine, the presentation of amino acid ratios in terms of one particular amino acid is open to criticism on a number of counts. First, it implies that all amino acids are biologically comparable, which they are not; secondly, it emphasizes the pattern of amino acids present rather than their physiological function; and, thirdly, it masks the actual changes in the intestinal amino acid pool to which the worms must respond. In addition to confirming the observations of E. S. Nasset, P. Schwartz ,and H. V. Weiss (/. Nutrition 56:83 [1955]) on the changing concentration of amino acids in different sections of the intestine, we have found also that there are considerable changes in the same region of the intestine (Fig. 1 ), and that these changes are dependent on the source of the dietary proteins. These observations are particularly significant as the degree of competition for an uptake locus depend on the absolute concentration of the amino acids involved as well as on their molar ratios (C. Gitler and D. Martinez-Rojas, in The Role of the Gastrointestinal Tract in Protein Metabolism-, A. Kilejian, /. Parásito!. 52:1108 [1966]). From these observations, I think it is clear that

any intestinal parasite must be able to tolerate considerable fluctuations in both the amounts and ratios of individual amino acids and in the absolute concentration of the intestinal amino acid pool. I should like to ask how Dr Read overcomes the problem of recycling in experiments using 14Clabelled casein. R. Dawson and J. W. G. Porter (Brit. J. Nutrition 16:27 [1962]), who also used a 14 C-labelled protein, measured the total amino acid flow in portal and systemic blood over a 6-hour period following a meal and found that the difference corresponded to absorption of five times the amount of labelled protein fed. The exogenous protein must therefore have been digested, absorbed, and recycled several times in this time interval. Finally, with reference to the digestive functions of mucosal cells, and the evidence for membrane (contact) digestion in tapeworms, it has recently been found that in H. diminuía, when carbohydrates of different quality are added in increasing quantity to a host's diet, a plateau is soon reached when any further increase in the amount of carbohydrate fed does not result in an increase in worm size. An exception is glucose, where increasing quantities result in increased worm size, and the growth rate follows a typical Bertalanffy growth curve (L. C. Duukley, unpublished). A possible explanation is that in the case of the other carbohydrates digestive enzyme production or availability is a limiting factor, but I would like to hear Dr Read's comments on this and the other points I have raised.

200

CLARK P. READ

N O N — P R O T E I N NITROGEN

REGION OF INTESTINE

FIGURE 1. Changes in the concentrations of nonprotein nitrogen (free amino acids) and protein nitrogen in the rat small intestine following the

PROTEIN NITROGEN

REGION OF INTESTINE

feeding of a casein diet (see D. F. Mettrick, Comp. Biochem. Physiol. 37 [1970] 517-41).

The movement of nematodes in the external environment

Previous papers on nematode movement have stressed the mechanics of undulatory propulsion and the influence of the environment on movement (Wallace 1968a). In this paper I want to discuss and speculate about movement in terms of form and function and at the same time point out the significance of movement in the ecology of plant and animal parasitic nematodes.

H. R. WALLACE SHAPE E

8gs

No generalizations can be made about the shape of nematode eggs. They vary from spherical to spheroidal and ellipsoidal, although the commonest shape is the ellipsoid. Closer examination of such eggs indicates that most are not symmetrical about the longitudinal axis through the poles, but in fact give the impression of having been bent about the axis (Fig. SA). How this shape arises is not understood, but it is possible that this asymmetrical shape is imposed on developing eggs as they move around bends in the oviduct and is retained when the membranes of the egg harden. I have mentioned the shape of the nematode egg because it has important consequences on the first phase of nematode movement, motility of the larva in the egg prior to hatching. The larvae of most plant-parasitic nematodes are coiled in a series of figure-eight curves within the egg (Fig. SA). Observations on the movement of Meloidogyne javanica suggest that the larva exerts pressure on the poles of the egg and thereby develops a propulsive force (Wallace 1968b) . By using a model, it was demonstrated that the asymmetrical shape of the egg just mentioned provided an environment for the most efficient movement of larvae. The asymmetry produces a change in curvature at the poles of the egg so, as the nematode glides along

202

in this region, the muscles progressively contract as the radius of curvature of the egg shell decreases, thus releasing potential energy for kinetic energy of propulsion. In eggs with hemispherical ends the curvature does not change and the muscles cannot, therefore, contract, making movement probably less efficient. It has been suggested that various functions are performed by movement of larvae in the eggs including the exertion of pressure by the larva to break through the egg-shell, the development of physiological efficiency before hatching (Wallace 1966), and the emulsification of the inner lipid layer (Wilson 1958). Later in this paper I shall describe another possible function. It has also been suggested that prehatching movement is likely to be most active in those nematodes that hatch in the environment external to the host and subsequently travel through the soil before reaching the host. Larvae and adults As Harris and Crofton (1957) have pointed out, there is close similarity in form and organization in nematodes probably because mechanical factors play so large a part. The elongate cylinder with no appendages is a characteristic shape of the great majority of nematodes and I would like to develop some speculative ideas on this aspect. My first point is that although there are short fat and long thin nematodes most species cover a rather narrow range of shape and size. Shape may be expressed by dividing the length (L) by the width ( a ) . In Figure le I have plotted the frequency distribution of L/a for the adult plant nematode species described in Goodey's (1963) textbook of soil and freshwater nematodes. It is clear that the majority fall within the range 25 to 32. Furthermore, the frequency distributions of L and a alone indicate that most adult plant nematodes that inhabit the soil are 500 to 1200 /¿m long and 15 to 45 /¿m wide (Fig. IA, B). Body width restricts the type of soil environment in which nematodes can move. First, nematodes do not move soil particles to any great ex-

H. R. WALLACE

tent so they are restricted to existing spaces. Hence only soils with pores greater than 15 ¡Jim in width, i.e. sandy soils and loam soils in good tilth, are penetrable by nematodes. The suction necessary to drain a pore of 15 ¡mm diameter is about 200 cm of water; above this value nematode mobility is restricted by lack of water (Wallace 1958a). Studies on various nematode species have shown in fact that the optimum pore diameter is about 1.5 times the nematode diameter (Wallace 1958b). Thus it seems that nematodes move quickest in soils with fairly large pores, such as coarse sandy soils, and at low suctions of 200 cm of water or less. Since nematodes soon become inactive when oxygen is depleted from their environment, good drainage which allows good aeration also promotes their activity. But sandy soils soon dry out at the surface by evaporation and drainage, so frequent rainfall or irrigation and a high water table are also required for good nematode mobility. Some of the most devastating damage to crops in Australia occurs in areas where there are light soils under irrigation or in places like Northern Queensland where the summers are hot and wet. Some nematodes, like Ditylenchus, which infect the plant at or above soil level occur more frequently on heavy soils possibly because the population is retained at the surface, thereby reducing the chance of their moving or being washed downwards away from infection sites (Wallace 1962). Perhaps the incidence of hookworm and other animal parasitic nematodes shows similar relationships to their environment. Incidentally, the fact that nematodes move through pores of 15 to 45 ¡¿m removes them from the environment of most micro-arthropods in the soil, so that prédation by these larger soil animals is unlikely to have any serious effect on nematode populations in the soil. Whether smaller organisms like fungi and bacteria cause any significant mortality among nematodes is another question. The influence of body length on movement is less clear. Experiments with different nematode species indicate that mobility is at a maximum

203

THE MOVEMENT OF NEMATODES

A

Length (fji)

Length/Width

when the length of the nematode is about three times the diameter of the soil crumbs (Wallace 1958c) but this relationship may depend on optimum pore diameter for movement since particle and pore diameter are themselves interdependent. However, in order to move, the nematode's body must be long enough to have at least one complete wave so that it can exert the neces-

B

Width (¿0

FIGURE 1. Frequency distribution of (A) the length, (B) the width, and (c) the ratio of length to width for adult plant parasitic nematodes.

sary forces against the surrounding soil crumbs and propel itself along through the pore spaces (Wallace 1968a). Studies on the wave patterns of nematodes moving between sand particles indicate that the wave form varies according to the amount of water in the pores. When there is just sufficient suction to empty the larger pores in the sand, the nematodes frequently glide with

204

H. R. WALLACE

waves of long wavelengths and low amplitude. As the soil gets drier the surface tension forces of the water opposing movement increase and the nematode responds by increasing its wave amplitude, which gives it greater propulsive thrust (Wallace 1958b). The geometry of the relationship between pore diameter (d) , soil particle diameter (D) , nematode diameter (a) , and nematode length (L) can now be considered by asking whether the pore and particle size of the soil and the wave requirements of the nematode demand a particular shape of nematode (L/a) for efficient movement. As stated previously, experiments show that d = (3/2) a and L = 3D for maximum speed. In a system of identical spheres D = 6d approximately; thus L _

o

3D

9D

54d

(2/3)d

2d

Id

= 27.

The hypothetical nematode with these requirements is shown in Fig. 2. Thus it can be argued that the ratio of length to width in most adults is 25 to 35 because this is the shape that gives maximum mechanical efficiency for movement between soil crumbs when the pores are drained of water. In this idealized soil-nematode model there are many assumptions that are open to question. For example, the soil geometry is never as simple as that described; there are several species that have an L/a ratio of 60 or more; the wave frequency or activity of nematodes varies considerably and so a soil which is too dry for movement in one species may be optimal for another species; nematodes may have some sensory response to their surroundings so that their tracks through the soil may not be random and some species may be better adapted than others in this respect; the relationship between pore and nematode width for maximum speed as described above is empirical and vague. However, it is these kinds of questions that the hypothesis evokes that may justify the hypothesis, however unsatisfactory it is.

MOVEMENT IN WATER FILMS AND DEEP WATER

In their movement through the soil nematodes occasionally enter spaces which are too large for their body length to utilize the soil crumbs as external resistances as illustrated in Fig. 2. Under these conditions, however, they can still move efficiently through the water films. Previous work (Wallace 1959) has shown that the propulsive forces are then obtained from pressure exerted by the nematodes against the surface tension of the water film. Indeed their movement can then be very efficient as can be seen from observations made over the surface of agar. The surface tension hypothesis of movement in films has recently been tested (Wallace 1969) by introducing surfactants into soil at different moisture contents to lower surface tension. The results indicate that motility and infection of plant nematodes could be reduced by as much as 60 per cent in soil by the introduction of surfactants; we are at present running a field trial in a pine seedling nursery to see whether such a procedure can be put to any practical use. Movement in films occurs over surfaces above ground as well as in the soil. The plant nematodes Anguina and Aphelenchoides, for example, crawl up plants in water films on their outer surface. The behaviour of hookworms and Trichostrongylus is similar. In fact, the similarity is so

FIGURE 2. The geometry of a hypothetical nematode moving between identical spheres.

205

THE MOVEMENT OF NEMATODES

great that it is useful to make comparisons. All these aerial nematodes are dependent on water for movement, and they are all very active as measured by their wave frequency. They can also survive desiccation far more than the soil nematodes can, but they have similar temperature requirements. These characteristics have survival value, for to escape from the soil and ascend a plant or any other surface the nematode must be able to swim, i.e. to move through water without touching other objects. To swim the nematode must form body waves of sufficient frequency, wavelength, and amplitude to elicit the required propulsive thrust. Soil nematodes are usually too sluggish to swim and sink downwards in deep water. A long thin nematode (high L/a) can develop the requisite waves more efficiently, so in plant nematodes at least the swimmers are mostly of the long thin variety. As the nematodes swim upwards in the thick water films evaporation reduces film thickness until eventually they start to crawl as the water film presses down on their bodies. At this stage nematodes like Aphelenchoides can develop enough thrust to invade the leaf surface, and hookworm can penetrate the skin. Further evaporation of the film immobilizes the nematodes and finally leads to desiccation by the surrounding atmosphere. The nematode then remains immobile until water again covers the surface. Comparisons of the characteristics of Aphelenchoides ritiemabosi and Trichostrongylus colubriformis emphasizes the similarities between nematodes whose hosts are quite different but which inhabit the same ecological niche (Wallace and Doncaster 1964). Indeed the similarities are close enough to suggest that animal and plant nematologists could have useful discussions in this general field of ecology in the external environment. I shall go even further and suggest that in the field of movement there is something to be gained from comparative studies of the movement of nematodes in narrow channels such as blood vessels and plant cells as well as in the external environment.

MOVEMENT AND FLEXIBILITY

Judging by the way nematodes bend during invasion of the host (Fig. SB ), during movement in the egg (Fig. SA) , and during copulation and movement through the soil, the body must be very flexible. Nematodes frequently form coils too, and considerations of simple geometry applied to a sector of the body show that the cuticle on the outside of the curve must stretch by at least 50 per cent and on the inside of the curve must shrink by a similar amount. It can also be shown on mechanical principles that it requires much less force to induce a given stress in a cylinder by bending than by stretching it along the longitudinal axis (Wallace 1970). The relationship between the percentage stretch (or contraction) of the cuticle, the diameter of the nematode's body ( a ) , and the radius of curvature of the bend (r) is given from Fig. 3 by AB - CD CD

100.

But AB =

(r + a/2) - r 100a x 100 = r 2r '

(1)

so the amount of stretch increases with increase in body width and decrease in radius of curvature of the wave. These considerations do not appear too helpful, since the nematode's cuticle must be able to increase in length by 50 per cent and it seems doubtful that the material from which it is made can be stretched by this amount and still retain its elastic properties after recovery. The nematode's cuticle is not smooth, however; it is annulated (Fig. 5c) so that the surface area is much greater than appears at first sight. When the nematode bends, the annuli on the outside of the curve tend to straighten and those on the inside of the curve to buckle. Hence, increase in length of cuticle is achieved by bending of the curved annuli and not by stretching of the cuticle. The geometry of the bent annulus illustrates this

206

H. R. WALLACE

(a)

(b)

(c)

FIGURE 3. (a) Diagram of a nematode showing annuli with stretching and contraction of the cuticle, (b) The geometry of an annulus at the crest of a wave, (c) Diagram of a nematode moving between spheres to show the relationship between the pitch of the waves (0), the wavelength (X), the body width ( a ) , and the radius of curvature of the wave

(r).

THE MOVEMENT OF NEMATODES

207

point (Fig. 3b). The chord of the unbent annulus has a width Y, whereas when fully stretched during bending the chord increases to Z although the actual length of the arc of the annulus does not change. The percentage change in the length of the annular chord is given by Z - y

y

x 100

(2)

But Z = PQ and Y = RS and so Z - y

y

PQ - RS x 100 RS

x 100 =

(r + a/2) - r

lOOfl x 100 -

2r

r

(3)

where a is the width of the body and r is the radius of curvature. Changes in the length of the annular chord thus enable the nematode to bend as much as it does. If a whole nematode (of length L and width a) is considered moving between identical spheres of radius r such that the body makes an angle 0 with the axis of progression XY and forms waves of length X, then the arc AB is 2rO radians (Fig. 3c). But AB is half a wave; thus if there are N waves per body length, L = 2N x arc AB = 2N x

29

360

x 2nr =

NnrQ

45

(4)

If C is the maximum percentage stretch of the cuticle, /^ __ a x 100

2r

so

r =

lOOfl

2C

Substituting in (4) give¡ L =

lOOfl NnQ x 45 2C

Hence a

C = 3.5JV6 —. L

(5)

From these calculations it can be concluded that

the percentage maximum stretch increases as the number and pitch (6) of the waves increase and as the shape factor (L/a) decreases. The stretch seldom exceeds 55 per cent in most of the nematode species examined, judging by the dimensions of the arcs and chords in the annuli (Z and Y in Fig. 3b). In larvae of Meloidogyne ¡avanica the pitch of the waves in thin films is about 55°, L/a == 29, and the number of body waves has not been seen to exceed 4. So the theoretical maximum extension of the cuticle at the crests and troughs of the waves is about 27 per cent, well within the measured maximum extensibility of 60 per cent in this species (Wallace 1970). Recent studies on the flexibility of larvae of Meloidogyne javanica within the egg suggest a possible mechanism by which the annuli are formed (Wallace 1970). In this species there is one moult within the egg and the annuli first appear in the newly formed second-stage cuticle before the first ecdysis. Observations suggested that the annuli were formed by a mechanical buckling of the cuticle as the larva moved around the poles of the egg. A model nematode was made by coating plastic tubing with latex and then bending the tube. Annuli of the same form and dimensions as those seen in nematodes were obtained. Moreover, with increasing width of tube and thickness of latex, the annular size increased. The mechanics of buckling of thin-walled tubes fully explains these observations, and so it seems likely that bigger nematodes will have bigger annuli because of their greater width and thicker cuticles. If the cuticular material of different nematodes is similar and if the ratio of body width to cuticular width is approximately the same, then there should be a fairly simple relationship between body width and annular size. To test this hypothesis a graph was plotted of the two variables obtained from photographs of 21 plant and free-living nematodes (Fig. 4). The significant correlation suggests that the hypothesis is probably valid. However, the species that do not fit this relationship are probably the most useful to consider. The order Dorylaimida,

208

for example, are usually described in textbooks as having smooth cuticles, but Xiphinema is annulated although the annuli are small and cannot be seen easily under the light microscope. They are large nematodes measuring 2 to 3 mm in length and 40 to 60 fcm in width so if the previous hypothesis is correct observations should show that the outer cortical layers of the cuticle which form the annuli are either much thinner than in the other nematodes or made of a material with different elastic properties. If the buckling hypothesis proves to be valid, it seems likely that annulation or striation in other nematodes is also formed prior to ecdysis by buckling of the newly developed cuticle during ordinary undulatory movement. Observations indicate that although the new annuli correspond to the annuli of the previous stage in position, they are larger and so permit extension

FIGURE 4. Relationship between body width (à) and annular width (P) in some plant-parasitic and freeliving nematodes.

H. R. WALLACE

of the cuticle following ecdysis and subsequent growth. Whether the hypothesis is valid for all nematodes is questionable. In Ascaris, for example, it is difficult to imagine how the extension of the annuli could account for the great increments in growth. There are other ways of achieving flexibility. Trichodorus, a plant nematode, buckles on bending (Fig. SD) but the changes in cuticular form are only temporary. Ascaris is similar in this respect. Flexibility in nematode movement, with its dependence on annulation, is one more function fulfilled by movement of larvae in the egg, for without such movement annuli would not be formed. Attempts to produce non-annulated larvae by stopping larval movement in the egg while allowing development to continue have not yet succeeded.

FIGURE 5. A, larva of Meloidogyne javanica coiled within the egg-shell; B, larvae of M. javanica invading a tomato root; c, annulation of the cuticle in

Heterodera avenae\ D, Trichodorus sp. showing buckling of the cuticle.

210

MOVEMENT AND THE ECOLOGY OF NEMATODES

In plant parasitic nematodes the newly hatched second-stage larva does not feed during its life in the soil until it reaches the root of the host plant. As the nematode migrates through the soil it uses up its energy reserves and within eight days or so it loses its ability to infect the root (Van Gundy, Bird, and Wallace 1967 ). This phase of the life cycle is critical, as it is of those animal nematodes that do not feed during all or part of their existence outside the host. However, plant nematodes appear to have evolved characteristics that increase their chances of reaching the host: eggs may hatch at a greater rate when exudates from the root stimulate them a short distance away and they may only respond to exudates of their particular host; the larvae tend to reduce the distance travelled between egg and root by orienting themselves in concentration gradients of chemicals released by the roots; when the soil dries and resistance to movement increases the nematode conserves its energy by reducing the power output for locomotion. The hypothesis of conservation of energy by nematodes in the soil environment seems useful because it indicates the kind of experiments that might be done to find other adaptations. Unlike the hosts of animal nematodes, the plant is almost immobile and it occupies the same environment in the soil as the nematode. The plant nematode can therefore actively and purposefully seek its host whereas the animal nematode must rely on the activities of its host to achieve infection, infection being largely by chance. A useful question for study would be to ask what adaptations animal nematodes could have to increase their chances of infection. Although studies on the factors that influence the movement of nematodes in their external environment could usefully include both animal and plant parasitic nematodes, there is one further aspect where they are different. In addition to the direct effect of the environment on the plant nematode there is the effect of the environment through the plant. A soil may have a par-

H. R. WALLACE

ticular structure that favours nematode mobility but it may not provide the best conditions for plant growth so it has a reduced root system and consequently fewer invasion sites. Infection is therefore reduced, and those nematodes that do invade the roots develop more slowly because of inadequate nutrition. In animal nematodes the interactions between host, parasite, and environment are less definite.

CONCLUSION In this paper I have tried to make the point that nematode movement can usefully be considered as a mechanical problem in which form and function play a major role. It is frequently very difficult to show satisfactorily that form and function are related but it is sometimes a useful and stimulating method of approach because it produces many worthwhile questions that can be answered by doing experiments. Furthermore, the mechanics of movement of nematodes, whatever their host, are so similar that it would be a useful exercise to compare the ecology of animal and plant parasites that occupy the same environment, e.g. the soil or the surface of plants. Studies on the dynamics of nematode locomotion, the influence of the environment on movement, and the characteristics of the nematode that produce efficient movement have enabled the plant nematologist to describe the sorts of conditions under which the parasite is likely to be most active and most infective. I see no reason why the epidemiology of animal parasitic nematodes could not be approached in the same way.

REFERENCES Goodey, T. 1963. Soil and Freshwater Nematodes, rev. éd., éd. J. B. Goodey. London: Methuen & Co. Ltd. Harris, J. E., and Crofton, H. D. 1957. Structure and function in the nematodes : internal pressure

211

THE MOVEMENT OF NEMATODES

and cuticular structure in Ascaris. J. ExptL Biol. 34:116-30 Van Gundy, S. D., Bird, A. F., and Wallace, H. R. 1967. Aging and starvation in larvae of Meloidogyne javanica and Tylenchulus semipenetrans. Phytopathology 57:559-71 Wallace, H. R. 1958a. Movement of eel worms, i. The influence of pore size and moisture content of the soil on the migration of larvae of the beet eelworm, Heterodera schachtii Schmidt. Ann. Appl. Biol. 46:74-85 - 1958b. Movement of eelworms. H. A comparative study of the movement in soil of Heterodera schachtii Schmidt and of Ditylenchus dipsaci (Kuhn) Filipjev. Ann. Appl. Biol. 46:86-94 — 1958c. Movement of eelworms. in. The relationship between eelworm length, activity and mobility. Ann. Appl. Biol. 46:662-8 - 1959. The movement of eelworms in water films. Ann. Appl. Biol. 47:366-70 - 1962. Observations on the behaviour of Ditylenchus dipsaci in soil. Nematologica 7:91-101 — 1966. Factors influencing the infectivity of plant parasitic nematodes. Proc. Roy. Soc. (London), B 164:592-614 - 1968a. The dynamics of nematode movement. Ann. Rev. Phytopathol. 6:91-114 - 1968b. Undulatory locomotion of the plant parasitic nematode Meloidogyne javanica. Parasitology 58:377-91 - 1969. The influence of a non-ionic detergent on the movement of larvae of Meloidogyne javanica. Nematologica 15:107—14 - 1970. Theflexibilityof larvae of Meloidogyne javanica within the egg. Nematologica 16:24957 Wallace, H. R., and Doncaster, C. C. 1964. A comparative study of the movement of some microphagous, plant-parasitic and animal-parasitic nematodes. Parasitology 54:313-26 Wilson, P. A. G. 1958. The effect of weak electrolyte solutions on the hatching rate of the eggs of Trichostrongylus retortaeformis (Zeder) and its interpretation in terms of a proposed hatching mechanism of strongyloid eggs. /. Exptl. Biol. 35:584-601

Discussion NORMAN D. LEVINE

College of Veterinary Medicine, University of Illinois, Urbana, 111., USA Dr Wallace's discussion has been extremely stimulating. However, it has dealt primarily with plantparasitic and free-living nematodes, so I think that I should open the discussion by saying something about how the ideas that he has brought out apply to nematode parasites of animals. There are several differences between plant-parasitic and animal-parasitic nematode larvae. Most of the former occur in the soil (Dr Wallace discussed these primarily), while most animal-parasitic nematode larvae live at first in the faeces and then migrate up onto the vegetation. Some, like hookworm larvae, penetrate the skin actively, while others, like trichostrongyle larvae, are ingested when the final host eats the vegetation. A film of water on the vegetation is necessary for animal-parasitic nematodes to climb up on the vegetation but, except for stating this, I am not sure that a study of the dynamics of nematode locomoion would be paricularly pertinent to their epidemiology. A second important difference between plantand animal-parasitic nematodes is that the infective stage of the former is usually a second-stage larva, while that of the latter is either an egg (containing a larva) or a third-stage larva. In those species in which the larva hatches from the egg, the first- and second-stage larvae feed on micro-organisms in the faeces before they become infective for their final hosts. Do the corresponding larvae of plant-parasitic larvae feed? I don't know. Dr Wallace said that there was a relationship between the length-width (LIa) ratio, the type of undulation, and the speed of progression in plantparasitic nematodes, and suggested in his written paper that a study of the movement of nematodes in narrow channels such as blood vessels would be worth while. The results of a comparison of these factors for microfilariae (first-stage larvae) in the blood should therefore be interesting (Table I).

212

H. R. WALLACE

TABLE I Relationship between size, length-width ratio, undulations, and movement of microfilariae (first-stage larvae) of mammals in blood Nematode species

Host

Length (microns)

Width (microns)

L\W

Dirofilaria immitis Dipetalonema reconditum Wuchereria bancrofti Brugia malayi

Dog Dog Man Man

313 271 280 220

6.5 5.2 7 5.5

48 52 40 40

Loa loa

Man

300

7

43

Mansonella ozzardi

Man

200

4.7

43

Frankly, I can't see any relationship between these factors. The type of undulation and the type of movement (whether progressive or not) appear characteristic of the species and cannot be correlated with each other or with any other factors. The function of the annuli in body-bending and the fact that the cuticle must stretch when the body bends are worth emphasizing. However, I'm not so sure about the suggestion that the annuli are formed by mechanical buckling of the cuticle as the worms bend within the egg. It is true that you can form annuli in this way by bending a piece of latex-covered tubing, but cuticle is an entirely different thing. Further research is needed; it should establish whether annulus formation is determined genetically.

Type of undulation Smooth Smooth Smooth Stiff, with irregular kinky curves Stiff, with secondary curves Smooth

Sheath

Progressive movement

No No Yes Yes

No Yes No Yes

Yes

Yes

No

No

The ecology of onchocerciasis in man and animals B. o. L. DUKE

The term onchocerciasis denotes those pathological manifestations which are produced in the vertebrate host following infection with filarial worms of the genus Onchocerca. Different species of Onchocerca are found in cattle, horses, and various wild ungulates, and their effects on meat and hides cause considerable veterinary problems. Without doubt, however, the human parasite O. volvulus, causing the medical and economic problems of human onchocerciasis, is the most notorious of all. Onchocerciasis affects some 20 million people in the world producing severe chronic pruritus and disfiguring changes in the skin as well as the more dreaded eye lesions that may culminate in River Blindness. The ecology of onchocerciasis is complex and fascinating. As O. volvulus is transmitted by simuliid vectors, it must be studied in relation to the internal environments of both the vertebrate and the invertebrate hosts. Furthermore, because the parasite's continuing existence depends upon the survival of its several hosts, it is intimately affected by the ecology of man and of the Simulium flies, both of whose behaviour may in turn be altered by the presence of the parasite. The complexities of the situation have led me to concentrate in this paper almost exclusively on the human parasite O. volvulus, for this species illustrates all the principles and paradoxes involved in onchocerciasis while at the same time posing many unsolved problems. To appreciate the ecology of O. volvulus an understanding of its life history is essential. In man the adult worms, males and females, live together in tangled masses, each comprising many worms bound together in a fibrous reaction of the host's tissue to form an onchocercal nodule. Nodules vary in size from that of a split pea to a walnut. Some lie superficially and can be palpated just under the skin over the bony prominences of the hips, trochanters, knees, ribs, and

214

scapulae, or over the scalp. Others are impalpable, lying deep in relation to the joint capsules and bones, especially around the hip joint. The females are viviparous, bringing forth large numbers of motile embryos known as microfilariae, and the adult worms may live for as long as 15 years. The formation of the microfilariae is the only stage in the life history at which any multiplication takes place. The microfilariae invade the skin, mainly in the same anatomical quarter as the nodule, and, lying in the subepidermal layer, they await ingestion by the feeding Simulium. They do not multiply or develop further in the human host, but merely accumulate for the duration of their normal life-span, which is of the order of one to two years. The microfilariae can only continue their development when they are ingested by a female Simulium of a suitable species which takes a blood meal from an infected area of skin. The microfilariae then penetrate the gut wall to reach the thoracic musculature of the fly, where development proceeds. There are two larval moults and after some 6 to 12 days (depending on the temperature) mature infective larvae are produced which move away through the haemocoelic fluid and finally come to lie in the head and proboscis. When the fly then bites a new host, the infective larvae emerge from the proboscis and enter the skin through the puncture wound. It should be emphasized again that there is no multiplication of parasites in the fly, one microfilariae developing to one infective larva. Following the entry of the infective larvae into the definitive host little is known of their development until they are found as adults in the usual sites. During this period also there is no multiplication, and the build-up of a heavy worm-load in man depends on the repeated inoculation of infective larvae from the bites of many infected Simulium, particularly those carrying several worms. The pre-patent interval between the inoculation of infective larvae and the appearance of microfilariae in the skin is of the order of 10 to 20months (commonly 15 to 18).

B. O. L. DUKE

THE ECOLOGY OF O. volvulus IN THE VERTEBRATE HOST AND ITS RELATIONSHIP TO THE ECOLOGY OF MAN

The site of entry of the infective larvae into man is determined largely by the areas of the body on which the particular vector species of Simulium feed by preference. Thus the most important African vector S. damnosum bites preferentially on the legs, whereas S. ochraceum in Central America attacks predominantly the upper parts of the body. Broadly speaking, under natural conditions the distribution of the nodules appears to be related to these feeding preferences. In Africa most nodules are found over the knees, hips, and ribs, while head nodules are rare. In Guatemala the majority of the nodules are on the head and shoulders with smaller numbers around the ribs and over the pelvic girdle. In experimentally infected chimpanzees, however, the distribution of the adult worms is not related to the site of inoculation of infective larvae. In these animals the adult worms are almost always found lying deep down in close relation to the posterior surface of the capsule of the hip joint, whether the infective larvae are inoculated into the legs or the arms or the head. The site of the adult worms in turn determines the parts of the body where the highest concentrations of microfilariae are found. A single active nodule on the head can lead to rapid microfilarial invasion of the eye with consequent greatly increased risk of blinding eye lesions. This state of affairs is especially frequent with the Central American strain of the parasite, but it may also occur in Africa, particularly among children. Normally, however, in African onchocerciasis transmitted by S. damnosum, the greatest concentrations of microfilariae are found in the lower half of the body and invasion of the upper parts and of the eyes usually only follows exposure to intense transmission extending over many years. After such prolonged and heavy exposure high concentrations of microfilariae will eventually build up throughout the body.

THE ECOLOGY OF ONCHOCERCIASIS

We have no knowledge as to how the adult worms find their way to their chosen locations in the body or what makes them select these particular sites. It is hard to see any physical, anatomical, or physiological condition which might account for their location. How does O. volvulus in man and chimpanzee come to settle over the bony prominences, or against the capsule of the hip joint? How does O. gutturosa come to find the ligamentum nuchae or the gastrosplenic ligament in the cow, or O. cervicalis in the horse come to lie in the wall of the aorta? It could be that a chemotactic response to products liberated by adult worms attracts infective larvae or immature worms towards an established nodule, but this does not explain how the first worms to arrive locate the favoured sites. In the early stages of its life in man O. volvulus excites no reaction in the host but remains masked and able to move freely through the tissues until such time as it reaches the adult sites. Only then does the remarkable process of nodule formation begin. The nodule is a fibrous structure produced by the host's body in response to the presence of the worm; but the fibrous tissue is produced only around the worm, and does not invade the body of the worm or kill it as it would in the normal foreign-body defence reaction. The fibrous wall of the nodule serves rather to protect the worm from physical damage. It is further remarkable that, following death of the worms from chemotherapeutic treatment, the nodules often resolve and disappear. The mechanisms by which helminth parasites in the tissues avoid being recognized as foreign bodies by the host's tissues are little understood and would be a stimulating field of investigation for immunologists. O. volvulus is of particular interest for it seems to go one step further than most worms and stimulates a controlled reaction in the host's tissue for its own good. Infection with O. volvulus is never fatal to man, if we exclude those rare deaths which occur during treatment of the disease, and which are thought to be due to the side-effects of the drugs

215

used. Most of the known pathological effects of the parasite on the host are produced by the microfilariae. These motile embryos invade the skin, passing, by means unknown, through the tissue spaces in the subepidermal layer. They may be so scanty that many skin snips must be examined to reveal a single microfilaria, and in such cases their presence in the skin may be revealed by treating the host with diethylcarbamazine, which causes a small papule to develop around each microfilaria. In heavily infected persons, however, with 10 or more nodules and harbouring perhaps 50 or more fertile female worms, the concentration of microfilariae in the skin may build up to as much as 200 per mg. The reaction of the individual host to the presence of microfilariae in the skin varies considerably but little is known of the immunological or suppressive mechanisms involved. In many persons who have been exposed to light or moderate transmission since early childhood, there is a high degree of tolerance. Large numbers of microfilariae survive in the skin without producing any symptoms or signs, and the individual is often unaware that he is infected. Obviously in such circumstances the parasite is to be regarded as well-adapted and efficient. The host continues to live a normal life which brings him into contact with the vector, and the probability is high that microfilariae will be ingested by Simulium and thus contribute to the spread of the parasite. In other persons the reaction of the host's tissues to the parasite may be severe, as is particularly common when infection is first contracted in adult life. The arrival of microfilariae in the skin may then set up a severe pruritus and give rise to unsightly rashes. Persons thus affected, especially those well educated and in positions of authority, are persistent in their demands for treatment, and are likely to urge the control and eradication of transmission. Attempts to control or eliminate transmission of O. volvulus also follow from its propensity to invade the eye and produce lesions which can proceed to blindness. Many of the eye lesions are

216

produced by the microfilariae which, especially in heavily infected persons or in those with head nodules, invade the anterior segment of the eye from the adjacent skin. When they die in the ocular tissues they produce minute foreign-body tissue reactions which eventually build up to form gross scarring of the cornea and of the anterior uveal tract( the ciliary body and the iris) which in turn may interfere with the aqueous drainage of the eyes and the nutrition of the lens so that glaucoma or cataract may follow. In addition, and also of great importance as causes of blindness, are the lesions of the posterior segment of the eye, chorioretinitis, choroidal sclerosis, and optic atrophy. These are undoubtedly associated with onchocercal infection, but their exact pathogenesis remains an obscure and interesting field for research. Were it not for its blinding capacities, O. volvulus would attract much less attention. In primitive communities a blindness rate of 1 to 3 per cent from this disease, affecting mainly the oldest age groups, is accepted fatalistically and does not seriously affect the economy of the community. Old persons are often confined to the environs of their huts for the declining years of their lives; consequently their parasites cannot come into contact with the vector. However, there are localities, particularly in the hot, arid, savanna regions of West Africa, where circumstances are such that intense infections develop earlier in life causing blindness to occur more rapidly and more frequently. In these circumstances blindness rates as high as 10 to 15 per cent may be encountered in certain small communties. Affecting particularly the male population at the peak of its working life, this situation seriously disturbs the economic viability of the community and leads eventually to abandonment of the S/raw//wra-infested area. This is unfavourable, of course, to the parasite. In addition to the voluntary removal of human hosts from close contact with the vector, their demands for relief from so obvious a scourge may lead again to the initiation of control measures. These examples illustrate how the pathological manifesta-

B. O. L. DUKE

tions of infection with O. volvulus, caused by a lack of perfect adaptation to the human host, may affect the behaviour of the latter to the detriment of the parasite. Survival of the parasite is favoured, however, by man's eternal need for water in order to satisfy the demands of agriculture, for fishing, and, in modern times, for hydroelectric energy. The Simulium vectors of O. volvulus all require fastflowing water for their larval sites, and wherever man approaches or creates a flowing body of water, be it a fertile river valley or the spillway of a modern dam, he will come in contact with Simulium flies and thus greatly enhance the opportunities for the propagation of its parasite. From our knowledge of the parasitology and transmission of onchocerciasis and from socioanthropological studies (Hunter 1966) it is possible to reconstruct the likely basic sequence of events in a community exposed to heavy transmission of O. volvulus of a strain and in an environment where blindness is a severe hazard. Let us consider the probable chain of events. We may imagine an uninhabited but fertile river valley in the hot savanna zone of West Africa. The river provides breeding sites for 5. damnosum, an insect catholic in its choice of hosts for blood meals, for it can well maintain itself by feeding on birds or animals other than man. Human beings then come to settle in this valley, cultivating the rich soil near the river and building their habitations nearby. Immediately the local S. damnosum are provided with a new animal on which to feed, one that is diurnal, plentiful, readily visible, of characteristic and fairly strong odour, which spends much of its time within easy reach of water, and almost the whole of whose smooth hairless skin provides a ready feeding platform from which to obtain blood meals. As the presence of man, the hunter, also tends to disperse other forms of animal life, the establishment of a high degree of anthropophily by S. damnosum rapidly ensues. Let us assume that O. volvulus is imported into the valley by some of the human invaders. We postulate a human origin for the parasite for to date no wild

THE ECOLOGY OF ONCHOCERCIASIS

reservoir-host of O. volvulus has been found in these regions, and the forest-dwelling great apes (chimpanzees and gorillas) which are susceptible to the infection are not found today in such areas. Initially the spread of the infection from the O. volvulus carrier will be very slow, for microfilariae will rarely be encountered in blood meals taken at random by the feeding fly population, and the establishment of new microfilariae carriers from a new generation of O. volvulus worms will probably take at least two to three years since there is no multiplication of the parasite in the vector. Gradually, however, and from a slow start, the number of infected persons will increase, as will the concentrations of microfilariae in their skins. As a result the infective density in the Simulium population will also increase and the intensity of transmission to which new individuals are exposed from infective Simulium will rise. It may well take 20 to 30 years before the parasite becomes established at its peak level of transmission in the area, and individuals in the community will have to undergo exposure to this level of transmission from birth before they may be expected to develop heavy infections sufficiently early in life for the production of serious eye lesions. On this time scale it may be 40 to 60 years before blindness becomes a problem in the community and a further 10 to 20 years before its prevalence begins to affect the economy of the population, rendering it no longer viable. There will then follow a gradual retreat from the environs of the river, because of its association with blindness in the minds of the people. The 5/mw//wm-infested areas are abandoned and the community moves elsewhere. For a generation or two thereafter tradition associates the area with these misfortunes; but as human memory is short, as the demand for good land increases, and as young men are ambitious, one day the potentially fertile valley will again be put under cultivation, at first with an apparent immunity to ocular troubles which seems to belie the old legends. Thus the cycle starts again. This succession of events is admittedly hypo-

217

thetical and probably oversimplified but I believe it approximates to what has happened in the past and to what may still happen today in the absence of control. However, the cycle is so long that a short-term visitor to these lands may not appreciate what is happening. He will see, and be struck by, those communities which are disintegrating in the last stages of the cycle. He may not appreciate that other apparently viable communities in onchocercal zones are heading slowly for the same fate unless control operations intervene. The question whether O. volvulus occurs naturally in other hosts than man is a vexed one. Among a wide range of experimental animals tested the parasite has only been successfully transmitted to chimpanzees (Duke 1962). It is not known whether these animals ever become infected in their natural habitat. There is no reason why they should not be, although it is doubtful whether their habits in virgin forest would bring them into close contact with the main anthropophilic Simulium. The question is not readily answerable for, as a closely protected animal, the adults are rarely shot and examined for infection. O. volvulus has been reported once in a wild gorilla. There is evidence that S. damnosum and S. neavei feed readily on other hosts as well as man. Both species have been found harbouring natural infections with filarial parasites other than O. volvulus (Nelson and Pester 1962; Duke 1967) and evidence from precipitin tests on blood meals from wild flies caught restingo on vegetation o c indicates that S. damnosum certainly feeds on birds (Disney and Boreham 1969). There is thus ample opportunity for infective larvae of O. volvulus from man to be inoculated by Simulium into non-human hosts, while at the same time man must from time to time be exposed to challenge with animal or avian parasites. In the present state of knowledge, although the search for possible reservoir-hosts should continue, it is best to regard O. volvulus as a parasite of man and his progenitors. On this basis its evolution in Africa may be reasonably

218

conjectured, but its occurrence in America is less easily explained. There are very considerable differences in the host-parasite-vector complexes of the African and American regions, indicating a remarkably close adaptation to the local vectors (de Leon and Duke 1966). There are also historical and anthropological records which indicate that the Central American parasite was already established in pre-Colombian times (Figueroa 1963 ). It is improbable that onchocerciasis in the New World is derived from infections carried by African slaves, and the parasite now found in the Americas would appear to have undergone a long period of evolution on that continent. THE ECOLOGY OF O. Volvulus IN THE INSECT HOST

The ecology of O. volvulus in the intermediate host is intimately linked with the necessity for the Simulium vectors to maintain a close association with running water suitable for the development of their eggs, larvae, and pupae. The distribution of the vectors being thus limited by the aquatic factor, the zones where the parasite is transmitted are confined to within the flight range of the fly from its breeding sites. Dissemination of the parasite to new areas usually takes place by movement of infected persons into places where previously uninfected Simulium populations were living, fishing being perhaps the human occupation which most commonly brings the parasite into contact with new sources of flies. It must further be remembered that many water courses in the tropics, especially in the more arid parts, dry up for several months of the year. They can only maintain actively breeding Simulium populations for a limited season. The transmission of O. volvulus is thus similarly confined, and the parasite must maintain itself in the vertebrate host until such time as the vector population is re-established. We have already noted that the microfilariae of O. volvulus tend to collect in the skin over

B. O. L. DUKE

those portions of the body from which they are most likely to be ingested by feeding Simulium. In Central America they are concentrated in the head, body, and arms where S. ochraceum commonly bites. In Africa they are mainly in the legs, the preferred biting site for S. damnosum and S. neavei. This selective spatial distribution, although striking, is not so remarkable as that of O. gut tur osa in the cow, where the adult worms live in the ligamentum nuchae and the microfilariae concentrate almost exclusively in a limited portion of the skin of the umbilicus which is the elected biting site for the vector S. ornatum (Professor G. S. Nelson, personal communication). Onchocerca microfilariae do not exhibit any marked temporal periodicity similar to that of the blood-dwellng microfilariae of Wuchereria bancrojti or Loa loa. Their location in the tissue spaces of the skin makes frequent and rapid population movements more difficult, but there is some evidence in African O. volvulus of a gentle fluctuation in concentration in the superficial parts of the subepidermal layers, which may coincide broadly with the peak biting densities of S. damnosum (Duke et al. 1967). Despite these possible spatial and temporal factors, the ingestion of microfilariae by feeding S. damnosum remains widely random and, although on heavily infected skin almost all flies will ingest some microfilariae, the range of intake may well be from zero to several hundred. In the American forms of the parasite there appears to be a definite and powerful attraction of microfilariae towards the feeding fly which tends to concentrate them for ingestion by S. ochraceum, S. metallicum, S. callidum, and S. exiguum. This concentration of microfilariae toward the feeding fly, which may be a chemotactic response to some substance in the fly's saliva, is confined to the American strains of the parasite and does not occur when the American vectors are fed on carriers of West African strains of the parasite. It is capable of multiplying the number of microfilariae ingested by a factor of 10 to 20 times.

THE ECOLOGY OF ONCHOCERCIASIS

The sudden change of environment encountered when microfilariae are taken from the skin into the gut of the fly provides the stimulus for their further development. For successful development the microfilariae must pierce the gut wall and enter the thoracic muscles. Moreover, the passage out of the gut must be accomplished rapidly before the chitinous peritrophic membrane is secreted around the ingested blood meal, imprisoning those microfilariae remaining inside (Lewis 1953). The peritrophic membrane appears to be one mechanism by which an excess of ingested microfilariae is reduced to supportable proportions. Many flies ingest more microfilariae than it would be physically possible for them to support to the infective stage. The increased mortality in the fly that is associated with the ingestion of excessive numbers of parasites takes place in the first 24 to 48 hours. Likewise most of the elimination of excess parasites occurs during this period. Thereafter there is no appreciable added mortality in the fly associated with the infection, and the great majority of the parasites which survive 24 to 48 hours may be expected to reach the infective stage, although some may die at each of the two further moults taking place during development. There is some evidence that the age of the microfilariae when they are ingested affects their ability to survive in the fly, those which are young at the time of ingestion having a slightly higher viability (Duke 1968a). The number of infective larvae produced in individual flies fed on heavily infected areas of skin varies considerably, but the total number of infective larvae maturing can be very great. Sixty-nine have been recorded from a single experimentally fed S. damnosum, and 20 to 30 is not an uncommon number to find in individual wild flies. The development of the parasite in the flight musculature of the thorax may theoretically impede the powers of flight of the infective fly but there is as yet no direct evidence indicating that this is an important factor limiting transmission in nature.

219

The length of time the parasite takes to develop in the fly varies with the ambient temperature to which the fly is exposed. In hot weather, with an average temperature through the 24 hours of 75° F, development may be completed in 6 days. In cool weather the cycle may extend to 10 to 12 days. It is, however, always linked accurately with the average length of the gonotrophic cycle of the insect, which is likewise dependent on temperature. The female Simulium digests the infecting blood meal, matures its ovaries, and deposits its eggs by the time the parasites are half-grown. It takes its next blood meal soon after and, having digested this and deposited its eggs, it comes in for its third blood meal at the time when the parasites are just mature and have reached the proboscis. Under normal circumstances it is at this and subsequent feeds that the parasite is passed on to its new vertebrate host. Transmission of the parasite by Simulium depends on the multitude of environmental factors which affect the vector itself. Among those to be considered as influencing the efficiency of transmission are the absolute numbers of the vectors available, their longevity, their degree of anthropophily, and the climatic conditions prevailing. In absolute terms the amount of transmission can perhaps best be measured by means of an index known as the transmission potential (Duke 1968b) which, for any given environment, is the number of infective larvae to which a man would be exposed under optimal biting conditions in unit time. To take account of possibly wide seasonal variations in the amount of transmission the ideal unit of time to consider is one year. Although such an index may give a figure much in excess of the real exposure of any individual in the environment under investigation, it offers nevertheless a reasonable basis for comparing transmission in different environments and for making comparisons in the same environment before and after vector control operations. The relationship between O. volvulus and its intermediate hosts is extremely intricate, and

220

recent work has shown that in West Africa there are at least two distinct parasite vector complexes involving O. volvulus and S. damnosum. It appears that there are different forms of S. damnosum adapted to the different climatic zones of West Africa which are encountered as one proceeds north from the humid equatorial rain-forests to the hot, arid savannas that give way eventually to the Sahara desert. Probably different physiological adaptations are required to ensure the survival of 5. damnosum populations at the two climatic extremes between which its range extends, and there are morphological differences affecting colour, size, and bristle patterns between populations taken from these different regions (Lewis and Duke 1966). Associated with these is the remarkable fact that each different form of the fly will only transmit the strain of O. volvulus from the same bioclimatic zone (Duke, Lewis, and Moore 1966). Forest S. damnosum from West Africa are equally well able to ingest microfilariae from human carriers coming either from forest areas or from the northern savannas, but whereas the forest microfilariae develop readily to infective larvae the microfilariae of the northern Sudan-savanna strains give rise to little or no development. Likewise Sudan-savanna S. damnosum will support the development of microfilariae from carriers living in their own zone but permit little or no development of the forest strain of O. volvulus. With minor local variations we have then in West Africa two main parasite-vector complexes. One extends for some 1,500 miles east-west in the forest and over the greater part of the Guinea-savanna zone; the other extends likewise over a great distance in an east-west direction but is confined to the Sudan-savanna zone and to the extreme north of the Guinea-savanna zone. The frontier between the two, as one proceeds from south to north, is extremely sharp at least in the places where it has been investigated in detail. To account for these findings we must postulate that there are at least two main strains of O. volvulus in West Africa, one in the northern savannas and the other covering the rest of the

B. O. L. DUKE

area; and the two strains do indeed also behave somewhat differently in the readiness with which they can be transmitted to chimpanzees. Strain difference in O. volvulus may well be one of the factors giving rise to the marked differences in the pattern of the disease in the different bioclimatic zones of West Africa. The most severe blinding onchocerciasis is found only in the area in which the Sudan-savanna strain is transmitted, and it may be that the strain of parasite, combined with the harsher environmental factors operating in the area, accounts for its greater pathogenicity. CONTROL OF ONCHOCERCIASIS

No remarks on the ecology of onchocerciasis would be complete without including some reference to the control of the disease. Mankind now appreciates the dangers of the parasite and seeks to control or eradicate it. Today this is perhaps the most important factor in the ecology of the worm. Using scientific methods man tries first to understand the complex relationships of the parasite with its varied environments and then to use this knowledge to find ways of breaking the transmission cycle. At present these efforts are concentrated mainly on vector control. The application of DDT as a larvicide to Simulium breeding sites in running water is a relatively cheap and easy way to deprive the parasite of its vector. This method has indeed been used successfully to eradicate S. neavei from Kenya (Garnham and McMahon 1954) and has thus brought transmission to a halt. Following treatment the parasite has slowly died out of the human population over a period of some 15 years. With the other main vector species, S. damnosum, S. ochraceum, and S. metallicum, it appears that control rather than eradication is all that can be hoped for from the use of larvicides. Such control must be maintained over very large areas so as to render insignificant the extent of peripheral invasion by Simulium flying in from outside the control area. Likewise the transmission must be reduced to insignificant levels (which are as yet

THE ECOLOGY OF ONCHOCERCIASIS

undetermined) for periods of at least 15 years before there can be any hope that the parasite will have died out of the human population. Such long-term insecticide treatment raises many ancillary problems. There is the possible deleterious action on other forms of aquatic life, including fish, which form an important item of human diet in these regions. There is also the everpresent threat that insecticide resistance might develop. The effect of vector control on transmission would be much more rapid and efficacious if it could be combined with measures against the parasite reservoir in man. By means of such a combined attack it might be possible to reduce transmission to the level of the "break-point," whose existence has been postulated on mathematical grounds by Macdonald (1965 ) as applying to all bisexual helminthic infections. The "break-point" is the level of transmission below which the parasite is no longer able to maintain itself in the community and thus becomes selfeliminating. It is of great practical interest in helminthology to determine where the "breakpoint" lies for any given parasitic infection. Does it lie so low as to be for all practical purposes synonymous with total eradication, or does it lie significantly above this at a level which can be achieved relatively early and economically in practice? For O. volvulus we have only two weapons for dealing with the parasite in man - nodulectomy and chemotherapy. Experience in Guatemala has shown that continued and thorough routine nodulectomy may reduce the severe ophthalmological complications of the infection, but has little effect on transmission. In the realm of chemotherapy there exists as yet no remedy which is sufficiently safe and practical to be used on a large scale for mass treatment in heavily infected areas. Unfortunately, therefore, our means of attacking the parasite are severely limited. It remains difficult to establish the level of the "break-point" in this infection, and impossible, without vast expense, to achieve more than a limited degree of control.

221

REFERENCES De Leon, J. R., and Duke, B. O. L. 1966. Experimental studies on the transmission of Guatemalan and West African strains of Onchocerca volvulus by Simulium ochraceum, S. metallicutn and S. callidum. Trans. Roy. Soc. Trop. Med. Hyg. 60:735 Disney, R. H. L., and Borham, P. F. L. 1969. Blood-gorged black flies in Cameroon and evidence of zoophily in Simulium damnosum. Trans. Roy. Soc. Trop. Med. Hyg. 63:292 Duke, B. O. L. 1962. Experimental transmission of Onchocerca volvulus to a chimpanzee. Trans. Roy. Soc. Trop. Med. Hyg. 56:271 - 1967. Infective filaria larvae, other than Onchocerca volvulus, in Simulium damnosum. Ann. Trop. Med. Parasitol 61:200 - 1968a. The intake and transmissibility of Onchocerca volvulus microfilariae by Simulium damnosum fed on patients treated with diethylcarbamazine, suramin or Mel W. Bull. Wld. Hlth. Org. 39:169 - 1968b. Studies on factors influencing the transmission of onchocerciasis. iv. The biting cycles, infective biting density and transmission potential of "forest" Simulium damnosum. Ann. Trop. Med. Parasitol. 62:95 Duke, B. O. L., Lewis, D. J., and Moore, P. J. 1966. Onchocerca-Simulium complexes. I. Transmission of forest and Sudan-savanna strains of Onchocerca volvulus, from Cameroon, by Simulium damnosum from various West African bioclimatic zones. Ann. Trop. Med. Parasitol. 60:318 Duke, B. O. L., Scheffel, P. D., Guyon, J., and Moore, P. J. 1967. The concentration of Onchocerca volvulus microfilariae in skin snips taken over 24 hours. Ann. Trop. Med. Parasitol. 61:206 Figueroa, H. M. 1963. Historia de la Enfermedad de Robles en America y de su descubrimiento en Guatemala. In Enfermedad de Robles. Academia de Ciencias Medicas, Físicas y Naturales de Guatemala Garnham, P. C. C.? and McMahon, I. P. 1954. Bull Entomol Research 45:175

222 Hunter, J. M. 1966. River blindness inNangodi, Northern Ghana : A hypothesis of cyclical advance and retreat. Geograph. Rev. 56:398 Lewis, D. J. 1953. Simulium damnosum and its relation to onchocerciasis in the Anglo-Egyptian Sudan. Bull. Entomol. Research 43:597 Lewis, D. J., and Duke, B. O. L. 1966. OnchocercaSimulium complexes, n. Variation in West African female Simulium damnosum. Ann. Trop. Med. Parasitol. 60:337 MacDonald, G. 1965. The dynamics of helminthic infection with special reference to schistosomes. Trans. Roy. Soc. Trop. Med. Hyg. 59:489 Nelson, G. S. and Pester, F. R. N. 1962. The identification of infective filarial larvae in Simuliidae. Bull. Wld. Hlth. Org. 27:473

Discussion R. C. ANDERSON

Department of Zoology, University of Guelph, Guelph, Ont. We are most grateful to Dr Duke for his lucid account of onchocerciasis. Dr Duke has explained the need in Africa for flowing water for use in irrigation. Yet when man creates flowing water in endemic areas he runs the risk of creating new breeding-sites for vectors of Onchocerca volvulus. When he moves into previously unpopulated areas he tends to replace other large animals and becomes a main source of blood for simuliids. Thus, onchocerciasis, like schistosomiasis, is an unexpectedly urgent problem of mid-20th-century man in Africa. At present control of onchocerciasis relies heavily on the use of DDT. As Dr Duke has mentioned, control of damnosum would require the use of DDT over extremely wide areas for long periods of time. I think many here, including Dr Duke, view with some apprehension the possible long-range effects of the use of insecticides on this scale. It is most encouraging, therefore, that broad aspects of onchocerciasis are being investigated with the hope that integrated control methods can be developed which will be ecologically acceptable.

B. O. L. DUKE

The discovery by Dr Duke and Dr Lewis that there are various forms of damnosum and at least two main strains of volvulus characteristic of major bioclimatic zones of Africa is most interesting. Perhaps damnosum refractory to one strain of volvulus could be induced to replace ecologically a form of damnosum which is a vector of that strain. On the other hand it might be as realistic to reverse the human population. I am curious to know what consideration has been given to the use of suitable clothing and fly repellents along with the use of insecticides, nodulectomy, and chemotherapy. Is it feasible in the context of Africa? Perhaps some of the points may come out in the discussion. Subcutaneous filarioids rarely elicit nodules and one wonders if the presence of nodules does not indicate a rather imperfect host-parasite relationship, perhaps a recent one. Dr Duke apparently thinks it may be a special adaptation. In Central America nodules occur mainly in the upper parts of the body whereas in Africa nodules are found in the lower or middle parts. These two distributions are generally attributed to the feeding habits of vectors. Does this mean infective larvae develop in the general region of the bite, which is to imply worms do not move extensively, or does it mean that even if worms were introduced elsewhere they would migrate to these regions - in other words are there genetically controlled behavioural differences between African and Central American volvulus which have evolved in response to the feeding behaviour of vectors? I am impressed by the discovery that volvulus tends to localize in the hip region of the chimpanzee regardless of where it is introduced. It is my understanding that it does not elicit a nodule. One immediately wonders if volvulus is not more adapted to the chimpanzee, which raises the further thought that this helminth has its origins in wild primates and that there exist reservoirs unrecognized as such. Dr Duke has thought much about these matters and perhaps he could be prevailed upon to expand on his previous remarks.

Mosquito vector and vertebrate host interaction: The key to maintenance of certain arboviruses WILLIAM C. REEVES

This research was supported in part by Research Grant AI 03028 from the National Institute of Allergy and Infectious Diseases and General Research Support Grant i-SOl-FR-05441 from the National Institutes of Health, United States Department of Health, Education and Welfare.

Arthropod-borne diseases are dependent on the blood-feeding relationship of an arthropod vector to a vertebrate host. Our present knowledge of mosquito-borne diseases, such as malaria, filariasis, and a number of arboviruses, have led us to focus attention on particular attributes of vector populations that influence the success of pathogen transmission. The characteristics of vector populations that are recognized to be most critical are: numerical abundance, longevity at temperatures that favour completion of extrinsic incubation of a pathogen, and affinity of the vector for a vertebrate species that can circulate the pathogen in its blood in an infective dose for the vector. These three variables with accompanying variables of the host and environment have been investigated in the greatest depth and sophistication with reference to malaria (MacDonald 1957), and the concepts and data for that disease have been transferred to a stochastic model that has considerable theoretical and applied interest. It is inevitable that in the near future a similar effort will be made to develop statistical models to express the epidemiologic insight that we have of arboviruses such as western equine encephalomyelitis (WEE), St. Louis encephalitis (SLE), eastern equine encephalomyelitis (EEE), Venezuelan equine encephalitis (VEE), and Japanese B encephalitis (JBE). An extensive base of research data has accumulated for each of these infections. These agents are of considerable public health and veterinary importance as they have been responsible for recurring epidemics in the Americas and Asia. My purpose is to discuss some of the problems that must be considered if we are to transfer concepts from the malaria model to an arbovirus such as WEE and particularly to focus attention on certain aspects of vector-host relationships that have come to our attention in studies in the western United States. I shall use WEE as the principal reference point for a mosquito-borne

224

arbovirus. Most of the data referred to were accumulated by my colleagues in this research effort. SIGNIFICANT BIOLOGIC DIFFERENCES BETWEEN THE BASIC CYCLES OF MALARIA AND WEE

Human malaria has been subject to extensive study because of its world-wide health importance. The disease is caused by three principal species of parasites; there is a single host species, man; and the maintenance cycle for these infections depends on an effective sequential transfer of parasites solely from man to man by Anopheles mosquitoes. Although the total epidemiology of malaria is indeed complex, the impact of a rise or fall in Anopheles populations, variations in the affinity of different Anopheles species for man, and the influence of control on vector populations have all been related directly to increases or decreases in the prevalence of infection and disease in the host population (MacDonald 1957). An arbovirus such as WEE presents a very different epidemiologic problem. The biologic maintenance of infection is dependent on infection in a wide range of avian species and the sequential transfer of infection is by Culex mosquitoes. The principal difference from malaria in this basic cycle is that many avian species can be effective hosts for the infection, which is usually inapparent. However, malaria ceases to be a useful reference point when one is concerned with the epidemiologic importance of WEE or similar arboviruses as a cause of clinical disease in man or domestic animals. The problem is that infection in the clinically susceptible hosts represents a tangential extension of the infection cycle into hosts that do not contribute significantly to maintenance of this cycle. In malaria, such transfers of parasites to aberrant hosts do not lead to infection or disease. Such occurrences are advantageous to man and have even been considered as a means of control under the designation "zooprophylaxis." For WEE, EEE, or SLE viruses, such an event is calamitous for man and

WILLIAM C. REEVES

is called an epidemic. Thus, our applied interests in the control of the arboviruses have two areas of concern with reference to vector populations. The first concern is similar to that for malaria in that we wish to know to what critical threshold level we must reduce a vector population to interrupt effective transmission of arboviruses through contact between the vector and basic vertebrate host population. We have an equally urgent interest to determine if there is a critical threshold level for the vector population below which there will be little or no significant risk of infection in the aberrant hosts, man and horse, even though the basic cycle continues (Reeves 1968, 1969). With the preceding background, I shall turn to some specific research data that may influence our considerations of vector-vertebrate host interactions with reference to the transmission of WEE virus. INTERACTIONS OF CuleX tarsulis AND VERTEBRATE HOST POPULATIONS

Identification of blood meals from Anopheles mosquitoes by the precipitin test provided a simple and useful technique for malariologists, as, in its simplest form, it provided data on the proportion of an Anopheles population that fed on a single host species, man (MacDonald 1957; Weitz 1952). In recent years the precipitin test has undergone extensive refinement and it now provides useful data to arbovirologists (Reeves elal. 1963;Tempelis andLofy 1963;Tempelis et al. 1965 ) in that the extent to which a Culex population has fed on a wide range of potential vertebrate sources of blood can be determined. Figure 1 presents data on the relative frequency of avian versus mammalian blood feeding by Culex tarsalis from four widely separated geographic areas. This mosquito is the principal vector of WEE virus in western North America. The proportion of blood meals from mammalian hosts relative to avian hosts increased by mid or late summer in each of these areas. The fact that this occurred so similarly in such diverse areas indicated that it was a consistent biologic pheno-

MOSQUITO VECTOR AND VERTEBRATE HOST INTERACTION

225

Hale County, Texas 1964-1966 Salt Lake County, Utah 1966 Weld County, Colorado 1964 Kern County, California 1960-1962

FIGURE 1. The monthly percentage of Culex tarsalis that fed on mammals in Kern County, California; Weld County, Colorado; Salt Lake County, Utah; and Hale County, Texas. (Source: Published and unpublished data from Dr C. H. Tempelis, School of Public Health, University of California, Berkeley, California. The percentage unaccounted for represented blood meals from birds.) menon and should be reflected in the patterns of virus transmission between, and from, avian host populations. Details of the species of hosts involved have been given in a series of publications (Reeves etal 19 63; Tempelis ¿f al 1967; Tempelis 1970; Andersen, Collett, and Winget 1967). The important biologic problem was to explain this change in vector feedings. The epidemiologic implications were obvious in that the mosquito vector fed almost exclusively on effective viral hosts in the spring and early summer,

which would favour establishment and acceleration of viral transmission. Then, as the summer advanced, there was a progressive increase of feeding on aberrant mammalian hosts that included clinically susceptible species. In the fall and winter in California (Tempelis et al. 1965 ), the vector returned to avian hosts as the principal source of blood. Similar changes from avian to mammalian blood sources were reported for Culex nigripalpus, the mosquito that was believed to be the principal epidemic vector of SLE

226

WILLIAM C. REEVES

virus in Florida (Edman and Taylor 1968). An unsuccessful effort was made in both the west and Florida to correlate the shifts in feeding patterns with changes in the abundance and availability of the two major host groups. The only factors that consistently correlated with the increase in mammalian blood meals were a numerical increase in the vector population and the height of temperature as the summer progressed. When these variables reverted to lower values in the fall, the proportion of blood meals from birds increased. It was difficult to advance a biologic hypothesis that temperature alone would alter host preference or availability or would select a vector population that favoured one host group over another. Superficial review of the literature failed to reveal similar observations on mosquito vectors of malaria, filariasis, or other arthropodborne diseases. It appeared most probable that increases or decreases in the population of vectors in some way influenced a significant shift in the hosts that the vector fed on. Earlier studies on the development of a "baitcan" mosquito trap to expose sentinel chickens to vector attacks (Bellamy and Reeves 1952) and to obtain viral infection and transmission rates of C. tarsalis populations (Reeves, Bellamy, and Scrivani 1961) provided data that would possibly explain the observed shifts in host feedings. Table i summarizes findings from the TABLE I Relation of engorgement of female Culex tarsalis to the total numbers caught in non-rotated traps baited with chickens* Range in total female C. tarsalis^

Number Per cent Per cent with of C. tarsalis trace of blood in catches with blood total with bloodt

1-100 101-200 201-350 351-500 501-1000 1565

16 13 16 10 11 1

80 72 60 53 36 12

4 6 10 10 13 33

*Source: Dow, Reeves, and Bellamy (1957). tExclusive of gravid specimens and those too damaged to permit the determination of the state of engorgement. {Calculated from the totals in each group of catches.

first summer when young chicks were exposed in bait cans (Dow, Reeves, and Bellamy 1957). Two significant trends were observed: as the number of mosquitoes attracted to an unrestrained host increased, the proportion that took blood decreased, and the proportion that took a partial blood meal increased. Two interpretations were possible : ( 1 ) as the number of mosquitoes increased, they interfered with each other's feeding effort; and (2) as the number of mosquitoes that attempted to feed increased, the host became intolerant and resisted further attacks. In subsequent years, the observations were expanded and some chicks were restrained in fine mesh cloth so their movement was restricted (Table n). As in previous studies, the proportion of mosquitoes that fed on unrestrained chicks decreased as the numbers of mosquitoes increased; when chicks were restrained, nearly all mosquitoes fed regardless of the number attracted. These observations showed that mosquitoes did not interfere with each other from sheer numbers alone. When chicks of different ages were exposed in bait cans (Table in), there again was an indication that younger and smaller chicks were less tolerant of mosquito feedings than were older birds, and restraint of the birds eliminated the differences. The only other study of age differential response of birds to mosquito attack (Blackmore and Dow 1958) indicated that nestlings of three species were more tolerant of mosquito feeding than were adult birds. If the simple assumption is made that most species of birds are like chickens and are intolerant to attack by large numbers of C. tarsalis, we have a simple biologic explanation for the common observation that an increased proportion of feedings by C. tarsalis is diverted to mammalian hosts in the period when the vector is most abundant. In the spring, early summer, fall, and winter when the vector population is small, the bird population accommodates the majority of blood-sucking C. tarsalis. The epidemiologic implications of this phenomenon are important if we are to identify the critical threshold level of the vector population that is necessary to main-

MOSQUITO VECTOR AND VERTEBRATE HOST INTERACTION

227

TABLE II Relation of Culex tarsalis population to success of feeding on chicks less than three weeks old; field exposures in Kern County, California Unrestrained chicks

Restrained chicks

Number of mosquitoes in trap

Number of tests

Per cent mosquitoes fed

Number of tests

Per cent mosquitoes fed

1-100 101-200 201-300 301-400 401-500 500 +

48 35 17 7 4 13

52 38 49 45 36 22

32 15 2 1 1 2

96 98 97 99 99 99

TABLE III Relation of chick age to success of mosquito feeding; field exposures in Kern County, California Unrestrained chicks

Restrained chicks

Age of chicks in weeks

Number of tests

Per cent mosquitoes fed

Number of tests

Per cent mosquitoes fed

12

124 25 43 2

36 53 65 80

53 10 10

98 95 96

tain viral transmission from bird to bird. In addition, the influence of an increased vector population, if it leads to divergence of feeding to aberrant mammalian hosts, can be reflected in an increased epidemic potential. One can even hypothesize that when there is an extremely large C. tarsalis population in an area where there also is an available population of large mammals such as cattle, the chances that an individual C. tarsalis will take repeated blood meals from a bird are so small that viral transmission will be seriously impaired. I believe we have seen this situation in some western localities. The situation described above is reminiscent of the concept of zooprophylaxis in malaria control. Who in this audience does not believe that man is intolerant of excessive mosquito bites as are birds? Thus, we have a somewhat analogous situation for malaria in that a species that is a preferred or at

least is a highly acceptable source of blood for the vector protects itself, Anopheles transfer their feeding to other mammals if they are available, and incidentally, or purposefully, the level of pathogen transmission is dampened. THRESHOLD LEVELS OF VECTOR POPULATIONS CRITICAL TO ARBOVIRUS TRANSMISSION

We can now consider the relationship of the preceding data on vector-host interactions to the several threshold levels of vector populations that will be reflected in different patterns of viral transmission. Over the past 30 years a number of epidemics of WEE, SLE, and EEE have been described in North America. In western North America, each epidemic resurgence of WEE has been associated with high populations of C. tarsalis, and there

228

WILLIAM C. REEVES

have been similar large populations of a principal vector associated with epidemics of EEE and SLE in other regions. The conclusion of the preceding considerations that C. tarsalis would shift feedings from avian to mammalian hosts in the presence of a large vector population is compatible with the expectation that these circumstances should lead to an increased risk of human and horse infection. The epidemics of SLE in Florida where C. nigripalpus was the epidemic vector could be explained similarly (Edman and Taylor 1968). One suspects that the epidemics of Japanese B encephalitis in Asia and urban epidemics of SLE in mid-western cities of North America may reflect a similar situation. In California, we have had a unique opportunity over the past 25 years to observe the effect of intensified C. tarsalis control on the incidence of clinical cases of WEE in man and horse and of SLE in man. We have seen a significant decrease in the occurrence of clinical cases, particularly in urban centres where the C. tarsalis control program was most effective (Reeves 1968,1969).

When one uses a seasonal index of C. tarsalis, as measured by females per light trap night, it appears that SLE disappeared as a clinical disease when the index was between an average of 10 and 20 per trap night and clinical cases of WEE disappeared when the index was reduced below an average of 10.1 now believe that indices in these ranges represent vector population levels for which the rate of feeding of C. tarsalis on, and virus transmission to, the aberrant hosts are so low that the chances of a clinical infection developing are practically zero, although there could still be a low incidence of inapparent infections in man. There was ample evidence that the two viruses continued to persist in their basic vertebrate host-vector cycles. In the period 1960 through 1963, we attempted to control the aquatic stages of C. tarsalis and to stop the basic transmission cycles of WEE and SLE viruses in rural areas that varied from 18 to 60 square miles in size. We succeeded in reducing the average seasonal light trap indices of female C. tarsalis to 21 to 36 per trap

TABLE IV Annual relationship of Culex tarsalis population indices to western equine encephalitis viral activity indices in urban, rural community, and agricultural environments, Kern County, California

Population index*

Infection rate per thousand

Year and area

Bait trap

Light trap

1965 Urban Rural community Rural agricultural