Diatom Gliding Motility: Biology and Applications [1 ed.] 1119526353, 9781119526353

Table of contents :
Cover
Half-Title Page
Series Page
Title Page
Copyright Page
Dedication to Jeremy D. Pickett-Heaps In Memoriam 1940–2021
Contents
Preface
1 Some Observations of Movements of Pennate Diatoms in Cultures and Their Possible Interpretation
1.1 Introduction
1.2 Kinematics and Analysis of Trajectories in Pennate Diatoms with Almost Straight Raphe along the Apical Axis
1.3 Curvature of the Trajectory at the Reversal Points
1.4 Movement of Diatoms in and on Biofilms
1.5 Movement on the Water Surface
1.6 Formation of Flat Colonies in Cymbella lanceolata
1.7 Conclusion
References
2 The Kinematics of Explosively Jerky Diatom Motility: A Natural Example of Active Nanofluidics
2.1 Introduction
2.2 Material and Methods
2.2.1 Diatom Preparation
2.2.2 Imaging System
2.2.3 Sample Preparation
2.2.4 Image Processing
2.3 Results and Discussion
2.3.1 Comparison of Particle Tracking Algorithms
2.3.2 Stationary Particles
2.3.3 Diatom Centroid Measurements
2.3.4 Diatom Orientation Angle Measurements
2.3.5 Is Diatom Motion Characterized by a Sequence of Small Explosive Movements?
2.3.6 Future Work
2.4 Conclusions
Appendix
Dynamics of Diatom in Low Reynolds Number Regime
References
3 Cellular Mechanisms of Raphid Diatom Gliding
3.1 Introduction
3.2 Gliding and Secretion of Mucilage
3.3 Cell Mechanisms of Mucilage Secretion
3.4 Mechanisms of Gliding Regulation
3.5 Conclusions
Acknowledgments
References
4 Motility of Biofilm-Forming Benthic Diatoms
4.1 Introduction
4.2 General Motility Models and Concepts
4.2.1 Adhesion
4.2.2 Gliding Motility
4.2.3 Motility and Environmental Responsiveness
4.3 Light-Directed Vertical Migration
4.4 Stimuli-Directed Movement
4.4.1 Nutrient Foraging
4.4.2 Pheromone-Based Mate-Finding Motility
4.4.3 Prioritization Between Co-Occurring Stimuli
4.5 Conclusion
References
5 Photophobic Responses of Diatoms – Motility and Inter-Species Modulation
5.1 Introduction
5.2 Types of Observed Photoresponses
5.2.1 Light Spot Accumulation
5.2.2 High-Intensity Light Responses
5.3 Inter-Species Effects of Light Responses
5.3.1 Inter-Species Effects on High Irradiance Direction Change Response
5.3.2 Inter-Species Effects on Cell Accumulation into Light Spots
5.4 Summary
References
6 Diatom Biofilms: Ecosystem Engineering and Niche Construction
6.1 Introduction
6.1.1 Diatoms: A Brief Portfolio
6.1.2 Benthic Diatoms as a Research Challenge
6.2 The Microphytobenthos and Epipelic Diatoms
6.3 The Ecological Importance of Locomotion
6.4 Ecosystem Engineering and Functions
6.4.1 Ecosystem Engineering
6.4.2 Ecosystem Functioning
6.5 Microphytobenthos as Ecosystem Engineers
6.5.1 Sediment Stabilization
6.5.2 Beyond the Benthos
6.5.3 Diatom Architects
6.5.4 Working with Others: Combined Effects
6.5.5 The Dynamic of EPS
6.5.6 Nutrient Turnover and Biogeochemistry
6.6 Niche Construction and Epipelic Diatoms
6.7 Conclusion
Acknowledgments
References
7 Diatom Motility: Mechanisms, Control and Adaptive Value
7.1 Introduction
7.2 Forms and Mechanisms of Motility in Diatoms
7.2.1 Motility in Centric Diatoms
7.2.2 Motility in Pennate Raphid Diatoms
7.2.3 Motility in Other Substrate-Associated Diatoms
7.2.4 Vertical Migration in Diatom-Dominated Microphytobenthos
7.3 Controlling Factors of Diatom Motility
7.3.1 Motility Responses to Vectorial Stimuli
7.3.2 Motility Responses to Non-Vectorial Stimuli
7.3.3 Species-Specific Responses and Interspecies Interactions
7.3.4 Endogenous Control of Motility
7.3.5 A Model of Diatom Vertical Migration Behavior in Sediments
7.4 Adaptive Value and Consequences of Motility
7.4.1 Planktonic Centrics
7.4.2 Benthic Pennates
7.4.3 Ecological Consequences of Vertical Migration
Acknowledgments
References
8 Motility in the Diatom Genus Eunotia Ehrenb.
8.1 Introduction
8.2 Accounts of Movement in Eunotia
8.3 Motility in the Context of Valve Structure
8.3.1 Motility and Morphological Characteristics in Girdle View
8.3.2 Motility and Morphological Characteristics in Valve View
8.3.3 Motility and the Rimoportula
8.4 Motility and Ecology of Eunotia
8.4.1 Substratum-Associated Environments
8.4.2 Planktonic Environments
8.5 Motility and Diatom Evolution
8.6 Conclusion and Future Directions
Acknowledgements
References
9 A Free Ride: Diatoms Attached on Motile Diatoms
9.1 Introduction
9.2 Adhesion and Distribution of Epidiatomic Diatoms on Their Host
9.3 The Specificity of Host-Epiphyte Interactions
9.4 Cost-Benefit Analysis of Host-Epiphyte Interactions
9.5 Conclusion
References
10 Towards a Digital Diatom: Image Processing and Deep Learning Analysis of Bacillaria paradoxa Dynamic Morphology
10.1 Introduction
10.1.1 Organism Description
10.1.2 Research Motivation
10.2 Methods
10.2.1 Video Extraction
10.2.2 Deep Learning
10.2.3 DeepLabv3 Analysis
10.2.4 Primary Dataset Analysis
10.2.5 Data Availability
10.3 Results
10.3.1 Watershed Segmentation and Canny Edge Detection
10.3.2 Deep Learning
10.4 Conclusion
Acknowledgments
References
11 Diatom Triboacoustics
11.1 State-of-the-Art
11.1.1 Diatoms and Their Movement
11.1.2 The Navier-Stokes Equation
11.1.3 Low Reynolds Number
11.1.4 Reynolds Number for Diatoms
11.1.5 Further Thoughts About Movement of Diatoms
11.1.6 Possible Reasons for Diatom Movement
11.1.7 Underwater Acoustics, Hydrophones
11.2 Methods
11.2.1 Estimate of the Momentum of a Moving Diatom
11.2.2 On the Speed of Expansion of the Mucopolysaccharide Filaments
11.2.3 Gathering Diatoms
11.2.4 Using a Hydrophone to Detect Possible Acoustic Signals from Diatoms
11.3 Results and Discussion
11.3.1 Spectrograms
11.3.2 Discussion
11.4 Conclusions and Outlook
Acknowledgements
References
12 Movements of Diatoms VIII: Synthesis and Hypothesis
12.1 Introduction
12.2 Review of the Conditions Necessary for Movements
12.3 Hypothesis
12.4 Analysis – Comparison with Observations
12.4.1 Translational Apical Movement
12.4.2 The Transapical Toppling Movement
12.4.3 Diverse Pivoting
12.5 Conclusion
Acknowledgments
References
13 Locomotion of Benthic Pennate Diatoms: Models and Thoughts
13.1 Diatom Structure
13.1.1 Ultrastructure of Frustules
13.1.2 Bending Ability of Diatoms
13.2 Models for Diatom Locomotion
13.2.1 Edgar Model for Diatom Locomotion
13.2.2 Van der Waals Force Model (VW Model) for Diatom Locomotion
13.3 Locomotion and Aggregation of Diatoms
13.3.1 Locomotion Trajectory and Parameters of Diatoms
13.4 Simulation on Locomotion, Aggregation and Mutual Perception of Diatoms
13.4.1 Simulation Area and Parameters
13.4.2 Diatom Life Cycle and Modeling Parameters
13.4.3 Simulation Results of Diatom Locomotion Trajectory with Mutual Perception
13.4.4 Simulation Results of Diatom Adhesion with Mutual Perception
13.4.5 Adhesion and Aggregation Mechanism of Diatoms
References
14 The Whimsical History of Proposed Motors for Diatom Motility
14.1 Introduction
14.2 Historical Survey of Models for the Diatom Motor
14.2.1 Diatoms Somersault via Protruding Muscles (1753)
14.2.2 Vibrating Feet or Protrusions Move Diatoms (1824)
14.2.3 Diatoms Crawl Like Snails (1838)
14.2.4 The Diatom Motor Is a Jet Engine (1849)
14.2.5 Rowing Diatoms (1855)
14.2.6 Diatoms Have Protoplasmic Tank Treads (1865)
14.2.7 Diatoms as the Flame of Life: Capillarity (1883)
14.2.8 Bellowing Diatoms (1887)
14.2.9 Jelly Powered Jet Skiing Diatoms (1896)
14.2.10 Bubble Powered Diatoms (1905)
14.2.11 Diatoms Win: “I Have No New Theory to Offer and See No Reason to Use Those Already Abandoned”12 (1940)
14.2.12 Is Diatom Motility a Special Case of Cytoplasmic Streaming? (1943)
14.2.13 Diatom Adhesion as a Sliding Toilet Plunger (1966)
14.2.14 Diatom as a Monorail that Lays Its Own Track (1967)
14.2.15 The Diatom as a “Compressed Air” Coanda Effect Gliding Vehicle (1967)
14.2.16 The Electrokinetic Diatom (1974)
14.2.17 The Diatom Clothes Line or Railroad Track (1980)
14.2.18 Diatom Ion Cyclotron Resonance (1987)
14.2.19 Diatoms Do Internal Treadmilling (1998)
14.2.20 Surface Treadmilling, Swimming and Snorkeling Diatoms (2007)
14.2.21 Acoustic Streaming: The Diatom as Vibrator or Jack Hammer (2010)
14.2.22 Propulsion of Diatoms Via Many Small Explosions (2020)
14.2.23 Diatoms Walk Like Geckos (2019)
14.3 Pulling What We Know and Don’t Know Together, about the Diatom Motor
14.4 Membrane Surfing: A New Working Hypothesis for the Diatom Motor (2020)
Acknowledgments
References
Appendix
References
Index
Also of Interest
Check out these published and forthcoming related titles from Scrivener Publishing

Citation preview

Diatom Gliding Motility

Scrivener Publishing 100 Cummings Center, Suite 541J Beverly, MA 01915-6106

Diatoms: Biology and Applications Series Editors: Richard Gordon ([email protected]) and Joseph Seckbach ([email protected]) Scope: The diatoms are a single-cell algal group, with each cell surrounded by a silica shell. The shells have beautiful attractive shapes with multiscalar structure at 8 orders of magnitude, and have several uses. 20% of the oxygen we breathe is produced by diatom photosynthesis, and they feed most of the aquatic food chain in freshwaters and the oceans. Diatoms serve as sources of biofuel and electrical solar energy production and are impacting on nanotechnology and photonics. They are important ecological and paleoclimate indicators. Some of them are extremophiles, living at high temperatures or in ice, at extremes of pH, at high or low light levels, and surviving desiccation. There are about 100,000 species and as many papers written about them since their discovery over three hundred years ago. The literature on diatoms is currently doubling every ten years, with 50,000 papers during the last decade (2006-2016). In this context, it is timely to review the progress to date, highlight cutting-edge discoveries, and discuss exciting future perspectives. To fulfill this objective, this new Diatom Series is being launched under the leadership of two experts in diatoms and related disciplines. The aim is to provide a comprehensive and reliable source of information on diatom biology and applications and enhance interdisciplinary collaborations required to advance knowledge and applications of diatoms.

Publishers at Scrivener Martin Scrivener ([email protected]) Phillip Carmical ([email protected])

Diatom Gliding Motility

Edited by

Stanley Cohn,

DePaul University, Chicago, IL, USA

Kalina Manoylov

Georgia College & State University, Milledgeville, GA, USA

and

Richard Gordon

Gulf Specimen Marine Laboratory & Aquarium, Panacea, FL, USA and Wayne State University, Detroit, MI, USA

This edition first published 2021 by John Wiley & Sons, Inc., 111 River Street, Hoboken, NJ 07030, USA and Scrivener Publishing LLC, 100 Cummings Center, Suite 541J, Beverly, MA 01915, USA © 2021 Scrivener Publishing LLC For more information about Scrivener publications please visit www.scrivenerpublishing.com. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, except as permitted by law. Advice on how to obtain permission to reuse material from this title is available at http://www.wiley.com/go/permissions. Wiley Global Headquarters 111 River Street, Hoboken, NJ 07030, USA For details of our global editorial offices, customer services, and more information about Wiley products visit us at www. wiley.com. Limit of Liability/Disclaimer of Warranty While the publisher and authors have used their best efforts in preparing this work, they make no rep­resentations or warranties with respect to the accuracy or completeness of the contents of this work and specifically disclaim all warranties, including without limitation any implied warranties of merchant-­ability or fitness for a particular purpose. No warranty may be created or extended by sales representa­tives, written sales materials, or promotional statements for this work. The fact that an organization, website, or product is referred to in this work as a citation and/or potential source of further informa­tion does not mean that the publisher and authors endorse the information or services the organiza­tion, website, or product may provide or recommendations it may make. This work is sold with the understanding that the publisher is not engaged in rendering professional services. The advice and strategies contained herein may not be suitable for your situation. You should consult with a specialist where appropriate. Neither the publisher nor authors shall be liable for any loss of profit or any other commercial damages, including but not limited to special, incidental, consequential, or other damages. Further, readers should be aware that websites listed in this work may have changed or disappeared between when this work was written and when it is read. Library of Congress Cataloging-in-Publication Data ISBN 9781119526353 Cover image: Thomas Harbich Cover design by Russell Richardson Set in size of 11pt and Minion Pro by Manila Typesetting Company, Makati, Philippines Printed in the USA 10 9 8 7 6 5 4 3 2 1

Dedication to Jeremy D. Pickett-Heaps In Memoriam 1940–2021 The editors of this volume would like to dedicate this collection to Dr. Jeremy D. Pickett-Heaps, with thanks and gratitude for his leadership and stalwart advocacy in advancing studies of diatoms and diatom motility. Sadly, Jeremy passed away just prior to this volume’s publication. One of the editors (SAC) had the pleasure and honor of studying and working with Jeremy and is proud to write this dedication. In addition, another editor (RG) visited Jeremy in Colorado where he was greatly influenced on his work in diatom morphogenesis. Jeremy was truly an international scholar. Born in Mumbai, India, he received his B.A. and Ph.D. in Cambridge, England, and did his postdoctoral work back in his home of Australia. He then worked for almost 20 years as a professor at the University of Colorado Boulder, after which he went back to Melbourne to the University of Melbourne until his retirement. His work on algae was prodigious, as witnessed by the large number of excellent publications listed at the end of this dedication. Jeremy was always a strong advocate for live observation of cell behaviors. While Jeremy always understood the value and use of theoretical and in vitro biochemical studies (in fact, early in my career I published a theoretical model of cell division along with him [1.180]), Jeremy would always tell everyone in the lab to let the in vivo living cells tell you what is really going on. While the in vitro studies and electron microscope structural studies could provide direction and constraints, Jeremy always relied on live cell observations to drive his understandings. His love of microscopy led Jeremy to not only record cells for research purposes, but to start a new company, Cytographics, in which he used 16 mm and video recordings to make educational materials displaying cellular processes (e.g., [1.128] [1.129] [1.161]). In Jeremy’s own words from his Cytographics site, “As this [electron microscope] work progressed, I became increasingly frustrated at trying to recreate dynamic cellular events solely from static images. A turning point in my career came when I first saw the extraordinary sight of a live diatom undergoing mitosis at high magnification. After borrowing a 16 mm time-lapse camera, I was soon filming algae doing all the things I had studied with the electron microscope. Since then, I have built up a laboratory devoted to the high-resolution video imaging and recording of all sorts of cells and microscopic organisms going about their complex and extraordinary lives. It’s the best peep show around!” Jeremy was a true trailblazer in the study of algae. Discovering the passage of cell wall material from the Golgi [1.51], the role of microtubules and microtubule organizing centers v

vi   (e.g., [1.90] [1.94] [1.119] [1.123]), and evolutionary relationships among algae (e.g., [1.93] [1.106] [1.111] [1.113]) Jeremy always tried to look at algae in new ways. His work, and that of his students and colleagues, was instrumental in using the highly organized mitotic spindle in diatoms to understand microtubule organization during cell division [1.61] [1.79] [1.218] [1.224] [1.228]. But among the algae, diatoms have always been special to Jeremy. Jeremy was fascinated by the early microscopy work by the botanist Robert Lauterborn and his exquisitely detailed drawings of algal phenomena. In 1984 he published a work with some co-authors on a translation of Lauterborn’s 1896 treatise, along with some modern microscopic observations of the same cells [1.167]. The publication displayed how modern optical and electron microscopy simply confirmed the excellent work of Lauterborn in understanding the dynamics of diatom mitosis. I had the great privilege of seeing a copy of the 1896 document when Jeremy had it briefly on loan to take copies of some of the original images for his publication, and the drawings truly were beautiful and amazingly detailed. I was lucky enough to be working in Jeremy’s lab in Boulder during an exciting time in diatom motility. Dr. Lesley Edgar was working in the lab, investigating the underlying ultrastructure of motile diatoms, leading her to develop a model of diatom gliding [1.20] [1.21], and where I had the honor of publishing with her on some aspects of diatom morphogenesis [1.11]. She searched through Boulder for some ponds containing the best diatoms for investigation, one of which I still use as my major source of cells for research. As a graduate student I began working with intracellular motility and the role of microtubules and microtubule organizing centers in forming the diatom valves used in motility. I, too, became enthralled with watching diatoms as they glided, and divided, and formed new cell walls. Using both video and film, under Jeremy’s tutelage the lab analyzed the motile behavior and intracellular movements of the cells, and using scanning electron microscopy I studied the forming raphe and valve structures of diatoms during development and reproduction. Jeremy always wanted to know what cell phenomena members of the lab were watching and seeing, helping us to contemplate both their mystery and their beauty as well as their biological importance. After a short time in Jeremy’s lab I was hooked on diatoms and their movement and have never looked back. During my time in his lab, Jeremy always filled the lab with joy and enthusiasm for science and exploration. Any time someone would come up with an idea or suggested an experiment, Jeremy would always encourage us to try it out and see what happens. He was a firm believer in the idea that science is about using new techniques and new approaches to poke at the cells and see what they were trying to tell you. And at every point in the work we were doing, Jeremy would strive for excellence in the microscopy coming out of the lab. Whether it was light microscopy using the newest optical techniques, electron microscopy using the best approaches for serial sectioning, or scanning electron microscopy finding the best angles for imaging, he wanted the cleanest, clearest images possible. He had an innate sense of the images that would not only be the best to show the processes or structures we were trying to explain, but would also be the most beautiful. He was worried far less about dogma or current trends, and far more about trying to find the truth. I also had the pleasure of working in his lab as a visiting colleague after he had gone back to the University of Melbourne. His enthusiasm was undiminished, and his love for microscopy and for developing educational materials had, if anything, only expanded. His encouragement to test and try new ideas led to investigations into some of the light-based

  vii responses of diatoms that I continue to this day. As always, he encouraged everyone in the lab to use the latest techniques to tease the truth out of the cells. His care for everyone who came into the lab, whether student, technician, visiting colleague, or postdoc, was always an inspiration. He constantly showed a love of life, a love of science, and a love for his lab personnel, all in equal measure. He helped us understand that science is a way to help organize and understand the world and the fabric of nature, and that the diatoms were a beautiful and glimmering thread in that fabric. This dedication would also be remiss if it did not mention the incredible diatom researchers from the Pickett-Heaps Lab who were remarkable colleagues and mentors, but have also unfortunately passed away far too soon. I owe my deepest thanks to the late Drs. Lesley Edgar, Cindy Troxell, and Timothy Spurck. Their friendship, knowledge, humor, and dedication helped foster and guide my research into diatoms. It is to their love of science and diatoms, and to Jeremy’s, that this book is dedicated.

Bibliography of Jeremy D. Pickett-Heaps [1.1] Ackland, J.C., West, J.A., Pickett-Heaps, J., Actin and myosin regulate pseudopodia of Porphyra pulchella (Rhodophyta) archeospores. J. Phycol., 43, 1, 129–138, 2007. [1.2] Beech, P.L., Wetherbee, R., Pickett-Heaps, J.D., Transformation of the flagella and associated flagellar components during cell division in the coccolithophorid Pleurochrysis carterae. Protoplasma, 145, 1, 37–46, 1988. [1.3] Beech, P.L., Wetherbee, R., Pickett-Heaps, J.D., Secretion and deployment of bristles in Mallomonas splendens (Synurophyceae). J. Phycol., 26, 1, 112–122, 1990. [1.4] Boyle, J.A., Pickett-Heaps, J.D., Czarnecki, D.B., Valve morphogenesis in the pennate diatom Achnanthes coarctata. J. Phycol., 20, 563–573, 1984. [1.5] Callow, M.E., Callow, J.A., Pickett-Heaps, J.D., Wetherbee, R., Primary adhesion of Enteromorpha (Chlorophyta, Ulvales) propagules: Quantitative settlement studies and video microscopy. J. Phycol., 33, 6, 938–947, 1997. [1.6] Callow, M.E., Callow, J.A., Pickett-Heaps, J.D., Wetherbee, R., Primary adhesion of enteromorpha propagules. Phycologia, 36, 4, 15, 1997. [1.7] Cohn, S.A., Nash, J., Pickett-Heaps, J.D., The effect of drugs on diatom valve morphogenesis. Protoplasma, 149, 130–143, 1989. [1.8] Cohn, S.A. and Pickett-Heaps, J.D., The effects of colchicine and dinitrophenol on the in vivo rates of anaphase A and B in the diatom Surirella. Eur. J. Cell Biol., 46, 3, 523–530, 1988. [1.9] Cohn, S.A., Schoeller, A., Spurck, T.P., Pickett-Heaps, J.D., Edgar, L.E., Sexual reproduction in the pennate diatom Navicula cuspidata. II. Auxo spore development and initial cell formation. J. Phycol., 20, Suppl, 27, 1984. [1.10] Cohn, S.A., Spurck, T.P., Pickett-Heaps, J.D., High energy irradiation at the leading tip of moving diatoms causes a rapid change of cell direction. Diatom Res., 14, 2, 193–206, 1999. [1.11] Cohn, S.A., Spurck, T.P., Pickett-Heaps, J.D., Edgar, L.A., Perizonium and initial valve formation in the diatom Navicula cuspidata (Bacillariophyceae). J. Phycol., 25, 15–26, 1989. [1.12] Cohn, S.A., Weitzell, R.E., Spurck, T.P., Pickett-Heaps, J.D., Characterization of motility and adhesion in pennate diatoms. Mol. Biol. Cell, 6, Suppl. S, 261a, 1995. [1.13] Coss, R.A., Bloodgood, R.A., Brower, D.L., Pickett-Heaps, J.D., MacIntosh, J.R., Studies on the mechanism of action of isopropyl N phenyl carbamate. Exp. Cell Res., 92, 2, 394–398, 1975.

viii   [1.14] Coss, R.A. and Pickett-Heaps, J.D., Gametogenesis in the green alga Oedogonium cardiacum. I. The cell divisions leading to formation of spermatids and oogonia. Protoplasma, 78, 1, 21–39, 1973. [1.15] Coss, R.A. and Pickett-Heaps, J.D., The effects of isopropyl N phenyl carbamate on the green alga Oedogonium cardiacum. I. Cell division. J. Cell Biol., 63, 1, 84–98, 1974. [1.16] Crawford, R., Round, F., Pickett-Heaps, J., Johnson, A., Obituary: Lesley Ann Edgar (1955 2006). Diatom Res., 22, 1, 237–240, 2007. [1.17] Crawford, S., Chiovitti, T., Pickett-Heaps, J., Wetherbee, R., Micromorphogenesis during diatom wall formation produces siliceous nanostructures with different properties. J. Phycol., 45, 6, 1353–1362, 2009. [1.18] Crawford, S.A., Chiovitti, A., Pickett-Heaps, J., Wetherbee, R., Micromorphpogeneis during diatom wall formation produces siliceous nanostructures with different properties. J. Phycol., 45, 6, 1353–1362, 2009. [1.19] Edgar, L.A. and Pickett-Heaps, J.D., Ultrastructural localisation of polysaccharides in the motile diatom Navicula cuspidata. Protoplasma, 113, 10–22, 1982. [1.20] Edgar, L.A. and Pickett-Heaps, J.D., The mechanism of diatom locomotion. I. An ultrastructural study of the motility apparatus. Proc. R. Soc. B: Biol. Sci., 218, 331–343, 1983. [1.21] Edgar, L.A. and Pickett-Heaps, J.D., Diatom locomotion. Prog. Phycol. Res., 3, 47–88, 1984. [1.22] Edgar, L.A. and Pickett-Heaps, J.D., Valve morphogenesis in the pennate diatom Navicula cuspidata. J. Phycol., 20, 47–61, 1984. [1.23] Forer, A. and Pickett-Heaps, J., Fibrin clots keep non adhering living cells in place on glass for perfusion or fixation. Cell Biol. Int., 29, 9, 721–730, 2005. [1.24] Forer, A. and Pickett-Heaps, J., Precocious (pre anaphase) cleavage furrows in Mesostoma spermatocytes. Eur. J. Cell Biol., 89, 8, 607–618, 2010. [1.25] Forer, A. and Pickett-Heaps, J.D., Checkpoint control in crane fly spermatocytes: Unattached chromosomes induced by cytochalasin D or latrunculin treatment do not prevent or delay the start of anaphase. Protoplasma, 203, 100–111, 1998. [1.26] Forer, A. and Pickett-Heaps, J.D., Cytochalasin D and latrunculin affect chromosome behaviour during meiosis in crane fly spermatocytes. Chromosome Res., 6, 7, 533–549, 1998. [1.27] Forer, A., Pickett-Heaps, J.D., Spurck, T., What generates flux of tubulin in kinetochore microtubules? Protoplasma, 232, 3-4, 137–141, 2008. [1.28] Forer, A., Spurck, T., Pickett-Heaps, J.D., Ultraviolet microbeam irradiations of spindle fibres in crane fly spermatocytes and newt epithelial cells: resolution of previously conflicting observations. Protoplasma, 197, 230–240, 1997. [1.29] Forer, A., Spurck, T., Pickett-Heaps, J.D., Actin and myosin inhibitors block elongation of kinetochore fibre stubs in metaphase crane fly spermatocytes. Protoplasma, 232, 1-2, 79–85, 2007. [1.30] Forer, A., Spurck, T., Pickett-Heaps, J.D., Wilson, P.J., Structure of kinetochore fibres in crane fly spermatocytes after irradiation with an ultraviolet microbeam: Neither microtubules nor actin filaments remain in the irradiated region. Cell Motil. Cytoskeleton, 56, 3, 173–192, 2003. [1.31] Fowke, L.C. and Pickett-Heaps, J.D., Electron microscope study of vegetative cell division in two species of Marchantia. Can. J. Bot.-Revue Can. Botanique, 56, 5, 467–475, 1978. [1.32] Hepler, P.K., Pickett-Heaps, J.D., Gunning, B.E.S., Some retrospectives on early studies of plant microtubules. Plant J., 75, 2, 189–201, 2013. [1.33] Hinz, I., Spurck, T.P., Pickett-Heaps, J.D., Metabolic inhibitors and mitosis. III. Effects of dinitrophenol on spindle disassembly in Pinnularia. Protoplasma, 132, 1-2, 85–89, 1986. [1.34] Hoffman, L.R., Vesk, M., Pickett-Heaps, J.D., The cytology and ultrastructure of zoospores of Hydrurus foetidus (Chrysophyceae). Nord. J. Bot., 6, 1, 105–122, 1986.

  ix [1.35] Jarman, M. and Pickett-Heaps, J., Cell division and nuclear movement in the saccoderm desmid Netrium interruptus. Protoplasma, 157, 1-3, 136–143, 1990. [1.36] Leslie, R.J. and Pickett-Heaps, J.D., Ultraviolet microbeam studies of diatom mitosis. J. Cell Biol., 91, 2, A311, 1981. [1.37] Leslie, R.J. and Pickett-Heaps, J.D., Ultraviolet microbeam irradiations of mitotic diatoms: Investigation of spindle elongation. J. Cell Biol., 96, 2, 548–561, 1983. [1.38] Leslie, R.J. and Pickett-Heaps, J.D., Spindle microtubule dynamics following ultraviolet microbeam irradiations of mitotic diatoms. Cell, 36, 3, 717–727, 1984. [1.39] Marchant, H.J. and Pickett-Heaps, J.D., Ultrastructure and differentiation of Hydrodictyon reticulatum. I. Mitosis in the coenobium. Aust. J. Biol. Sci., 23, 6, 1173–1186, 1970. [1.40] Marchant, H.J. and Pickett-Heaps, J.D., Ultrastructure and differentiation of Hydrodictyon reticulatum. II. Formation of zooids within the coenobium. Aust. J. Biol. Sci., 24, 3, 471–486, 1971. [1.41] Marchant, H.J. and Pickett-Heaps, J.D., Ultrastructure and differentiation of Hydrodictyon reticulatum. III. Formation of vegetative daughter net. Aust. J. Biol. Sci., 25, 2, 265–278, 1972. [1.42] Marchant, H.J. and Pickett-Heaps, J.D., Ultrastructure and differentiation of Hydrodictyon reticulatum. IV. Conjugation of gametes and development of zygospores and azygospores. Aust. J. Biol. Sci., 25, 2, 279–292, 1972. [1.43] Marchant, H.J. and Pickett-Heaps, J.D., Ultrastructure and differentiation of Hydrodictyon reticulatum. V. Development of polyhedra. Aust. J. Biol. Sci., 25, 6, 1187–1197, 1972. [1.44] Marchant, H.J. and Pickett-Heaps, J.D., Ultrastructure and differentiation of Hydrodictyon reticulatum. VI. Formation of germ net. Aust. J. Biol. Sci., 25, 6, 1199–1213, 1972. [1.45] Marchant, H.J., Pickett-Heaps, J.D., Jacobs, K., An ultrastructural study of zoosporogenesis and the mature zoospore of Klebsormidium flaccidum. Cytobios, 8, 29, 95–107, 1973. [1.46] McDonald, K., Pickett-Heaps, J.D., McIntosh, J.R., Tippit, D.H., On the mechanism of anaphase spindle elongation in Diatoma vulgare. J. Cell Biol., 74, 2, 377–388, 1977. [1.47] McDonald, K.L. and Pickett-Heaps, J.D., Mitosis and cytokinesis in Cladophora glomerata. J. Phycol., 11, 8–9, 1975. [1.48] McDonald, K.L. and Pickett-Heaps, J.D., Ultrastructure and differentiation in Cladophora glomerata. I. Cell division. Am. J. Bot., 63, 5, 592–601, 1976. [1.49] McIntosh, K., Pickett-Heaps, J.D., Gunning, B.E.S., Cytokinesis in Spirogyra: Integration of cleavage and cell plate formation. Int. J. Plant Sci., 156, 1, 1–8, 1995. [1.50] McIntosh, K.J., Pickett-Heaps, J.D., Gunning, B.E.S., Cytokinesis in Spirogyra: Integration of cleavage and cell plate formation. Phycologia, 36, 4, Suppl. 5, 71, 1997. [1.51] Northcote, D.H. and Pickett-Heaps, J.D., A function of the Golgi apparatus in polysaccharide synthesis and transport in the root cap cells of wheat. Biochem. J., 98, 1, 159–167, 1966. [1.52] Ott, D.W. and Pickett-Heaps, J.D., Mitosis in Sphaeroplea. J. Phycol., 11, 8, 1975. [1.53] Pickett-Heaps, J., Reproduction by zoospores in Oedogunium. I. Zoosporogenesis. Protoplasma, 72, 2-3, 275–314, 1971. [1.54] Pickett-Heaps, J., Cell division in Cosmarium botrytis. J. Phycol., 8, 4, 343–360, 1972. [1.55] Pickett-Heaps, J., Cell division in Cyanophora paradoxa. New Phytol., 71, 4, 561–567, 1972. [1.56] Pickett-Heaps, J., Cell division in Klebsormidium subtilissimum (formerly Ulothrix subtilissima), and its possible phylogenetic significance. Cytobios, 6, 23, 167–183, 1972. [1.57] Pickett-Heaps, J., Cell division and wall structure in Microspora. New Phytol., 72, 2, 347–355, 1973. [1.58] Pickett-Heaps, J., The evolution of mitosis and eukaryotic condition. BioSystems, 6, 1, 37–48, 1974. [1.59] Pickett-Heaps, J., Cell division in eukaryotic algae. Bioscience, 26, 7, 445–450, 1976.

x   [1.60] Pickett-Heaps, J., Cell division and evolution of branching in Oedocladium (Chlorophyceae). Cytobiologie, 14, 3, 319–337, 1977. [1.61] Pickett-Heaps, J., The diatom spindle: A useful model for studying mitosis. Proceedings of the International Botanical Congress, vol. 13, p. 29, 1981. [1.62] Pickett-Heaps, J., Mitotic mechanisms: An alternative view. Trends Biochem. Sci., 11, 12, 504–507, 1986. [1.63] Pickett-Heaps, J., Morphogenesis of the labiate process in the araphid pennate diatom Diatoma vulgare. J. Phycol., 25, 1, 79–85, 1989. [1.64] Pickett-Heaps, J., Hidden Worlds: Pond Life. https://www.youtube.com/watch?v=xXaTyYpGkTU, 1990. [1.65] Pickett-Heaps, J., Cell division in diatoms, in: International Review of Cytology-A Survey of Cell Biology, vol. 128, pp. 63–108, 1991. [1.66] Pickett-Heaps, J., The cytoskeleton in diatom valve morphogenesis. Phycologia, 40, 4 Supplement, 18, 2001. [1.67] Pickett-Heaps, J., Bush Stone-Curlew. https://www.youtube.com/watch?v=SY_fMZDplrc, 2013. [1.68] Pickett-Heaps, J., Dictyostelium - a cellular slime mold. https://www.youtube.com/ watch?v=5h8WOWEqP6o, 2013. [1.69] Pickett-Heaps, J., Grey Fantail. https://www.youtube.com/watch?v=Q6nQYbVoIpI, 2013. [1.70] Pickett-Heaps, J., Rainbow Lorikeets. https://www.youtube.com/watch?v=ZBnI_ObpJyY, 2013. [1.71] Pickett-Heaps, J., Rufous Fantail. https://www.youtube.com/watch?v=gG7qsrfQ9vk, 2013. [1.72] Pickett-Heaps, J., Spurwing Plovers at their Nest. https://www.youtube.com/watch?v=-uUjXf0M2xE, 2013. [1.73] Pickett-Heaps, J., Spurwinged Plover hatching chicks. https://www.youtube.com/ watch?v=DSx-PbWQpas, 2013. [1.73] Pickett-Heaps, J., Superb Lyrebird. https://www.youtube.com/watch?v=Iju8yIEycqU, 2013. [1.75] Pickett-Heaps, J., Euglenoid Flagellates. https://www.youtube.com/watch?v=WDWbId0MAmU, 2014. [1.76] Pickett-Heaps, J. and Carpenter, J., Spine formation in diatoms. J. Phycol., 31, 3 SUPPL., 21, 1995. [1.77] Pickett-Heaps, J., Carpenter, J., Koutoulis, A., Spine morphogenesis in the diatom Chaetoceros peruvianum. Mol. Biol. Cell, 4, SUPPL., 403A, 1993. [1.78] Pickett-Heaps, J. and Forer, A., Mitosis: Spindle evolution and the matrix model. Protoplasma, 235, 1-4, 91–99, 2009. [1.79] Pickett-Heaps, J. and Kowalski, S.E., Valve morphogenesis and the microtubule center of the diatom Hantzschia amphioxys. Eur. J. Cell Biol., 25, 1, 150–170, 1981. [1.80] Pickett-Heaps, J., McNiven, M., Porter, K., Translocation of pigment in erythrophores: Cine analysis of responses to various drugs. J. Cell Biol., 99, 4, A122, 1984. [1.81] Pickett-Heaps, J., Spurck, T., Tippit, D., Chromosome motion and the spindle matrix. J. Cell Biol., 99, 1 Pt 2, 137s–143s, 1984. [1.82] Pickett-Heaps, J. and Tippit, D., Time lapse cine analysis of mitosis in two diatoms. J. Cell Biol., 79, 2, A286, 1978. [1.83] Pickett-Heaps, J.D., Effects of colchicine on ultrastructure of dividing plant cells xylem wall differentiation and distribution of cytoplasmic microtubules. Dev. Biol., 15, 3, 206–236, 1967. [1.84] Pickett-Heaps, J.D., Further observations on the Golgi apparatus and its functions in cells of the wheat seedling. J. Ultrastruct. Res., 18, 3, 287–303, 1967.

  xi [1.85] Pickett-Heaps, J.D., Preliminary attempts at ultrastructural polysaccharide localization in root tip cells. J. Histochem. Cytochem., 15, 8, 442–455, 1967. [1.86] Pickett-Heaps, J.D., Ultrastructure and differentiation in Chara sp. I, 1967. [1.87] Pickett-Heaps, J.D., Ultrastructure and differentiation in Chara sp. II, 1967. [1.88] Pickett-Heaps, J.D., The use of adioautography for investigating wall secretion in plant cells. Protoplasma, 64, 1, 49–66, 1967. [1.89] Pickett-Heaps, J.D., Further ultrastructural observations on polysaccharide localization in plant cells. J. Cell Sci., 3, 1, 55–64, 1968. [1.90] Pickett-Heaps, J.D., Microtubule like structures in growing plastids or chloroplasts of two algae. Planta, 81, 2, 193–200, 1968. [1.91] Pickett-Heaps, J.D., Ultrastructure and differentiation in Chara (fibrosa). IV. Spermatogenesis. Aust. J. Biol. Sci., 21, 4, 655–690, 1968. [1.92] Pickett-Heaps, J.D., Ultrastructure and differentiation in Chara sp. III, 1968. [1.93] Pickett-Heaps, J.D., The evolution of the mitotic apparatus: An attempt at comparative ultrastructural cytology in dividing plant cells. Cytobios, 1, 3, 257–280, 1969. [1.94] Pickett-Heaps, J.D., Preprophase microtubule bands in some abnormal mitotic cells of wheat. J. Cell Sci., 4, 2, 397–420, 1969. [1.95] Pickett-Heaps, J.D., Preprophase microtubules and stomatal differentiation in Commelina cyanea. Aust. J. Biol. Sci., 22, 2, 375–392, 1969. [1.96] Pickett-Heaps, J.D., Preprophase microtubules and stomatal differentiation; some effects of centrifugation on symmetrical and asymmetrical cell division. J. Ultrastruct. Res., 27, 1, 24–44, 1969. [1.97] Pickett-Heaps, J.D., Mitosis and autospore formation in green alga Kirchneriella lunaris. Protoplasma, 70, 3-4, 325–347, 1970. [1.98] Pickett-Heaps, J.D., Some ultrastructural features of Volvox, with particular reference to phenomenon of inversion. Planta, 90, 2, 174–190, 1970. [1.99] Pickett-Heaps, J.D., Autonomy of centrioles: fact or fallacy? Cytobios, 3, 12, 205–214, 1971. [1.100] Pickett-Heaps, J.D., Bristly cristae in algal mitochondria. Planta, 100, 4, 357–359, 1971. [1.101] Pickett-Heaps, J.D., Cell division in Tetraedron. Ann. Bot., 36, 693–701, 1972. [1.102] Pickett-Heaps, J.D., Possible virus infection in green alga Oedogonium. J. Phycol., 8, 1, 44–47, 1972. [1.103] Pickett-Heaps, J.D., Reproduction by zoospores in Oedogonium. II. Emergence of zoospore and motile phase. Protoplasma, 74, 1-2, 149–167, 1972. [1.104] Pickett-Heaps, J.D., Reproduction by zoospores in Oedogonium. III. Differentiation of germling. Protoplasma, 74, 1-2, 169–193, 1972. [1.105] Pickett-Heaps, J.D., Reproduction by zoospores in Oedogonium. IV. Cell division in germling and evidence concerning possible evolution of wall rings. Protoplasma, 74, 1-2, 195– 212, 1972. [1.106] Pickett-Heaps, J.D., Variation in mitosis and cytokinesis in plant cells: Its significance in phylogeny and evolution of ultrastructural systems. Cytobios, 5, 17, 59–77, 1972. [1.107] Pickett-Heaps, J.D., Cell division in Bulbochaete .1. Divisions utilizing wall ring. J. Phycol., 9, 4, 408–420, 1973. [1.108] Pickett-Heaps, J.D., Cell division in Tetraspora. Ann. Bot., 37, 153, 1017–1026, 1973. [1.109] Pickett-Heaps, J.D., Stereo scanning electron microscopy of desmids. J. Microsc., 99, SEP, 109–116, 1973. [1.110] Pickett-Heaps, J.D., Scanning electron microscopy of some cultured desmids. Trans. Am. Microsc. Soc., 93, 1, 1–23, 1974. [1.111] Pickett-Heaps, J.D., Aspects of spindle evolution. Ann. N. Y. Acad. Sci., 253, 352–361, 1975. [1.112] Pickett-Heaps, J.D., Cell division and evolution in Bulbochaete .3. Sexual reproduction and evolution of branched habit. Cytobiologie, 12, 1, 28–51, 1975.

xii   [1.113] Pickett-Heaps, J.D., Green Algae. Structure, Reproduction and Evolution in Selected Genera, Oxford University Press, Oxford, UK, 1975. [1.114] Pickett-Heaps, J.D., Electron microscopy and the phylogeny of green algae and land plants. Am. Zool., 19, 2, 545, 1979. [1.115] Pickett-Heaps, J.D., Cell division in Surirella, a tribute to Robert Lauterborn. J. Cell Biol., 95, 2, A307, 1982. [1.116] Pickett-Heaps, J.D., New Light on the Green Algae, Carolina Biology Reader 115, Carolina Biological Supply Company, Burlington, North Carolina, 1982. [1.117]  Pickett-Heaps, J.D., Cell division and morphogenetic movements in the diatom Cymatopleura. J. Cell Biol., 97, 5, A248, 1983. [1.118] Pickett-Heaps, J.D., Morphogenesis in desmids: Our present state of ignorance, in: Spatial Organization of Eukaryotic Cells: Proceedings of a Symposium Held in Honor of Keith R. Porter, Boulder, Colorado, April 30-May 2, 1982, pp. 241–258, 1983. [1.119] Pickett-Heaps, J.D., Valve morphogenesis and the microtubule center in three species of the diatom Nitzschia. J. Phycol., 19, 3, 269–281, 1983. [1.120] Pickett-Heaps, J.D., Diatom mitosis: implications of a model system, in: Cell Walls and Surfaces, Reproduction, Photosynthesis, W. Wiessner, R.C. Starr, D. Robinson (Eds.), pp. 28–33, Springer Verlag, Berlin, 1987. [1.121] Pickett-Heaps, J.D., Morphogenesis of the labiate process in the centric diatom Ditylum brightwellii. Protoplasma, 143, 139–149, 1989. [1.122] Pickett-Heaps, J.D., Microtubule cytoskeletons in cellular reorganization of the diatom Cymatopleura. J. Phycol., 26, 2 SUPPL, 12, 1990. [1.123] Pickett-Heaps, J.D., Post mitotic cellular reorganization in the diatom Cymatopleura solea: The role of microtubules and the microtubule center. Cell Motil. Cytoskeleton, 18, 4, 279– 292, 1991. [1.124] Pickett-Heaps, J.D., The kinetochore fiber in Oedogonium: An extended component visible when microtubules are removed. Mol. Biol. Cell, 3, A345, 1992. [1.125] Pickett-Heaps, J.D., Cell division and morphogenesis of the centric diatom Chaetoceros decipiens (Bacillariophyceae) I. Living cells. J. Phycol., 34, 6, 989–994, 1998. [1.126] Pickett-Heaps, J.D., Cell division and morphogenesis of the centric diatom Chaetoceros decipiens (Bacillariophyceae) II. Electron microscopy and a new paradigm for tip growth. J. Phycol., 34, 6, 995–1004, 1998. [1.127] Pickett-Heaps, J.D., Rapid, highly efficient method for collecting, fixing, and embedding planktonic and other small cells for electron microscopy. J. Phycol., 34, 6, 1088–1089, 1998. [1.128] Pickett-Heaps, J.D., Remarkable Plants: The Oedogoniale [DVD], Cytographics, http://cytographics.com, 2004. [1.129] Pickett-Heaps, J.D., The Kingdom Protista: The Dazzling World of Living Cells [DVD], Cytographics, http://cytographics.com, 2006. [1.130]  Pickett-Heaps, J.D., Foreword: The enigma of morphogenesis - a personal view, in: Handbook of Biomineralization, vol. 1, E. Bäuerlein (Ed.), pp. vii–ix, Wiley-VCH, 2009. [1.131] Pickett-Heaps, J.D., Volvox carterii, 2014, https://www.youtube.com/watch?v=n7 RggKhWD8g. [1.132] Pickett-Heaps, J.D., 2020. Jeremy D. Pickett Heaps. https://www.researchgate.net/profile/ Jeremy_Pickett-Heaps. [1.133]  Pickett-Heaps, J.D., 2020. Prof Jeremy Pickett-Heaps Honorary. https://findanexpert. unimelb.edu.au/profile/13098-jeremy-pickett-heaps. [1.134] Pickett-Heaps, J.D. and Bajer, A.S., Mitosis: An argument for multiple mechanisms achieving chromosomal movement. Cytobios, 19, 75-76, 171–180, 1977.

  xiii [1.135] Pickett-Heaps, J.D. and Carpenter, J., An extended corona attached to metaphase kinetochores of the green alga Oedogonium. Eur. J. Cell Biol., 60, 2, 300–307, 1993. [1.136] Pickett-Heaps, J.D., Carpenter, J., Koutoulis, A., Valve and seta (spine) morphogenesis in the centric diatom Chaetoceros peruvianus Brightwell. Protoplasma, 181, 1-4, 269–282, 1994. [1.137] Pickett-Heaps, J.D., Cohn, S., Schmid, A.M., Tippit, D.H., Valve morphogenesis in Surirella (Bacillariophyceae). J. Phycol., 24, 35–49, 1988. [1.138] Pickett-Heaps, J.D. and Forer, A., Pac Man does not resolve the enduring problem of anaphase chromosome movement. Protoplasma, 215, 1-4, 16–20, 2001. [1.139] Pickett-Heaps, J.D., Forer, A., Spurck, T., Rethinking anaphase: Where “Pac Man’’ fails and why a role for the spindle matrix is likely. Protoplasma, 192, 1-2, 1–10, 1996. [1.140] Pickett-Heaps, J.D., Forer, A., Spurck, T., Traction fibre: Toward a “tensegral’’ model of the spindle. Cell Motil. Cytoskeleton, 37, 1, 1–6, 1997. [1.141] Pickett-Heaps, J.D. and Fowke, L.C., Cell division in Oedogonium. I. Mitosis, cytokinesis, and cell elongation. Aust. J. Biol. Sci., 22, 4, 857–894, 1969. [1.142] Pickett-Heaps, J.D. and Fowke, L.C., Cell division in Oedogonium. II. Nuclear division in O. cardiacum. Aust. J. Biol. Sci., 23, 1, 71–92, 1970. [1.143] Pickett-Heaps, J.D. and Fowke, L.C., Mitosis, cytokinesis, and cell elongation in desmid, Closterium littorale. J. Phycol., 6, 2, 189–215, 1970. [1.144] Pickett-Heaps, J.D. and Fowke, L.C., Conjugation in the desmid Closterium littorale. J. Phycol., 7, 1, 37–50, 1971. [1.145] Pickett-Heaps, J.D., Gunning, B.E.S., Brown, R.C., Lemmon, B.E., Cleary, A.L., The cytoplast concept in dividing plant cells: cytoplasmic domains and the evolution of spatially organized cell division. Am. J. Bot., 86, 2, 153–172, 1999. [1.146] Pickett-Heaps, J.D., Hill, D.R.A., Blaze, K., Active gliding motility in an araphid marine diatom, Ardissonea (formerly Synedra) crystallina. J. Phycol., 27, 718–725, 1991. [1.147] Pickett-Heaps, J.D., Hill, D.R.A., Wetherbee, R., Cellular movement in the centric diatom Odontella sinensis. J. Phycol., 22, 334–339, 1986. [1.148] Pickett-Heaps, J.D. and Klein, A.G., Tip growth in plant cells may be amoeboid and not generated by turgor pressure. Proc. R. Soc. B-Biol. Sci., 265, 1404, 1453–1459, 1998. [1.149] Pickett-Heaps, J.D. and Marchant, H.J., Phylogeny of green algae: New proposal. Cytobios, 6, 24, 255–264, 1972. [1.150] Pickett-Heaps, J.D. and Martin, A., An early, neglected description of the fibers (micro­ tubules) in the eukaryotic flagellum. Protoplasma, 176, 1-2, 14–16, 1993. [1.151] Pickett-Heaps, J.D. and McDonald, K.L., Cylindrocapsa: Cell division and phylogenetic affinities. New Phytol., 74, 2, 235–241, 1975. [1.152] Pickett-Heaps, J.D., McDonald, K.L., Tippit, D.H., Cell division in pennate diatom Diatoma vulgare. Protoplasma, 86, 1-3, 205–242, 1975. [1.153] Pickett-Heaps, J.D., McDonald, K.L., Tippit, D.H., Cell division in the pennate diatom Diatoma vulgare. Protoplasma, 86, 1-3, 205–242, 1975. [1.154] Pickett-Heaps, J.D., McDonald, K.L., Tippit, D.H., Spindle structure and function in diatoms. J. Cell Biol., 67, 2, A336, 1975. [1.155] Pickett-Heaps, J.D. and Northcote, D.H., Cell division in formation of stomatal complex of young leaves of wheat. J. Cell Sci., 1, 1, 121–128, 1966. [1.156] Pickett-Heaps, J.D. and Northcote, D.H., Cell division in the formation of the stomatal complex of the young leaves of wheat. J. Cell Sci., 1, 1, 121–128, 1966. [1.157] Pickett-Heaps, J.D. and Northcote, D.H., Organization of microtubules and endoplasmic reticulum during mitosis and cytokinesis in wheat meristems. J. Cell Sci., 1, 1, 109–120, 1966.

xiv   [1.158] Pickett-Heaps, J.D. and Northcote, D.H., Relationship of cellular organelles to formation and development of plant cell wall. J. Exp. Bot., 17, 50, 19–?, 1966. [1.159]  Pickett-Heaps, J.D. and Ott, D.W., Ultrastructural morphology and cell division in Pedinomonas. Cytobios, 11, 41, 41–58, 1974. [1.160] Pickett-Heaps, J.D. and Pickett-Heaps, J., From Egg to Tadpole: Early Morphogenesis in Xenopus [VHS and DVD], Sinauer, 1999. [1.161] Pickett-Heaps, J.D. and Pickett-Heaps, J., Diatoms: Life in Glass Houses [DVD], Cytographics, 2003. [1.162] Pickett-Heaps, J.D. and Pickett-Heaps, J., Teacher’s Guide, Diatoms: Life in Glass Houses [http://www.cytographics.com/resource/catalog/tapes/pg dia.htm + video tape], Cytographics, Melbourne, 2003. [1.163] Pickett-Heaps, J.D. and Pickett-Heaps, J.F., 3’ 21” Bacillaria paradoxa, a colonial marine diatom; 2X: Living Cells: Structure and Diversity. Instructor’s Guide. http://www.cytographics. com/resource/catalog/tapes/in-lc.htm, 1996. [1.164] Pickett-Heaps, J.D. and Pickett-Heaps, J.F., Colonial diatom Bacillaria paradoxa (5X), Microscopic Life Instructor’s Guide, Obtaining Living Micro Organisms for Light Microscopy. http://www.cytographics.com/resource/catalog/tapes/in-ml.htm, 2002. [1.165] Pickett-Heaps, J.D., Schmid, A.M.M., Edgar, L.A., The cell biology of diatom valve formation. Prog. Phycol. Res., 7, 1–168, 1990. [1.166] Pickett-Heaps, J.D., Schmid, A.M.M., Tippit, D.H., Cell division in diatoms. Protoplasma, 120, 1-2, 132–154, 1984. [1.167] Pickett-Heaps, J.D., Schmid, A.M.M., Tippit, D.H., Cell division in diatoms: A translation of part of Robert Lauterborn’s treatise of 1896 with some modern confirmatory observations. Protoplasma, 120, 1-2, 132–154, 1984. [1.168] Pickett-Heaps, J.D. and Spurck, T., Effects of drugs on the diatom spindle. J. Cell Biol., 91, 2, A316, 1981. [1.169] Pickett-Heaps, J.D. and Spurck, T.P., Studies on kinetochore function in mitosis. I. The effects of colchicine and cytochalasin on mitosis in the diatom Hantzschia amphioxys. Eur. J. Cell Biol., 28, 1, 77–82, 1982. [1.170] Pickett-Heaps, J.D. and Spurck, T.P., Studies on kinetochore function in mitosis. II. The effects of metabolic inhibitors on mitosis and cytokinesis in the diatom Hantzschia amphioxys. Eur. J. Cell Biol., 28, 1, 83–91, 1982. [1.171] Pickett-Heaps, J.D. and Staehelin, L.A., Ultrastructure of Scenedesmus (Chlorophyceae). II. Cell division and colony formation. J. Phycol., 11, 2, 186–202, 1975. [1.172] Pickett-Heaps, J.D. and Tippit, D.H., Desmid morphogenesis. Brookhaven Symp. Biol., 25, 191–205, 1973. [1.173] Pickett-Heaps, J.D. and Tippit, D.H., The diatom spindle in perspective. Cell, 14, 3, 455–467, 1978. [1.174] Pickett-Heaps, J.D., Tippit, D.H., Andreozzi, J.A., Cell division in pennate diatom Pinnularia. I. Early stages in mitosis. Biol. Cell., 33, 1, 71–78, 1978. [1.175] Pickett-Heaps, J.D., Tippit, D.H., Andreozzi, J.A., Cell division in the pennate diatom Pinnularia I. Early stages in mitosis. Biol. Cell., 33, 1, 71–78, 1978. [1.176] Pickett-Heaps, J.D., Tippit, D.H., Andreozzi, J.A., Cell division in the pennate diatom Pinnularia. II. Later stages in mitosis. Biol. Cell., 33, 1, 79–84, 1978. [1.177] Pickett-Heaps, J.D., Tippit, D.H., Andreozzi, J.A., Cell division in the pennate diatom Pinnularia. III. The valve and associated cytoplasmic organelles. Biol. Cell., 35, 2, 195–198, 1979. [1.178] Pickett-Heaps, J.D., Tippit, D.H., Andreozzi, J.A., Cell division in the pennate diatom Pinnularia. IV. Valve morphogenesis. Biol. Cell, 35, 2, 199–203, 1979.

  xv [1.179] Pickett-Heaps, J.D., Tippit, D.H., Andreozzi, J.A., Cell division in the pennate diatom Pinnularia. V. Observations on live cells. Biol. Cell., 35, 3, 295–304, 1979. [1.180] Pickett-Heaps, J.D., Tippit, D.H., Cohn, S.A., Spurck, T.P., Microtubule dynamics in the spindle. Theoretical aspects of assembly/disassembly reactions in vivo. J. Theor. Biol., 118, 2, 153–169, 1986. [1.181] Pickett-Heaps, J.D., Tippit, D.H., Leslie, R., Cell division in the diatom Hantzschia amphioxys. J. Cell Biol., 87, 2, A234, 1980. [1.182] Pickett-Heaps, J.D., Tippit, D.H., Leslie, R., Light and electron microscopic observations on cell division in two large pennate diatoms, Hantzschia and Nitzschia. I. Mitosis in vivo. Eur. J. Cell Biol., 21, 1, 1–11, 1980. [1.183] Pickett-Heaps, J.D., Tippit, D.H., Leslie, R., Light and electron microscopic observations on cell division in two large pennate diatoms. Hantzschia and Nitzschia. II. Ultrastructure. Eur. J. Cell Biol., 21, 1, 12–27, 1980. [1.184] Pickett-Heaps, J.D., Tippit, D.H., Porter, K.R., Rethinking mitosis. Cell, 29, 3, 729–744, 1982. [1.185] Pickett-Heaps, J.D. and West, J., Time lapse video observations on sexual plasmogamy in the red alga Bostrychia. Eur. J. Phycol., 33, 1, 43–56, 1998. [1.186] Pickett-Heaps, J.D., West, J.A., Wilson, S.M., McBride, D.L., Time lapse videomicroscopy of cell (spore) movement in red algae. Eur. J. Phycol., 36, 1, 9–22, 2001. [1.187] Pickett-Heaps, J.D. and Wetherbee, R., Spindle function in the green alga Mougeotia. Absence of anaphase a correlates with postmitotic nuclear migration. Cell Motil. Cytoskeleton, 7, 1, 68–77, 1987. [1.188] Pickett-Heaps, J.D., Wetherbee, R., Hill, D.R.A., Cell division and morphogenesis of the labiate process in the centric diatom Ditylum brightswellii. Protoplasma, 143, 139–149, 1988. [1.189] Pickett-Heaps, J.D.P.-H.J., Living Cells: Structure, Diversity, and Evolution [12” NTSC videodisc], Sinauer Associates, Sunderland, MA, 1994. [1.190] Pollock, F., Pickett-Heaps, J., Schmid, A.M., Diatom protoplasts are amoeboid during recovery from osmotic shock. J. Phycol., 34, 3 SUPPL., 48–48, 1998. [1.191] Pollock, F., Pickett-Heaps, J., Schmid, A.M., Diatom protoplasts are affected differently by cytochalasin D, latrunculin B and oryzalin. J. Phycol., 35, 3 SUPPL., 24–25, 1999. [1.192] Pollock, F.M. and Pickett-Heaps, J.D., Spatial determinants in morphogenesis: Recovery from plasmolysis in the diatom Ditylum. Cell Motil. Cytoskeleton, 60, 2, 71–82, 2005. [1.193] Pollock, F.M. and Pickett-Heaps, J.D., Valve formation without mitosis in the diatom Ditylum recovering from plasmolysis. Nova Hedwigia, Suppl. 130, 119–125, 2006. [1.194] Reymond, O.L. and Pickett-Heaps, J.D., A routine flat embedding method for electron microscopy of microorganisms allowing selection and precisely orientated sectioning of single cells by light microscopy. J. Microsc., 130, Pt 1, 79–84, 1983. [1.195] Sampson, K. and Pickett-Heaps, J.D., Phallacidin staining of the mitotic spindle in the green alga Oedogonium. Mol. Biol. Cell, 11, Suppl. 5, 370A, 2000. [1.196] Sampson, K. and Pickett-Heaps, J.D., Phallacidin stains the kinetochore region in the mitotic spindle of the green algae Oedogonium spp. Protoplasma, 217, 4, 166–176, 2001. [1.197] Sampson, K. and Pickett-Heaps, J.D., Phallacidin stains the kinetochore region in the mitotic spindle of the green algae Oedogonium spp. (vol 217, pg 166, 2001). Protoplasma, 218, 3-4, 237, 2001. [1.198] Sampson, K., Pickett-Heaps, J.D., Forer, A., Cytochalasin D blocks chromosomal attachment to the spindle in the green alga Oedogonium. Protoplasma, 192, 3-4, 130–144, 1996. [1.199] Schibler, M.J. and Pickett-Heaps, J.D., Mitosis in Oedogonium: spindle microfilaments and the origin of the kinetochore fiber. Eur. J. Cell Biol., 22, 2, 687–698, 1980.

xvi   [1.200] Schibler, M.J. and Pickett-Heaps, J.D., Kinetochore bundle structure in the green alga, Oedogonium. J. Cell Biol., 91, 2, A316, 1981. [1.201] Schibler, M.J. and Pickett-Heaps, J.D., The kinetochore fiber structure in the acentric spindles of the green alga Oedogonium. Protoplasma, 137, 1, 29–44, 1987. [1.202] Schoeller, A., Cohn, S.A., Spurck, T.P., Pickett-Heaps, J.D., Edgar, L.E., Sexual reproduction in the pennate diatom Navicula cuspidata. I. Gametogenesis and zygote formation. J. Phycol., 20, Suppl, 28, 1984. [1.203] Schoeller, A., Pickett-Heaps, J.D., Gilkey, J., The cytoplast in algal cell structure. J. Cell Biol., 99, 4, A196, 1984. [1.204]  Snyder, J.A., Armstrong, L., Stonington, O.G., Spurck, T.P., Pickett-Heaps, J.D., UV microbeam irradiations of the mitotic spindle: Spindle forces and structural analysis of lesions. Eur. J. Cell Biol., 55, 1, 122–132, 1991. [1.205] Soranno, T. and Pickett-Heaps, J., Directionally controlled spindle disassembly after mitosis in the diatom Pinnularia. Eur. J. Cell Biol., 26, 2, 234–243, 1982. [1.206] Spurck, T., Forer, A., Pickett-Heaps, J., Ultraviolet microbeam irradiations of epithelial and spermatocyte spindles suggest that forces act on the kinetochore fibre and are not generated by its disassembly. Cell Motil. Cytoskeleton, 36, 2, 136–148, 1997. [1.207] Spurck, T., Pickett-Heaps, J., Klymkowsky, M., Metabolic inhibitors and mitosis .1. Effects of DNP DOG and nocodazole on live cells. J. Cell Biol., 99, 4, A240, 1984. [1.208] Spurck, T., Pickett-Heaps, J., Klymkowsky, M., Metabolic inhibitors and mitosis .2. Effect of DNP DOG and nocodazole on microtubules. J. Cell Biol., 99, 4, A443, 1984. [1.209] Spurck, T.P. and Pickett-Heaps, J.D., On the mechanism of anaphase A: evidence that ATP is needed for microtubule disassembly and not generation of polewards force. J. Cell Biol., 105, 4, 1691–1705, 1987. [1.210] Spurck, T.P. and Pickett-Heaps, J.D., The effects of diazepam on mitosis and the microtubule cytoskeleton. I. Observations on the diatoms Hantzschia amphioxys and Surirella robusta. J. Cell Sci., 107, 2643–2651, 1994. [1.211] Spurck, T.P., Pickett-Heaps, J.D., Klymkowsky, M.W., Metabolic inhibitors and mitosis. I. Effects of dinitrophenol deoxyglucose and nocodazole on the live spindle. Protoplasma, 131, 1, 47–59, 1986. [1.212] Spurck, T.P., Pickett-Heaps, J.D., Klymkowsky, M.W., Metabolic inhibitors and mitosis. II. Effects of dinitrophenol deoxyglucose and nocodazole on the microtubule cytoskeleton. Protoplasma, 131, 1, 60–74, 1986. [1.213] Spurck, T.P., Stonington, O.G., Snyder, J.A., Pickett-Heaps, J.D., Bajer, A., Molebajer, J., UV microbeam irradiations of the mitotic spindle. II. Spindle fiber dynamics and force production. J. Cell Biol., 111, 4, 1505–1518, 1990. [1.214] Staehelin, L.A. and Pickett-Heaps, J.D., Ultrastructure of Scenedesmus (Chlorophyceae). I. Species with reticulate or warty type of ornamental layer. J. Phycol., 11, 2, 163–185, 1975. [1.215] Stonington, O.G., Spurck, T.P., Snyder, J.A., Pickett-Heaps, J.D., UV microbeam irradiations of the mitotic spindle. I. The UV microbeam apparatus. Protoplasma, 153, 1-2, 62–70, 1989. [1.216] Storey, E., Spurck, T., Pickett-Heaps, J., Beyreuther, K., Masters, C.L., The amyloid precursor protein of Alzheimer’s disease is found on the surface of static but not actively motile portions of neurites. Brain Res., 735, 1, 59–66, 1996. [1.217] Storey, E., Spurck, T., Pickett-Heaps, J., Beyreuther, K., Masters, C.L., Surface app is not found on rapidly changing portions of neurites, including growth cones. Neurobiol. Aging, 15, S63, 1994.

  xvii [1.218] Tippit, D.H., Fields, C.T., O’Donnell, K.L., Pickett-Heaps, J.D., McLaughlin, D.J., The organization of microtubules during anaphase and telophase spindle elongation in the rust fungus Puccinia. Eur. J. Cell Biol., 34, 1, 34–44, 1984. [1.219] Tippit, D.H., McDonald, K.L., Pickett-Heaps, J.D., Cell division in the centric diatom Melosira varians. Cytobiologie, 12, 52–73, 1975. [1.220] Tippit, D.H. and Pickett-Heaps, J.D., Apparent amitosis in the binucleate dinoflagellate ˆ. J. Cell Sci., 21, 2, 273–289, 1976. [1.221] Tippit, D.H. and Pickett-Heaps, J.D., Mitosis in pennate diatom Surirella ovalis. J. Cell Biol., 73, 3, 705–727, 1977. [1.222]  Tippit, D.H. and Pickett-Heaps, J.D., Reconstruction of spindle microtubules during anaphase elongation in the rust fungus Puccinia. J. Cell Biol., 97, 5, A187, 1983. [1.223] Tippit, D.H., Pickett-Heaps, J.D., Leslie, R., Cell division in two large pennate diatoms Hantzschia and Nitzschia. III. A new proposal for kinetochore function during prometaphase. J. Cell Biol., 86, 2, 402–416, 1980. [1.224]  Tippit, D.H., Pillus, L., Pickett-Heaps, J., Organization of spindle microtubules in Ochromonas danica. J. Cell Biol., 87, 3, 531–545, 1980. [1.225]  Tippit, D.H., Pillus, L., Pickett-Heaps, J.D., Interactions of spindle microtubules in Ochromonas danica. Eur. J. Cell Biol., 22, 1, 309–309, 1980. [1.226] Tippit, D.H., Pillus, L., Pickett-Heaps, J.D., Near neighbor analysis of spindle microtubules in the alga Ochromonas. Eur. J. Cell Biol., 30, 1, 9–17, 1983. [1.227] Tippit, D.H., Schulz, D., Pickett-Heaps, J.D., Changes in spindle structure during mitosis in Fragilaria. J. Cell Biol., 75, 2, A279, 1977. [1.228] Tippit, D.H., Schulz, D., Pickett-Heaps, J.D., Analysis of the distribution of spindle microtubules in the diatom Fragilaria. J. Cell Biol., 79, 3, 737–763, 1978. [1.229] Tippit, D.H., Smith, H., Pickett-Heaps, J.D., Cell form mutants in Micrasterias. Protoplasma, 113, 3, 234–236, 1982. [1.230] Troutt, L.L. and Pickett-Heaps, J.D., Reactivating prometaphase movement in permeabilized animal cells. Protoplasma, 170, 1-2, 22–33, 1992. [1.231] Troutt, L.L., Spurck, T.P., Pickett-Heaps, J.D., The effects of diazepam on mitosis and the microtubule cytoskeleton. II. Observations on newt epithelial and Ptk1 cells. Protoplasma, 189, 1-2, 101–112, 1995. [1.232] Troxell, C.L., Scheffey, C., Pickett-Heaps, J.D., Ionic currents during wall morphogenesis in Micrasterias and Closterium. Prog. Clin. Biol. Res., 210, 105–112, 1986. [1.233] van de Meene, A.M.L. and Pickett-Heaps, J.D., Spine morphogenesis in three marine centric diatoms. Phycologia, 36, 4, 115, 1997. [1.234] van de Meene, A.M.L. and Pickett-Heaps, J.D., Cytoplasmic rotation during valve morphogenesis of the marine centric diatom Rhizosolenia setigera. Phycologia, 40, 4 Supplement, 40, 2001. [1.235] van de Meene, A.M.L. and Pickett-Heaps, J.D., Valve morphogenesis in the centric diatom Proboscia alata Sundstrom. J. Phycol., 38, 2, 351–363, 2002. [1.236] van de Meene, A.M.L. and Pickett-Heaps, J.D., Valve morphogenesis in the centric diatom Rhizosolenia setigera (Bacillariophyceae, Centrales) and its taxonomic implications. Eur. J. Phycol., 39, 1, 93–104, 2004. [1.237] Vesk, M., Hoffman, L.R., Pickett-Heaps, J.D., Mitosis and cell division in Hydrurus foetidus (Chrysophyceae). J. Phycol., 20, 4, 461–470, 1984. [1.238] Weatherill, K., Lambiris, I., Pickett-Heaps, J.D., Deane, J.A., Beech, P.L., Plastid division in Mallomonas (Synurophyceae, Heterokonta). J. Phycol., 43, 3, 535–541, 2007. [1.239] Weatherill, K.J., Lambiris, I., Pickett-Heaps, J., Beech, P.L., Chloroplast division in the chromophyte alga, Mallomonas. Phycologia, 36, 4, Suppl. 5, 121, 1997.

xviii   [1.240]  West, J.A., Zuccarello, G.C., Scott, J., Pickett-Heaps, J., Kim, G.H., Observations on Purpureofilum apyrenoidigerum gen. et sp. nov. from Australia and Bangiopsis subsimplex from India (Stylonematales, Bangiophyceae, Rhodophyta). Phycol. Res., 53, 1, 49–66, 2005. [1.241] Wetherbee, R., Andersen, R.A., Pickett-Heaps, J.D. (Eds.), The Protistan Cell Surface (“Protoplasma”), Springer, 1994. [1.242] Wetherbee, R., Andersen, R.A., Pickett-Heaps, J.D., Untitled. Protoplasma, 181, 1-4, R3, 1994. [1.243] Wetherbee, R., Platt, S.J., Beech, P.L., Pickett-Heaps, J.D., Flagellar transformation in the heterokont Epipyxis pulchra (Chrysophyceae): Direct observations using image enhanced light microscopy. Protoplasma, 145, 1, 47–54, 1988. [1.244] Wilson, S.M., Pickett-Heaps, J.D., West, J.A., Fertilization and the cytoskeleton in the red alga Bostrychia moritziana (Rhodomelaceae, Rhodophyta). Eur. J. Phycol., 37, 4, 509–522, 2002. [1.245] Wilson, S.M., Pickett-Heaps, J.D., West, J.A., Vesicle transport and the cytoskeleton in the unicellular red alga Glaucosphaera vacuolata. Phycol. Res., 54, 1, 15–20, 2006. [1.246] Wilson, S.M., West, J.A., Pickett-Heaps, J.D., Time lapse videomicroscopy of fertilization and the actin cytoskeleton in Murrayella periclados (Rhodomelaceae, Rhodophyta). Phycologia, 42, 6, 638–645, 2003.

Contents Preface xxvii 1 Some Observations of Movements of Pennate Diatoms in Cultures and Their Possible Interpretation Thomas Harbich 1.1 Introduction 1.2 Kinematics and Analysis of Trajectories in Pennate Diatoms with Almost Straight Raphe along the Apical Axis 1.3 Curvature of the Trajectory at the Reversal Points 1.4 Movement of Diatoms in and on Biofilms 1.5 Movement on the Water Surface 1.6 Formation of Flat Colonies in Cymbella lanceolata 1.7 Conclusion References

1 2 3 9 13 16 23 29 29

2 The Kinematics of Explosively Jerky Diatom Motility: A Natural Example of Active Nanofluidics 33 Ahmet C. Sabuncu, Richard Gordon, Edmond Richer, Kalina M. Manoylov and Ali Beskok 2.1 Introduction 34 2.2 Material and Methods 35 2.2.1 Diatom Preparation 35 2.2.2 Imaging System 35 2.2.3 Sample Preparation 36 2.2.4 Image Processing 36 2.3 Results and Discussion 41 2.3.1 Comparison of Particle Tracking Algorithms 41 2.3.2 Stationary Particles 42 2.3.3 Diatom Centroid Measurements 43 2.3.4 Diatom Orientation Angle Measurements 46 2.3.5 Is Diatom Motion Characterized by a Sequence of Small Explosive Movements? 49 2.3.6 Future Work 50 2.4 Conclusions 51 Appendix 52 References 59 xix

xx  Contents 3 Cellular Mechanisms of Raphid Diatom Gliding Yekaterina D. Bedoshvili and Yelena V. Likhoshway 3.1 Introduction 3.2 Gliding and Secretion of Mucilage 3.3 Cell Mechanisms of Mucilage Secretion 3.4 Mechanisms of Gliding Regulation 3.5 Conclusions Acknowledgments References

65

4 Motility of Biofilm-Forming Benthic Diatoms Karen Grace Bondoc-Naumovitz and Stanley A. Cohn 4.1 Introduction 4.2 General Motility Models and Concepts 4.2.1 Adhesion 4.2.2 Gliding Motility 4.2.3 Motility and Environmental Responsiveness 4.3 Light-Directed Vertical Migration 4.4 Stimuli-Directed Movement 4.4.1 Nutrient Foraging 4.4.2 Pheromone-Based Mate-Finding Motility 4.4.3 Prioritization Between Co-Occurring Stimuli 4.5 Conclusion References

77

65 67 68 71 72 72 73

77 86 87 89 91 93 94 94 97 99 99 100

5 Photophobic Responses of Diatoms – Motility and Inter-Species Modulation Stanley A. Cohn, Lee Warnick and Blake Timmerman 5.1 Introduction 5.2 Types of Observed Photoresponses 5.2.1 Light Spot Accumulation 5.2.2 High-Intensity Light Responses 5.3 Inter-Species Effects of Light Responses 5.3.1 Inter-Species Effects on High Irradiance Direction Change Response 5.3.2 Inter-Species Effects on Cell Accumulation into Light Spots 5.4 Summary References

111

6 Diatom Biofilms: Ecosystem Engineering and Niche Construction David M. Paterson and Julie A. Hope 6.1 Introduction 6.1.1 Diatoms: A Brief Portfolio 6.1.2 Benthic Diatoms as a Research Challenge 6.2 The Microphytobenthos and Epipelic Diatoms 6.3 The Ecological Importance of Locomotion 6.4 Ecosystem Engineering and Functions 6.4.1 Ecosystem Engineering 6.4.2 Ecosystem Functioning

135

112 112 112 114 118 119 123 123 131

135 135 136 136 137 139 139 140

Contents  xxi 6.5 Microphytobenthos as Ecosystem Engineers 6.5.1 Sediment Stabilization 6.5.2 Beyond the Benthos 6.5.3 Diatom Architects 6.5.4 Working with Others: Combined Effects 6.5.5 The Dynamic of EPS 6.5.6 Nutrient Turnover and Biogeochemistry 6.6 Niche Construction and Epipelic Diatoms 6.7 Conclusion Acknowledgments References 7 Diatom Motility: Mechanisms, Control and Adaptive Value João Serôdio 7.1 Introduction 7.2 Forms and Mechanisms of Motility in Diatoms 7.2.1 Motility in Centric Diatoms 7.2.2 Motility in Pennate Raphid Diatoms 7.2.3 Motility in Other Substrate-Associated Diatoms 7.2.4 Vertical Migration in Diatom-Dominated Microphytobenthos 7.3 Controlling Factors of Diatom Motility 7.3.1 Motility Responses to Vectorial Stimuli 7.3.1.1 Light Intensity 7.3.1.2 Light Spectrum 7.3.1.3 UV Radiation 7.3.1.4 Gravity 7.3.1.5 Chemical Gradients 7.3.2 Motility Responses to Non-Vectorial Stimuli 7.3.2.1 Temperature 7.3.2.2 Salinity 7.3.2.3 pH 7.3.2.4 Calcium 7.3.2.5 Other Factors 7.3.2.6 Inhibitors of Diatom Motility 7.3.3 Species-Specific Responses and Interspecies Interactions 7.3.4 Endogenous Control of Motility 7.3.5 A Model of Diatom Vertical Migration Behavior in Sediments 7.4 Adaptive Value and Consequences of Motility 7.4.1 Planktonic Centrics 7.4.2 Benthic Pennates 7.4.3 Ecological Consequences of Vertical Migration 7.4.3.1 Motility-Enhanced Productivity 7.4.3.2 Carbon Cycling and Sediment Biostabilization Acknowledgments References

141 141 143 144 144 145 145 146 149 150 150 159 159 160 160 161 162 163 164 164 164 165 166 166 167 167 167 168 168 168 169 169 169 170 170 172 172 173 175 175 176 176 176

xxii  Contents 8 Motility in the Diatom Genus Eunotia Ehrenb. Paula C. Furey 8.1 Introduction 8.2 Accounts of Movement in Eunotia 8.3 Motility in the Context of Valve Structure 8.3.1 Motility and Morphological Characteristics in Girdle View 8.3.2 Motility and Morphological Characteristics in Valve View 8.3.3 Motility and the Rimoportula 8.4 Motility and Ecology of Eunotia 8.4.1 Substratum-Associated Environments 8.4.2 Planktonic Environments 8.5 Motility and Diatom Evolution 8.6 Conclusion and Future Directions Acknowledgements References

185

9 A Free Ride: Diatoms Attached on Motile Diatoms Vincent Roubeix and Martin Laviale 9.1 Introduction 9.2 Adhesion and Distribution of Epidiatomic Diatoms on Their Host 9.3 The Specificity of Host-Epiphyte Interactions 9.4 Cost-Benefit Analysis of Host-Epiphyte Interactions 9.5 Conclusion References

211

10 Towards a Digital Diatom: Image Processing and Deep Learning Analysis of Bacillaria paradoxa Dynamic Morphology Bradly Alicea, Richard Gordon, Thomas Harbich, Ujjwal Singh, Asmit Singh and Vinay Varma 10.1 Introduction 10.1.1 Organism Description 10.1.2 Research Motivation 10.2 Methods 10.2.1 Video Extraction 10.2.2 Deep Learning 10.2.3 DeepLabv3 Analysis 10.2.4 Primary Dataset Analysis 10.2.5 Data Availability 10.3 Results 10.3.1 Watershed Segmentation and Canny Edge Detection 10.3.2 Deep Learning 10.4 Conclusion Acknowledgments References

185 188 194 194 196 198 198 199 201 202 203 204 205

211 213 215 217 219 219 223 224 224 227 228 228 230 234 234 235 235 235 236 243 245 245

Contents  xxiii 11 Diatom Triboacoustics Ille C. Gebeshuber, Florian Zischka, Helmut Kratochvil, Anton Noll, Richard Gordon and Thomas Harbich Glossary 11.1 State-of-the-Art 11.1.1 Diatoms and Their Movement 11.1.2 The Navier-Stokes Equation 11.1.3 Low Reynolds Number 11.1.4 Reynolds Number for Diatoms 11.1.5 Further Thoughts About Movement of Diatoms 11.1.6 Possible Reasons for Diatom Movement 11.1.7 Underwater Acoustics, Hydrophones 11.1.7.1 Underwater Acoustics 11.1.7.2 Hydrophones 11.2 Methods 11.2.1 Estimate of the Momentum of a Moving Diatom 11.2.2 On the Speed of Expansion of the Mucopolysaccharide Filaments 11.2.2.1 Estimation of Radial Expansion 11.2.2.2 Sound Generation 11.2.3 Gathering Diatoms 11.2.3.1 Purchasing Diatom Cultures 11.2.3.2 Diatoms from the Wild 11.2.4 Using a Hydrophone to Detect Possible Acoustic Signals from Diatoms 11.2.4.1 First Setup 11.2.4.2 Second Setup 11.3 Results and Discussion 11.3.1 Spectrograms 11.3.2 Discussion 11.4 Conclusions and Outlook Acknowledgements References

249

12 Movements of Diatoms VIII: Synthesis and Hypothesis Jean Bertrand 12.1 Introduction 12.2 Review of the Conditions Necessary for Movements 12.3 Hypothesis 12.4 Analysis – Comparison with Observations 12.4.1 Translational Apical Movement 12.4.2 The Transapical Toppling Movement 12.4.3 Diverse Pivoting  12.5 Conclusion Acknowledgments References

283

249 251 251 252 253 254 254 255 256 256 257 257 257 258 258 261 266 267 267 269 269 271 272 272 277 277 279 279

283 284 285 288 288 290 290 291 292 292

xxiv  Contents 13 Locomotion of Benthic Pennate Diatoms: Models and Thoughts Jiadao Wang, Ding Weng, Lei Chen and Shan Cao 13.1 Diatom Structure 13.1.1 Ultrastructure of Frustules 13.1.2 Bending Ability of Diatoms 13.2 Models for Diatom Locomotion 13.2.1 Edgar Model for Diatom Locomotion 13.2.2 Van der Waals Force Model (VW Model) for Diatom Locomotion 13.2.2.1 Locomotion Behavior of Diatoms 13.2.2.2 Moving Organelles and Pseudopods 13.2.2.3 Chemical Properties of Mucilage Trails 13.2.2.4 Mechanical Properties of Mucilage Trails 13.2.2.5 VW Model for Diatom Locomotion 13.3 Locomotion and Aggregation of Diatoms 13.3.1 Locomotion Trajectory and Parameters of Diatoms 13.4 Simulation on Locomotion, Aggregation and Mutual Perception of Diatoms 13.4.1 Simulation Area and Parameters 13.4.2 Diatom Life Cycle and Modeling Parameters 13.4.3 Simulation Results of Diatom Locomotion Trajectory with Mutual Perception 13.4.4 Simulation Results of Diatom Adhesion with Mutual Perception 13.4.5 Adhesion and Aggregation Mechanism of Diatoms References

295

14 The Whimsical History of Proposed Motors for Diatom Motility Richard Gordon 14.1 Introduction 14.2 Historical Survey of Models for the Diatom Motor 14.2.1 Diatoms Somersault via Protruding Muscles (1753) 14.2.2 Vibrating Feet or Protrusions Move Diatoms (1824) 14.2.3 Diatoms Crawl Like Snails (1838) 14.2.4 The Diatom Motor Is a Jet Engine (1849) 14.2.5 Rowing Diatoms (1855) 14.2.6 Diatoms Have Protoplasmic Tank Treads (1865) 14.2.7 Diatoms as the Flame of Life: Capillarity (1883) 14.2.8 Bellowing Diatoms (1887) 14.2.9 Jelly Powered Jet Skiing Diatoms (1896) 14.2.10 Bubble Powered Diatoms (1905) 14.2.11 Diatoms Win: “I Have No New Theory to Offer and See No Reason to Use Those Already Abandoned” (1940) 14.2.12 Is Diatom Motility a Special Case of Cytoplasmic Streaming? (1943) 14.2.13 Diatom Adhesion as a Sliding Toilet Plunger (1966) 14.2.14 Diatom as a Monorail that Lays Its Own Track (1967)

335

295 295 297 300 300 302 302 304 307 310 314 319 319 323 323 323 326 327 331 332

336 338 338 338 342 344 346 350 354 355 355 358 360 360 365 366

Contents  xxv 14.2.15 The Diatom as a “Compressed Air” Coanda Effect Gliding Vehicle (1967) 368 14.2.16 The Electrokinetic Diatom (1974) 371 14.2.17 The Diatom Clothes Line or Railroad Track (1980) 372 14.2.18 Diatom Ion Cyclotron Resonance (1987) 374 14.2.19 Diatoms Do Internal Treadmilling (1998) 375 14.2.20 Surface Treadmilling, Swimming and Snorkeling Diatoms (2007) 376 14.2.21 Acoustic Streaming: The Diatom as Vibrator or Jack Hammer (2010) 378 14.2.22 Propulsion of Diatoms Via Many Small Explosions (2020) 379 14.2.23 Diatoms Walk Like Geckos (2019) 380 14.3 Pulling What We Know and Don’t Know Together, about the Diatom Motor 381 14.4 Membrane Surfing: A New Working Hypothesis for the Diatom Motor (2020) 393 Acknowledgments 397 References 397 Appendix 420

Index 421

Preface Anyone who has peered into a microscope and observed the movement of diatoms knows they have witnessed an intriguing example of cellular biology. Unlike most other of their sister algae, this movement involves neither swimming through solution (like Euglena or Chlamydomonas) or amoeboid crawling of membrane and cytoplasm (like Synchroma). Surrounded by a hardened silicified cell wall, motile diatoms are still able to glide gracefully along surfaces while the cell protoplast remains contained within these ornate cell walls. As such, the mysteries involving this curious form of movement have been of interest for well over a hundred years, and models of many sorts have been proposed to explain it (see [1.20]). Our hope is that this volume will help to not only convey our excitement about research in diatoms, but also demonstrate a variety of techniques and approaches currently used to understand some of the aspects of diatom movement. We have included chapters centering on a number of areas: detailed observation of movements [1.23] [1.43], cellular aspects of motility [1.5] [1.8], ecology and environmental interactions [1.13] [1.40] [1.44], more passive and epiphitic movements [1.18] [1.42], new and novel methodologies [1.2] [1.51] and potential models of motility [1.7] [1.20] [1.47]. Our goal is not to vigorously promote and defend any one particular model, but rather to present the reader with the variety of experimental approaches that are currently being used to address the problem. In this way readers will be able to assess for themselves the areas of diatom motility that require further exploration, and the predictions of various models that still need to be tested. For example, the exact mechanism of force production for diatom motility is still unresolved. While models of force generation arising from motor proteins interacting with the cytoskeleton and coupled to secreted mucilage strands are favored by some, others currently favor models generating motile force generated by the explosive release and hydration of mucilage regulated by the localization of the secretory site directed by the underlying cytoskeleton. There are certainly areas of diatom motility that were unfortunately not able to be included in this volume, and we encourage readers to explore these areas if they wish to be more fully aware of important work in the field. In particular, the editors want to note a number of areas of diatom motility that are not fully addressed in the current volume or are open areas and questions needing more research: Chemotaxis: Understanding the chemical triggers that can stimulate and help regulate diatom movement, especially during cell pairing during reproduction, is crucial to a full understanding of the process. Important work on diatom chemoattractants and pheromones has been done in recent years (e.g., diatom pheromones [1.19] [1.39]), although

xxvii

xxviii  Preface the mechanisms by which these chemical stimulants interact with and help to regulate the motility generating process are still poorly understood. Tube-dwelling diatoms: A number of species have the ability to specialize their extracellular secretions to provide their own surfaces for movement [1.27] [1.48], generating types of stalks and tubes through which the diatoms can move, but providing three-dimensional structures important for attachment and ecology of other organisms [1.17] [1.28]. Centric diatoms: While centric diatoms have little or no direct substratum motility as seen with many of the pennate diatoms, they can modify their position in the water column [1.36] and there has been some great recent work demonstrating there is direct regulation of diatom buoyancy [1.16] [1.34]. We encourage readers to explore this topic as well if they wish to be further engaged in current approaches regarding functional regulation of centric movement. Composition of diatom mucilage: Understanding the chemical and physical nature of diatom mucilages and secretions is important to understanding the way that diatoms can use mucilage for a variety of functions. It is likely that different materials are secreted for purposes such as protection of cells during reproduction, and holding the two halves of their frustule together, stalk production, as well as making connections that can move their position relative to the frustule. A number of prior investigations have begun to look into this (e.g., [1.26] [1.27]) and it seems like a great opportunity for continued future work. It has practical impact in the study of biofouling [1.50] and underwater adhesives. Photoreception: A number of labs have begun to investigate the types of molecules responsible for photoreception in algae. While numerous types of promising candidates have been described (e.g. [1.12] [1.29] [1.31] [1.35] [1.37]), there have been no definitive studies pointing to specific molecules driving the diurnal, light aggregation, or light avoidance behaviors. Better knowledge of the specific light and chemical receptors in diatoms, and how they alter the processes of force generation and directional bias in cells will be needed too. Light piping in the colonial pennate diatom Bacillaria has been postulated [1.21], but not yet tested. Effect of morphogenetic alterations on motility: Numerous diatom species have alternative morphologies based on the environmental conditions (e.g., [1.9] [1.30]). In addition, while numerous pennate diatoms are basically symmetric about the transapical plane dividing the two raphe branches (e.g., Navicula spp.), there are also numerous other species (e.g., Gomphonema spp.) in which the raphe runs down the apical axis, but the morphology at the two ends is decidedly different. There are also species where the raphe is displaced along valvar wings and the break between branches is at one end (e.g., Surirella spp.). The characterization of such species, correlating the valve morphology and raphe morphology with motility characteristics, seems like a productive line of research to better determine the relationship between wall structure and movement, and whether the motility associated with the ends of raphe branches can be regulated independently. Cytoskeletal organization: The actin cables comprised of large bundles of actin filaments underlying the raphe in motile raphid diatoms appear essential to active,

Preface  xxix well-regulated motility [1.41]. But the connection between the filaments of the cables and the raphe mucilage fibers remains poorly understood. More research definitely needs to be done on more detailed organization of the actin filaments within the larger ultrastructure and orientation of the cell and frustule. The polarity of the two actin cables (parallel? antiparallel?) and relative placement to the membrane are crucial details to resolve. Evolutionary relationships of motility: While diatom gliding is a somewhat unique form of motility, a clearer understanding of the movement would also arise from a better knowledge of its evolutionary basis. For example, some algae such as desmids or filamentous bacteria can move via direct mucilage secretion through specialized pores [1.10] [1.14] [1.15], some algae like Chara use membrane-associated actin cables to generate intracellular movement and cytoplasmic streaming [1.25] [1.46], and in some cases algae that normally swim (e.g., Chlamydomonas) can glide over a surface using the membranous intracellular transport powered by motor proteins using the underlying cytoskeleton [1.45]. Gliding in myxobacteria [1.49] is similar to that of diatoms. Studying how some of these types of components might be related to diatom gliding could yield important insights into both the evolution and physiology of these types of movements. In addition to these, there are numerous areas that are ripe for fresh research. Diatom species are a foundational food component to many aquatic ecosystems, and are quite sensitive (in motility, metabolism and reproduction) to temperature fluctuations. Thus, detailed ecological studies of effects of diatoms in changing temperatures would be crucially important to understand the ecological impact of diatoms related to temperature change, daily and long-term. We know that mucilage strength and resilience, and subsequent motile abilities, are all related to temperature and could be severely affected by small temperature changes. Knowing how these motility and adhesion attributes change, and which species are more sensitive, would be a great boost to understanding the ecological ramifications of climate change. Another interesting topic not fully resolved and of interest is determining how expensive such motility really is for the diatoms, as they leave large amounts of extruded carbohydrate externally to the frustule and need to constantly synthesize materials associated with motility. While such mucilage also becomes connected to other ecological issues with a tight-knit algal community, the energy costs for an individual cell is an intriguing question. Diatom secretions are also strongly related to many ecological structures of aquatic ecosystems [1.4]. They affect soil and sediment stability, food access during diurnal movements, food accessibility in stalked versus benthic forms, and connection and stability within complex algal communities where diatoms can work to interconnect various forms of algae. It would be worthwhile to investigate the details of the mucilage from various forms and species of diatoms in different communities (and, as above, their changes as a function of temperature) in an attempt to understand which secretions (e.g., motility related versus non-motility related) are most important for different aspects of an algal community. Investigating the changes in these secretions, and in the resulting motile characteristics of the cells, will provide a much stronger understanding of the ways diatoms are functionally integrated within an algal community. Moreover, understanding the energetics of motility is crucial to understanding the constraints placed on a cell in its generation of movement. While a recent work has begun to consider the energetics of diatom movement during

xxx  Preface diurnal migration [1.38], there is much work to be done in understanding the energy consumed by diatoms under different ecological situations. There is also a strong need for additional work into high resolution forms of microscopy to determine, in living cells, where mucilage is being secreted, the characteristics of the mucilage secreted from the raphe, and the way in which raphe connections to the substratum are correlated with stimuli (e.g., light irradiation) affecting the direction and motility. Understanding the molecular controls on motility within the cell needs additional research to identify the receptors and molecules responsible for regulation and synthesis of the different enzymes and substances needed in the motile machinery. Despite these many open areas of diatom research into motility, it is also important to reiterate what we do know about diatom motility: Adhesion and motility are closely coupled: The requirement for raphe secreted mucilage to adhere to a solid or semi-solid substratum in order to move is well supported. Diatom secretions are crucial for a number of cell processes, including protection of protoplasts during cell conjugation, attachment and integrity of cell wall components, formation of stalks, and motility of cells. Inhibitors of diatom secretion inhibit motility, and diatom motility is strongly correlated with the degree to which it can adhere to a substratum. In some research, it has been shown that diatoms on the underside of a surface can also pull themselves back up to a motile confirmation after briefly remaining adhered by only a single end of the cell. Motile characteristics are species specific: Numerous lines of research have shown that path curvature, mean path lengths, light wavelength stimuli, speed, and strength of adhesion during motility during movement are all species specific. In this way, species determination can be made by distinct characteristics of motile behavior, in addition to more typically used aspects such as frustule ornamentation and detail and life history. Thus, diatoms, like most other organisms, are not just a set of morpho-species separated by evolutionary diversification of frustule design, but have clear physiological differences that relate not just to reproductive ability, but also to everyday photo- and chemo-responsive behaviors. Motile characteristics of cells are crucial components of local ecology: Every aspect of diatom motility is directly connected to aspects of local aquatic ecology. The amount of secretion has direct effect on sediment stability, rates and sensitivities of photoresponses directly affect diurnal rhythms through sediment and thereby access of diatoms to higher level consumers, movement and mucilage secretion can provide surface conditioning of rocks and surfaces for immigration and colonization of other organisms, differential motile responses can lead to niche partitioning and increased species diversity, and movement of diatoms via epiphytic attachment can drive influx and retention of other species. Motile characteristics are sensitive to physical and biological ecological conditions: Many studies have shown that pH, ionic composition, temperature, and surface conditions all play a role in motile characteristics of diatom cells. This not only allows surveys of species distributions to help determine the ecological conditions present in various ecosystems, but helps to understand geographical distributions and immigration/emigration characteristics. Moreover, motile characteristics can also be modulated by the presence or absence of

Preface  xxxi other organisms or diatoms, leading to increased movements into areas that allow for lower competition and increased ecological success. Motility in mudflats along coasts have been studied in detail: Common diatoms in those habitats like Cylindrotheca closterium (Ehrenberg) Reimann & J.C. Lewin have been shown to have four types of movement modalities: gliding (smooth or corkscrew), non-gliding (pivots and rollover), gliding pirouettes and detaching movements [1.3]. Considering the cohesive nature of the mudflat sediment [1.6], corkscrew gliding was reported to help with mechanisms for movement through the fine layers. Responses to salinity included non-­ gliding movements like rollover and detachment (probably associated with polysaccharide synthesis) [1.1]. Changes in the chemical gradients with the mudflats stimulates pirouettes and pivot movements, helping the cells to escape unfavorable conditions [1.24]. Actin filaments underlying the raphe are crucial to raphid pennate motility: While the way in which the actin filaments contribute to motility is still not fully understood, it is clear that inhibition of actin coordinately inhibits motility. While possibly used for mucilage placement, orientation, or coupling to motor protein force generation, actin importance is undeniable. Localizations of diatoms during movement is due to directional bias: While many other types of algae and protists can maneuver in elaborate two-dimensional or three-­ dimensional movements, diatoms mainly are constrained locally to a one-dimensional axis defined by the raphe. Within that local area, movement is essentially regulated by biasing the cell in the direction of movement along that axis. For example, while the intensity and wavelength vary by species, virtually all motile diatoms are biased along the axis to move away from very high irradiance light, and towards more moderate light levels. Similarly, cells triggered to undergo reproduction tend to find other cells to pair with by biased forward/back movements along with random rotations, rather than any kind of true directional reorientation. Questions raised 20-30 years ago, like whether migration rhythms of sigmoid and nitzschioid biraphid diatoms responding to different stimuli like tides or light [1.22] [1.33] or chemical motility inhibitors working through changes is photosynthetic activity or not [1.11], have been partially answered. The rhythms of diatom movements appear synchronous with tides for large motile representatives of genera like Pleurosigma, Gyrosigma and Navicula. At low tide, movement and speed on the surface of the sediment was observed, ensuring the cells good access to light. At high tide, movement was minimal, probably due to sheer pressure of the water layer above the sediment, making it impossible for microbes to move. Individuals within a colony of Bacillaria paxillifera (O.F. Müller) T. Marsson followed diurnal rhythms and moved only when light was available [1.32]. Chemicals inhibiting myosin-based motility in animals or actin-binding chemical from marine sponges were shown to inhibit diatom motility [1.11] [1.41]. We would like to thank all the authors and contributors to this volume for bringing their joy of diatoms to share with the readers. We hope that this volume will help reinforce the enthusiasm of all those interested in diatom motility, and help them in the search for better understanding of a truly fascinating phenomenon.

xxxii  Preface The editors would also like to thank all the authors for sharing their knowledge and ideas on diatom motility and for their patience in the process. Also, the editors would like to gratefully acknowledge our external reviewers for agreeing to critique the initial drafts of these manuscripts. These reviewers include: Małgorzata Bąk, Karen Bondoc, Manfred Drack, Natalie Hicks, Kai Lu, James Nienow, Chris Peterson, Tom Portegys, Nicole Poulsen, and Johannes Srajer. Their work has contributed immensely to the quality of the manuscripts. Stanley A. Cohn Kalina M. Manoylov Richard Gordon June 2021

Preface  xxxiii

References [1.1] Abdullahi, A.S., Underwood, G.J.C. and Gretz, M.R. (2006) Extracellular matrix assembly in diatoms (Bacillariophyceae). V. Environmental effects on polysaccharide synthesis in the model diatom, Phaeodactylum tricornutum. Journal of Phycology 42(2), 363-378. [1.2] Alicea, B., Gordon, R., Harbich, T., Singh, A., Varma, V., Mehan, P. and Singh, U. (2020) Towards a digital diatom: Image processing and deep learning analysis of Bacillaria paradoxa dynamic morphology In: Diatom Gliding Motility [DIGM, Volume 2 in the series: Diatoms: Biology & Applications, series editors: Richard Gordon & Joseph Seckbach] S.A. Cohn, K.M. Manoylov and R. Gordon, (eds.) Wiley-Scrivener, Beverly, MA, USA: This volume. [1.3] Apoya-Horton, M.D., Yin, L., Underwood, G.J.C. and Gretz, M.R. (2006) Movement modalities and responses to environmental changes of the mudflat diatom Cylindrotheca closterium (Bacillariophyceae). Journal of Phycology 42(2), 379-390. [1.4] Aumeier, C. and Menzel, D. (2012) Secretion in the diatoms. In: Secretions and Exudates in Biological Systems. J.M. Vivanco and F. Baluška, (eds.) Springer: 221-250. [1.5] Bedoshvili, Y.D. (2020) Cellular mechanisms of raphid diatom motility. In: Diatom Gliding Motility [DIGM, Volume 2 in the series: Diatoms: Biology & Applications, series editors: Richard Gordon & Joseph Seckbach] S.A. Cohn, K.M. Manoylov and R. Gordon, (eds.) Wiley-Scrivener, Beverly, MA, USA. [1.6] Bellinger, B.J., Abdullahi, A.S., Gretz, M.R. and Underwood, G.J.C. (2005) Biofilm polymers: Relationship between carbohydrate biopolymers from estuarine mudflats and unialgal cultures of benthic diatoms. Aquatic Microbial Ecology 38(2), 169-180. [1.7] Bertrand, J. (2020) Diatom movements VIII: Synthesis and hypothesis. Translation of: Bertrand, J. (2008). Mouvements des diatomées VIII: synthèse et hypothèse. Diatom Research 23(1), 19-29, by: Richard Gordon, Martin Laviale & Karen K. Serieyssol in consultation with Jean Bertrand. In: Diatom Gliding Motility [DIGM, Volume 2 in the series: Diatoms: Biology & Applications, series editors: Richard Gordon & Joseph Seckbach] S.A. Cohn, K.M. Manoylov and R. Gordon, (eds.) Wiley-Scrivener, Beverly, MA, USA: This volume. [1.8] Bondoc-Naumovitz, K. and Cohn, S.A. (2020) Motility of biofilm-forming benthic diatoms. In: Diatom Gliding Motility [DIGM, Volume 2 in the series: Diatoms: Biology & Applications, series editors: Richard Gordon & Joseph Seckbach] S.A. Cohn, K.M. Manoylov and R. Gordon, (eds.) Wiley-Scrivener, Beverly, MA, USA: This volume. [1.9] Borowitzka, M.A. and Volcani, B.E. (1978) The polymorphic diatom Phaeodactylum tricornutum: Ultrastructure of its morphotypes. Journal of Phycology 14(1), 10-21. [1.10] Burchard, R.P. (1981) Gliding motility of prokaryotes: Ultrastructure, physiology, and genetics. Annu. Rev. Microbiol. 35, 497-529. [1.11] Cartaxana, P., Brotas, V. and Serodio, J. (2008) Effects of two motility inhibitors on the photosynthetic activity of the diatoms Cylindrotheca closterium and Pleurosigma angulatum. Diatom Research 23(1), 65-74. [1.12] Coesel, S., Mangogna, M., Ishikawa, T., Heijde, M., Rogato, A., Finazzi, G., Todo, T., Bowler, C. and Falciatore, A. (2009) Diatom PtCPF1 is a new cryptochrome/photolyase family member with DNA repair and transcription regulation activity. EMBO Reports 10, 655–661. [1.13] Cohn, S.A., Warnick, L. and Timmerman, B. (2020) Photophobic responses of diatoms Motility and inter-species modulation. In: Diatom Gliding Motility [Volume 2 in the series: Diatoms: Biology & Applications, series editors: Richard Gordon & Joseph Seckbach]. S.A. Cohn, K.M. Manoylov and R. Gordon, (eds.) Wiley-Scrivener, Beverly, MA, USA: This volume. [1.14] Domozych, C.R., Plante, K., Blais, P., Paliulis, L. and Domozych, D.S. (1993) Mucilage processing and secretion in the green alga Closterium. I. Cytology and biochemistry. Journal of Phycology 29(5), 650-659.

xxxiv  Preface [1.15] Domozych, D.S. and Domozych, C.R. (1993) Mucilage processing and secretion in the green alga Closterium. II. Ultrastructure and immunocytochemistry. Journal of Phycology 29(5), 659-667. [1.16] Du Clos, K.T., Karp-Boss, L., Villareal, T.A. and Gemmell, B.J. (2019) Coscinodiscus wailesii mutes unsteady sinking in dark conditions. Biol Lett 15(3), 20180816. [1.17] Fricke, A., Kihara, T.C., Kopprio, G.A. and Hoppenrath, M. (2017) Anthropogenically driven habitat formation by a tube dwelling diatom on the Northern Patagonian Atlantic coast. Ecological Indicators 77, 8-13. [1.18] Furey, P. (2020) Motility in the diatom genus Eunotia Ehr. In: Diatom Gliding Motility [DIGM, Volume 2 in the series: Diatoms: Biology & Applications, series editors: Richard Gordon & Joseph Seckbach] S.A. Cohn, K.M. Manoylov and R. Gordon, (eds.) Wiley-Scrivener, Beverly, MA, USA: This volume. [1.19] Gillard, J., Frenkel, J., Devos, V., Sabbe, K., Paul, C., Rempt, M., Inze, D., Pohnert, G., Vuylsteke, M. and Vyverman, W. (2013) Metabolomics enables the structure elucidation of a diatom sex pheromone. Angewandte Chemie-International Edition 52(3), 854-857. [1.20] Gordon, R. (2020) The whimsical history of proposed motors for diatom motility. In: Diatom Gliding Motility [DIGM, Volume 2 in the series: Diatoms: Biology & Applications, series editors: Richard Gordon & Joseph Seckbach]. S.A. Cohn, K.M. Manoylov and R. Gordon, (eds.) Wiley-Scrivener, Beverly, MA, USA: This volume. [1.21] Gordon, R., Losic, D., Tiffany, M.A., Nagy, S.S. and Sterrenburg, F.A.S. (2009) The Glass Menagerie: Diatoms for novel applications in nanotechnology. Trends in Biotechnology 27(2), 116-127. [1.22] Happey-Wood, C.M. and Jones, P. (1988) Rhythms of vertical migration and motility in intertidal benthic diatoms with particular reference to Pleurosigma angulatum. Diatom Research 3(1), 83-93. [1.23] Harbich, T. (2020) Some observations of movements of pennate diatoms in cultures and their possible interpretation. In: Diatom Gliding Motility [DIGM, Volume 2 in the series: Diatoms: Biology & Applications, series editors: Richard Gordon & Joseph Seckbach]. S.A. Cohn, K.M. Manoylov and R. Gordon, (eds.) Wiley-Scrivener, Beverly, MA, USA: This volume. [1.24] Harper, M.A. (1977) Movements. In: The Biology of Diatoms. D. Werner, (ed.) Blackwell Scientific Publications, Oxford, UK: 224-249. [1.25] Higashi-Fujime, S., Ishikawa, R., Iwasawa, H., Kagami, O., Kurimoto, E., Kohama, K. and Hozumi, T. (1995) The fastest actin-based motor protein from the green-algae, Chara, and its distinct mode of interaction with actin. FEBS Letters 375(1-2), 151-154. [1.26] Higgins, M.J., Molino, P., Mulvaney, P. and Wetherbee, R. (2003) The structure and nanomechanical properties of the adhesive mucilage that mediates diatom-substratum adhesion and motility. Journal of Phycology 39(6), 1181-1193. [1.27] Hoagland, K.D., Rosowski, J.R., Gretz, M.R. and Roemer, S.C. (1993) Diatom extracellular polymeric substances: function, fine structure, chemistry, and physiology. Journal of Phycology 29(5), 537-566. [1.28] Houpt, P.M. (1994) Marine tube-dwelling diatoms and their occurrence In the Netherlands. Netherlands Journal of Aquatic Ecology 28(1), 77-84. [1.29] Ishikawa, M., Takahashi, F., Nozaki, H., Nagasato, C., Motomura, T. and Kataoka, H. (2009) Distribution and phylogeny of the blue light receptors aureochromes in eukaryotes. Planta 230, 543–552. [1.30] Iwasa, K. and Shimizu, A. (1972) Motility of the diatom, Phaeodactylum tricornutum. Experimental Cell Research 74(2), 552-558. [1.31] Jaubert, M., Bouly, J.-P., d’Alcalà, M. R. and Falciatore, A. (2017) Light sensing and responses in marine microalgae. Current Opinions Plant Biology 37, 70–77.

Preface  xxxv [1.32] Kapinga, M.R.M. (1989) Observations on the Growth and Motile Behavior of the Colonial Diatom Bacillaria paradoxa in Culture [M.Sc. Thesis, Supervisor: R. Gordon]. University of Manitoba, Winnipeg. [1.33] Kapinga, M.R.M. and Gordon, R. (1992) Cell motility rhythms in Bacillaria paxillifer. Diatom Research 7(2), 221-225. [1.34] Krishnamurthy, D., Li, H., du Rey, F.B., Cambournac, P., Larson, A. and Prakash, M. (2019) Scale-free Vertical Tracking Microscopy: Towards Bridging Scales in Biological Oceanography. https://www.biorxiv.org/content/10.1101/610246v1.full [1.35] Kroth, P. G., Wilhelm, C. and Kottke, T. (2017) An update on aureochromes: Phylogeny – mechanism – function. Journal of Plant Physiology 217, 20–26. [1.36] Lewin, J. and Hruby, T. (1973) Blooms of surf-zone diatoms along the coast of the Olympic Peninsula, Washington: II. A diel periodicity in buoyancy shown by the surf-zone diatom species, Chaetoceros armatum T. West. Estuarine and Coastal Marine Science 1, 101-105. [1.37] Lin, C. and Todo, T. (2005) The cryptochromes. Genome Biology (online) 6, 220. https://doi. org/10.1186/gb-2005-6-5-220. [1.38] Marques da Silva, J., Duarte, B. and Utkin, A.B. (2020) Travelling expenses: The energy cost of diel vertical migrations of epipelic microphytobenthos. Frontiers in Marine Science 7(433). [1.39] Moeys, S., Frenkel, J., Lembke, C., Gillard, J.T.F., Devos, V., Van den Berge, K., Bouillon, B., Huysman, M.J.J., De Decker, S., Scharf, J., Bones, A., Brembu, T., Winge, P., Sabbe, K., Vuylsteke, M., Clement, L., De Veylder, L., Pohnert, G. and Vyverman, W. (2016) A sex-inducing pheromone triggers cell cycle arrest and mate attraction in the diatom Seminavis robusta. Scientific Reports 6, #19252. [1.40] Paterson, D.M. and Hope, J.A. (2020) Diatom biofilms: Ecosystem engineering and niche construction. In: Diatom Gliding Motility [DIGM, Volume 2 in the series: Diatoms: Biology & Applications, series editors: Richard Gordon & Joseph Seckbach] S.A. Cohn, K.M. Manoylov and R. Gordon, (eds.) Wiley-Scrivener, Beverly, MA, USA. [1.41] Poulsen, N.C., Spector, I., Spurck, T.P., Schultz, T.F. and Wetherbee, R. (1999) Diatom gliding is the result of an actin-myosin motility system. Cell Motility and the Cytoskeleton 44(1), 23-33. [1.42] Roubeix, V. and Laviale, M. (2020) A free ride: Diatoms epiphytic on motile diatoms. In: Diatom Gliding Motility [DIGM, Volume 2 in the series: Diatoms: Biology & Applications, series editors: Richard Gordon & Joseph Seckbach]. S.A. Cohn, K.M. Manoylov and R. Gordon, (eds.) Wiley-Scrivener, Beverly, MA, USA: This volume. [1.43] Sabuncu, A.C., Gordon, R., Richer, E., Manoylov, K.M. and Beskok, A. (2020) The kinematics of explosively jerky diatom motility: A natural example of active nanofluidics. In: Diatom Gliding Motility [Volume 2 in the series: Diatoms: Biology & Applications, series editors: Richard Gordon & Joseph Seckbach]. S.A. Cohn, K.M. Manoylov and R. Gordon, (eds.) Wiley-Scrivener, Beverly, MA, USA: This volume. [1.44] Serôdio, J., Paterson, D.M. and Hope, J.A. (2020) Diatom motility: Mechanisms control an adaptive value. In: Diatom Gliding Motility [DIGM, Volume 2 in the series: Diatoms: Biology & Applications, series editors: Richard Gordon & Joseph Seckbach] S.A. Cohn, K.M. Manoylov and R. Gordon, (eds.) Wiley-Scrivener, Beverly, MA, USA: This volume. [1.45] Shih, S.M., Engel, B.D., Kocabas, F., Bilyard, T., Gennerich, A., Marshall, W.F. and Yildiz, A. (2013) Intraflagellar transport drives flagellar surface motility. eLife 2, e00744-e00744. [1.46] Tominaga, M., Kimura, A., Yokota, E., Haraguchi, T., Shimmen, T., Yamamoto, K., Nakano, A. and Ito, K. (2013) Cytoplasmic streaming velocity as a plant size determinant. Developmental Cell 27(3), 345-352. [1.47] Wang, J., Weng, D., Chen, L. and Cao, S. (2020) Locomotion of benthic pennate diatoms. In: Diatom Gliding Motility [DIGM, Volume 2 in the series: Diatoms: Biology & Applications,

xxxvi  Preface series editors: Richard Gordon & Joseph Seckbach] S.A. Cohn, K.M. Manoylov and R. Gordon, (eds.) Wiley-Scrivener, Beverly, MA, USA: This volume. [1.48] Williams, R.B. (1965) Unusual motility of tube-dwelling pennate diatoms. Journal of Phycology 1(4), 145-146. [1.49] Wolgemuth, C., Hoiczyk, E., Kaiser, D. and Oster, G. (2002) How myxobacteria glide. Curr Biol 12(5), 369-377. [1.50] Zeriouh, O., Reinoso-Moreno, J.V., Lopez-Rosales, L., Ceron-Garcia, M.D., Sanchez-Miron, A., Garcia-Camacho, F. and Molina-Grima, E. (2017) Biofouling in photobioreactors for marine microalgae. Critical Reviews in Biotechnology 37(8), 1006-1023. [1.51] Zischka, F., Kratochvil, H., Noll, A., Gordon, R., Harbich, T. and Gebeshuber, I.C. (2020) Diatom triboacoustics. In: Diatom Gliding Motility [DIGM, Volume 2 in the series: Diatoms: Biology & Applications, series editors: Richard Gordon & Joseph Seckbach] S.A. Cohn, K.M. Manoylov and R. Gordon, (eds.) Wiley-Scrivener, Beverly, MA, USA: This volume.

1 Some Observations of Movements of Pennate Diatoms in Cultures and Their Possible Interpretation Thomas Harbich

*

Independent Researcher, Am Brüdenrain, Weissach im Tal, Germany

Abstract

This chapter presents observations on the motility of pennate diatoms with interpretations and hypotheses. They refer to movement in different environments, on a solid substrate, in a biofilm, on the water surface and in the context of colonies. The trajectories of certain species on solid substrate are analyzed including the orientation of the apical axis and two hypotheses are made which establish a relationship between motion and the location of the contact point on the diatom to the substrate. The obtained location is validated by observing the movement with a viewing direction parallel to the substrate. Using this method, the author also investigates the change in location of the contact point during reversal of motion. It is furthermore shown that the analysis of the path curvature of certain species allows conclusions to be drawn about the location of the propulsion. Observations of the movement of Pinnularia viridiformis in and on a biofilm confirm its viscoelasticity. It is shown that elastic deformations of the biofilm, made visible by marking with particles, reveal details of the activity of the raphe branches. Investigations were carried out on cultivated diatoms which are found on the surface of the water. There, they show movements that partly deviate considerably from the movement patterns on substrate. In Nitzschia sigmoidea, hydrophobicity is assumed to be responsible for buoyancy and its role in structure formation is examined. Finally, the movement patterns in connection with colony formation in Cymbella lanceolata are studied. Qualitative and quantitative methods are used to show how typical colonies develop on flat substrates due to changing movement activity during the day-night cycle and to the influence of the adhesion force of the gelatinous excretions. As all observations are based on diatom cultures in a laboratory environment, the possibilities of occurrence of the observed phenomena in nature are discussed in the context of the respective observations. Keywords:  Movements, kinematics, contact point, biofilm, floatability, hydrophobicity, pattern formation, colony formation

Email: [email protected] Thomas Harbich: https://www.researchgate.net/profile/Thomas_Harbich, https://diatoms.de/en/ Stanley Cohn, Kalina Manoylov and Richard Gordon (eds.) Diatom Gliding Motility, (1–32) © 2021 Scrivener Publishing LLC

1

2  Diatom Gliding Motility

1.1 Introduction Diatoms form an ecologically important class of single-celled photosynthetic algae found in freshwater and seawater. They are characterized by a solid skeleton (frustule) consisting of amorphous hydrated silicon dioxide and organic substance. It is composed of two halves, the epitheca and the hypotheca. The structure is similar to a Petri dish, with the epitheca overlapping the hypotheca. Each theca consists of a more or less arched valve and the cingulum, a number of associated siliceous bands (girdle bands). According to their shape, diatoms are divided into centric diatoms that are radially symmetric and pennate diatoms that are bilaterally symmetric. Among the pennate diatoms there are many species which are able to glide over a substratum. An essential morphological feature of these motile diatoms is one or two distinct slits in the valve called raphe (Figure 1.1). Through the raphe, strands of mucilage are secreted, which enable the cells to adhere to the substrate and are involved in motility. The various theories on motility will not be discussed here. Reference is made to the paper by Edgar and Pickett-Heaps [1.12] as well as to a short compilation by Häder and Hoiczyk [1.17]. The paths of pennate diatoms having a raphe system on a solid smooth substrate like a microscopic slide are often almost circular, spiral, straight and occasionally sigmoid. These forms of movement are related to the form of the raphe system (Nultsch [1.25], Cohn [1.7], overview in Round et al. [1.31]). The gliding movement is interrupted by sudden back and forth movement and in many species also by complex motion sequences such as horizontal rotation around a point of the cell, vertical pivoting or pirouettes. A systematic survey of the movement patterns of 135 raphid diatom species from 35 genera was given by Bertrand [1.2] in 1992. Movement can be advantageous for diatoms in many ways. Some diatoms, which inhabit sand deposits in intertidal zones, show a vertical migration [1.14]. As a consequence of wind, tides and currents, sediment is being constantly deposited, so that the diatoms have to migrate to the top of the sediment continuously to photosynthesize (review article in [1.20]). Diatoms use phototaxis for this purpose [1.19] and allegedly geotaxis [1.29] [1.30]. In addition, under certain lighting conditions diatoms show a phobotaxis, a kind of shock reaction to local light exposure with a high intensity gradient [1.27]. Diatoms reverse when they are partially illuminated, for example, when crossing a light-dark boundary. There were early indications of chemotaxis [1.25]. Recent studies show a movement towards a silicate source [1.5] and a pheromone-directed movement [1.15] [1.6]. Pennate diatoms, Raphe

Epivalve

Epitheca Frustule Hypotheca

Girdle bands

Figure 1.1  Drawing of a pennate diatom with two raphe branches on its valve.

Hypovalve

Movements of Pennate Diatoms in Cultures  3 possessing a raphe system, also show motility under isotropic conditions, on which we will focus here. Influences by inhomogeneous illumination, chemical gradients, thermal gradients, flow, unevenness or other anisotropy of the substrate, such as inclination to the horizontal, cannot be completely excluded in the experiment, but were avoided where possible. Such controlled conditions can also be achieved over a longer observation period by the use of diatoms cultures. All observations described below were performed either directly using batch cultures in Petri dishes or after transfer of the diatoms from the culture into a suitable test vessel. The analysis of the trajectories of diatoms has long been of qualitative nature. This has changed with the use of video technology, manual or automatic tracking of motion and the use of computers to analyze movement data. Edgar [1.10] determined speeds and accelerations of diatoms and the speed of particles transported along the dorsal raphe of Nitzschia sigmoidea. A statistical analysis of the trajectory of Navicula sp. was published by Murase et al. [1.23] whereby the movement of the diatoms was confined by a micro chamber. Murguía et al. [1.24] performed a time series analysis of the Hurst exponent. The above-mentioned experiments on chemotaxis [1.5] [1.6] are based on the statistical evaluation of frames of video recordings.

1.2 Kinematics and Analysis of Trajectories in Pennate Diatoms with Almost Straight Raphe along the Apical Axis In the following, pennate diatoms will be considered whose raphe system is almost straight and is located centrally between the apices. Furthermore, it is assumed that the valve has a convex surface, so that the diatom contacts a flat smooth substrate with only one point P of its valve. When the diatom moves on such a substrate, it can therefore be assumed that the driving force acts at the contact point P or in its immediate vicinity. Observations on diatoms of the genera Navicula, Craticula or Stauroneis show at least slightly curved and often circular or spiral paths in valve view. The radii of the trajectories in these genera are large compared to the size of the diatoms. Particularly in smaller species, the orientation of the apical axis shows random fluctuations around the direction of movement, which are visually perceived as “wagging.” The question arises whether and in what way the position of P has an influence on this movement and what consequences this has for the analytical description of the movement. If there is such an influence, then changes in the position of P are also relevant. Two hypotheses are formulated which connect the position of point P with the movement. The first hypothesis states that the alignment of the diatom is tangential to the trajectory of the diatom at the point of contact P, except for random fluctuations. It is based on the assumption that at the contact point the direction of the propulsion force lies in the direction of the raphe with only minor deviations. In Figure 1.2 the hypothesis is visualized with a point P near the leading apex. Since P is not located in the center of the cell, the radii of the trajectories of the apices A1 and A2 differ. An analysis of the movement of the diatoms from a view perpendicular to the valves allows answering the question whether there is a point B on the connecting line of the apices, with the property that the apical axis lies tangentially to its trajectory on statistical average. A match between the kinematic point B and the contact point cannot be proven in this way.

4  Diatom Gliding Motility A2 A1

P

Figure 1.2  Hypothesis that there is a point P between apices A1 and A2, so that the apical axis is tangential to the trajectory of P.

Figure 1.3  Traces of two trackers attached close to the apices of a diatom of Navicula sp.

To investigate the hypothesis, individual diatoms from a culture were transferred to a Petri dish of polystyrene and their movement was recorded under an inverted microscope. Using the Video Spot Tracker1 software tool, movement data can be obtained from the video files, which include the orientation of the diatom. For this purpose, two circular trackers can be attached to the apices of the diatom in the first analyzed frame. Sometimes the use of a rectangular tracker proves to be more suitable. The tracker coordinates determined for each frame are imported into a spreadsheet program for analysis, using scripts for complex evaluations. Figure 1.3 shows a diatom of the genus Navicula with two trackers and their trajectories. As in Figure 1.2, the curve of the tracker trailing behind has the larger radius. Fluctuations †

1 †

I like to thank Computer Integrated Systems for Microscopy and Manipulation (CISMM) at UNC Chapel Hill, supported by the NIH NIBIB (NIH 5-P41-RR02170) for the use of Video Spot Tracker free of charge.

Movements of Pennate Diatoms in Cultures  5 are clearly visible, mainly due to random changes in the orientation of the apical axis. Because of these fluctuations, long paths between reversal points were preferred for analysis. The numerically determined trajectory of a hypothetical point x between the trackers is smoothed by means of a finite impulse response (FIR) filter (low pass). Then the angle between the tangent in x with the apical axis can be determined. The root-mean-square deviation (RMSD) of these angles over all frames of the included video sequence serves as a measure for the deviation from meeting the tangential condition. When x is varied, the curve shown in Figure 1.4 is obtained in this example. The positions of the trackers are used as a reference for the position of x. The value x = 0 on the abscissa corresponds to a match of P with the tracker trailing behind, the value x = 1 corresponds to the leading tracker and the value x = 0.5 represents the center of the diatom. In accordance with the hypothesis, the variance has a minimum, which is about 0.65 in this case. As the trackers do not sit exactly on the apices, but are slightly shifted inwards, a correction must be made to determine the position of the minimum with respect to the apices. On the normalized line segment A₁A₂, the minimum and thus the position of B is at x = 0.62. B is located on the side of the leading apex, as shown in Figure 1.2. To illustrate the correctness of the evaluation, the frequencies of the angular deviation between the smoothed trajectory in B and the apical axis were calculated (Figure 1.5). According to the hypothesis, the density function is in good approximation symmetrical to the origin. The standard deviation amounts to 4.15 degrees. The minima of the RMSD for 10 analyses of the same Navicula sp. were between 0.58 and 0.77, thus yielding similar values. In contrast, in Craticula cuspidata the minimum RMSD is typically at around 0.2, so that B is close to the trailing apex. To clarify the question of whether the determined point B actually corresponds to the contact point P, the diatoms were viewed from an angle that is almost parallel to the substrate. For observation, a coverslip can be placed almost vertically in a Petri dish with diatoms and examined with an inverse microscope. If there are enough diatoms in the Petri

Root mean square deviation (degree)

9.0°

8.0°

7.0°

6.0°

5.0°

4.0°

3.0° 0.5

0.55

0.6

0.65

0.7

0.75

0.8

0.85

0.9

x

Figure 1.4  Root-mean-square deviation of the angle between the apical axis and the smoothed trajectory of the point x located between the trackers.

6  Diatom Gliding Motility

Frequencies of angular deviations

0.12

-35.0°

0.08

0.04

-25.0°

-15.0°

-5.0°

5.0°

15.0°

25.0°

35.0°

Figure 1.5  Histogram of the frequencies of the angular difference between the direction of the diatom (apical axis) and the smoothed curve in P.

Figure 1.6  Craticula cuspidata observed from an almost horizontal view.

dish, after some time diatoms will have migrated onto the coverslip. Otherwise, the coverslip is first laid flat on the bottom of the Petri dish, diatoms are placed on its surface and then carefully tilted vertically. Figure 1.6 shows a diatom of the species Craticula cuspidata and its mirror image from an almost horizontal view. On longer sections the diatom moves to the right at this inclination. The inclination of the apical axis to the substrate in this image is about 7.5°. However, a considerable fluctuation of the inclination occurs, so that the contact point P also changes. According to the hypothesis, one finds that the numerically determined value of B is within the range of the variation of P. Due to the fluctuations of P, the position of the minimum of RMSD yields just an average value for P. The second hypothesis exclusively concerns the fluctuations of the apical axis around the direction of motion. It says that this is a rotation about the point P (Figure 1.7). From the viewing direction perpendicular to the substrate, only the existence of a pivot point T can be concluded, although a match of a pivot point with the contact point seems to be obvious.

Movements of Pennate Diatoms in Cultures  7

P

Figure 1.7  Hypothesis that there is a point P between apices A1 and A2, so that the diatom performs stochastic rotary motions around P.

0.90

Root mean square deviation (µn/s)

0.85 0.80 0.75 0.70 0.65 0.60 0.55 0.50 0.45 0.40 0.5

0.55

0.6

0.65

0.7

0.75

0.8

0.85

0.9

x

Figure 1.8  Root-mean-square deviation of the transverse component of the fluctuations of the hypothetical pivot point.

In order to determine the pivot point, a smoothed trajectory is calculated for a hypothetical point x between the apices and the motion of x is split into a component in the direction of the smoothed curve and a perpendicular, i.e., transversal, direction. The transversal component disappears in the ideal case when x coincides with the pivot point. Due to the random fluctuations of T this is not exactly the case. For the trajectory shown in Figure 1.3, the RMSD of the transverse fluctuations against x is plotted (Figure 1.8). In accordance with the hypothesis, a distinct minimum is apparent, whose position is in good agreement with the obtained value of B. The two approximation methods for determining the contact point evaluate different information and differ in their applicability. The algorithm described first is based on the orientation of the tangents and cannot be used for straight sections. On the other hand, the method of using the rotation does not make any assumptions about the shape of the paths. However, it requires a sufficiently well observable fluctuation of the orientation of the diatom. None of the methods is suitable for almost straight paths without significant

8  Diatom Gliding Motility directional fluctuations. In addition, the statistical methods require sufficiently long path sections between the reversal points. In principle, both numerical methods could be generalized to more universal forms of raphes. When determining P, only hypothetical points lying on a numerically modeled raphe are to be used. In all observed species of the genus Navicula the point P was closer to the leading than to the trailing apex. Within the proposed interpretation, the driving force is acting in the leading half of the diatom. To put it simply, the diatom is pulled. In contrast, in the case of Stauroneis sp. and Craticula cuspidata, point P is near the trailing apex. These diatoms are pushed. In some species there is no clear statement regarding P, which could be due to frequently changing positions of the contact point. From the observations it cannot be concluded in which respect the different positions of the propulsion have significance for a species and which advantages or disadvantages they possess. When the diatom stops moving and then reverses, the point P in the following section of the path is located again on the same side in relation to the direction of movement. With respect to the cell, it changes to a point on the other branch of the raphe, which lies approximately symmetrically to the center. There are two alternatives for changing the position of P when reversing direction: Either the diatom tilts around the transapical axis at the reversal point so that the opposite raphe branch comes into contact with the substrate. Then the reversal of direction would be the result of an opposite activity of the raphe branches. Or the direction of activity of the driving raphe changes first, so that the direction is reversed at an identical contact point. Then the contact point should shift after the reversal. These alternatives are shown in Figure 1.9 for a diatom of the species Craticula cuspidata, which touches the substrate on longer paths at a point near the trailing apex. The observation of Craticula cuspidata from a horizontal point of view reveals that at reversal points a change of direction occurs first. Within a few valve lengths the apical axis then tilts to the other raphe branch. It obviously has the same direction of activity; otherwise there would be another reversal of direction, this time according to the alternative scheme. It is conceivable that the rapid back and forth jerking observed in some diatom species is caused by such an alternation of driving raphes having opposite activity. When viewed from a horizontal perspective, it can be seen that the transapical axis is usually inclined against the substrate (Figure 1.10). In this case, the raphe is located at the

(a)

(b)

Figure 1.9  The left side (a) illustrates the sequence of steps for reversal of direction, in which the tilting takes place after the direction of motion has been changed. In alternative (b), tilting takes place before reversing the direction.

Movements of Pennate Diatoms in Cultures  9

Figure 1.10  Craticula cuspidata viewed from a horizontal perspective. The transapical axis is inclined against the substrate.

edge of the contact area. This causes an asymmetric friction with respect to the raphe, with adherent extracellular polymeric substances (EPS) being relevant. The often surprisingly small radii of curvature of the trajectories compared to the curvature of raphe could be a consequence of this asymmetry. In the natural habitat of diatoms there are usually no comparably flat substrates. Trajectories of Craticula cuspidata on a stone from the habitat of diatoms exhibit to a certain extent the typical orbital curvatures, but also disturbances in the direction of motion and frequent reversals, so that they often showed more of the appearance of a random walk. It is to be expected that a substrate of high roughness compared to the size of the diatom will lead to changing and even to simultaneous contact points in all motile diatom species. In the case of large irregularities, the methods presented for determining P fail. The relevance of observing trajectories under laboratory conditions may be considerable, but conclusions about the natural environment are limited. After observing the circular random motion of Nitzschia communis, Gutiérrez-Medina et al. [1.16] came to the conclusion that this motion is not optimized for long distances but for covering a limited area. The authors assume that the biological role is the efficient formation of a biofilm. This conclusion is based on comparable isotropic surfaces in nature. In the case in which the trajectories of the diatoms are severely disturbed in their habitat, a movement in the form of a typical random walk would allow bridging larger distances.

1.3 Curvature of the Trajectory at the Reversal Points Additional information on the relationship between the shape of the raphe and the propulsion can be obtained from the behavior of the curvature of the trajectory during reversal of motion. For certain shapes of the raphe, it is possible to derive plausible statements about the location of propulsion from the curvature of the path. In many cases at least the assignment to the driving raphe branch is feasible. The absolute value of the curvature will not be the focus of the discussion, but the direction or sign of the curvature. The curvature is defined as positive if the diatom moves counterclockwise from the observer’s point of view. If it moves clockwise, the curvature is negative. An important prerequisite for the following considerations therefore is that the trajectories show clearly recognizable curved and smooth path sections, which implies sufficiently long paths. Furthermore, only movements

10  Diatom Gliding Motility in which the raphe is in direct contact with the substrate are regarded. We limit ourselves to uncomplicated changes of direction in which the diatoms do not rotate around the apical axis or pervalvarous axis and do not straighten up [1.2]. Many motile diatoms meet these requirements over a sufficiently long observation time. Diatoms whose valves are symmetrical with respect to the transapical plane shall be considered, such as diatoms of the genera Navicula or Rhopalodia, but not sigmoid forms. Figure 1.11 shows the movement of such a diatom with reversal points. The curvature of the raphe is intentionally exaggerated compared to the curvature of the trajectory. The division into two raphe branches shown in the picture and the possibly changing position of point P does not play a significant role here. Furthermore, it is irrelevant whether P is on the side of the leading or trailing apex. But it is essential that the position of P changes only between points whose centers of curvature are on the same side of the diatom. If one assumes that the curvature of the path follows the curvature of the raphe or that a constant lateral oblique position determines the path curvature due to friction effects, then the direction of curvature always changes at the reversal points. This behavior can be well observed in Navicula sp. (Figure 1.12). The behavior becomes more complex when regions of the raphe with different directions of curvature can act as the center of the propulsion. A Cymbella (Figure 1.13) does not meet the requirement of a simple numerical analysis for the determination of P. According to the predominant curvature of the raphe, it is reasonable to assume that the diatom would move on curved paths in such a way that the center of curvature is located on the ventral side. This can be observed in many Cymbella spp. and in these cases the sign of curvature alternates at the reversal points. In Cymbella cistula and probably other Cymbella spp. the center of curvature is surprisingly often on the dorsal side. A change of the center of curvature between dorsal and ventral often occurs at the reversal points, but also during an uninterrupted movement (point of inflection). Here, both alternating and identical directions of curvature

P

P

P

Figure 1.11  Trajectory of a diatom with reversal points. The point P never changes to a place of the raphe with opposite curvature.

Movements of Pennate Diatoms in Cultures  11

Figure 1.12  Path of a diatom of the genus Navicula with reversal points. The direction of curvature changes at each reversal point.

dorsal

ventral

Figure 1.13  Outline and raphe of a Cymbella.

exist. When looking at the path of the leading apex at the dorsal center of curvature, the alignment of the diatom is approximately tangential to it (Figure 1.14). There seems to be a point near the apex from which the diatom is drawn. A dorsal center of curvature corresponds to the direction of curvature of the helictoglossa, so that the propulsion should originate from it. Often the movement at the dorsal center of curvature changes into a rotation around the apex in the same direction of rotation (Figure 1.15). A movement in which the diatom is pushed from the distal end of the raphe could not be observed, but occasionally a reversal of the direction of rotation. As a last example, the behavior at direction reversal of Surirella biseriata will be presented. Figure 1.16 shows that two raphe sections symmetrical to the apical axis can cause the movement when the diatom lies on its valve. Depending on the lateral inclination, one or the other raphe couples to a flat substrate. Accordingly, paths with a center of curvature occur on one side or the other of the diatom. Occasionally, the coupling can change between the sides, leading to sigmoid trajectories. At the reversal points, a conservation of the direction of curvature is most frequently observed. In this case the driving side of the

12  Diatom Gliding Motility

Figure 1.14  Overlay of four images from a video showing a trajectory of a diatom of the species Cymbella cistula. The yellow line shows the trajectory of the leading apex. The line segment between the apices is marked in white.

(a)

(b)

Figure 1.15  On the left (a) the superimposition of images of a Cymbella rotating around a point near the helictoglossa is shown, on the right (b) a sketch of the diatom with raphe.

100 µm

Figure 1.16  Surirella biseriata in valve view.

Movements of Pennate Diatoms in Cultures  13

Figure 1.17  Trajectory of a Surirella biseriata. The driving side changes at the reversal points.

Figure 1.18  Paths of Surirella biseriata in a culture. They were visualized by overlaying frames of a video.

diatom changes. Figure 1.17 shows the principle and Figure 1.18 an image that was created by superimposing frames from a video. The direction of movement only rarely reverses with the driving side being retained, which leads to a changing direction of curvature.

1.4 Movement of Diatoms in and on Biofilms Diatoms often live in biofilms and contribute to their formation by secreting EPS. Such biofilms, consisting of numerous cell types enmeshed within a secreted polysaccharide matrix

14  Diatom Gliding Motility [1.9], often form their own complex ecosystems and are important contributors to biofouling [1.22] [1.1]. While most descriptions of diatom motility have referred to movement on solid substrates, Consalvey et al. [1.8] suggest that diatoms move in a biofilm during the vertical migration in the mudflats. The appearance of biofilms was observed in several batch cultures. In contrast to biofilms in nature, these biofilms are only colonized by one diatom species and bacteria (no axenic cultures). Such biofilms are not generally suitable for making statements about biocenosis. They allow, however, studying the possible movements of pennate diatoms in such a film. To accomplish this, observations were made on cultures in Petri dishes in which the focal plane was chosen between the plane of the substrate and the surface of the biofilm. With the focal plane fixed, time-lapse recordings were made and analyzed. A biofilm of Pinnularia viridiformis (Figure 1.19) adhering stably to the substrate was used as a model system to investigate the movement. In the 30-day-old culture the thickness of the biofilm ranged from about 150 µm to 380 µm. Diatoms were found directly on the substrate, on the surface of the biofilm and occasionally in between (Figure 1.20). In most cases, the surface of the biofilm was more densely populated than the substrate. On both surfaces, the diatoms showed motility, whereby the diatoms, which are in contact with the substrate, have a significantly lower speed, often move jerkily, perform pivoting movements and frequently stand vertically, which they do not do in free water. On the substrate, the diatoms lie predominantly on their girdle bands. On the surface of the biofilm you can see the diatoms in valve view and girdle band view and you can find the movement patterns that are typical for motion on a substrate, but the movement occurs somewhat more erratically and slower. In girdle band view there are only back and forth movements [1.18]. Lumps of trail substance enable the transmission of force from the raphe to the biofilm. The relatively rare diatoms between the bottom of the Petri dish and the surface of the film often rotate around changing axes.

50 µm

Figure 1.19  Pinnularia viridiformis with a length of approx. 90 µm.

water body

biofilm substratum

Figure 1.20  Places within and on a biofilm where Pinnularia viridiformis can be found. The typical movement patterns are indicated by arrows. Shunting movements are marked with short arrows at both apices.

Movements of Pennate Diatoms in Cultures  15 The biofilm exhibits viscoelasticity, i.e., it is neither purely viscous nor purely elastic, which is typical for biofilms [1.13]. The viscous property of the gel allows the movement of the diatoms at the bottom of the Petri dish and within the biofilm. Because of the elasticity the diatoms are able to stay on the surface without sinking to the bottom of the Petri dish. From the observations, no statement can be made as to whether diatoms from the interior of the biofilm can reach its surface. At slight vibrations of the culture, the gel gets into a horizontal oscillation, which can be recognized by the co-swinging diatoms. A description of the movement of Pinnularia within the biofilm must take into account the viscoelastic properties, because elastic forces occur which do not exist in Newtonian, purely viscous liquids. To describe movement in this environment, Edgar’s [1.11] analysis of movement in biofilms would have to be extended. These observations do not provide any information about the structure of the biofilm. Lauterborn [1.21] already reported in 1896 that Pinnularia maior extrudes gelatinous filaments during movement. The elasticity of these structures, which often contract into lumps, can easily be seen in the ink coloring used by Lauterborn. Presumably these gelatinous formations stick together to form a biofilm that is smooth at the top. The surface of the transparent biofilm is not visible under the light microscope. When small particles are placed from above, they remain on the gel and mark the surface of the biofilm to the water. This can be used to observe the activity of the raphes, because due to the small modulus of elasticity of the film, the diatoms cause easily observable elastic deformations. Mineral particles with a linear expansion of typically less than 3 µm were distributed on the gel layer. Diatoms with uniform gliding motion do not cause local tensions on the surface of the gel. Frequently, however, the movement of the diatoms stops and it can be seen that the gel around the diatom is compressed (negative strain) towards the proximal raphe ends. As long as the opposite activity of the raphe continues, this elastic deformation in the direction of the apical axis persists. Figure 1.21 shows a superimposition of the frames of a video during the standstill of the diatom. The deformation has a visible effect at a distance of several diatom lengths. After a while the raphes no longer work against each other and the movement is continued in the same or opposite direction. The biofilm returns

Figure 1.21  Superimposed frames of a video during the standstill of a diatom. A tension has built up in the biofilm.

16  Diatom Gliding Motility to its equilibrium position within a short time, which is noticeable in the movement of the gel surface in some distance from the diatom. An elastic elongation (tensile strain) resulting from simultaneous activity towards the distal raphe ends is not observed. The question remains whether the opposing activity of the raphe branches provides an advantage or whether there is a lack of coordination. The investigation of the elastic deformations of the biofilm by marking can therefore serve the qualitative study of the relationship between movement and activity of the raphes, since both can be observed simultaneously. The method provides the possibility of quantitatively determining the forces exerted on the substrate by the raphe. For this purpose the elasticity module and the displacement of the particles have to be determined. Furthermore, a model is to be developed which relates the deformations on the surface to the forces exerted. These forces correspond to the driving force of the diatom. If there is no biofilm, it can be replaced by a synthetic elastic layer. This method complements the measurement of the force between the diatom and the substrate by deflection of a glass fiber introduced by Harper and Harper [1.18].

1.5 Movement on the Water Surface When examining drops hanging from a cover glass, the diatoms contained sink to the water-air interface. They can show movements there even without a solid substrate [1.26] [1.3]. Occasionally, living diatoms also float on the water surface of densely populated cultures. This was relatively common in Nitzschia sigmoidea and much rarer in Cymatopleura solea, as well as in certain species of the genera Cymbella, Rhopalodia and Pinnularia. Since only a small number of diatom species have been cultivated, it can be assumed that this phenomenon occurs in many other genera and species. These questions arise in particular: • How do the diatoms get to the water surface? • Why do the diatoms not sink immediately and how long do the diatoms remain on the water surface? • Can the diatoms survive on the water surface for a long time? • Which movement patterns and interactions between the diatoms occur and how can they be explained? • Do diatoms have an advantage from their ability to float and to move on the water surface? It is anticipated that there will probably be no valid answers for all species, because the differences are considerable, especially with respect to their motility. The best observations to be interpreted are available in Nitzschia sigmoidea. For this reason, these are to be presented first. This is followed by comments on Pinnularia. The cultivated Nitzschia sigmoidea came from various waters, a small reservoir (Aichstruter Stausee), its outflow (Lein) and a pond (in Stuttgart-Hohenheim). In a newly created culture, the phenomenon of floating diatoms is apparent after about 2 to 3 weeks, whereby the first isolated floating diatoms can often be found after only one week. In a fully developed culture, connected structures can form that cover a large part of the water surface. There are cultures in which the number of diatoms on the water surface is far greater

Movements of Pennate Diatoms in Cultures  17 than on the substrate. The diatoms are not located in a flake connected by EPS, which gets buoyancy by oxygen bubbles, something that occasionally occurs in this species at very high population density, but lie horizontally on the water surface. Whereas this species can occasionally be found in valve view when looking vertically at a solid substrate, on the water surface it is always found in girdle band view, apparently the equilibrium position. A possibility to rotate on the water surface around the apical axis does not seem to exist. According to current knowledge, a buoyancy of the diatoms due to a lower specific weight than water can be excluded. The cultures were repeatedly carried a few meters to the microscopes for examination. In many cultures, one could observe that diatoms were whirled up and sedimented slowly. Apparently, even small water flows cause Nitzschia sig­ moidea to detach from the substrate and to accumulate in the water. It can be assumed that diatoms also reach the water surface. As the diatoms obviously do not sink again, but show a remarkable floatability, they accumulate on the water surface. Further observations are required to substantiate this mechanism. In the phase contrast, pronounced brightening is visible on most apices. Significant changes in brightness also occur in differential interference contrast (DIC) or differential interference contrast for plastic receptacles (PlasDIC). Figure 1.22 shows a typical appearance. Under the stereomicroscope you can see with an oblique view that the water surface is arched around the apices. There is a more or less distinct convex meniscus, which explains the appearance in phase contrast, DIC or PlasDIC. The diatom lies deep in the water (Figure 1.23). The ends of the diatom have obviously hydrophobic properties, which lead to the deformation of the water surface and give it buoyancy due to the surface tension of the water, as is known from the water strider. In Figure 1.24 a diatom is sketched from a horizontal view. Wang et al. [1.33] have found that cleaned valves of Coscinodiscus sp. float on water surfaces. Here too, hydrophobicity is the cause. After examination of cleaned valves, the authors conclude that the hydrophobicity is based on the convex form and 40 nm sieve pores. I consider it an open question whether this explanation applies to floating living Nitzschia sigmoidea. Apart from the other structure of the valve, it can be assumed that living diatoms are surrounded by a layer of organic material. This could prevent the influence

Figure 1.22  Nitzschia sigmoidea on the water surface viewed with PlasDIC.

18  Diatom Gliding Motility

Figure 1.23  Nitzschia sigmoidea with a stereomicroscope in oblique view.

Figure 1.24  Sketch of a Nitzschia sigmoidea on the water surface seen from the horizontal direction.

of the pores on one side and on the other side it could have hydrophobic properties itself. In this context, it should be mentioned that the cell lines of Nitzschia sigmoidea lost the ability to float after a few months and never regained it. At first the typical patterns of connected diatoms on the water surface became less regular and finally disappeared completely. The floating diatoms did not always lie in the same plane but frequently crossed, and occasionally they hung with only one end on the surface of the water. Later, the proportion of diatoms on the water surface decreased noticeably. An explanation of the origin of hydrophobia should also clarify this. Nitzschia sigmoidea can survive many days on the water surface. This is probably made possible by the high proportion of wetted surface. The rapid increase of diatoms on the water surface and the sometimes high density of floating diatoms compared to benthic living diatoms suggest that they reproduce asexually on the water surface. There is an attractive interaction between hydrophobic bodies on the water surface. When floating hydrophobic bodies move towards each other due to this force, energy is released into the environment. The system strives for a state of minimal energy. Lycopodium spores scattered on a water surface, for example, bond together and form two-dimensional structures with local order. In this process, restructuring takes place in which contacts are broken up and are closed at other locations until a local minimum of energy and thus a stable equilibrium is reached. The global minimum will only be reached in the case of very few particles. Wang et al. [1.33] report on the formation of regular structures at Coscinodiscus sp. on the surface of a water droplet. This self-­assembled monolayer is a consequence of the short-range attractive interaction between the hydrophobic frustules. In Nitzschia sigmoidea, the attractive force acts at the ends of the diatoms. Figure 1.25 shows two diatoms in the valvar plain with the resulting water surface. The water surface has a lower energy than with two separate diatoms according to Figure 1.24. When several diatoms drift with hydrophobic apices on the water surface, patterns are formed in which the ends preferably stick together. This results in locally star-shaped and polygonal

Movements of Pennate Diatoms in Cultures  19

Figure 1.25  Sketch of two adjacent Nitzschia sigmoidea on the water surface seen from the horizontal direction.

Figure 1.26  Very regular structure of a diatom cluster on the water surface (dark field).

structures. Not all diatoms have a pronounced hydrophobic apex. It is enough to stay on the water surface, but the attractive interaction is barely recognizable in the movement. There are systematic differences in the strength of the hydrophobicity in diatoms from different localities that affect the patterns. A low interaction leads to loose arrangements (Figure 1.22), a strong interaction to patterns where the apices are close together (Figure 1.26). It will probably not be possible to give an analytical representation of the attractive force. A simplified modeling consists in the replacement of the diatom by two rotationally symmetric hydrophobic particles, which are connected by a rod having the length of the diatom. For the experimental determination of the attractive interaction one can examine the motion of diatoms, which move toward each other, or the movement of a hydrophobic particle like spores of Lycopodium in the vicinity of a diatom. Under the plausible assumption that the inertia force can be neglected in the equation of motion, and Stokes’ law holds, the velocity is proportional to the force. In Figure 1.27 the speed of a diatom, which is proportional to the force, is plotted versus the distance between two approaching diatoms. It was determined from a video recorded at 30 fps. This is not sufficient to measure the high speeds shortly before the collision. The fluctuations in velocity at longer distances are probably due to the different drift velocities of the diatoms. This image is primarily intended to illustrate the method. A remarkable aspect is the dynamics of movement. Different forms of movement occur, which indicate different mechanisms. Single floating diatoms are able to move slowly in

20  Diatom Gliding Motility

Speed (µm/s)

8000

6000

4000

2000

0 0

100

200

300 400 500 Distance (µm)

600

700

800

Figure 1.27  Relative speed of two diatoms plotted versus their distance.

the direction of the apical axis. This is similar to the movement on substrate, but hardly any longer distances are covered. Adhering lumps of EPS are typically not visible. Since no axenic cultures are used, bacteria can usually be seen on the water surface in phase contrast, making water movement observable. One recognizes changing currents of the water along the raphes, which run opposite to the movement of the diatoms. Apparently, the activity of the raphes can couple to a liquid medium. I do not want to exclude the possibility that the bacteria have an influence on the movement by mechanically coupling to the raphes, but in view of their small size and their lack of mutual contact, I consider their influence to be small. Theories of motility that require adhesion to a solid substrate cannot adequately explain this phenomenon. The movement of diatoms in hanging drops at the interface to air was interpreted by Nultsch in 1957 on the basis of the since disproved cytoplasmic streaming theory. Wavelike movements of microfibrils, as suggested by Bertrand [1.4] in 2008 (translation in Chapter 12 of this book), could cause the observed water transport along the raphe. They strengthen this hypothesis. I could not observe this form of movement with certainty in clusters of several connected diatoms, possibly because they contribute little to movement compared to the other forces acting there. It is obvious in Nitzschia sigmoidea cultures with high population density that the diatoms secrete EPS. These excretions can accumulate and lead to the formation of flakes of living diatoms. Sometimes it can be noticed that more or less large lumps of EPS adhere to individual diatoms on the water surface. When two diatoms come into contact through the mediation of an EPS lump, the transport of this lump along the raphes of both diatoms leads to a fast and remarkable movement. As there is no adhesion to a solid substrate, the

Movements of Pennate Diatoms in Cultures  21 two diatoms “dance” around each other. This movement pattern often occurs without the EPS lumps being recognizable. Diatoms that have come together in structures such as those shown in Figure 1.22 and Figure 1.26 show a peculiar dynamic. The attractive interaction of the hydrophobic apices is of great importance here. If only this attractive interaction existed, a pattern would be formed on the water surface, whereby the movement would come to rest as soon as a local minimum of energy is reached. However, the activity along the raphe and in the area of the apical pore field leads to continuous changes of the pattern. According to the simple model of two hydrophobic particles in the distance of the apices one expects structures with few unbound ends. For example, when there are three diatoms in a cluster, energetically favorable configurations in which all ends have bonds to other diatoms should prevail (Figure 1.28). A chain of three diatoms should close to a triangle because of the attractive interaction of the free ends. But the diatoms are not passive floating objects, but show active movement. They release energy and thus temporarily produce patterns with higher energy. There are elementary processes which are observable locally in the entire structure pattern. The end of one or more parallel diatoms can be moved by another diatom along its raphe (Figure 1.29). At the beginning of this movement, an existing connection at an apex of the moved diatom can be broken. Energy is released locally from the diatom into the structure. Angular changes at connected apices (Figure 1.29b) are very frequent. Diatoms that adhere with the apex on a substrate are capable of performing tilting movements. Probably it is the same mechanism as observed on the water surface. It is by no means certain that all motion sequences can be explained thereby, because the simultaneously occurring changes in this many-body system are very complex. One striking feature is the separation of connections at the apex without subsequent transport along a raphe. It could be due to opposing forces within the cluster. The static attractive force of the apices, the active movement along the raphe, the active change of the angle between two connected diatoms and possibly other mechanisms can generate a very large number of patterns and sequences in a system of several diatoms (Figure 1.30). These processes are fundamentally different from those on a solid substrate. In nature, the currents in the water are also likely to whirl up diatoms that do not have a high adhesion to the ground. They may occasionally reach the water surface. The ability to stay there for a longer period of time could lead to dispersal by drifting, i.e., hydrochory.

(a)

(b)

Figure 1.28  Energetically favorable patterns of three diatoms on the water surface: all diatoms parallel (a) and diatoms form a triangle (b).

22  Diatom Gliding Motility (a)

(b)

Figure 1.29  Frequently observed movement patterns: movement along the raphe (a) and angular changes at connected apices (b).

Figure 1.30  Image sequence showing the temporal development of seven connected diatoms. The time between the first and last image is 170 seconds.

100 µm

Figure 1.31  Pinnularia gentilis.

I am not convinced of this mechanism and without evidence it is nothing but speculation. A benefit from the mobility at the surface is not evident. Structures of interconnected Nitzschia sigmoidea, as described above, require a very calm water surface in addition to a high population density. Under light winds and waves, these fragile structures will certainly not be able to form or preserve themselves. I consider these patterns and their dynamics to be an artifact that only occurs in cultures, but it allows an insight into the motility of this species. The observations of Craticula cuspidata, Cymbella spp., Rhopalodia spp. and Pinnularia spp. on the water surface differ in several aspects from those of Nitzschia sigmoidea. In the following I will restrict myself to comments on Pinnularia gentilis (Figure 1.31) from a small pond in Stuttgart-Hohenheim. At the time of observation, the diatoms were already six months in culture and had a typical length of 200 µm. When Pinnularia cultures are prepared, a fast sedimentation of inserted diatoms is usually observed. In the case of diatoms

Movements of Pennate Diatoms in Cultures  23 from these cultures, it is noticeable that many diatoms are stirred up when the Petri dish is carried to the microscope and settle relatively slowly on the substrate. Immediately after swirling up the diatoms, one regularly finds a few to several tens of diatoms on the water surface. Already in the first minutes many of the diatoms sink to the ground. Others remain on the surface for hours and only a few for days. In all observed cases, the sinking starts with the diatom taking a position perpendicular to the water surface. Often it remains on the water surface in this orientation for a while. Either the diatom drops to the ground in this orientation or it rotates around the transapical axis or pervalvarous axis as it sinks. It was not possible to recognize which axis it is. During one observation, a diatom returned from the vertical position back to the horizontal position on the surface, which requires an energy supply by the diatom. Pinnularia floating horizontally on the surface are almost completely enclosed by water. A deformation of the water surface is not visible in phase contrast. There is also no formation of regular patterns due to an attractive interaction. Nevertheless, I consider a slight hydrophobicity to be possible. Bacteria can often be found on the water. If these form a coherent turf, this probably has a significant influence on the movement of the diatoms. Also, with these observations the water surface showed only a low bacterial density, so that I consider its influence negligible. When looking vertically at the water surface, the diatoms appear on the water surface in both valve view and girdle band view, with the valve view dominating. Occasionally there is a 90° rotation around the apical axis and thus a change between the two views. Presumably an activity of the raphe in the area of the helictoglossa is responsible for it. As on substrate, Pinnularia in valve view have a high mobility and cover longer distances, while in girdle band view back and forth movements are carried out. The movement is very similar to that on a solid substrate. Occasionally Pinnularia in valve view show rotations around the pervalvarous axis, which cannot be found on substrate. The adhesion to the substrate will probably prevent such rotations. In practice, this movement is often accompanied by a drift movement. In this context, it should be noted that the collective movement of two Pinnularia also appeared. I suspect the coupling by adhering EPS lumps. The observations of Pinnularia gentilis, which move actively on the water surface, whereby the driving raphe is completely below the water surface, again support Bertrand’s [1.4] hypothesis of a wavelike movement of microfibrils.

1.6 Formation of Flat Colonies in Cymbella lanceolata Cymbella species either are tube dwelling, develop stalks that often branch into tree-like structures, form colonies directly on substrates or are free-living [1.28]. The transition between free-living and colony-forming is smooth, as diatoms often leave colonies and develop new ones elsewhere. This is the topic of the observations described subsequently. First, Cymbella lanceolata with a length of approx. 190 µm is discussed (Figure 1.32). The adhesive EPS excretions are clearly visible in phase contrast, DIC or PlasDIC (Figure 1.33). A longer observation of accumulations reveals these processes: 1. Diatoms detach from a colony and move away from the colony. This typically happens at the edge of accumulations.

24  Diatom Gliding Motility 2. 3. 4. 5.

Diatoms move within the space between the colonies. Diatoms meet an existing colony and remain in this cluster. Diatoms stop their movement and attach themselves to the substrate. Diatoms reproduce asexually inside and outside colonies.

Short-term contacts between diatoms are not mentioned here, as they are transient and of minor importance for the establishment of structures. The relevant steps for the structure formation are exemplarily illustrated in Figure 1.34. Diatoms that attach to colonies usually

100 µm

Figure 1.32  Cymbella lanceolata.

Figure 1.33  Two small colonies photographed with PlasDIC.

2

3 1

5

4

Figure 1.34  Elementary steps that contribute to structure formation.

Movements of Pennate Diatoms in Cultures  25 remain on the edge of the colony. As they themselves secrete EPS, the area produced by EPS deposits is continuously increasing. Events 1, 2 and 3 do not allow the emergence of new colonies. A colony can develop from individual adherent diatoms according to 4 by following cell divisions and attachment of diatoms. To give an impression of the movement activity of the freely moving diatoms, images were taken in artificial daylight for over one hour (one picture every 10 s) and superimposed (Figure 1.35). The movement activity of diatoms requires sufficient light intensity. With increasing intensity, the raphes become active in the observed Cymbella, regardless of whether they move freely or are in a colony. As a result, more and more diatoms detach from colonies. The driving force then exceeds the adhesion to the substrate caused by EPS. For demonstration, two small colonies were irradiated with a light intensity between 7000 lx and 9000 lx. This is significantly above the intensities used in cultivation, which were at about 500 lx. The microscope illumination with a color temperature of approx. 3000 K was used for irradiation. The light intensity at the location of the diatom under observation is essential for the magnitude of the driving force. A high degree of homogeneity of the illumination in the area under observation is not required for this observation. Figure 1.36 shows the colonies at the beginning of irradiation and after about two hours. A considerable reduction in size can be recognized. The removal of diatoms from a colony and migration require sufficient light intensity. Correspondingly, the activity of the movement decreases when the intensity becomes low. When diatoms encounter existing colonies or deposits of EPS on the substrate at low

Figure 1.35  Movement activity of diatoms between colonies.

(a)

(b)

Figure 1.36  Colonies at the beginning of intensive light irradiation (a) and after about two hours (b).

26  Diatom Gliding Motility

(a)

(b)

Figure 1.37  Cymbella culture in the light phase (a) and dark phase (b).

brightness, they adhere to the colonies because they cannot overcome the adhesion (process 3). At very low light intensity or darkness, the free movement comes to rest. The diatoms then excrete an EPS pad, which they use to adhere to the substrate. At light intensities, as utilized in cultivation, the effect of colony reduction is less than demonstrated in Figure 1.36, but still clearly recognizable. For Figure 1.37 a culture in light and dark phase was photographed with a daytime cycle of about 12.5 hours of light per day. To take pictures in the dark phase, the culture was illuminated with low intensity from below through a diffusing screen with a white LED. The light intensity in the bright phase was about 200 lx in the dark phase 15 lx. Probably because of this remaining brightness the movement never comes to rest completely. It is striking that all diatoms moving during the day have reconnected to a colony with a few exceptions. To qualitatively and quantitatively study the formation of colonies and the daytime variations of diatoms in and outside the colonies, a culture was observed over 24 days from the time of inoculation, with lighting conditions corresponding to those just mentioned. Another cell line with a diatom size slightly larger than 100 µm was used. Pictures were taken every 10 seconds. The visible area amounted to 8.27 mm × 6.21 mm, corresponding to 2.6% of the cultivated area. A new colony can develop when a single diatom attaches itself to the substrate. It creates an adhesive area where other diatoms can get stuck. Many locations where single or a few diatoms adhere during the night are left by all diatoms in daylight. Due to the low water solubility of the excreted jelly, these areas are often an “anchorage” in the following dark phases. After a few days without settlement, the adhesion of such areas seems to decrease. Larger colonies do not disintegrate completely in the light phase. To quantify the growth of culture and colonies, 60 temporally equidistant images per day were converted to binary images after setting a brightness threshold. With the help of Fiji open-source software [1.32] particle analysis, the sizes of all connected image parts were determined for each image. The result is a list of areas ranging from individual diatoms to the largest colony. To estimate the number of freely moving diatoms and the diatoms bound in colonies, a classification was made: • Areas smaller than a lower threshold of a few pixels (about 1 to 4 pixels) are caused by small particles and unclean boundaries and are sorted out. • All objects larger than an upper threshold (in this evaluation, 70 pixels) are considered as colonies. • All objects in between are interpreted as individual diatoms.

Movements of Pennate Diatoms in Cultures  27 This classification may be incorrect, for example, if several diatoms overlap in the image and, due to their projection, occupy such a small area that they are categorized as one cell. An exemplary validation shows that this does not lead to significant distortions of the result. The total area occupied by diatoms shows notable daytime fluctuations. Diatoms bound in colonies occupy on average a different area than the diatoms between the colonies, because diatoms in colonies can overlap in the image and often do not lie horizontally. The area occupied by diatoms in colonies also does not grow linearly with the size of the colonies. In rough approximation a linear conversion factor of the areas has been determined. The criterion used was the minimization of the daytime fluctuations of the total number of diatoms. Apparently, the areas of the individual diatoms outside the colonies appear surprisingly small, as can be seen when explicitly comparing images in darkness and subsequent brightness (Figure 1.38). With the beginning of the bright phases, the number of individual diatoms increases rapidly. It falls off again in the course of the bright phase. Conversely, the area of the colonies is reduced. Figure 1.39 shows the total number of diatoms resulting from the sum of the two curves discussed. Compared to Figure 1.38, the daytime fluctuations are smaller, but still considerable. On the one hand, this is due to the assumed and only roughly fulfilled proportionality of the area of the colonies and the number of diatoms contained therein, and on the other hand due to fluctuations of the number of moving diatoms caused by the leaving and entering of the region of interest. The second effect dominates in the first days, in which the formation of the colonies just begins. The culture is in good approximation in exponential growth. The number of diatoms between the colonies can be determined by the described classification of the sizes of the connected clusters of an image. This is not identical to the number of diatoms that move, as some diatoms remain in place, especially at low brightness. When one superimposes two images whose recording times differ so much that moving diatoms have covered a distance of at least their own length, then they can be seen twice in this 550 500

Number of diatoms

450 400 350 300 250 200 150 100 50 0

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 Time (days)

Figure 1.38  Number of free diatoms (blue) and the number of diatoms bound in colonies (red) over 24 days. A yellow bar indicates the phases of bright light.

28  Diatom Gliding Motility 600

Total number of diatoms

500 400 300 200 100 0 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 Time (days)

Figure 1.39  Total number of diatoms (red) with exponential fitting (blue).

Number of diatoms in motion

50 40 30 20 10 0 0

1

2

3

4

5 6 Time (days)

7

8

9

10

Figure 1.40  Number of diatoms in motion over the last 10 observation days.

picture. The respective number of individual diatoms is obtained by particle analysis of one of the initial images and the superimposed image. The number of moving diatoms results from the difference. To improve the numerical quality, not only two, but six images were superimposed at one-minute intervals and the difference in the number of individual diatoms compared to the first image was divided by five, as moving diatoms appear five times in addition. Errors in this analysis mainly occur when diatoms do not continually move during the analyzed period (for example, by connecting to a colony) or when they cross the boundaries of the region of interest. Figure 1.40 shows the number of moving diatoms over the last 10 days of the observation time. Apparently, the movement activity drops off quickly

Movements of Pennate Diatoms in Cultures  29 after the beginning of the light phase. One can observe Cymbella species, which show high movement activity during the whole light phase. It can be assumed that there is a strong dependence on the light intensity. Colonies in Petri dishes are growing on a flat surface. On leaf surfaces or stones with unevenness which is smaller than the size of the diatoms, similar conditions could exist. In three-dimensional fibrous nettings, heaps or spherical colonies would be more likely to form. Diatoms escaping from dense populations must then move along thin filaments. Such netting was found in a sample from a pond where nutrient solution was added. A migration from and to colonies over the fibers could be observed. In this topology, a much lower fluctuation and less exchange between colonies are to be expected, as the colonies only offer possibilities for moving in and out at a few points.

1.7 Conclusion The present study consolidates observations on motile pennate diatoms in different environments. With respect to the movement on a smooth solid substrate, it is demonstrated that the inclusion of the orientation of the apical axis as a degree of freedom is useful in the analysis of trajectories, as it provides information about the point of the valve where the diatom touches the substrate. This point is subject to statistical fluctuations and is at different positions in Navicula sp. and Craticula cuspidata. Methodically independent observations confirm these results and allow investigating the question of the processes of motion reversal. In and on biofilms, pennate diatoms move with significantly different motion patterns and speeds. Adhesion to the surface of the viscoelastic film is sufficient to enable movement similar to that on a solid substrate. When the activity of the raphe branches is opposite, distortions of a biofilm marked with particles can be observed. Diatoms inside the biofilm without contact to the substrate rotate around changing axes. At the water-air interface, Pinnularia gentilis can perform active movements without coupling to a substrate, which is interpreted as the interaction between raphe activity and the surrounding water. Significantly different phenomena occur in Nitzschia sigmoidea. Hydrophobia of unknown origin at the apices gives this species buoyancy and leads to the formation of connected dynamic patterns. A deeper understanding of hydrophobia and dynamic pattern formation requires further investigation. The formation of colonies in Cymbella lanceolata can be described on the basis of elementary processes and their light dependence. They enable the exchange of diatoms between colonies, whose frequency depends on the topology of the environment.

References [1.1] Aumeier, C. and Menzel, D., Secretion in the Diatoms, in Secretions and Exudates, in: Bio­ logical Systems, J.M. Vivanco and F. Baluška (Eds.), pp. 221–250, Springer, Berlin Heidelberg, 2012. [1.2] Bertrand, J., Mouvements des diatomées. II - Synthèse des mouvements. Cryptogam. Algol., 13, 49–71, 1992.

30  Diatom Gliding Motility [1.3] Bertrand, J., Les Diatomées et la tension superficielle: un outil de recherche et de démonstration [movie], in: Association des Diatomistes de Langue Francaise, du 12 au 17 Septembre 1999, 18eme Collogue, Nice, France, 1999. [1.4] Bertrand, J., Mouvements des diatomées VIII: synthèse et hypothèse. Diatom Res., 23, 1, 19–29, 2008. [1.5] Bondoc, K.G., Heuschele, J., Gillard, J., Vyverman, W., Pohnert, G., Selective silica-directed motility in diatoms. Nat. Commun., 1–6, 2016. [1.6] Bondoc, K.G., Lembke, C., Vyverman, W., Pohnert, G., Searching for a Mate: PheromoneDirected Movement of the Benthic Diatom Seminavis robusta. Microb. Ecol., 72, 287–294, 2016. [1.7] Cohn, S.A., Photo-stimulated effects on diatom motility, in: Photomovement, D.P. Häder and M. Lebert (Eds.), pp. 375–401, Elsevier, Amsterdam, 2001. [1.8] Consalvey, M., Paterson, D.M., Underwood, G.J.C., The ups and downs of life in a benthic biofilm: Migration of benthic diatoms. Diatom Res., 19, 2, 181–202, 2004. [1.9] Donlan, R.M., Biofilms: microbial life on surfaces. Emerging Infect. Dis., 8, 9, 881–890, 2002. [1.10] Edgar, L.A., Diatom locomotion, computer assisted analysis of cine film. Br. Phycol. J., 14, 83–101, 1979. [1.11] Edgar, L., Diatom locomotion: a consideration of movement in a highly viscous situation. Br. Phycol., 17, 243–251, 1983. [1.12] Edgar, L.A. and Pickett-Heaps, J.D., The mechanism of diatom locomotion. I. An ultrastructural study of the motility apparatus. Proc. R. Soc. B, 218, 331–343, 1983. [1.13] Fabbri, S. and Stoodley, P., Mechanical properties of biofilms, in: The perfect slime, H.-C. Flemming, T. Neu, J. Wingender (Eds.), pp. 153–177, IWA Publ., London, 2016. [1.14] Fauvel, P. and Bohn, G., Le rythme des marées chez les diatomée littorales. C. R. Séances Soc. Biol., 62, 121–123, 1907. [1.15] Frenkel, J., Vyverman, W., Pohnert, G., Pheromone signaling during sexual reproduction in algae. Plant J.: Cell Mol. Biol., 79, 632–644, 2014. [1.16] Gutiérrez-Medina, B., Guerra, A.J., Maldonado, A.I.P., Rubio, Y.C., Meza, J.V.G., Circular random motion in diatom gliding under isotropic conditions. Phys. Biol., 1–10, 2014. [1.17] Häder, D.-P. and Hoiczyk, E., Gliding motility, in: Algal cell motility, M. Melkonian (Ed.), pp. 1–38, Chapman and Hall, New York, 1992. [1.18] Harper, M.A. and Harper, J.T., Measurements of diatom adhesion and their relationship with movement. Br. Phycol. Bull., 3, 195–207, 1967. [1.19] Harper, M.A., Movement and migrations of diatoms on sand grains. Br. Phycol. J., 4, 97–103, 1969. [1.20] Harper, M.A., Movements, in: The Biology of Diatoms, D. Werner (Ed.), pp. 224–249, Blackwell, Oxford, 1977. [1.21] Lauterborn, R., Untersuchungen über Bau, Kernteilung und Bewegung der Diatomeen, W. Englemann, Leipzig, 1896. [1.22] Molino, P.J. and Wetherbee, R., The biology of biofouling diatoms and their role in the development of microbial slimes. Biofouling, 24, 365–379, 2008. [1.23] Murase, A., Kubota, Y., Hirayama, S., Kumashiro, Y., Okano, T., Mayama, S., Umemura, K., Two-dimensional trajectory analysis of the diatom Navicula sp. using a micro chamber. J. Microbiol. Methods, 87, 3, 316–319, 2011. [1.24] Murguía, J.S., Rosu, H.C., Jimenez, A., Gutiérrez-Medina, B., García-Meza, J.V., The Hurst exponents of Nitzschia sp. diatom trajectories observed by light microscopy. Physica A: Stat. Mechanics its Appl., 417, 176–184, 2015. [1.25] Nultsch, W., Studien über die Phototaxis der Diatomeen. Arch. Protistenkunde, 101, 1–68, 1956.

Movements of Pennate Diatoms in Cultures  31 [1.26] Nultsch, W., Die Bewegung der Diatomeen. Mikrokosmos, 46, 220–227, 1957. [1.27] Nultsch, W., Phototactic and photokinetic action spectra of the diatom Nitzschia communis. Photochem. Photobiol., 14, 705–712, 1971. [1.28] Rimet, F. and Bouchez, A., Life-forms, cell-sizes and ecological guilds of diatoms in European rivers. Knowl. Manage. Aquat. Ecosyst., 406, 1–14, 2012. [1.29] Round, F.E. and Happey, C.M., Persistent, vertical-migration rhythms in benthic microflora. IV. A diurnal rhythm of the epipelic diatom association in non-tidal flowing water. Br. Phycol. Bull., 2, 463–471, 1965. [1.30] Round, F.E. and Eaton, J.W., Persistent, vertical-migration rhythms in benthic microflora. III. The rhythm of epipelic algae in a freshwater pond. J. Ecol., 54, 609–616, 1966. [1.31] Round, F.E., Crawford, R.M., Mann, D.G., The diatoms: biology and morphology of the gen­ era, pp. 104–106, Cambridge University Press, Cambridge, 1990. [1.32] Schindelin, J., Arganda-Carreras, I. et al., Fiji: an open-source platform for biological-image analysis. Nat. Methods, 9, 676–682, 2012. [1.33] Wang, Y., Pan, J., Cai, J., Zhang, D., Floating assembly of diatom Coscinodiscus sp. microshells. Biochem. Biophys. Res. Commun., 420, 1, 1–5, 2012.

2 The Kinematics of Explosively Jerky Diatom Motility: A Natural Example of Active Nanofluidics Ahmet C. Sabuncu1*, Richard Gordon2,3, Edmond Richer4, Kalina M. Manoylov5 and Ali Beskok4 Department of Mechanical Engineering Worcester Polytechnic Institute Worcester, Massachusetts, USA 2 Gulf Specimen Marine Laboratory, Panacea, Florida, USA 3 C.S. Mott Center for Human Growth & Development, Department of Obstetrics & Gynecology, Wayne State University, Detroit, Michigan, USA 4 Department of Mechanical Engineering, Southern Methodist University, Dallas, Texas, USA 5 Department of Biological and Environmental Sciences, Georgia College and State University, Milledgeville, Georgia, USA 1

Abstract

Motile diatoms move in the low Reynolds number regime without any apparent organelle for motility. While several possible hypotheses have been proposed, the diatom motility mechanism is not fully understood. In this study, the kinematics of the jerky motion of individual diatoms was investigated at 1.2 millisecond temporal and 9 nm subpixel spatial resolution of species: Nitzschia palea, Navicula cryptocephala, Navicula sp. Centroids of the diatoms were measured from 821 fps digital movies using algorithms for particle tracking. The examination of the displacement data indicated that: 1) for all the diatoms investigated, speed along its trajectory varied from one frame to the next; 2) the motion was not unidirectional; 3) the displacement data included rare large velocities that were as large as 250 µm/s for the Navicula diatoms. The calculated spontaneous accelerations are on the order of 5 × 105 µm/s2. Rapid changes in the orientation angle as large as 2.66 rad/s were observed for Nitzschia palea. The jerky diatom motion could be related to elastic snapping and recoil at the yield point and/or explosive discharge of the mucilage secreted and left on the glass substrate as the “diatom trail.” The flow of mucilage in the slits in the diatom silica valve, called raphes, could be a first example of “active nanofluidics,” which we define as self-propulsion of a fluid. Active nanofluidics could be useful in designing micro- and nanorobots that require pumps for fluid flow. Keywords:  Diatom motility, particle tracking, image processing, active materials

*Corresponding author: [email protected] Richard Gordon: [email protected] Edmond Richer: [email protected] Kalina M. Manoylov: [email protected] Ali Beskok: [email protected] Stanley Cohn, Kalina Manoylov and Richard Gordon (eds.) Diatom Gliding Motility, (33–64) © 2021 Scrivener Publishing LLC

33

34  Diatom Gliding Motility

2.1 Introduction “Of the Oat-Animal (later identified as the diatom Craticula cuspidate [2.75]).... it can change its place by jerks or leaps, which it makes by the action of some strong muscles in the two protruded parts, whose spring throws it to the distance at least of its own shell’s length every time they are exerted. These leaps however have long intervals between, and are never made till the animal is perfectly undisturbed” [2.3].

Algae are important primary producers in aquatic habitats. Limitations to light and nutrient availability can be overcome by motility. Representatives of all algal groups, with the exception of Rhodophyta, can move. Diatoms have a cell wall made from hydrated silica with specialized slits and openings for nutrient supply and motility. Diatoms do not have apparent specialized organelles for locomotion, such as cilia or flagella, and their motility depends on continuous release of fibrils from the raphe slits (“thin strand (probably mucopolysaccharide)” in [2.46], “membranous profiles” in [2.25], “mucilage strands” in [2.29]), leaving behind a “diatom trail” [2.24] [2.49] consisting of polysaccharides [2.54]. As the material is a viscous liquid and sticky, it is a mucopolysaccharide. As the fibrils are aligned within the raphe perpendicular to their flow [2.29], the “raphe fluid” is a smectic liquid crystal. Observations indicate that diatoms move only when the raphe system slits are adjacent to a substrate. The continuous swelling release of the mucopolysaccharide could be responsible for the diatom motility [2.30] [2.46]. A capillarity model by Gordon and Drum proposes that the relatively hydrophobic mucopolysaccharide wets the hydrophobic raphe slit as it is released from the cytoplasmic side. It undergoes a reaction with water, converting it to a hydrophilic form that cannot wet the raphe walls and leaves the raphes at the trailing end [2.40]. (Raphe silica is chemically distinct from valve silica [2.23]). If this hypothesis is correct and raphe fluid is not propelled by motor proteins on the cytoplasmic side of the slits [2.30], the fluid and the motility mechanism of diatoms could be the first example of active nanofluids, where the propulsion of the fluid only depends on the fluid and slit chemistry and does not involve any moving mechanism or external field. Driven mostly by the studies on motor proteins, such as myosin, research on active matter, which are materials capable of selfmanipulation, has increased in the recent years [2.72]. Active nanofluidics, for instance, could be used as a self-propulsion mechanism in micro and nano swimmers in the low Reynolds number regime in the field of micro- and nanorobotics [2.12] or as a selfpumping fluid in microfluidic devices. A remarkable paper [2.28] by the late Leslie Edgar [2.22] and her colleagues showed that, when motile pennate diatoms are analyzed at 10 frames per second (10 fps), there are large changes in position from one frame to the next. This indicates that there are repeated events occurring that change the speed and acceleration in less than a tenth of a second. This kind of motion has been called “jerky” [2.3] [2.28] [2.71]. Here, we present an attempt to detect these events by using a camera/microscope/software system capable of up to 869 fps, and then interpret their meaning in the context of trying to understand the mechanism of raphebased motility. For this purpose, particle tracking experiments on four motile diatoms were made. Different image processing algorithms were compared for the best accuracy using a numerical test object. The centroid data from the recordings of three motile diatom species were extracted using the best algorithm and analyzed. Turning characteristics in addition to translational motion of diatoms were analyzed.

The Kinematics of Explosively Jerky Diatom Motility  35

2.2 Material and Methods 2.2.1 Diatom Preparation Diatoms were collected by scraping the brick wall lining of a man-made, unnamed urban lake in Dallas, Texas at 32°51’25”N - 96°45’34”W. The lake has a continuous flow via culverts, a fountain in the middle, and is inhabited by mallard and other ducks. Two samples were collected from the lake and carried to the laboratory in centrifuge tubes. Lake water samples were stored in sealed glass 20 ml vials in a north view windowsill starting in January 2015 in a room kept at 22 °C. All experiments were conducted at that room temperature. Whole algal community assessment was performed on the diatom samples that were collected from the lake. The community was dominated by diatoms (48% of the community), followed by green algae (42%) and single representatives of Cyanobacteria, Euglenophyta, and Synurophyta. From the diatoms documented, 50% were motile. Selecting the diatoms to work with was based on physiological observations of chloroplast structure and color. In addition, the camera’s field of view was a narrow rectangle 300 pixels (30 µm) wide, which restricted the study to small Navicula-like species. This study used 4 diatom specimens and each was given a number: diatoms #1 and #4 (Nitzschia palea (Kützing) W. Smith), diatom #2 (Navicula cryptocephala Kützing), and diatom #3 (Navicula sp. broadly lanceolate with broadly rostrate apices, length 12 µm).

2.2.2 Imaging System We used a Hamamatsu C11440 ORCA-Flash4.0 Digital Camera to record diatom motion. The imaging software used was CellSens Dimension (v1.12, Olympus). Timing resolution was within 1 µsec by use of an Olympus U-RTC, Real Time Controller. Movies were recorded using a 40x phase objective (Olympus LUCPlanFLN) adjusted for the thickness of the glass slide in an inverted microscope (Olympus IX81). The microscope was on a vibrationcanceling optical table to minimize drift in the experiments. Since we did not know in advance what spatial and temporal frequencies would characterize diatom motility, we adjusted the parameters of the camera for smallest pixel size and smallest time interval between frames. Although the camera CMOS sensor consists of 20482 pixels, only a narrow strip can be imaged at high frame rates with the software provided. We avoided pixel binning and used the free running mode, which gives us the tradeoffs in Table 2.1. In each case the length of the strip can be up to 2048 pixels, but its position must be in the middle of the 20482 array of detectors. According to Table 2.1, decreasing the width of the strip below 100 pixels did not have any effect on the acquisition speed. For the diatom recordings, the width of the imaging strip was chosen so that it is slightly larger than the width of the diatom under test. Imaging data was retained when paths were straight enough so that the diatom remained within the strip for at least a few seconds. To avoid scratches that could deviate the path of a diatom, a clean, previously unused glass slide was used for each preparation. The magnification changer (U-ECA1.6x) was set at 1.6x for the experiments, giving an effective magnification of 64x. At this magnification the width of one pixel is 102 nm. This was crosschecked by using a stage micrometer. The diatoms used in this

36  Diatom Gliding Motility Table 2.1  The acquisition speed of the camera in frames per second (fps) as a function of width of the region being imaged. Width of strip (pixels)

fps

300

589

200

821

100

869

50

869

study were 4-7 µm wide, allowing the highest possible acquisition speeds (Table 2.1). The microscope is on a vibration isolation platform (Newport RS 2000).

2.2.3 Sample Preparation A mini-channel was fabricated to accommodate the samples (Figure 2.1). The channel is composed of a polycarbonate block, double sided tape, and a clean glass slide. The channel was engraved into the polycarbonate piece using a computer controlled mill. The depth of the channel is 1 mm while the width is 2 mm. Double-sided tape was used to adhere the glass slide to the polycarbonate block. The outer surface of the assembly was cleaned with a lint free tissue paper soaked with isopropyl alcohol prior to the experiments. The glue of the tape came in contact with the medium containing the diatoms. There was no evidence of toxicity as motile algal species were present in the channel even after 1 hour of introduction of samples. All the recordings were acquired in less than 1 hour after the samples were introduced to the chamber. Samples were fed into the channel using a pipette through the inlet on the glass side. The assembly was turned upside down after the suspension was fed into the channel and placed under the microscope so that the glass slide faced the objective lens of the inverted microscope. Diatoms were thus imaged after they had settled on and adhered to the glass slide. As the microscope stage is horizontal, the diatoms should not have experienced a horizontal component to Earth’s gravity. The assembly was open to air through the inlet and outlet holes. No significant evaporation was observed during the experiments.

2.2.4 Image Processing In each run between 1000 to 4000 consecutive 16-bit gray-scale frames were recorded at 821 fps using the camera software. The frames were exported in TIFF (Tagged Image File Format). The timestamps of the frames were recorded manually and used in the analysis. All image processing was performed off-line using a commercial software package (MATLAB 8.4, The MathWorks Inc., Natick, MA, 2014, with Image, Computer Vision, and Signal Processing Toolboxes). Eight different particle tracking algorithms were tested for accuracy and precision in detecting subpixel displacements prior to application to diatoms. We followed a similar approach to the study by Cheezum et al. to test the tracking algorithms [2.13].

The Kinematics of Explosively Jerky Diatom Motility  37 In order to develop a computational model for studying the tracking algorithms a 1000 by 1000 high resolution matrix bearing a noise free, centered, 3 µm radius circular object was simulated with a pixel size of 10 nm. The circular object was given a white level of 216 and the background a black level of 0. In order to account for the diffraction of light in the microscope, we convolved the high resolution binary image (circle 216 with background 0) with an Airy disk point spread function [2.13] [2.90] [2.92]:

PSF (r ) =



2

 2 J1 (ar )  ,  r 

(2.1)

where a = (2π NA)/λ.

In the above equations J1 is the Bessel function of the first kind, NA is the numerical aperture of the objective (0.6 for the objective used), λ is the wavelength of light (chosen as 570 nm), r is the radial distance from the center. The image at the camera was simulated by integrating 10 by 10 square regions in the high resolution matrix. Accordingly, one pixel in the simulated camera matrix is 100 nm, close to the camera imaged pixel width (102 nm/ px). Following the integration, the camera matrix was scaled so that the black level corresponds to 10 and the white level 1000, per [2.13]. One of the dominant noise sources in imaging systems is the shot noise that occurs due to the uncertainty of photon arrival rate at each photodiode. The number of incoming photons at each recording follows a Poisson distribution, where the variance is equal to the mean [2.42]. In this study the shot noise was modeled using artificial noise, neglecting other sources such as pixel crosstalk and pixel gain mismatch. The artificial noise was added to each element in the simulated camera matrix using random numbers obeying the Poisson distribution with mean equal to the local value of the pixel in the original matrix. Repetition of this step simulates a time varying microscopy movie of a spherical particle in focus at its center. By assigning center displacements less than 10 units (100 nm) in the high resolution image, subpixel movements in the camera matrix were simulated. As a metric for the performance of eight different particle tracking algorithms the mean error of estimation introduced by the algorithm and its standard deviation were calculated:



B = 〈|s − sinput |〉 and σ = 〈(s − 〈s 〉)2 〉0.5 ,

(2.2)

where s is the displacement, subscript input is for the input displacement, brackets and || denote averaging and absolute value, respectively: 1) “Grayscale Centroid”: Center of Mass (Centroid) Calculation using Grayscale Images: Centers of mass of two consecutive images were compared to estimate the distance an object has moved. For a given grayscale intensity image matrix of Iij(t) that contains the object, which appears brighter than its background, the centroid (Cx, Cy) on x and y axes is given by:



Cx =

∑ ∑ n

m

i =1

j =1

xi Iij (t )

∑ ∑ n

m

i =1

j =1

Iij (t )



(2.3a)

38  Diatom Gliding Motility and

Cy =



∑ ∑ n

m

i =1

j=1

y jIij (t)

∑ ∑ n

m

i =1

j=1

Iij (t)



(2.3b)

2) “Grayscale Centroid with Threshold”: The effect of prior thresholding the image on centroid extraction was also tested as a modification to “Grayscale Centroid” algorithm. All pixels that had intensity values less than 10%, 25%, or 50% of the maximum intensity value were set to zero, and the remaining pixels were untouched. Note that in thresholding no distinction is made between pixels inside or outside the diatom. This is addressed in the next centroid method. 3) “Binary Centroid”: Center of Binary Mass Calculation using Edge Detection: It is desirable to have good contrast between the object and the background, where the background is assumed to have nearly constant intensity values. Therefore, prior to calculating the centroid, image thresholding and image segmentation were used to determine the boundary of the test object, and the object was represented with the pixel value 1, while the background was assigned value 0, i.e., the grayscale image is converted to a binary image. The thresholding made here applies to the gradient of the grayscale image. Depending on the magnitude of the gradient some edges are preserved and some are discarded. The edge detection algorithm implemented allows two thresholds for the gradient. One is used for detecting strong edges, and the other for detecting weak edges. Weak edges are included in the analysis only if they are connected to the strong edges. The following steps were undertaken to find the centroid location: 1) Detection of all edges in the image; 2) Boundary dilation to satisfy edge connectivity; 3) Filling in the closed boundaries; 4) Erosion of the pixels at the edges, where the magnitude of erosion is equal to the values used in dilation; 5) High pass filter of the image so that only the object of interest is present in the image and other smaller artifacts are eliminated; 6) Centroid calculation following Eq. (2.3), but with Iij(t) = 0 or 1. An advantage of this algorithm is that the internal structure of the object has no effect on the centroid, and need not be constant over time. For example, chloroplasts in diatoms can move [2.63]. 4) “Binary Centroid with Median Filtering”: As a modification to the “Binary Centroid” algorithm, effects of a median filter before detecting the edges (step #1 in “Binary Centroid” algorithm) were tested. The operation replaces the pixels in 3 by 3 neighborhoods in the original image by the median value of these pixels. After the median filter process all other steps are the same as those in “Binary Centroid.” 5) “Template Correlation”: Calculating Correlation between a Template and Target Images: In this method, image I was shifted by one pixel increments in the target image K, and the product X x , y =

∑ ∑ n −1

m −1

i =0

j =0

I x +i , y + j K i , j, was calculated for each increment. The correlation

matrix Xx,y is maximum where I and K are most similar. The template image was formed by cropping the first frame in a movie at a size a little larger than the object. The template image was correlated with each image in the movie. The correlation technique returns one-pixel resolution on centroid extraction as increments are single pixels in calculating the correlation. In order to achieve subpixel resolution, a quadratic surface, which has the following form: z = a + bx + cy + dx2 + ey2, was fit to the correlation matrix, where a, b, c,

The Kinematics of Explosively Jerky Diatom Motility  39 d, e are constants. MATLAB’s Curve Fitting Toolbox was used to calculate the coefficients. b c The centroid location on the x and y axes was calculated using   and   , respec­  2d   2e  tively. The fit was done using the maximum point in the correlation matrix and the closest 24 neighboring points to the maximum. Note that this method does not include rotation of the template, and therefore presumes little deviation of the diatom from a straight-line trajectory. 6) “Template Matching”: Using Template Matching Algorithms: The template matching algorithm in MATLAB’s Computer Vision Toolbox was also tested to determine distance moved by the object. The algorithm finds the best match between the template image and the target image. This is achieved by shifting the template pixel by pixel on the target image and minimizing the Sum of Absolute Differences: SADx ,y =

∑ ∑ n −1

m −1

i =0

j =0

| I x +i , y + j − K i , j |,

where the template image I has translated x and y on the target image K. The template matching algorithm is dependent on the sum of the differences of pixel values between the template and the target while the correlation technique is dependent on the summation of the scalar product of the pixel values in the target and the template. Besides also being sensitive to rotation of the diatom, this template matching algorithm had one-pixel resolution, and therefore it was omitted in the rest of the study. 7) “Ellipse Fitting”: We tried fitting an ellipse to the edge points for the diatom that were detected in the Binary Centroid method [2.55], and deemed this approach to be ineffective for centroid calculations. However, such fitting gave us orientation data. 8) “Image Registration”: We employed the IGOR PRO 6.3.7 (WaveMetrics Inc., Lake Oswego, Oregon, USA) implementation of an image registration algorithm that minimizes the mean square intensity difference between a reference and a test image. The minimization is performed according to a variation (ML*) of the Marquardt–Levenberg algorithm for nonlinear least-square optimization [2.83]. The algorithm is capable of subpixel resolution and produces an affine transformation for the relative rotation and translation as well as for isometric scaling or skew parameters. The algorithm was restricted to rigid-body motion (rotation and translation) and the resultant offset was used to calculate relative displacement between successive images. Polystyrene particles that are 10 µm in diameter were used to determine the noise in the measurement system. The particles were loaded in the mini-channel, and in order to have immobile particles, avoiding Brownian motion, the channel was placed in an oven for 10 minutes at 200 oC to sinter the polystyrene particles to the channel. The channel was then cooled down and filled with deionized (DI) water and placed under the microscope. The particles were imaged at the above-mentioned conditions as the Brownian motion was deemed to be absent, and particles were not dislodged following gentle flow of DI water. The centroids were calculated, and plotted versus time. A Fast Fourier algorithm was used to find the frequency components of the centroid data. Following the system calibration with stationary particles, the centroid extraction techniques were used to track diatom motion. Each frame underwent the following steps before calculating displacements: 1) Image cropping to include only the area of interest. Cropping was done using the same area for all the frames in a recording; 2) Contrast enhancement using a mapping, where the pixels in the top and bottom 1% intensity range were saturated. The following relationship was used for the contrast enhancement of a grayscale image I,

40  Diatom Gliding Motility



I −l  I out =  in in [hout − lout ] + lout ,  hin − lin 

(2.4)

where h and l stand for high and low intensity values for mapping, respectively. The subscripts out and in stand for the output and input quantities, respectively. The input image Iin was modified before applying the mapping to have the pixels having intensity values higher than hin and pixels having intensity values lower than lin equal to hin and lin, respectively. The Canny and Sobel algorithms, which are two different algorithms for edge detection, were tested for their performance with the Binary Centroid algorithm to extract diatom boundaries and centroid locations. The Canny algorithm [2.10] was found to be superior to the Sobel edge detection algorithm for these images. The diatom is composed of weak and strong edges caused by differential light scattering and diffraction around the diatom. Strong and weak edges correspond to different magnitudes of image intensity gradients at the diatom edge. The Canny algorithm searches for local maxima of the intensity gradient, is able to detect strong and weak edges, and includes the weak edges in the output only if their connectivity is satisfied, whereas, the Sobel algorithm depends on the global gradient field, and is therefore more prone to be affected by noise and by the presence of weak and false edges. In the analysis of the diatom motion, several parameters of the Binary Centroid algorithm, such as the thresholding factor for edge detection and the dilation and erosion kernel sizes, were selected by analyzing the first frame of a run. While keeping the threshold low for edge detection caused additional unwanted particles to appear in the region of interest, keeping the threshold high resulted in a clear background but exhibited unconnected segments in the diatom boundary. The parameters were varied and optimized by visual inspection, and then used for all frames of the run. Orientation angle is defined here as the angle between the major axis of the diatom and the x-axis (Figure 2.1). In order to calculate orientation angle, an ellipse that has the same

Figure 2.1  The polycarbonate channel used to image diatom motion, photographed on the stage of the inverted microscope. The depth, width, and length of the channel are 1 mm, 2 mm and 60 mm, respectively. A glass slide was bound to 2.5 mm × 7.5 mm polycarbonate block using two pieces of double-sided tape placed on the polycarbonate block. The coordinate origin coincides with the origin for the pixels. Diatoms move on the glass slide at the bottom of the chamber. The inset figure shows a schematic diatom, the coordinate axes, and the origin, in the plane of the top of the slide.

The Kinematics of Explosively Jerky Diatom Motility  41 second moments as the diatom was fit to the threshold image. The second moments are calculated as:



Mx =

∫∫

A

i 2 dA M y =

∫∫

A

j 2 dA M x , y =

∫∫

A

ij dA.



(2.5)

where A is the area in the image occupied by the diatom, and i and j are the column and row indices for the pixels inside the diatom boundary. MATLAB’s Image Processing Toolbox was used to fit the ellipse to the binary images of diatoms obtained by the Binary Centroid algorithm. For some of the diatom recordings contour smoothing was used to suppress large fluctuations in detected diatom boundaries. Smoothing was done by convolving a Gaussian kernel with the binary image that includes only the edge information from the Binary Centroid algorithm. The kernel parameters were dependent on the size and structure of the diatom. Different Gaussian kernel sizes were tested, and better accuracy was achieved for sizes comparable to the width of the diatom. However, Gaussian smoothing to this extent resulted in overestimation of the diatom boundary.

2.3 Results and Discussion 2.3.1 Comparison of Particle Tracking Algorithms A circular test object was formed in a 100 by 100 pixel simulated camera matrix, where the object radius was 30 pixels, and the center was at (50, 50) pixels. In the high resolution matrix (1000 by 1000 pixels) the following displacements were applied to the circular object: 0 pixel; 0.1 pixel, 0.5 pixel, and 1 pixel. Series of images were generated by letting the object translate in the camera matrix in the positive x direction and adding random noise per frame to each pixel. For the stationary case the frames differed only by the random noise. For each case the apparent displacement was calculated using the 8 different algorithms. Following the calculation of the apparent displacement, the mean error of estimation and the standard deviation were calculated (Eq. 2.2). The mean error of estimation and the standard deviation are tabulated for the 8 algorithms in Appendix Table AI. The mean error and standard deviation were calculated for different input displacements: 0, 0.1, 0.5, and 1 pixel. According to the results, as the input moving distance increased the mean error of estimation decreased for all the techniques. Overall, the minimum bias values were observed for the Binary Centroid algorithm, the Grayscale Centroid algorithm with 25% threshold, and the Image Registration algorithm. The algorithm used for the Canny technique in Binary Centroid algorithm chose the threshold values automatically based on the highest value of the gradient magnitude of the image. Consequently, it can be deduced that thresholding is necessary for subpixel particle tracking. The effect of a median filter in the Binary Centroid algorithm was also tested. While a reduction in the standard deviation was achieved when the median filter was present for displacements 50 nm and 100 nm, overall, using median filter increased the bias. The Template Correlation algorithm with interpolation to find subpixel resolution yielded higher bias for non-zero displacements than those of techniques with thresholding. Rare displacements that are of 1 pixel magnitude were present in the displacements calculated by the Template Correlation algorithm. The number of images

42  Diatom Gliding Motility generated decreased as the input moving distance ascribed to the test object increased, as the centroid was not calculated if the test object was partially out of the camera matrix. While the Grayscale Centroid algorithm returned similar values for the bias and the standard deviation, thresholding diatom images resulted in loss of a high percentage of pixels inside the diatom as diatoms are transparent to light. Therefore, finding edges preserved a greater amount of information for diatoms. While the Image Registration algorithm yielded low error for the numerical test object, as it was applied to diatom motion it had occasional large errors finding the centroid. Thus, for tracking the diatoms, the Binary Centroid algorithm was used in the rest of the study.

2.3.2 Stationary Particles An ellipsoidal polystyrene particle with major and minor projected radii of 10 and 8 µm, respectively, was imaged in deionized water. Images were processed using the Binary Centroid algorithm. The Canny algorithm was used for edge detection. Gaussian convolution and median filtering were not used. The 4000 images were acquired at 821 fps, and the centroid position was calculated for each frame. Using the centroid data, the displacements along the x and y axes and the total displacement per frame pair were calculated. The following formulae were used to calculate x and y displacement: dxi = (xi+1 – xi), dyi = (yi+1 – yi), where subscript i is for the frame number. The calculated displacements are per frame pairs that are separated by 1.217 ms. The direction of the motion was assumed to lie either along the x or y axis depending on the direction of the major displacement, and any displacement causing a negative displacement along the x or y axis was deemed negative. For each recording, the origin for the coordinate system was the top left-hand side of the image. The scatter of the recorded coordinates of the simulated test object and the stationary polystyrene particle centroids are shown in Figure 2.2. The scatter in the polystyrene particle coordinates is as large as 20 nm, whereas the scatter in the simulated test object coordinates is less than 10 nm. The x, y, and Gaussian fits to x, y displacement of the stationary particle are also given in the supporting information (Appendix Figure A1). The particle’s x and y displacement histograms follow a normal distribution centered around the origin. The standard deviation of the x and y displacement histograms are 5.7 nm and 7.3 nm,

(a)

4.991

4.985 4.985

(b) y coordinate [µm]

x coordinate [µm]

4.997

11.22

11.2

11.18 4.991 x coordinate [µm]

4.997

15.3

15.32 15.34 x coordinate [µm]

Figure 2.2  (a-b) Scatter plots of the simulated test object and the stationary polystyrene particle centroids, respectively. The stationary polystyrene particle was imaged at 821 fps. The number of points plotted is 1000 for the simulated particle and 4000 for the polystyrene particle.

The Kinematics of Explosively Jerky Diatom Motility  43 respectively. These values could be used as an indicator of the uncertainty in the diatom centroid measurements. A motionless particle is desirable to evaluate the performance of the particle tracking algorithms; however, there are two main obstacles in evaluating a motionless particle on a microscope stage. First, it is necessary to ensure that the particle is actually motionless relative to the microscope stage. This requires an alternative measurement system to independently measure the relative positions of the microscope stage and the object that is viewed. In this study, a motionless particle was approximated by sintering polystyrene particles on glass slides. Second, the motion of the imaging system (camera) relative to the microscope stage needs to be measured. Interestingly, Fourier analysis of the centroid position data that was used to calculate Appendix Figure A2 revealed vibration peaks at 67 Hz, 196 Hz, and 333 Hz in both x and y coordinates (Appendix Figure A2). The magnitude of the vibration in the x and y directions were 4 nm and 8 nm, respectively. The vibration was considered to be due to the illuminator and/or camera fans causing motion of the microscope stage relative to the camera. However, Fourier analysis of diatom #3 centroid position data at 821 fps did not return any vibrational modes (Appendix Figure A2 c and d), perhaps because of viscous damping of the raphe fluid by which it is attached to the glass slide. Therefore, this vibration was disregarded in the rest of the study, and assumed not to affect the observed kinematics of diatom motility. Furthermore, we measured the amount of drift in the stationary particle. The calculated x and y drift velocities are 0.9 × 10-8 µm/s and -0.4 × 10-7 µm/s, respectively. The drift is neglected in the rest of the study. Overall, the standard deviation of the centroid data of the polystyrene particle in x and y coordinates were 5 nm and 9 nm, respectively, whereas the simulated test object exhibited the standard deviation in x and y coordinates as low as 1 nm. The difference is probably due to mechanical vibrations in the imaging system as detailed above. The larger standard deviation in the y direction is consistent with the larger vibration detected in the y axis.

2.3.3 Diatom Centroid Measurements Each of the three diatoms (#1 Nitzschia palea, diatom #2 Navicula cryptocephala, and diatom #3 Navicula sp.) was imaged at 821 fps, and at least 500 frames were recorded at a single experiment in which a diatom was motile. Such observations were repeated at least 2 times with the same diatom, with much time passing in between because the diatom often left the long, narrow field of view of the camera in the high-speed mode. As a result, at least 3000 frames were acquired for each diatom, leading to 3000 displacement data points for each diatom. The centroid of each diatom was extracted using the “Binary Centroid” algorithm. For diatoms #1, #2, and #3 the edges were clearly defined using the Canny edge detection algorithm without extra Gaussian smoothing. The diatoms moved distances on the order of 5 µm along their traveling directions. The average speeds of the diatoms were calculated as follows: #1: 2.54 µm/s; #2: 8.28 µm/s; #3: 2.93 µm/s. The average speeds of the diatoms were calculated by determining the total distance that the diatom moved during the observation. The Reynolds number corresponding to this average speed is about 10-4. A qualitative examination of the diatom trajectories led us to the following conclusions: (1) diatom speed varied continually along the trajectory; (2) the motion was not unidirectional; (3) displacements as large as 290 nm were present in both the x and y directions. Three representative trajectories are given in the supporting information (Appendix Figures A3 and A4).

44  Diatom Gliding Motility Considering 290 nm displacement, the Reynolds number of the motion increases around 24-fold, while still remaining in the low-Reynolds-number flow regime. A natural consequence of the observation of nonuniform speed is that the force on the diatom necessary for gliding motion was not constant during the observation. Also, reverse motion and turning of the diatom occurred along its trajectory. The total displacement of the diatoms was calculated using the following formula: dsi = ( xi +1 − xi )2 + ( yi +1 − yi )2 . The diatoms investigated in this study moved in +x direction during the measurements. Therefore, any displacement resulting in a negative dxi was assigned a negative total displacement. Displacement histograms for the diatom and the stationary particle are shown in Figure 2.3. All image sequences of each diatom were used to make the diatom displacement histograms. The polystyrene particle displacement distribution is bimodal, centered around the origin, and symmetrical. Negative displacements of the diatoms are also evident from Figure 2.3, which are fewer in occurrence compared to positive displacements. While stationary particle image displacement distribution is limited to ±25 nm (presumably due to photon fluctuations and electronic noise), the diatoms exhibited displacements as large as 300 nm. The standard deviation of the polystyrene particle displacement distribution is 0.09 pixels. The estimate of the standard error in centroid estimation of polystyrene particles was also calculated assuming thresholding and photon noise errors in the system following the method of Patwardhan [2.68]. Using the configuration parameters of this study, the calculated estimated error in determining centroid position was 0.0018 pixels, which is around two orders of magnitude smaller than the standard error observed for the stationary particle. Several other factors, such as vibration of the microscope stage could have contributed to this disagreement. In order to validate the existence of large displacements observed in Figure 2.3, the frames pertaining to these displacements were individually analyzed. In order to illustrate these efforts, a detailed analysis of a large displacement in a representative trajectory of diatom #2 is given here. The plot of the total displacement in a recording of the diatom #2 is shown

Diatom #3 Diatom #2 Diatom #1 Stationary Particle

Count

102

101

100 -0.1

-0.05

0

0.05 0.1 0.15 Displacement [µm]

0.2

0.25

0.3

Figure 2.3  Displacement histogram of the diatoms (symbols) and stationary polystyrene particle (continuous line) imaged at 821 fps.

The Kinematics of Explosively Jerky Diatom Motility  45 in Figure 2.4a. An overlay image of two calculated boundary images corresponding to the large displacements at around 2000 ms in Figure 2.4a is shown Figure 2.4b. The captured frames pertaining to this large displacement are shown in Figure 2.4c and 2.4d. Although the diatom boundary does not follow the true boundary of the diatom due to the thresholding used, the boundary detection algorithm was consistent in determining the boundaries among the frames. The overlay of the boundary of one frame onto the other frame generated a visible abnormality. Therefore, we conclude that at this temporal scale a diatom translates with a broad range of velocities including rare explosive steps and backward motion. Backward explosive steps are also observed (Figure 2.3, diatom #3). Furthermore, we obtained similar explosive jerky movement using the “Image Registration” algorithm. We next sought a kinematic model to represent diatom motion. A quantitative analysis of the diatom trajectory can be made using the mean-squared displacement (MSD) data of the diatom trajectories. The MSD is:

∫

ρ(τ ) = |r (τ + t ) − r (t )|2 dt ,



(2.6)



where r(t) = [x(t), y(t)] represents a trajectory. For simple diffusion ρ(τ) = 4Dτ in 2 dimensions, if a flow at a constant velocity V is imposed on the diffusion the MSD takes the following form [2.74]:

ρ(τ) = 4Dτ + V2τ2.



(2.7)

Total Displacement [µm]

0.3

(b)

(a) 0.2 0.1 0 0

500

1000 1500 Time [ms]

2000

2500

(c)

(d)

50

50

100

100

150

50

100

150

200

250

300

350

150

50

100

150

200

250

300

350

Figure 2.4  (a) Plot of total displacement as a function of time in a diatom centroid measurement. The frames around the arrow are investigated in detail. (b) Overlay image of the diatom boundary for frames where the displacement was 293 nm at 1986 ms. White pixels show the overlapped points in the diatom boundary, while colors indicate change in boundary location; (c) and (d) show two consecutive frames of the diatom (diatom #2) for the displacement shown by an arrow in Figure 2.4 (a). The boundaries were calculated using the “Binary Centroid” algorithm. In b, green = frame c and magenta = frame d.

46  Diatom Gliding Motility Table 2.2  Drift velocity and diffusion coefficient of all three diatoms. Drift velocity was calculated using (1) MSD data, and (2) coordinates. Diatom volumes were estimated assuming rectangular prisms with 5 µm thickness, and equivalent diameters calculated as if they were spheres.

Diatom

Diffusion coefficient [µm2/s]

Drift velocity – MSD [µm/s]

Drift velocity – coordinates [µm/s]

Equivalent stokes diameter [µm]

Equivalent diatom diameter [µm]

Diatom #1

0.0915

2.57

2.54

2.65

11.6

Diatom #2

0.1951

8.68

8.28

1.24

11.8

Diatom #3

0.1162

3.14

2.93

2.08

8

Diffusion with a constant flow is usually described as Brownian motion with drift. We tested if diatom motion can be represented using the Brownian motion with drift model. For this purpose, the diatom trajectories were divided into n segments that are 500 frames long (n=6 for diatom #1, n=12 for diatom #2, n=11 diatom for diatom #3) and the MSD of each segment was calculated using a MATLAB routine [2.82]. The MSD data of all three diatoms are given in the supporting information (Appendix Figure A5). The average MSD curve was fit to Eq. (2.7) and best fit parameters (D and V) were extracted (Table 2.2). The best fit parameters and the average speed calculated using the initial and final coordinate points are tabulated in Table 2.2. There is a good agreement between the average speed found from MSD curve fitting and the average speed found using the coordinates. Thus, diatom motion can best be described as Brownian motion with drift. Distribution of the instantaneous velocities of the diatoms follows a Gaussian distribution. The large rare displacements observed for the diatoms are part of the Gaussian distribution. The rare explosive displacements do not generate a bimodal distribution. They are in the tails of the distribution. Autocorrelation of the velocity data of the diatom trajectory segments did not return any patterns in the signal (Appendix Figure A6), which could imply that each displacement event was independent of the others. The correlation was positive, which is indicative of a directed motion.

2.3.4 Diatom Orientation Angle Measurements The binary centroid algorithm in combination with second moment calculations (Eq. 2.5) were used to find the orientation angles versus time of the diatoms. The orientation angle of diatoms #1-3 were extracted as the diatoms traversed the imaging strip horizontally. The angular velocity was calculated using the orientation angle data. Figure 2.5 shows the angular velocity histogram of the diatoms. In the figure, rare explosive rotation with speeds over 5 rad/s of the diatoms are present. In order to elucidate the diatom rotation mechanism individual trajectories are investigated. In Figure 2.6 two representative orientation angle graphs of diatoms #2 and #3, as well as the orientation angle graph of diatom #4, which exhibited a rotation around its center without translation, are given. In Figures 2.6a, 2.6c, and 2.6e, fluctuations, smooth and steep changes in the orientation angle were present.

The Kinematics of Explosively Jerky Diatom Motility  47 105

Count Density

104 103 102 101 100 -20

-10

0 10 Angular Velocity [rad/s]

20

Figure 2.5  Angular velocity histogram of the diatoms #1-3 that were imaged at 821 fps. The diatoms exhibited changes in their orientation angles as they traversed the imaging strip horizontally.

Orientation [degree]

-6 -7

(a)

121 ms

-8 -9

230 ms (b)

-10 0

Orientation [degree]

20

500

1000 1500 Time [ms]

2000

0 ms

2500

(c) 1301 ms

15

(d)

10 0

500 100

Orientation [degree]

80

1000 1500 Time [ms]

2000

1269 ms

2500

t = 1439 ms

(e)

t = 409 ms

60 40 20

t = 349 ms

0 -20 -40 0

500

1000 Time [ms]

1500

(f)

t = 0 ms

Figure 2.6  In Figures 2.6a and 2.6c the orientation angle corresponding to the diatoms whose trajectories were previously investigated are shown (diatoms #2 and #3 in a and c, respectively). In Figure 2.6e the orientation angle of another diatom (diatom #4) is shown. Orientation angle is the angle between the major axis of the diatom and the x-axis. Figures (b), (d), and (f) are overlay images of the diatom boundary for different frames of the data shown in (a), (c), and (e), respectively. White pixels show the stationary points in diatom boundary, while colors indicate change in boundary location. An extra Gaussian smoothing was included in the image processing algorithm for the diatom #4, whose orientation angle is shown in Figure 2.6e. Otherwise, it was not possible to extract orientation data for diatom #4.

48  Diatom Gliding Motility In Figure 2.6a a relatively smoother change in orientation angle of the diatom is evident starting from t = 0 ms until t = 230 ms. The diatom boundaries in the frames pertaining to this smooth change are shown in Figure 2.4b by different colors. The change in the orientation angle occurred in 230 ms in this interval, where the change in the angle was 2.7°. This corresponds to an angular velocity of 0.206 rad/s. Here the changes in the orientation of the diatom occurred in combination with translation, as evident from Figure 2.6b, where the tangent angle to the trajectory varied with the orientation angle. In Figure 2.6c, the orientation angle for a recording of diatom #3 is plotted. A steep 6° change in the orientation angle of the diatom is evident at 1275 ms. The change occurred in 33 ms, where the corresponding angular velocity was 2.66 rad/s. The diatom boundaries pertaining to this motion are shown in Figure 2.6d. The white regions in Figure 2.6d denote stationary points during rotation while the colored lines indicate moving diatom boundaries. Here we show that rotation of the diatom while keeping a pivot point on the diatom is possible. Therefore, the diatom analyzed here had the ability to keep the tangent angle to the trajectory different from the orientation angle. In Figure 2.6e, a larger change of the orientation angle was recorded for diatom #4, where the angle changed from -30° to 80° in 1.5 s. The angular velocity corresponding to this motion is 1.28 rad/s. The frames pertaining to these steep changes were extracted and overlay images of the calculated diatom boundaries were formed (Figure 2.6f). Figures 2.6d and 2.6f indicate no translation during these steep changes in the orientation angle. The pivot point in Figure 2.6d was along the trailing end of diatom #3, while examination of Figure 2.6f indicates a pivot point in the central region of diatom #4. For Figure 2.6d, we can test whether or not the pivot is a single point, by assuming that it is the arc distance center of the contiguous white pixels. If the earlier of the two images is rotated about this point, as shown in Figure 2.7a, the boundary matches very well with the observed boundary. For the rotation about the center in Figure 2.6f, an analysis can be

t = 409 ms

t = 1439 ms

(a)

t = 349 ms

t = 0 ms

(b)

Figure 2.7  (a) The diatom in Figure 2.6d is rotated around the arc center of the white pixel region in Figure 2.6d; (b) the consecutive intersections of the major axis of the diatoms during rotation are shown as yellow dots. In (b), while the green dots at the centerlines are close to one another, they would coincide exactly if the rotation were precisely around the center of the diatom.

The Kinematics of Explosively Jerky Diatom Motility  49 made using the locations of the intersections of the diatom major axes. If the intersections coincided exactly at the same point, this point is the pivot point. However, we can see that there is some drift (Figure 2.7b) for the rotation shown in Figure 2.6f. Furthermore, another rotational mode of motion is observed occasionally with the motile diatoms. In this rotational motion, diatom rolls around the axis that is parallel to the direction of motion. This could also cause the projected area of the diatom to fluctuate. Generation of a torque with no translation is possible in two conditions: (1) distribution of forces, which create a pure moment with zero resultant force; (2) keeping a point stationary and applying a distributed force with a resultant force that is not collinear with the fixed point. Observations indicate the motion in Figures 2.6d and 2.6f could be a result of the second type of the torque generation, while the motion in Figure 2.6b could be a result of the first type of the torque generation. Furthermore, steep changes in the orientation angle indicate a mechanism similar to the one responsible for the explosive peaks in the diatom translation (Figures 2.3 and 2.4).

2.3.5 Is Diatom Motion Characterized by a Sequence of Small Explosive Movements? Moving diatoms secrete mucilage when adhered to a substrate [2.40] [2.49] [2.54], and this mucilage that is left behind as the diatom trail [2.24] is somehow responsible for diatom gliding. The explosive steps in diatom displacement might be due to elastic snapping of the mucilage fibrils left behind the diatom as the trail [2.24]. The negative displacements in the same observations might be due to the elastic recoil of the mucilage fibrils at the trail. Alternatively, the motion may be due to the hypothesized hydration and swelling of the raphe fibrils [2.40] [2.46], which may occur suddenly for individual fibrils. This could be analogous to the explosive discharge of discobolocysts in Ochromonas, which takes O(0.1 µsec), though these do not swell [2.39]. Temporal resolution of diatom motion due to individual fibrils would therefore require 108 frames per second, compared to 103 fps in this study. Individual fibrils are 50 to 100 nm in diameter [2.50]. A 10 µm long diatom would have O(105) fibrils along the length of its raphes. If it moves at a speed of 10 µm/sec there would be O(105) fibrils discharged per second. To pick up individual contributions we would need “only” O(106 fps), 1000 times faster than the camera we used. Although not mentioned in its caption, elastic snapping was also evident from a movie of diatom motion that was made to visualize the mucilage in the diatom trail using silica beads [2.87]. Unfortunately, the movie frame did not include the full diatom, and therefore, the particle tracking algorithms used in this study could not be used to extract diatom centroid data. However, we could use an automated feature detection and tracking algorithm that is available in MATLAB’s computer vision toolbox that implements the Speeded-Up Robust Features (SURF) algorithm [2.4]. This algorithm calculates the translation in between two frames by tracking the features detected by the algorithm. The features could be a portion of the diatom or the silica particles that were used to visualize the trail. The analysis returned peaks both for the diatom and the silica particles in the calculated displacement (Appendix Figure A7a,b). Thus, the explosive motion could be related to elastic

50  Diatom Gliding Motility snapping of diatom trail material as feature tracking only in this region resulted in peaks in the displacement. Using the diffusion coefficients given in Table 2.2 we were able to calculate the equivalent Stokes diameters of the three diatoms using the Stokes-Einstein equation [2.38] (Table 2.2). The diatoms exhibited normally distributed excursions around a mean drift. The equivalent diameter estimates the diameter of a sphere that would undergo the same size of excursions in Brownian motion. The Stokes diameters are much smaller than the equivalent diameters of spherical particles having the same volume as the diatoms. Consequently, it can be deduced that the excursions around the mean drift are much greater than what could be ascribed to Brownian motion. In other words, while diatom motility fits a model of drift plus Brownian motion, the excursions are far greater than can be ascribed to passive Brownian motion, i.e., excursions of that magnitude require active force generation. Furthermore, if a stationary diatom is given an instantaneous force that could be described by the unit impulse function in the low Reynolds number regime, its velocity will rise sharply and will decay following the impulse at a rate given by m/f, where m is the diatom mass and f is the drag constant (Appendix). The drag constant on a sphere in the low Reynolds number regime is given by the Stokes law: f = 6πμr, where μ is the dynamic viscosity and r is the sphere radius. Using the above equivalent spherical radius, the characteristic time for velocity decay (m/f) is O(10-5) s. Therefore, a temporal accuracy on the order of µs is needed to resolve acceleration and deceleration of diatoms by pulses. Another consequence of this fact is that the acceleration pattern of the diatom at 821 fps should be indicative of the motility force acting on the diatom. The results in Figures 2.2, 2.3, and 2.4 indicate rapid fluctuations in the velocity and acceleration profiles of the diatoms as opposed to a smooth velocity profile and thus imply a fluctuating force. The calculated spontaneous accelerations for the diatom at 821 fps are on the order of 5×105 µm/s2, which is approximately 5000 times larger than the values at 10 fps [2.28]. At 821 fps the diatom motion is still jerky.

2.3.6 Future Work We would, of course, like to know how diatoms generate the force by which they move so gingerly in a low Reynolds number regime. This will require getting beyond the kinematics to an analysis of the dynamics of their motion. Furthermore, in order to better analyze diatom forces, the rotational motion of the diatom around the axis parallel to the direction of motion should be better understood. We propose in later work to examine the motility response to some of the various physical and chemical regimes that are known to or may alter or interfere with diatom motility: 1. 2. 3. 4. 5.

Bending of a small rod by a moving diatom [2.46] Motion with or against a flow [2.1] Vertical motion subjected to gravity [2.34] [2.43] Motility in microgravity [2.41] [2.61] Motility while being centrifuged [2.11] [2.44] [2.45] [2.47] [2.51] [2.59]

The Kinematics of Explosively Jerky Diatom Motility  51 6. Manipulation with optical tweezers [2.64] [2.67] [2.70] [2.78] [2.80] [2.81]. While motile diatoms have been used as nonspherical probes, most are too strong to be constrained by optical tweezers [2.69]. 7. Manipulation with acoustic tweezers [2.6] [2.21] [2.37] [2.77] [2.85] 8. Manipulation with atomic force microscopy [2.36] 9. Temperature effects [2.15] [2.16] [2.17] [2.18] [2.26] 10. Effects of hydrophobicity and other chemical properties of the substrate [2.1] [2.2] [2.14] [2.20] [2.33] [2.35] [2.48] [2.52] [2.53] [2.58] [2.60] [2.62] [2.65] [2.79] [2.84] [2.86] [2.91] 11. Effects of patterned substrates [2.76] 12. Effects of vibrating substrates [2.57] 13. Effects of cytoskeletal and other inhibitors [2.31] [2.73] [2.88] As motile diatoms are involved in biofouling [2.53], such research also has practical implications. Once the dynamics is understood, the control of that dynamics, i.e., diatom behavior, would be open to quantitative analysis.

2.4 Conclusions In this study, diatom motion was observed at 1.21 ms temporal resolution and 9 nm spatial resolution, the latter using subpixel centroid estimating algorithms. We found that the jerky motion of diatoms, noted in 1753 [2.3] and observed by Lesley Edgars at 10 fps [2.28], is still jerky at our higher spatiotemporal resolution, and that the excursions are actively driven for the Navicula-like species tested here. There was a good fit between the Brownian motion with drift model and the diatoms’ displacement data. Peaks as large as 290 nm and negative (backwards) displacements were observed in the calculated diatom displacement graphs. The characteristic displacement of the diatoms associated with the Brownian motion is on the order of 10 nm in 1.21 ms, and equivalent Stokes diameters are much smaller than the equivalent spherical diameters of the diatoms. Therefore, the excursions around the mean drift are not due to passive Brownian motion. The calculated spontaneous accelerations for a diatom are up to 5×105 µm/s2. Explosive motion and negative displacement of diatoms might be related to elastic snapping and recoil of diatom trail material, respectively. In addition, jerky motion might be attributable to explosive release of raphe mucopolysaccharide material from the raphe to the diatom trail. The result is an “acceleration noise” far greater than that experienced with most human vehicles [2.56]. Thus, studies with even higher speed methods are warranted. It may be possible to distinguish elastic recoil from explosive release by analyzing the motion of particles that move along the raphe. For example, in a published movie two polystyrene beads attached to the canal raphe of Nitzschia make a large jump as they move along the raphe, well before they are released into the diatom trail [2.71]. While it is possible that the beads were snagged by nearby material, this observation is suggestive of explosive release, perhaps at the end of the raphe.

52  Diatom Gliding Motility Orientation angle of the diatoms was calculated in addition to the translation. Steep changes in the orientation angle were observed, which are possibly related to diatom chemotaxis [2.19]. Frame-by-frame examination of the rotation indicated pivot points in the diatoms, confirming previous results [2.7] [2.8] [2.9]. The versatility and degrees of freedom of diatom motion could one day be useful in developing micro autonomous robots and self-pumping fluids. A future study will investigate dynamics of diatom motion, and will test different theories of diatom motility. Some seeds upon drying are explosively dispersed [2.32], an analog to the catapult. Individual Tetrahymena and bacterial cells can explode, probably due to sudden release of turgor pressure [2.89]. Explosive discharge of a hot, reacting liquid is also observed in the bombardier beetle, through a nozzle of 200 µm diameter [2.5], which thus qualifies as a microfluidic version of a “flame thrower.” If diatoms move by explosive hydration of raphe mucopolysaccharide, they have harnessed multiple explosions for motility, rather than these single explosion mechanisms. The only similar device is the design for a spaceship propelled by multiple nuclear bombs [2.27]. As with the wheel [2.66] and the gun [2.39] and the above devices, this spaceship may have also been presaged in nature, by the diatom raphe.

Appendix Dynamics of Diatom in Low Reynolds Number Regime Newton’s second law for a diatom that is given a pulse at to with magnitude Fdiatom in the low dV + fV − Fdiatom δ(t − to ) = 0, where V is the diatom Reynolds number regime is given as: m dt velocity. This equation can be solved for an initially stationary diatom using the Laplace f Fdiatom − m (t −to ) e uto (t ), where transform. The expression for the velocity of the diatom is V (t ) = m u is the unit step function. to

Binary centroid

0.019 (0.0107)

0.0025 (0.023)

7.55x10-5 (0.0321)

1.03x10-5 (0.0171)

Distance moved

0/0 (1000)

0.1/10 (39)

0.5/50 (39)

1/100 (20)

0.0028 (0.0129) 0.0162 (0.0141) 0.0366 (0.0151)

3.41x10-4 (0.0196)

6.78x10-4 (0.0147)

0.0161 (0.0085)

Grayscale centroid

0.0025 (0.0245)

0.0247 (0.0134)

Binary centroid (MF)

2.39x10-4 (0.0094)

4.22x10-4 (0.0138)

0.001 (0.0116)

0.0153 (0.0081)

Grayscale centroid (10% TH)

5.34 x10-4 (0.0141)

3.34x10-4 (0.0148)

4.05x10-4 (0.016)

0.016 (0.0084)

Grayscale centroid (25% TH)

0.0011 (0.0184)

3.02x10-4 (0.006)

0.2804 (0.3248)

0.1121 (0.3662)

9.47x10-4 (0.016) 7.73x10-4 (0.0201)

0.0064 (0.0034)

Template correlation

0.0171 (0.0089)

Grayscale centroid (50% TH)

8.84x10-5 (0.006)

5.839x10-4 (0.0281)

2.816x10-4 (0.0086)

0.01211 (0.004)

Image registration

Table AI  The mean error of estimation and the standard deviation (in pixels) introduced by the particle tracking algorithms and variations in the algorithms applied to the numerical test object. The numbers in the first column give the input displacement in pixels/nm. The numbers in parentheses in the first column denote the number of simulated images. In all other columns the mean error and standard deviation (in parentheses) pertaining to different algorithms are given. (MF: Median Filter, TH: Threshold).

The Kinematics of Explosively Jerky Diatom Motility  53

54  Diatom Gliding Motility (a)

Count

30 20 10 0 -0.02

-0.01

0 0.01 x displacement [µm]

0.02

Count

30

(b)

20 10 0

-0.02

-0.01 0 0.01 y displacement [µm]

0.02

Figure A1  Displacement histograms of the image of the stationary polystyrene particle situated in the micromachined channel; (a) and (b) show x and y displacement histograms, respectively. Dots represent microscopy measurement while continuous lines are the Gaussian fits to the measurements. Two Gaussian terms are used for fitting total displacement data. The particle was imaged at 821 fps.

|Y(f )|

0.08

(a)

0.06 0.04 0.02 0

0

100

|Y(f )|

0.08

200 300 Frequency [Hz]

400 (b)

0.06 0.04 0.02 0

0

100

|Y(f )|

0.08

200 300 Frequency [Hz]

400 (c)

0.06 0.04 0.02 0

0

100

|Y(f )|

0.08

200 300 Frequency [Hz]

400 (d)

0.06 0.04 0.02 0

0

100

200 300 Frequency [Hz]

400

Figure A2  Fast Fourier Transform (FFT) of the x and y coordinates of the centroid location for the recordings made with the stationary polystyrene particle and a diatom; (a) and (c) are the FFTs of the x centroid data and (b) and (d) are the FFTs of y centroid data of the polystyrene particle and diatom #3, respectively. Prior to the FFT the centroid location mean was subtracted from the data.

y coordinate [µm]

The Kinematics of Explosively Jerky Diatom Motility  55 8

(a)

7 6

y coordinate [µm]

19

6

20

21

22

23 x coordinate [µm]

24

25

26

27

26

27

28

29 x coordinate [µm]

30

31

32

33

(b)

5 4 25

y coordinate [µm]

6

(c) 5 4 3

9

10

11

12 13 x coordinate [µm]

14

15

16

8 7 6

y coordinate [µm]

y coordinate [µm]

y coordinate [µm]

Figure A3  Representative trajectories of the centroids of diatoms #1, #2, and #3 are given in (a), (b), and (c), respectively, over 500 frames. The diatoms were imaged at 821 fps.

(a) 20.6

20.8 21 x coordinate [µm]

21.2

5 (b) 4

5 4

26

26.5 x coordinate [µm]

27

9.4

9.6 9.8 x coordinate [µm]

10

(c)

9.2

10.2

Figure A4  Zoomed in views of the large displacements in the representative trajectories given in Figure A3. The plots (a), (b), and (c) show the centroids of diatoms #1, #2, and #3, respectively.

56  Diatom Gliding Motility 5

7 (a)

6

4

(b)

3

MSD (µm2)

MSD (µm2)

5

2

4 3 2

1 0

1 0

0.2

0.4 Delay (s)

0 0

0.6

0.2

0.4 Delay (s)

0.6

50 (c)

MSD (µm2)

40 30 20 10 0

0

0.2

0.4 Delay (s)

0.6

Figure A5  Mean square displacement (MSD) data of diatom trajectories. Diatom motion recordings were divided into segments that are 500 frames long and MSD data of each segment was calculated. The black dashed line represents the mean MSD and red line represents the best fit to the MSD for all diatoms (diatom #1 – (a), diatom #2 – (b), diatom #3 – (c)). All MSD data falls into the range shown by gray shaded areas in the figures.

The Kinematics of Explosively Jerky Diatom Motility  57 1 (a)

0.8 0.6 0.4 0.2 0 -0.2 -0.4

0

0.2

0.4 Delay (s)

0.6

0.2

0.4 Delay (s)

0.6

Normalized velocity autocorrelation

Normalized velocity autocorrelation

1

(c)

0.8 0.6 0.4 0.2 0 -0.2 -0.4

0

0.2

0.4 Delay (s)

0.6

1 Normalized velocity autocorrelation

(b) 0.8 0.6 0.4 0.2 0 0

Figure A6  Normalized velocity autocorrelation of all the diatoms (diatom #1 – (a), diatom #2 – (b), diatom #3 – (c)). The same trajectory segments as in calculation of diatom MSD data were used. All repetitions were used in calculating the autocorrelation.

58  Diatom Gliding Motility 25

(a)

Displacement [pixel]

20

15

10

5

0

0

100

200

300 400 Frame Number

500

600

Matching points (inliers only) ptsOriginal ptsDistorted

(b)

Diatom

Silica Particles

Figure A7  The publicly available video [2.87] was analyzed using the SURF feature detection algorithm (please see the main text for the details); (a) displacement calculated in pixels as a function of the frame number; (b) the overlay of the features detected at frame numbers 422 (crosses) and 423 (circles). pts = points. Scale bar was not present in the original video.

The Kinematics of Explosively Jerky Diatom Motility  59

References [2.1] Alles, M. and Rosenhahn, A., Microfluidic detachment assay to probe the adhesion strength of diatoms. Biofouling, 31, 5, 469–480, 2015. [2.2] Arce, F.T., Avci, R., Beech, I.B., Cooksey, K.E., Wigglesworth-Cooksey, B., A live bioprobe for studying diatom-surface interactions. Biophys. J., 87, 6, 4284–4297, 2004. [2.3] Baker, H., Employment for the Microscope. In Two Parts: I. An examination of Salts and Saline Substances, their amazing Configurations and Crystals, as formed under the Eye of the Observer. II. An Account of various Animalcules never before described, and of many other Microscopical Discoveries, R. Dodsley, London, England, 1753. [2.4] Bay, H., Ess, A., Tuytelaars, T., Van Gool, L., Speeded-Up Robust Features (SURF). Comput. Vis. Image Und., 110, 3, 346–359, 2008. [2.5] Beheshti, N. and McIntosh, A.C., A biomimetic study of the explosive discharge of the bombardier beetle. Int. J. Des. Nat., 1, 1, 61–69, 2007. [2.6] Bernassau, A.L., MacPherson, P.G.A., Beeley, J., Drinkwater, B.W., Cumming, D.R.S., Patterning of microspheres and microbubbles in an acoustic tweezers. Biomed. Microdevices, 15, 2, 289–297, 2013. [2.7] Bertrand, J., Mouvements des diatomées. V. Le pivotement polaire de Gomphonema acuminatum Ehrenberg [Movements of diatoms V. The median polar pivoting of Gomphonema acuminatum Ehrenherg]. Ann. Limnol.-Int. J. Lim., 33, 4, 211–222, 1997. [2.8] Bertrand, J., Mouvements des diatomées VIII: synthèse et hypothèse [Diatom movements VIII: Synthesis and hypothesis, French]. Diatom Res., 23, 1, 19–29, 2008. [2.9] Bertrand, J., Diatom movements VIII: Synthesis and hypothesis. 2019. Translation of: Bertrand, J., Mouvements des diatomées VIII: synthèse et hypothèse. Diatom Res., 23, 1, 19–29, 2008, by: Gordon, R., Laviale, M., Serieyssol, K.K., in consultation with, Jean Bertrand, In: Diatom Gliding Motility [DIGM, Volume in the series: Diatoms: Biology & Applications, series editors: Richard Gordon & Joseph Seckbach], S.A. Cohn, K.M. Manoylov, R. Gordon (Eds.), Wiley-Scrivener, Beverly, MA, USA, 23, 19–29. [2.10] Canny, J., A computational approach to edge-detection. IEEE Trans. Pattern Anal. Mach. Intell., 8, 6, 679–698, 1986. [2.11] Chaen, S., Inoue, J., Sugi, H., Force-velocity relation in the sliding movement of Chara myosin coated beads on actin cables studied with a centrifuge microscope. J. Muscle Res. Cell Motil., 13, 4, 483, 1992. [2.12] Cheang, U.K. and Kim, M.J., Self-assembly of robotic micro- and nanoswimmers using magnetic nanoparticles. J. Nanopart. Res., 17, 3, 145, 2015. [2.13] Cheezum, M.K., Walker, W.F., Guilford, W.H., Quantitative comparison of algorithms for tracking single fluorescent particles. Biophys. J., 81, 4, 2378–2388, 2001. [2.14] Cho, Y., Sundaram, H.S., Finlay, J.A., Dimitriou, M.D., Callow, M.E., Callow, J.A., Kramer, E.J., Ober, C.K., Reconstruction of surfaces from mixed hydrocarbon and PEG components in water: responsive surfaces aid fouling release. Biomacromolecules, 13, 6, 1864–1874, 2012. [2.15] Cohn, S.A. and Disparti, N.C., Analysis of environmental-influences on diatom cell motility. Mol. Biol. Cell, 3, Suppl. S, A361, 1992. [2.16] Cohn, S.A., Farrell, J., Munro, J., Rauschenberg, C., Schulze, J., The effect of light and temperature on diatom aggregation and adhesion. Mol. Biol. Cell, 11, 1980, 2000. [2.17] Cohn, S.A., Farrell, J.F., Munro, J.D., Ragland, R.L., Weitzell, R.E., Wibisono, B.L., The effect of temperature and mixed species composition on diatom motility and adhesion. Diatom Res., 18, 2, 225–243, 2003. [2.18] Cohn, S.A., Zelner, D., Crea, J., Wibisono, B., Silverman, M., Analysis of diatom motility using light avoidance and fluorescent bend assays. Mol. Biol. Cell, 10, Suppl., 1527, 1999.

60  Diatom Gliding Motility [2.19] Cooksey, K.E. and Wigglesworthcooksey, B., Adhesion of bacteria and diatoms to surfaces in the sea: a review. Aquat. Microb. Ecol., 9, 1, 87–96, 1995. [2.20] Cordeiro, A.L., Pettit, M.E., Callow, M.E., Callow, J.A., Werner, C., Controlling the adhesion of the diatom Navicula perminuta using poly(N-isopropylacrylamide-co-N-(1-phenylethyl) acrylamide) films. Biotechnol. Lett., 32, 4, 489–495, 2010. [2.21] Courtney, C.R.P., Demore, C.E.M., Wu, H., Grinenko, A., Wilcox, P.D., Cochran, S., Drinkwater, B.W., Independent trapping and manipulation of microparticles using dexterous acoustic tweezers. Appl. Phys. Lett., 104, 15, 154103, 2014. [2.22] Crawford, R., Round, F., Pickett-Heaps, J., Johnson, A., Obituary: Lesley Ann Edgar (19552006). Diatom Res., 22, 1, 237–240, 2007. [2.23] Crawford, S., Chiovitti, T., Pickett-Heaps, J., Wetherbee, R., Micromorphogenesis during diatom wall formation produces siliceous nanostructures with different properties. J. Phycol., 45, 6, 1353–1362, 2009. [2.24] Drum, R.W. and Hopkins, J.T., Diatom locomotion: an explanation. Protoplasma, 62, 1, 1–33, 1966. [2.25] Drum, R.W. and Pankratz, H.S., Pyrenoids, raphes and other fine structure in diatoms. Am. J. Bot., 51, 405–418, 1964. [2.26] Du, G.Y., Li, W.T., Li, H., Chung, I.K., Migratory responses of benthic diatoms to light and temperature monitored by chlorophyll fluorescence. J. Plant Biol., 55, 2, 159–164, 2012. [2.27] Dyson, G., Project Orion: The True Story of the Atomic Spaceship, Henry Holt and Company, New York, NY, 2003. [2.28] Edgar, L.A., Diatom locomotion: computer assisted analysis of cine film. Br. Phycol. J., 14, 1, 83–101, 1979. [2.29] Edgar, L.A., Mucilage secretions of moving diatoms. Protoplasma, 118, 44–48, 1983. [2.30] Edgar, L.A. and Pickett-Heaps, J.D., The mechanism of diatom locomotion. I. An ultrastructural study of the motility apparatus. Proc. R. Soc B: Biol. Sci., 218, 331–343, 1983. [2.31] Edgar, L.A. and Zavortink, M., The mechanism of diatom locomotion. II. Identification of actin. Proc. R. Soc B: Biol. Sci., 218, 345–348, 1983. [2.32] Evangelista, D., Hotton, S., Dumais, J., The mechanics of explosive dispersal and self-burial in the seeds of the filaree, Erodium cicutarium (Geraniaceae). J. Exp. Biol., 214, 4, 521–529, 2011. [2.33] Finlay, J.A., Callow, M.E., Ista, L.K., Lopez, G.P., Callow, J.A., The influence of surface wettability on the adhesion strength of settled spores of the green alga Enteromorpha and the diatom Amphora. Integr. Comp. Biol., 42, 6, 1116–1122, 2002. [2.34] Frankenbach, S., Pais, C., Martinez, M., Laviale, M., Ezequiel, J., Serôdio, J., Evidence for gravitactic behaviour in benthic diatoms. Eur. J. Phycol., 49, 4, 429–435, 2014. [2.35] Gawne, B., Wang, Y., Hoagland, K.D., Gretz, M.R., Role of bacteria and bacterial exopolymer in the attachment of Achnanthes longipes (Bacillariophyceae). Biofouling, 13, 2, 137–156, 1998. [2.36] Gebeshuber, I.C., Thompson, J.B., Del Amo, Y., Stachelberger, H., Kindt, J.H., In vivo nanoscale atomic force microscopy investigation of diatom adhesion properties. Mater. Sci. Technol., 18, July, 763–766, 2002. [2.37] Gesellchen, F., Bernassau, A.L., Dejardin, T., Cumming, D.R.S., Riehle, M.O., Cell patterning with a heptagon acoustic tweezer - application in neurite guidance. Lab. Chip, 14, 13, 2266–2275, 2014. [2.38] Gillespie, D.T. and Seitaridou, E., Simple Brownian Diffusion: An Introduction to the Standard Theoretical Models, OUP Oxford, UK, 2012. [2.39] Gordon, R., A retaliatory role for algal projectiles, with implications for the mechanochemistry of diatom gliding motility. J. Theor. Biol., 126, 419–436, 1987.

The Kinematics of Explosively Jerky Diatom Motility  61 [2.40] Gordon, R. and Drum, R.W., A capillarity mechanism for diatom gliding locomotion. Proc. Natl. Acad. Sci. U. S. A., 67, 1, 338–344, 1970. [2.41] Gordon, R., Hoover, R.B., Tuszynski, J.A., de Luis, J., Camp, P.J., Tiffany, M.A., Nagy, S.S., Fayek, M., Lopez, P.J., Lerner, B.E., Diatoms in space: Testing prospects for reliable diatom nanotechnology in microgravity. Proc. SPIE, 6694, V1–V15, 2007. [2.42] Gow, R.D., Renshaw, D., Findlater, K., Grant, L., McLeod, S.J., Hart, J., Nicol, R.L., A comprehensive tool for modeling CMOS image-sensor-noise performance. IEEE Trans. Electron Devices, 54, 6, 1321–1329, 2007. [2.43] Häder, D.-P., Hemmersbach, R., Lebert, M., Gravity and the Behavior of Unicellular Organisms, Cambridge University Press, New York, 2005. [2.44] Hall, K., Cole, D., Yeh, Y., Baskin, R.J., Kinesin force generation measured using a centrifuge microscope sperm-gliding motility assay. Biophys. J., 71, 6, 3467–3476, 1996. [2.45] Hall, K.W., Cole, D.G., Yeh, Y., Scholey, J.M., Baskin, R.J., Force generation in kinesin measured in a centrifuge microscope-based motility assay. Biophys. J., 68, 4 Suppl, 71S, 1995. [2.46] Harper, M.A. and Harper, J.T., Measurements of diatom adhesion and their relationship with movement. Br. Phycol. Bull., 3, 2, 195–207, 1967. [2.47] Hemmersbach-Krause, R., Briegleb, W., Häder, D.P., Swimming behavior of Paramecium– first results with the low-speed centrifuge microscope (NIZEMI). Adv. Space Res., 12, 1, 113–116, 1992. [2.48] Henriques Vieira, A.A., Coelho Ortolano, P.I., Giroldo, D., Dellamano Oliveira, M.J., Bittar, T.B., Lombardi, A.T., Sartori, A.L., Role of hydrophobic extracellular polysaccharide of Aulacoseira granulata (Bacillariophyceae) on aggregate formation in a turbulent and hypereutrophic reservoir. Limnol. Oceanogr., 53, 5, 1887–1899, 2008. [2.49] Higgins, M.J., Crawford, S.A., Mulvaney, P., Wetherbee, R., The topography of soft, adhesive diatom ‘trails’ as observed by atomic force microscopy. Biofouling, 16, 2-4, 133–139, 2000. [2.50] Higgins, M.J., Molino, P., Mulvaney, P., Wetherbee, R., The structure and nanomechanical properties of the adhesive mucilage that mediates diatom-substratum adhesion and motility. J. Phycol., 39, 6, 1181–1193, 2003. [2.51] Hiramoto, Y. and Kamitsubo, E., Centrifuge microscope as a tool in the study of cell motility, in: International Review of Cytology - A Survey of Cell Biology, vol. 157, pp. 99–128, 1995. [2.52] Hodson, O.M., Monty, J.P., Molino, P.J., Wetherbee, R., Novel whole cell adhesion assays of three isolates of the fouling diatom Amphora coffeaeformis reveal diverse responses to surfaces of different wettability. Biofouling, 28, 4, 381–393, 2012. [2.53] Holland, R., Dugdale, T.M., Wetherbee, R., Brennan, A.B., Finlay, J.A., Callow, J.A., Callow, M.E., Adhesion and motility of fouling diatoms on a silicone elastomer. Biofouling, 20, 6, 323–329, 2004. [2.54] Hopkins, J.T., The diatom trail. Microscopy, 30, 209–217, 1967. [2.55] Hunyadi, L., Estimation methods in the errors-in-variables context [PhD dissertation], Budapest University of Technology and Economics, Budapest, Hungary, 2013. [2.56] Jones, T.R. and Potts, R.B., The measurement of acceleration noise - a traffic parameter. Oper. Res., 10, 6, 745–763, 1962. [2.57] Kanda, K., Matsuda, T., Oka, T., Orientation response of vascular cells to periodic stretch. Jpn. J. Artif. Organs, 22, 2, 483–487, 1993. [2.58] Klein, G.L., Pierre, G., Bellon-Fontaine, M.N., Zhao, J.M., Breret, M., Maugard, T., Graber, M., Marine diatom Navicula jeffreyi from biochemical composition and physico-chemical surface properties to understanding the first step of benthic biofilm formation. J. Adhes. Sci. Technol., 28, 17, 1739–1753, 2014. [2.59] Kuroda, K. and Kamiya, N., Behavior of cytoplasmic streaming in Nitella during centrifugation as revealed by the television centrifuge-microscope. Biorheology, 18, 3-6, 633–641, 1981.

62  Diatom Gliding Motility [2.60] Leflaive, J. and Ten-Hage, L., Effects of 2E,4E-decadienal on motility and aggregation of diatoms and on biofilm formation. Microb. Ecol., 61, 363–373, 2010. [2.61] Lewis, M.L., Cellular Responses to Low-Gravity: Pilot Studies on Suborbital Rockets and Orbiting Spacecraft [NASA-CR-194402], Consortium for Materials Development in Space, University of Alabama, Huntsville, 1993. [2.62] Li, Y., Gao, Y.H., Li, X.S., Yang, J.Y., Que, G.H., Influence of surface free energy on the adhesion of marine benthic diatom Nitzschia closterium MMDL533. Colloids Surf. B-Biointerfaces, 75, 2, 550–556, 2010. [2.63] Mann, D.G., Chloroplast morphology, movements and inheritance in diatoms, in: Cytology, Genetics and Molecular Biology of Algae, B.R. Chaudhary and S.B. Agrawal (Eds.), pp. 249– 274, SPB Academic Publishing, Amsterdam, 1996. [2.64] Merola, F., Miccio, L., Memmolo, P., Di Caprio, G., Coppola, G., Netti, P., Ferraro, P., 3D manipulation and visualization of in-vitro cells by optical tweezers and digital holographic microscopy. Proc. SPIE, 8947, 89471a, 2014. [2.65] Molino, P.J., Hodson, O.A., Quinn, J.F., Wetherbee, R., The quartz crystal microbalance: a new tool for the investigation of the bioadhesion of diatoms to surfaces of differing surface energies. Langmuir, 24, 13, 6730–6737, 2008. [2.66] Mussill, M. and Jarosch, R., Bacterial flagella rotate and do not contract. Protoplasma, 75, 4, 465–469, 1972. [2.67] Olof, S.N., Grieve, J.A., Phillips, D.B., Rosenkranz, H., Yallop, M.L., Miles, M.J., Patil, A.J., Mann, S., Carberry, D.M., Measuring nanoscale forces with living probes. Nano Lett., 12, 11, 6018–6023, 2012. [2.68] Patwardhan, A., Subpixel position measurement using 1D, 2D and 3D centroid algorithms with emphasis on applications in confocal microscopy. J. Microsc.-Oxford, 186, 246–257, 1997. [2.69] Phillips, D.B., Applications of Closed-Loop Feedback Control with Holographic Optical Tweezers [PhD Thesis], H. H. Wills Physics Laboratory, University of Bristol, Bristol, UK, 2012. [2.70] Phillips, D.B., Simpson, S.H., Grieve, J.A., Gibson, G.M., Bowman, R., Padgett, M.J., Miles, M.J., Carberry, D.M., Position clamping of optically trapped microscopic non-spherical probes. Opt. Express, 19, 21, 20622–20627, 2011. [2.71] Pickett-Heaps, J.D. and Pickett-Heaps, J., Diatoms: Life in Glass Houses [DVD], Cytographics, 2003. [2.72] Popkin, G., The physics of life. Nature, 529, January 7, 16–18, 2016. [2.73] Poulsen, N.C., Spector, I., Spurck, T.P., Schultz, T.F., Wetherbee, R., Diatom gliding is the result of an actin-myosin motility system. Cell Motil. Cytoskeleton, 44, 1, 23–33, 1999. [2.74] Qian, H., Sheetz, M.P., Elson, E.L., Single particle tracking. Analysis of diffusion and flow in two-dimensional systems. Biophys. J., 60, 4, 910–921, 1991. [2.75] Round, F.E., Crawford, R.M., Mann, D.G., The Diatoms, Biology & Morphology of the Genera, Cambridge University Press, Cambridge, 1990. [2.76] Scardino, A.J., Zhang, H., Cookson, D.J., Lamb, R.N., de Nys, R., The role of nano-roughness in antifouling. Biofouling, 25, 8, 757–767, 2009. [2.77] Shi, J., Ahmed, D., Mao, X., Lin, S.-C.S., Lawit, A., Huang, T.J., Acoustic tweezers: patterning cells and microparticles using standing surface acoustic waves (SSAW). Lab. Chip, 9, 20, 2890–2895, 2009. [2.78] Sonek, G.J., Liu, Y., Iturriaga, R.H., Spectral fluorescence and scattering of cyanobacteria and diatoms held by optical tweezers. Proc. SPIE, 2258, 568–574, 1994. [2.79] Stanley, M.S. and Callow, J.A., Whole cell adhesion strength of morphotypes and isolates of Phaeodactylum tricornutum (Bacillariophyceae). Eur. J. Phycol., 42, 2, 191–197, 2007.

The Kinematics of Explosively Jerky Diatom Motility  63 [2.80] Tanaka, Y., Hirano, K., Nagata, H., Ishikawa, M., Real-time three-dimensional orientation control of non-spherical micro-objects using laser trapping. Electron. Lett., 43, 7, 412–414, 2007. [2.81] Tanaka, Y., Kawada, H., Hirano, K., Ishikawa, M., Kitajima, H., Non-contact micromanipulation system with computer vision. 4th European Conference of the International Federation for Medical and Biological Engineering, vol. 22(1-3), pp. 2400–2404, 2009. [2.82] Tarantino, N., Tinevez, J.Y., Crowell, E.F., Boisson, B., Henriques, R., Mhlanga, M., Agou, F., Israel, A., Laplantine, E., TNF and IL-1 exhibit distinct ubiquitin requirements for inducing NEMO-IKK supramolecular structures. J. Cell Biol., 204, 2, 231–245, 2014. [2.83] Thévenaz, P., Ruttimann, U.E., Unser, M., A pyramid approach to subpixel registration based on intensity. IEEE Trans. Image Process., 7, 1, 27–41, 1998. [2.84] Thompson, S.E.M., Taylor, A.R., Brownlee, C., Callow, M.E., Callow, J.A., The role of nitric oxide in diatom adhesion in relation to substratum properties. J. Phycol., 44, 4, 967–976, 2008. [2.85] Tran, S.B.Q., Marmottant, P., Thibault, P., Fast acoustic tweezers for the two-dimensional manipulation of individual particles in microfluidic channels. Appl. Phys. Lett., 101, 11, 114103, 2012. [2.86] Umemura, K., Yamada, T., Maeda, Y., Kobayashi, K., Kuroda, R., Mayama, S., Regulated growth of diatom cells on self-assembled monolayers. J. Nanobiotechnol., 5, 2, 2007. [2.87] Wetherbee, R., Diatom Trail Formation [movie], 2004, http://www.botany.unimelb.edu.au/ RW/media/trails.html. [2.88] Wetherbee, R., Lind, J.L., Poulsen, N.C., Spurck, T.P., Cell motility in marine raphid diatoms appears to be actin-based. Mol. Biol. Cell, 7, Suppl., 560a, 1996. [2.89] Wheatley, D., Undeniable cell demise by explosion. Cell Biol. Int., 27, 7, 503–505, 2003. [2.90] Wikipedia, Airy disk https://en.wikipedia.org/wiki/Airy_disk. https://en.wikipedia.org/ wiki/Airy_disk, 2016. [2.91] Willis, A., Pacifico, J., Dugdale, T.M., Wetherbee, R., Characterisation of the adhesion of fouling diatoms onto test surfaces. Diatom Res., 22, 2, 457–471, 2007. [2.92] Young, I.T., Quantitative microscopy. Eng. Med. Biol. Mag., IEEE, 15, 1, 59–66, 1996.

3 Cellular Mechanisms of Raphid Diatom Gliding Yekaterina D. Bedoshvili* and Yelena V. Likhoshway Limnological Institute, Russian Academy of Sciences, Irkutsk, Russia

Abstract

This brief review is devoted to a discussion of the possible gliding mechanisms of raphid pennate diatoms. This locomotion method is carried out without the participation of flagella, cilia and any deformations of the plasmalemma. One of the main hypotheses considered in recent years assumes participation of the actin-myosin system in the gliding. Such a system would require a connection by transmembrane components connecting with polymeric substances excreted through the raphe, and such transmembrane complexes for diatoms have not yet been directly observed, in contrast to a similar type of motility system in the parasitic protozoa Apicomplexa. However, there is little doubt that the actin-myosin system of the diatoms is required for motility, at the very least for transporting mucilage-containing vesicles that are essential and necessary for successful adhesion and further gliding of diatom cells along a substrate. Clearly, elucidating the mechanism of gliding will require the understanding of various chemical and mechanical properties of the mucilage, as well as in better determining the control systems for the secretion of mucilage strands and the factors that regulate them. Keywords:  Gliding, pennate, secretion of mucilage, vesicle transport, actin, myosin

3.1 Introduction Diatoms are successful and ubiquitous microalgae, interesting for their silica frustules containing intricate micro- and nano-structures. Many of the features of the diatoms are closely related to the silica frustules protecting their cells from mechanical damage, including being eaten by consumers [3.29]. Over millions of years of evolution, diatoms have adapted to some inconveniences of life in “glass houses.” A distinctive feature of many diatoms is their ability to attach to a substrate and glide smoothly over a surface; the gliding is mainly characteristic of the diverse group of pennate raphid diatoms. The valves of raphid diatoms are distinguished by their bilateral symmetry and the presence of a special slit on the valve, the raphe. During locomotion, raphid diatom cells leave trails of adhesive substances behind the moving cells [3.17] [3.22] [3.35] [3.36] [3.49] [3.50]. It is known that many species are able to move along the substrate at a fairly high speed (2-6 µm/s [3.5] [3.28]) without the participation of flagella, cilia and any cell membrane deformities. *Corresponding author: [email protected] Stanley Cohn, Kalina Manoylov and Richard Gordon (eds.) Diatom Gliding Motility, (65–76) © 2021 Scrivener Publishing LLC

65

66  Diatom Gliding Motility Since 1966 [3.17], attempts have been made to explain the mechanism of diatom locomotion. According to one of the main existing hypotheses [3.22] [3.50], the gliding of diatoms is a modification of a more generalized type of cellular motility, carried out via the force generation of myosin over actin microfilaments. Under this hypothesis it is assumed that the vesicles located in the cytoplasm secrete mucus into the raphe slit. This mucilage, comprising both polysaccharides and proteoglycans, is released as strands that attach cells to the substrate. Such extracellular mucilage fibers, extending through the raphe and connected to transmembrane intermediary complexes, would interact with actin cables via myosin, creating a force for cell movement by pulling on the adhering fibers. Gliding motility, in general, is typical for a wide variety of cells [3.31] mainly parasitic microorganisms apicomplexa, chlamydomonads and even many bacteria [3.41]. The mechanism of apicomplexan gliding, actively studied in recent years, also requires a combination of cellular adhesion and force generation using actin-myosin. The shape and integrity of cells’ apicomplexan depend on the intricate network of the cytoskeleton, which supports the trilaminar membranous pellicle consisting of the plasma membrane and a subjacent platform of one or more flattened vesicles, the alveolar membranes. Together with the cytoskeleton, alveolar membranes make up the internal membrane complex of the pellicle (Figure 3.1b [3.18] [3.16] [3.17]; Figure 3.1bii [3.15]) – with clearly visible structures directly under the plasmalemma, associated with its special structures. This structure is required for anchoring the gliding motor complex (for review see [3.32]). The structure of the inner membrane complex is supported by a subpellicular network containing, at least partially, proteins from the alveolin family [3.27]. More recently, several new families of proteins were identified in inner membrane complex of Apicomplexa [3.4] [3.8] [3.26]. Apicomplexa are able to secrete glycoproteins for cell adhesion to the substrate [3.9], and the gliding is an actin-myosin dependent process, as is the case for diatoms. The actin-myosin complex apicomplexan is referred to as the glycosome [3.45] and for Toxoplasma two proteins, TgGAP45 and TgGAP50, have been identified, presumably as components of the complex, linking integral membrane (a)

(b) Chl

Chl

m

Figure 3.1  Vesicles carrying mucilage (black arrows) on sections of cells of Pleurosigma sp. near the girdle bands (a) and Encyonema ventricosum (b) in the area of areolae (TEM). Around the cells Encyonema ventricosum are visible fibers between the plasmalemma and the frustule and a dense layer of mucilage fibers on the surface of the frustule, however, the connections between them are absent. Chl - chloroplasts; m mitochondrion. Scale – 200 nm.

Cellular Mechanisms of Raphid Diatom Gliding  67 glycoproteins and myosin [3.25]. These proteins, as well as their orthologs, found in related apicomplexes, constitute a family of gliding associated proteins (GAP). The model of apicomplexan locomotion, including the molecular mechanism for anchoring glycoproteins and the cytoskeleton through the cell membrane, can be considered well supported. However, for diatoms any transmembrane complexes that could connect adherent mucilage fibers to the cytoskeleton and glycoproteins on the plasmalemma have yet to be described. Despite many years of studies of diatoms cytology, including analysis of structures under the plasma membrane, such transmembrane connections which could be responsible for gliding have not been observed. We believe that in this regard it is necessary to outline and discuss the existing gaps in our knowledge, as well as possible shortcomings of the existing hypothesis, so that ideas surrounding diatom gliding could be better revised and supplemented as new experiments are developed.

3.2 Gliding and Secretion of Mucilage The movement of raphid diatoms over a substratum is associated with abundant mucilage or more complex substance secretion [3.20] [3.39]. As soon as the diatom cell comes in contact with substratum, cellular adhesion occurs and the cell begins gliding in a direction oriented by the raphe [3.50]. Studies of the mechanical properties of mucilage strands have shown that diatom cells are able to secrete at least two types of mucilage with various mechanical properties. Using atomic force microscopy, it was shown that the soft outer mucous layer of Craspedostauros australis revealed an adhesive force of 3.58 nN, while the adhesive mucilage released through the raphe as filaments were found to be resistant to breaking forces up to 60 nN [3.34]. The combination of carbohydrate chemical analysis of polysaccharide polymers and atomic force microscopy of Pinnularia viridis (Nitzsch) Ehrenberg showed that the compositional differences are associated with significant changes in the morphology and properties of mucilage [3.10]. Such studies show the extreme diversity of the secreted substances by diatoms and offer inspiring perspectives of investigations of their locomotion. The mucilage secretion occurs due to the exocytosis of the mucilage through specialized vesicles. It is still unclear whether there are stocks of these vesicles in the cells, and the secretion of their contents regulated as necessary, or whether mucilage synthesis and exocytosis occur continuously. However, using transmission electron microscopy, it has been shown that under the plasmalemma in the cytoplasm of diatoms capable of gliding, there are a large number of vesicles filled with electron-dense content. Often on some ultrathin sections it is clear that the contents of the vesicles appear to have some structural organization [3.21] [3.22] [3.28] (Figure 3.1a). Most likely, the secreted fibers are synthesized and simultaneously packed inside the vesicle, and their deployment occurs after the vesicle fuses with the plasmalemma and the content is released outside the cell, or in the space between the plasmalemma and the frustule. However, the localization of vesicle formation and secretion during this process is unclear. Many complex matrix polysaccharides of the plant cell wall [3.16] and secreted substances transported by vesicles [3.24] are synthesized in the Golgi apparatus, with the exception of cellulose, which is synthesized on the plasmalemma, and glycoproteins, the protein frameworks of which are generated in the endoplasmic reticulum [3.47]. A number of enzymes

68  Diatom Gliding Motility involved in the synthesis of polysaccharides are known for diatoms [3.3] [3.37]; however, specific studies regarding the manners of synthesis of secreted mucilage in diatoms have yet to be carried out. Considering that many of the extracellular polymeric substances secreted by diatoms can be a proteoglycan [3.10] [3.19] [3.39], the main synthesis pathways and vesicle formation (through the Golgi or the endoplasmic reticulum) remain unclear.

3.3 Cell Mechanisms of Mucilage Secretion Exocytosis and subsequent secretion of mucilage strands into the environment outside the frustule can occur not only in the raphe area, but also in the areolae (Figure 3.1b) [3.5]. The morphology of mucilage secreted through the raphe may differ significantly in different species [3.5] [3.35]. Mucilage strands of Pinnularia viridis [3.23] have the form of thick filaments, while mucilage secreted through the frustule of Encyonema ventricosum is presented

(a)

(b)

(c)

(d)

Figure 3.2  The surface cells of Nitzschia sp. (a, b) and Pleurosigma sp. (c, d) after removal of frustules (SEM). At the tips of the precisely repeated contour of the frustules in the area of the raphe system of Nitzschia sp. in some cases separated fragments (arrows) are visible; it may be the place where the cell is firmly attached to the valve. The frustule of Pleurosigma does not have so extremely relief, but the contours of the cell here is very accurately repeated especially in the area of the raphe (arrows). It is possibly that most of the surface is not diatotepum, as this polysaccharide layer is very firmly attached to the leaf, which is very well illustrated by the E. ventricosum (Figure 3.1b). Scale: (a, d) – 10 µm; (b) – 5 µm; (c) – 1 µm.

Cellular Mechanisms of Raphid Diatom Gliding  69 in the form of small isolated fragments [3.5]). It is not yet clear whether these differences are due to species diversity in mucilage chemical composition, or due to the characteristic cell sizes or raphe shape of the studied species. Interestingly, plasmalemma and an organic coating (diatotepum) line the inner surface of the frustule and precisely repeat its shape (Figure 3.2). For an illustration of this phenomenon, cell cultures of Pleurosigma sp. and Nitzschia sp. were fixed by glutaraldehyde and after further treatment dried at a critical point. With the help of double-sided carbon tape, the silica valves were removed from the cell surfaces. Scanning electron microscopy of the obtained samples showed no noticeable signs of mucilage strands associated with the plasmalemma and diatotepum. Moreover, neither transmission nor scanning electron microscopy have shown any evidence of appreciable mucilage strands passing through the silica frustule that connect with the underlying cell membrane.

(a)

(b)

Figure 3.3  Secretion of mucilage (arrow) through the raphe of the E. ventricosum (a) and vesicle (arrow), containing mucus near the raphe of Pleurosigma sp. (b) on ultra-thin cross sections (TEM) Scale: (a) – 200 nm; (b) – 500 nm.

70  Diatom Gliding Motility Close examination of the sections of Encyonema ventricosum demonstrated the traces of mucilage in the area of the raphe; however, there are also no signs of its attachment to the plasmalemma (Figure 3.3a). Particularly worth noting are the sections of the raphe Pleurosigma sp., which has a complex twisted structure (Figure 3.3b) and the mucilage goes through a difficult path prior to reaching the substrate surface; despite this issue, the cells of Pleurosigma are able to very actively move [3.51]. The raphe on the frustules of Bacillaria paxilifer has a similar construction [3.51], while in the raphe area beneath the plasmalemma there is a significant composite strand of electron-dense material — actin microfilaments — as revealed later using specific staining. These cables of actin microfilaments can be found in Navicula cuspidata ([3.22], Figures 14–18) and Stauroneis amphioxis [3.42]. It can be assumed that this difference in the ultrastructure of the plasmalemma in the region of the raphe is due to different methods of preparation of samples for TEM. In the present work, no signs of a developed polysaccharide layer is associated with the cytoskeleton through the plasma membrane. There is evidence for both actin microfilaments and myosin to be required in diatom gliding [3.6] [3.23] [3.46] [3.49] [3.52]. However, it has still not been directly determined if they are responsible for direct force production, or whether their main function in gliding might be the delivery of the vesicles carrying mucilage to the plasmalemma. (a)

(b’) (b)

(c)

(d)

Figure 3.4  Staining of actin microfilaments (phalloidin Alexa Fluor 488, green fluorescence) and nuclei (DAPI, blue fluorescence) in Pleurosigma sp. (a-b’) and Nitzschia sp. (c, d). Scale: (a, b) – 15 µm; (b’) – 5 µm; (c, d) – 10 µm.

Cellular Mechanisms of Raphid Diatom Gliding  71 The most prominent localization of actin filaments in cells of raphid diatoms is the raphe area (Figure 3.4). For many species, it has been shown that two cables of actin microfilament are located on both sides of the raphe (Craticula cuspidata [3.1]; Pleurosigma sp. [3.51]; Bacillaria paxillifer [3.52]; Figure 3.4a, b, b’), and possibly Nitzschia sp. along the raphe of the keel. (Figure 3.4c, d). It is well known that eukaryotic cells can express several myosins of different classes [3.30] and diatoms are no exception in this regard [3.33] [3.43]. Analysis of proteins associated with actin showed that actin-dependent intracellular transport in diatoms is carried out only with myosins, and the membrane extension form of actin-dependent transport associated with the complex actin-related proteins Arp2/3 is likely to be absent [3.2] [3.44]. Inventory of myosins expressed in Thalassiosira pseudonana [3.43] and Phaeodactylum tricornutum [3.33] revealed a large variety of this protein. Eight of the ten myosins expressed in P. tricornutum have some similarities with myosins of T. pseudonana, the genome of which includes 11 predicted myosins [3.43].

3.4 Mechanisms of Gliding Regulation The molecular mechanisms underlying regulation of diatom gliding including mucilage secretion is currently one of the most mysterious issues of diatom movement. It is known that the rate of movement and adhesion can be influenced by a change in temperature [3.12]. If the decrease in adhesion with increasing temperature could be explained by a change in the physical properties of extracellular polymers of diatoms, the change in the rate of locomotion can only be partially explained in this way—after all, the speed of movement still decreases and even stops when the temperature reaches a certain threshold [3.12]. The most likely reason for this dependence is possible myosin denaturation. While the denaturation of pig and chicken myosins reach a maximum at 40-50 °C [3.40] [3.48], similar to the temperature at which there is a diatom loss of motile function, these experiments with the myosin denaturation were carried out in extracellular systems with myosins that are evolutionarily very distant. A direct analysis of diatom myosins, either biochemically or in silico based on the available sequences in the databases, could reveal whether diatom myosins behave similarly. However, in the absence of experimental data, it would be impossible to compare the structure and function of myosins. Diatom movement can be influenced by light of varying quality and intensity and some diatoms may show obvious photophobic reactions in the direction of movement. Using different light irradiation levels, it was shown that cells of Craticula cuspidata can react quite quickly upon the high intensity light irradiation [3.11] [3.14] [3.38]—in which they change the direction of their movement. Moreover, vertical migration under high intensity light irradiation was shown in experiments with diatom communities inhabiting sand and mud [3.7]. For some diatoms it appears that the critical part of diatom light-sensitive systems is concentrated in the cell tips [3.11] [3.13]. While some experiments have indicated that these light sensitive components are independent of the chloroplasts [3.14], chloroplastrelated photoreceptors cannot be completely ruled out, since in many species that have been studied for movement the chloroplasts are large enough to reach the tips of the cells. It was recently suggested that movement of one of the species of Navicula sp. depends on the round structures found within the cell [3.49]. From the experience of the authors,

72  Diatom Gliding Motility the observed rounded structures have a morphology of lipid droplets, and the conclusions of the researchers may be premature. Not all processes in the cell occur because of cell gliding, but they are all interconnected simply because they occur in the same cell. Small fluctuations of lipid droplets occur even in immobile cells because the process of formation of droplets may take some time. It should not exclude the influence of the detected structures on the movement of diatoms, but most likely it is indirect.

3.5 Conclusions The molecular biology responsible for diatom gliding remains an unsolved problem, despite significant long-term interest. The most prevalent hypothesis, in which the gliding requires mucilage secretion coordinated with actin-myosin activity, has never been fully confirmed, although it has been circumstantially supported over time by new data. According to available data, adhesion and gliding of diatoms is carried out with the participation of mucilage consisting of polysaccharides and glycoproteins and secreting through the raphe system and areolae. The composition and properties of the mucilage may differ in the same cell, with different mucilages performing varying functions. For example, the softer ones may play a role of the main adhesive material that facilitates cell movement along the substrate, with the main motor function assumed by the mucilage fibers secreted through the raphe. At the present time, there is no doubt that the cellular actin-myosin system is an essential participant for gliding. However, it is still unclear whether the main role of actin and myosin is the delivery of vesicles carrying mucilage to the plasmalemma, since the precise localization of the synthesis of polymeric substances and the formation of vesicles is still unknown. The most unexplored issue is the regulation of the movement of mucilage at both cellular and molecular levels. As for the model in which the motile force is generated by myosin operating on the mucilage via a transmembrane connection, so far, transmembrane complexes connecting the cytoskeleton and glycoproteins (both of which ensure diatom sliding along a substrate) have yet to be discovered. As such, it is as yet undetermined whether changes in the properties of secreted mucus, or internal proteins acting on the mucus, may be more important for the implementation of gliding. In either case, explicit phototaxis, the direction of which depends on the quality of light, indicates a connection between gliding and photoreceptors that may exist in the terminal membrane areas of diatoms, or within their chloroplast photosystems.

Acknowledgments This work was supported by the topic of the Ministry of Education and Science of the Russian Federation 0279-2021-0008 “From a cell to an ecosystem: Investigation of the ultrastructure of aquatic organisms and their communities in the evolutionary and ecological aspect by methods of cell biology and genomics”. The author is grateful to Elena Nevrova for providing the culture of Nitzschia sp. and Pleurosigma sp., to Nikolay Sudakov and Igor Klimenkov for actin staining and to Stanley Cohn for editorial guidance. All microscopy was carried out in the Collective Instrumental Center “Ultramicroanalysis” at the Limnological Institute of the Siberian Branch of RAS.

Cellular Mechanisms of Raphid Diatom Gliding  73

References [3.1] Aumeier, C., The cytoskeleton of diatoms structural and genomic analysis, (Doctoral dissertation), Mathematisch-Naturwissenschaftlichen Fakultät Rheinischen Friedrich-WilhelmsUniversität Bonn, Germany, 2012. [3.2] Aumeier, C., Polinski, E., Menzel, D., Actin, actin-related proteins and profilin in diatoms: a comparative genomic analysis. Mar. Genomics, 23, 133–142, 2015. [3.3] Baïet, B., Burel, C., Saint-Jean, B., Louvet, R., Menu-Bouaouiche, L., Kiefer-Meyer, M.C., Bardor, M., N-Glycans of Phaeodactylum tricornutum diatom and functional characterization of its N-acetylglucosaminyltransferase I Enzyme. J. Biol. Chem., 286, 8, 6152–6164, 2011. [3.4] Beck, J.R., Rodriguez-Fernandez, I.A., Cruz de Leon, J., Huynh, M.-H., Carruthers, V.B., Morrissette, N.S., Bradley, P.J., A novel family of Toxoplasma IMC proteins displays a hierarchical organization and functions in coordinating parasite division. PloS Pathog., 6, e1001094, 2010. [3.5] Bedoshvili, Y.D., Gneusheva, K.V., Popova, M.S., Avezova, T.N., Arsentyev, K.Yu., Likhoshway, Y.V., Frustule morphogenesis of raphid pennate diatom Encyonema ventricosum (Agardh) Grunow. Protoplasma, 255, 911–921, 2018. [3.6] Cartaxana, P., Brotas, V., Serôdio, J., Effects of two motility inhibitors on the photosynthetic activity of the diatoms Cylindrotheca closterium and Pleurosigma angulatum. Diatom Res., 23, 1, 65–74, 2008. [3.7] Cartaxana, P., Ruvio, M., Hubas, C., Davidson, I., Serôdio, J., Jesus, B., Physiological versus behavioral photoprotection in intertidal epipelic and epipsammic benthic diatom communities. J. Exp. Mar. Biol. Ecol., 405, 120–127, 2011. [3.8] Chen, A.L., Kim, E.W., Toh, J.Y., Vashisht, A.A., Rashoff, A.Q., Van, C., Huang, A.S., Bradley, P.J., Novel components of the Toxoplasma inner membrane complex revealed by BioID. mBio, 6, e0235714, 2015. [3.9] Chiovitti, A., Heraud, P., Dugdale, T.M., Hodson, O.M., Curtain, R.C.A., Dagastine, R.R., Wetherbee, R., Divalent cations stabilize the aggregation of sulfated glycoproteins in the adhesive nanofibers of the biofouling diatom Toxarium undulatum. Soft Matter, 4, 811–820, 2008. [3.10] Chiovitti, A., Higgins, M.J., Harper, R.E., Wetherbee, R., Bacic, A., The complex polysaccharides of the raphid diatom Pinnularia viridis (Bacillariophyceae). J. Phycol., 39, 543–554, 2003. [3.11] Cohn, S.A., Bahena, M., Davis, J.T., Ragland, R.L., Rauschenberg, C.D., Smith, B.J., Characterization of the diatom photophobic response to high irradiance. Diatom Res., 19, 167–179, 2004. [3.12] Cohn, S.A., Farrell, J.F., Munro, J.D., Ragland, R.L., Weitzell, R.E., Wibisono, B.L., The effect of temperature and mixed species composition on diatom motility and adhesion. Diatom Res., 18, 2, 225–243, 2003. [3.13] Cohn, S.A., Halpin, D., Hawley, N., Ismail, A., Kaplan, Z., Kordes, T., Zapata, Y., Comparative analysis of light-stimulated motility responses in three diatom species. Diatom Res., 30, 213– 225, 2015. [3.14] Cohn, S.A., Spurck, T.P., Pickett-Heaps, J.D., High energy irradiation at the leading tip of moving diatoms causes a rapid change in cell direction. Diatom Res., 14, 193–206, 1999. [3.15] Dearnley, M.K., Yeoman, J.A., Hanssen, E., Kenny, S., Turnbull, L., Whitchurch, C.B., Dixon, M.W.A., Origin, composition, organization and function of the inner membrane complex of Plasmodium falciparum gametocytes. J. Cell Sci., 125, 8, 2053–2063, 2011.

74  Diatom Gliding Motility [3.16] Driouich, A., Follet-Gueye, M.-L., Bernard, S., Kousar, S., Chevalier, L., Vicré-Gibouin, M., Lerouxel, O., Golgi-mediated synthesis and secretion of matrix polysaccharides of the primary cell wall of higher plants. Front. Plant Sci., 3, 79, 2012. [3.17] Drum, R.W. and Hopkins, J.T., Diatom locomotion: an explanation. Protoplasma, 62, 1–33, 1966. [3.18] Dubremetz, J.F. and Torpier, G., Freeze fracture study of the pellicle of an eimerian sporozoite (Protozoa, Coccidia). J. Ultrastruct. Res., 62, 94–109, 1978. [3.19] Dugdale, T.M., Willis, A., Wetherbee, R., Adhesive modular proteins occur in the extracellular mucilage of the motile, pennate diatom Phaeodactylum tricornutum. Biophys. J., 90, 8, L58–60, 2006. [3.20] Edgar, L.A., Mucilage secretions of moving diatoms. Protoplasma, 118, 44–48, 1983. [3.21] Edgar, L.A. and Pickett-Heaps, J.D., Ultrastructural localization of polysaccharides in the motile diatom Navicula cuspidata. Protoplasma, 113, 10–22, 1982. [3.22] Edgar, L.A. and Pickett-Heaps, J.-D., The mechanism of diatom locomotion. I. An ultrastructural study of the motility apparatus. Proc. R. Soc London, 218, 331–343, 1983. [3.23] Edgar, L.A. and Pickett-Heaps, J.-D., The mechanism of diatom locomotion. II. Identification of actin. Proc. R. Soc London, 218, 345–348, 1983. [3.24] Fahy, J.V. and Dickey, B.F., Airway mucus function and dysfunction. New Engl. J. Med., 363, 23, 2233–2247, 2010. [3.25] Gaskins, E., Gilk, S., DeVore, N., Mann, T., Ward, G., Beckers, C., Identification of the membrane receptor of a class XIV myosin in Toxoplasma gondii. J. Cell Biol., 165, 383–393, 2004. [3.26] Gomez de Leon, C.T., Martin, R.D.D., Hernandez, G.M., Pozos, S.G., Ambrosio, J.R., Flores, R.M., Proteomic characterization of the subpellicular cytoskeleton of Toxoplasma gondii tachyzoites. J. Proteomics, 111, 86–99, 2014. [3.27] Gould, S.B., Tham, W.-H., Cowman, A.F., McFadden, G.I., Waller, R.F., Alveolins, a new family of cortical proteins that define the protist infrakingdom Alveolata. Mol. Biol. Evol., 25, 1219–1230, 2008. [3.28] Gupta, S. and Agrawal, S.C., Survival and motility of diatom Navicula grimmeiand and Nitzschia palea affected by some physical and chemical factors. Folia Microbiol., 52, 2, 127– 134, 2007. [3.29] Hamm, C., Merkel, R., Springer, O., Jurkojc, P., Maier, C., Prechtel, K., Smetacek, V., Architecture and material properties of diatom shells provide effective mechanical protection. Nature, 421, 841–843, 2003. [3.30] Hammer, J.A. and Sellers, J.R., Walking to work: roles for class V myosins as cargo transporters. Nat. Rev. Mol. Cell Biol., 13, 13–26, 2012. [3.31] Heintzelman, M.B., Cellular and molecular mechanics of gliding locomotion in eukaryotes. Int. Rev. Cytol., 251, 79–129, 2006. [3.32] Heintzelman, M.B., Gliding motility in apicomplexan parasites. Semin. Cell Dev. Biol., 46, 135–142, 2015. [3.33] Heintzelman, M.B. and Enriquez, M.E., Myosin diversity in the diatom Phaeodactylum tricornutum. Cytoskeleton, 67, 142–151, 2010. [3.34] Higgins, M.J., Crawford, S.A., Mulvaney, P., Wetherbee, R., Characterization of the adhesive mucilages secreted by live diatom cells using atomic force microscopy. Protist, 153, 25–38, 2002. [3.35] Higgins, M.J., Molino, P., Mulvaney, P., Wetherbee, R., The structure and nanomechanical properties of the adhesive mucilage that mediates diatom-substratum adhesion and motility. J. Phycol., 39, 1181–1193, 2003. [3.36] Hoagland, K.D., Rowoski, J.R., Gretz, M.R., Roener, S.C., Diatom extracellular polymeric substances: function, fine structure, chemistry, and physiology. J. Phycol., 29, 537–66, 1993.

Cellular Mechanisms of Raphid Diatom Gliding  75 [3.37] Huang, W., Río Bártulosa, C., Kroth, P.G., Diatom vacuolar 1,6-β-transglycosylases can functionally complement the respective yeast mutants. J. Eukaryot. Microbiol., 63, 4, 536–46, 2016. [3.38] Kapinga, M.R.M. and Gordon, R., Cell motility rhythms in Bacillaria paxillifer. Diatom Res., 7, 221–225, 1992. [3.39] Lind, J.L., Heimann, K., Miller, E.A., Vliet, C.V., Hoogenraad, N.J., Wetherbee, R., Substratum adhesion and gliding in a diatom are mediated by extracellular proteoglycans. Planta, 203, 213–221, 1997. [3.40] Liu, J., Puolanne, E., Ertbjerg, P., Temperature induced denaturation of myosin: Evidence of structural alterations of myosin subfragment-1. Meat Sci., 98, 124–128, 2014. [3.41] Mauriello, E.M.F., Mignot, T., Yang, Z., Zusman, D.R., Gliding motility revisited: how do the myxobacteria move without flagella? Microbiol. Mol. Biol. Rev., 74, 229–249, 2010. [3.42] McConville, M.J., Wetherbee, R., Bacic, A., Subcellular location and composition of the wall and secreted extracellular sulphated polysaccharides/proteoglycans of the diatom Stauroneis amphioxys Gregoty. Protoplasma, 206, 188–200, 1999. [3.43] Montsant, A., Allen, A.E., Coesel, S., De Martino, A., Falciatore, A., Mangogna, M., Bowler, C., Identification and comparative genomic analysis of signaling and regulatory components in the diatom Thalassiosira pseudonana. J. Phycol., 43, 585–604, 2007. [3.44] Morozov, A.A., Bedoshvili, Y.D., Popova, M.S., Likhoshway, Y.V., Novel subfamilies of actinregulating proteins. Mar. Genomics, 37, 128–134, 2018. [3.45] Opitz, C. and Soldati, D., ‘The glideosome’: a dynamic complex powering glidingmotion and host cell invasion by Toxoplasma gondii. Mol. Microbiol., 45, 597–604, 2002. [3.46] Poulsen, N.C., Spector, I., Spurck, T.P., Shultz, T.F., Wetherbee, R., Diatom gliding is the result of an actin-myosin motility system. Cell Motil. Cytoskeleton, 44, 23–33, 1999. [3.47] Tuvim, M.J., Mospan, A.R., Burns, K.A., Chua, M., Mohler, P.J., Melicoff, E., Dickey, B.F., Synaptotagmin 2 couples mucin granule exocytosis to Ca2+ signaling from endoplasmic reticulum. J. Biol. Chem., 284, 9781–9787, 2009. [3.48] Wang, S.F. and Smith, D.M., Heat-induced denaturation and rheological properties of chicken breast myosin and f-actin in the presence and absence of pyrophosphate. J. Agric. Food Chem., 42, 2665–2670, 1994. [3.49] Wang, J., Cao, S., Du, C., Chen, D., Underwater locomotion strategy by a benthic pennate diatom Navicula sp. Protoplasma, 250, 5, 1203–1212, 2013. [3.50] Wetherbee, R., Lind, J.L., Burke, J., The first kiss: establishment and control of initial adhesion in raphid diatoms. J. Phycol., 34, 9–15, 1998. [3.51] Yamaoka, N., The gliding mechanism of diatoms. (Doctoral dissertation). Department of Picobiology, Graduate School of Life Science, University of Hyogo, 2015. [3.52] Yamaoka, N., Suetomo, Y., Yoshihisa, T., Sonobe, S., Motion analysis and ultrastructural study of a colonial diatom, Bacillaria paxillifer. Microscopy, 65, 3, 211–221, 2016.

4 Motility of Biofilm-Forming Benthic Diatoms Karen Grace Bondoc-Naumovitz1* and Stanley A. Cohn2 1

Department of Marine and Coastal Sciences, Rutgers University, New Brunswick, New Jersey, USA 2 Department of Biological Sciences, DePaul University, Chicago, Illinois, USA

Abstract

Microscale processes shape aquatic communities and collectively affect microbial trophodynamics, nutrient cycling, and ecosystem functioning. This chapter highlights previous research devoted to understanding the mechanics and behavior of motility in biofilm-forming pennate diatoms and the relationship between their motility and the associated mucilage secretion. Events happening at the microscale are often overlooked due to traditional bulk approaches in understanding ecosystem dynamics, but with the advent of technologies to analyze single-cell behaviors, we can begin to probe the complex lives of individual microbes. The ability of diatoms to colonize specific surfaces, establish additional biofilm, and migrate towards beneficial stimuli such as light, nutrients, and potential reproductive partners, highlights how these unicellular organisms can exploit fine-scale spatial and temporal heterogeneity. Compared to other motile organisms such as ciliated, flagellated, or amoeboid cells, diatoms are unique, as their rigid silicified structure renders them relatively nonflexible, providing a unique and mechanistically interesting system for understanding movement and adaptability within sophisticated environments. As benthic diatoms contribute up to half of the primary productivity on submerged habitats, their collective behavior and interactions, including the contribution of their secreted materials, can potentially affect larger-scale processes, such as elemental cycling and nutrient fluxes, along with the overall ecological success of the ecosystem. Keywords:  Benthic diatoms, pennates, biofilm, adhesion, motility, behavior, microscale, gradients

4.1 Introduction Directed motility towards or away from a stimulus has been among the most well-investigated behaviors of different organisms [4.10] [4.51]. In contrast to most other single-celled organisms, however, diatom protoplasts are constrained within hardened silicified cell walls, limiting the direct membranous interactions that the cell can make with its physical environment. Nonetheless, many diatoms have developed mechanisms for active motility involving movement coordination with mucilage secretions through specialized openings *Corresponding author: [email protected] Karen Grace Bondoc-Naumovitz: https://scholar.google.com/citations?user=g0eGu7EAAAAJ&hl=en, https://www.researchgate.net/profile/Karen_Grace_Bondoc Stanley A. Cohn: [email protected] Stanley Cohn, Kalina Manoylov and Richard Gordon (eds.) Diatom Gliding Motility, (77–110) © 2021 Scrivener Publishing LLC

77

78  Diatom Gliding Motility in the cell wall. Like other motile organisms, diatoms have been shown to use their movement to better exploit new environments for greater energy (light) capture, avoidance of predation and refuge into the sediment, protection from photodamage, and detection and alignment with appropriate reproductive cells [4.8] [4.13] [4.28] [4.37]. Diatom gliding has been suggested to have evolutionarily arisen first in pennates without specialized slits in the cell wall (i.e., araphid, Figure 4.1a) [4.117] [4.135], with a process that became further enhanced over time through the evolution of a system using fissures (i.e., raphes, Figure 4.1b,c) that allow for better localization and regulation of adhesive mucilage secretion critical for cell gliding [4.73] [4.126]. This morphological development in raphid (i.e., raphe-containing) pennates (Figures 4.1b,c) is considered to be a crucial factor in their rapid diversification into different niches and ecosystems within submerged habitats, and the enhancement of their genetic diversity [4.98] [4.111] [4.141] [4.155]. Beyond their individual behavior, pennate diatoms are dominant constituents of the microphytobenthos or benthic biofilms, both in marine and freshwater realms. Together with prokaryotes and other photosynthetic microeukaryotes, they are embedded within a matrix of various extracellular components, including mucilaginous polysaccharides known as extracellular polymeric substances (EPS) (Figure 4.2) [4.47] [4.98] [4.155]. Benthic biofilms are one of the most productive ecosystems, thought to contribute up to half of the fixed carbon to estuarine communities [4.155], and are significant links between the terrestrial and aquatic domains [4.158]. Pennate diatoms dominate 40% of the topmost 2 mm layer of this environment, outperforming the productivity of phytoplankton in the overlying water column [4.131] [4.157]. This high biogeochemical turnover in the sediment (benthos) can lead to nutrient flux regulation in the sediment-water interface, impacting not just benthic but also primary production for organisms in the water column [4.140] [4.147] [4.148] [4.155]. Population densities of benthic diatoms reinforce their importance to the aquatic communities, demonstrating ranges from 105–107 cells cm-3 depending on sediment properties, location, and season [4.98], with a scale-dependent spatial distribution. While microscale (4 to 100 cm2) patches are frequently homogeneous [4.6] [4.132], centimeter-scale patches often form random-mosaic patterns that are heterogeneous when viewed on a larger scale, (a)

(b)

(c)

Figure 4.1  Morphology of an araphid diatom (a), Staurosira construens var. venter (scale bar: 2 µm) and a raphid diatom (b), Navicula radiosa (scale bar: 10 µm), diatom on valve view. The raphe (indicated by a white arrow) runs through the whole valve and is a primary structure for adhesion to surfaces and moving. (c) A close-up of the raphe (indicated by a white arrow) is shown (scale bar: 1 µm). (SEM images downloaded from diatoms.org with permission [4.107] [4.120] [4.143].)

Motility of Biofilm-Forming Benthic Diatoms  79

Algae Ino nu rga tri nic en ts

EPS

DO

M

L (U igh t V, PA R)

Bacteria Fungi Protozoan Extracellular enzyme

Substrate

Figure 4.2  Schematic of a benthic biofilm. Algae, predominantly pennate biofilms together with bacteria, protists, and fungi, are embedded within a protective matrix of extracellular polymeric substances (EPS). A benthic biofilm experiences high variability of spatial and temporal gradients of environmental factors such as inorganic nutrients, dissolved organic matter (DOM), and light. (Figure from Sabater et al. [4.129]. Reprinted under CC-BY license.)

a common characteristic of fractal spatial distributions [4.6]. Interspecific interactions (e.g., [4.35]) are thought to shape the microscale, while environmental conditions such as nutrient resources, tidal currents, or wave actions influence larger-scale distributions [4.20]. However, these data generally describe the more lateral and horizontal aspects of populations and usually neglect the vertical distribution of cells that can also develop discrete distributions [4.14]. In general, benthic diatoms live in a complex environment subjected to spatial and temporal fluctuating conditions such as irradiance, wave and tide action, temperature and salinity variations, desiccation, nutrient availability, and immigration/emigration of other algal species [4.20] [4.95] [4.96–4.99] [4.114] [4.131] [4.155] (Figure 4.3). Nonetheless, across the biofilm, benthic diatoms tend to dominate a specific niche. In the upper story, stalked and attached immotile pennates (i.e., epipsammic) are present along with the sediment particles. Motile epipelic diatoms (i.e., those residing at the water-sediment interface) tend to cover the rest of the upper story, while non-motile epipsammic araphids that are attached to sand particles dominate the understory [4.8] [4.98]. Physiological trade-offs mediate the positions and niche distributions in these cells, along with their behavioral adaptations [4.8] [4.12]. For example, diatoms in the understory tend to be successfully adapted to low light and nutrient levels in this part of the biofilm [4.116] [4.124], while those in the upper story often have physiological photoprotective mechanisms or motile regulation to help select beneficial conditions [4.8] [4.12] [4.13] [4.92] [4.116] [4.124].

80  Diatom Gliding Motility Ca2+

SHEAR NO

+

NH4

PO4

Vegetative and sexual reproduction

DEPTH

APHOTIC ZONE

NO2-

Si(OH)4

EPS production 3-

dSi and dP uptake

NO3

-

Vertical migration

PHOTIC ZONE

Surface wettability

Micro-niche segration

Figure 4.3  A summary of some factors and gradients affecting diatom adhesion and motility on intertidal sediment. While adhering to the surface, cells sense their wettability through the production of intracellular nitric oxide (NO), which in turn mediates EPS production. Pennates generally have a preference for hydrophobic surfaces for attachment. Shear force or water flow affects substrate attachment with weakly attached cells easily dislodged by stronger shear. Once attached, the gliding capability and reversal control are mediated by the availability of extracellular and intracellular Ca2+, respectively. As light only penetrates a few mm on the sediment, the substrate is divided into the photic and aphotic zone. Cells undergo diel vertical migration to the photic zone to photosynthesize on the surface with a taxa-specific temporal rhythm, thus leading to micro-niche segregation. Nutrient concentrations vary with depth, and motile raphids take up dissolved silicate (dSi) and phosphate (dP) on the aphotic zone as there are higher concentrations of these mineral nutrients with depth. The aphotic zone also provides a protected and stable environment for vegetative and sexual reproduction. (Figure modified and redrawn from Saburova et al. [4.132] and adapted from Fenchel [4.59]. Reprinted with permission.)

However, it is important to note the importance of cell movement within the biofilm. Although cells are embedded within an extracellular polymeric substance (EPS) matrix that acts as a protection, extracellular conditions are only partially modulated and stabilized [4.47]. Hence, active motility can be an adaptive behavior to minimize or avoid stressful situations during locally fluctuating conditions [4.2] [4.86] [4.124] [4.127] [4.137] [4.155]. In the last decade, studies on the specific aspects of motility, including mechanics of adhesion and gliding and responses to diverse environmental or chemical stimuli, have been steadily increasing. These studies opened up discussions on understanding how diatoms gained dominance in benthic communities and potentially affected large-scale processes such as biogeochemical cycles and primary productivity [4.138] [4.142] [4.145] [4.146]. This chapter will focus on consolidating advances in understanding pennate diatom motility, including both araphids and raphids, made over the last twenty years (1999–2019) and highlighting three distinct sets of contributions, namely general motility concepts and advances, current understanding of light-directed vertical migration, and considerations of stimuli-directed motility (Table 4.1). Since most of the research about benthic diatoms has been either community-based using large-scale samples, or studies on the microscale or single-cell behavior, with little work directly connecting the two, this chapter will tend to encompass behavioral and mechanistic studies on a single-cell level and infer how these individual responses collectively affect larger-scale dynamics.

Motility of Biofilm-Forming Benthic Diatoms  81 Table 4.1  Summary of advances in understanding diatom motility over the last 20 years (1999-2019). Author

Species

Findings

General motility concepts and advances Poulsen et al. 1999 [4.123]

Craspedostauros australis, Nitzschia sp., Pinnularia sp., Craticula spp.

Motility model: actin-myosin complex mediates gliding activity

WigglesworthCooksey et al. 1999 [4.162]

Amphora coffeaeformis

Similar adhesion profiles for both hydrophobic and hydrophilic surfaces; no motility when contact angles ≥40°; Homogeneously hydrophobic surfaces inhibit motility; Attach-detach behavior is more consistent on hydrophilic substrates

Schultz et al. 2000 [4.136]

Amphora coffeaeformis var. purpusilla

Longer pre-settlement time leads to higher shear resistance; weakly attached cells easily dislodged by water flow

Finlay et al. 2002 [4.60]

Amphora coffeaeformis var. purpusilla

Cells have stronger adhesion but reduced motility on hydrophobic surfaces (with contact angles ≥60°)

Higgins et al. 2002 [4.72]

Craspedostauros australis

Mucilage layer with low adhesive properties encases a living diatom cell wall; Mucilage excreted near the raphe are strongly adhesive and are composed of highly elastic polymers

Cohn et al. 2003 [4.36]

Craticula cuspidata, Stauroneis phoenicenteron, Nitzschia linnearis, Pinnularia viridis

Cell speed and adhesion affected by temperature and presence of other diatom species; cell speed and adhesion ability independent from each other

Arce et al. 2004 [4.5]

Navicula sp.

Similar adhesion strength for hydrophobic and hydrophilic surfaces; Adhesion profiles depend on individual cells and not growth phases; EPS biopolymers have different lengths

Holland et al. 2004 [4.77]

Amphora coffeaeformis var. purpusilla, Craspedostauros australis, Navicula perminuta

Cells have stronger adhesion on hydrophobic surfaces; motility not affected by surface properties; motility and adhesion not correlated with each other; Cells adhered to hydrophilic surfaces are more resistant to shear stress; resistance to shear stress is species-specific (Continued)

82  Diatom Gliding Motility Table 4.1  Summary of advances in understanding diatom motility over the last 20 years (1999-2019). (Continued) Author

Species

Findings

WiggslesworthCooksey and Cooksey 2005 [4.161]

Amphora coffeaeformis, Navicula sp.

Reduced cell motility due to the possible lectin/agglutinin chemical from Pseudoalteromonas sp. that binds biopolymers critical for adhesion and motility; EPS production depends on growth phase (Amphora sp.), Mode of EPS extrusion different for both species

Sato and Medlin 2006 [4.135]

Licmophora hyalina

Gliding motility towards the broader head pole on individual cells; the continuous rotational movement of colonies

Stanley and Callow 2007 [4.144]

Phaeodactylum tricornutum

Benthic oval cells have similar adhesion strength on hydrophobic surfaces comparable to other raphid diatoms

Thompson et al. 2008 [4.151]

Seminavis robusta

Cells strongly adhere to hydrophobic surfaces; stress response involving nitric oxide (NO) is used for sensing surface wettability and subsequently substrate selection and attachment

McLachlan et al. 2012 [4.104]

Navicula perminuta

Inhibition of Ca2+ influx did not affect photophobic response but triggered decreased cell speed; switching mechanism for reversal motion might be Ca2+-mediated

Murase et al. 2011 [4.108]

Navicula sp.

Use of a microchamber to contain cells for cell motility observations

Hodson et al. 2012 [4.76]

Amphora coffeaeformis

Removal rate of adhered cells under shear is the same for hydrophobic and hydrophilic surfaces; A highly motile strain is resistant to shear (i.e., not easily detached); Higher shear resistance when cells have longer settlement time before shear exposure

Murase et al. 2012 [4.109]

Navicula sp.

Time interval is essential in analyzing track data

Cao et al. 2013 [4.24]

Navicula sp.

Combined computational and experimental method showed that cellular interaction and reproduction causes aggregation/ biofilm formation (Continued)

Motility of Biofilm-Forming Benthic Diatoms  83 Table 4.1  Summary of advances in understanding diatom motility over the last 20 years (1999-2019). (Continued) Author

Species

Findings

Finlay et al. 2013 [4.61]

Navicula incerta

Developed an open channel flow to determine adhesion strength in adjustable shear stress; Cells have similar adhesion strength on both hydrophobic and hydrophilic surfaces under static and low shear stress; Cells attached to hydrophobic substrates can tolerate higher shear stress and form a biofilm under shear

Umemura et al. 2013 [4.152]

Navicula pavillardii

Use of semicircular microgrooves to confine cells for motility analysis; maximum cell speed is mechanically limited

Wang et al. 2013 [4.159]

Navicula sp.

Motility model: gliding behavior mediated by alternately pressing actin pseudopods; EPS provides an increased adhesive force to drive motility; cells can bend while moving

Gutiérrez-Medina et al. 2014 [4.68]

Nitzschia communis

Circular run-reverse motility in isotropic conditions; circular runs have a constant speed and arc diameter

Klein et al. 2014 [4.84]

Navicula jeffreyi

Diatom cells are moderately hydrophobic (contact angle >68°) and adhere more to hydrophobic surfaces; EPS contains 2.5 times more proteins than sugars; Decreased cell attachment when the total surface energy of the substrate is higher (>40 mJ/m²)

Murguía et al. 2015 [4.110]

Nitzschia sp.

Circular-like trajectories; Active cell motility is persistent and memory-associated

Umemura et al. 2015a [4.153]

Navicula pavillardii

Temperature affects cell speed

Umemura et al. 2015b [4.154]

Navicula pavillardii, Seminavis robusta

Angular velocity differs between the two species due to cell symmetry; Use of microchamber for characterizing angular velocity

Yamaoka et al. 2015 [4.166]

Bacillaria paxillifer

Coordinated and bidirectional gliding oscillations of cell pairs or multiple pairs (i.e., colony) is driven by an actomyosin complex; electron-dense/protein-rich regions along the raphe connecting two adjacent cells might be involved in gliding (Continued)

84  Diatom Gliding Motility Table 4.1  Summary of advances in understanding diatom motility over the last 20 years (1999-2019). (Continued) Author

Species

Findings

Chen et al. 2019 [4.26]

Navicula sp.

Mucilage (EPS) trails increases the adhesive force of a diatom on different substrates; developed a method for mapping mucilage trails

Helliwell et al. 2019 [4.71]

Phaeodactylum tricornutum

Gliding motility is activated by Ca2+dependent signaling through EukCatAs (single-domain voltage-gated channels) in diatoms

Laviale et al. 2019 [4.91]

Nitzschia palea

Cells adhere strongly to hydrophobic surfaces and adhesion is mediated by surface-associated glycolipids as the main component of EPS

Light-directed vertical migration Cohn et al. 1999 [4.38]

Craticula cuspidata

Phototaxis receptors have a specific optimum wavelength for response and might be located at the tips of the cells

McLachlan et al. 2009 [4.103]

Navicula perminuta, Cylindrotheca closterium

Different light wavelengths mediate phototactic and photokinetic response that leads to accumulation and dispersal of cells; phototactic response is exclusive only for “true” benthic diatoms

Cohn et al. 2015 [4.37]

Craticula cuspidata, Stauroneis phoenicenteron, Pinnularia viridis

Species-specific motility response towards light varies for different wavelengths and intensities

Ezequiel et al. 2015 [4.57]

Navicula cf. recens

Photoacclimated cells can actively select their preferred optimal light exposure

Cohn et al. 2016 [4.35]

Stauroneis phoenicenteron, Craticula cuspidata, Pinnularia viridis

Species-specific response to light irradiance and wavelength; presence of other diatom species can mediate cell direction and cell accumulation

Taira et al. 2018 [4.150]

Navicula sp.

Speed of diatom cells is influenced by light wavelength and not light intensity

Stimuli-directed movement Chepurnov and Mann 2004 [4.27]

Licmophora communis

Motile male gametes search for stationary female gametes; mechanism for motility unclear but requires contact with surfaces (Continued)

Motility of Biofilm-Forming Benthic Diatoms  85 Table 4.1  Summary of advances in understanding diatom motility over the last 20 years (1999-2019). (Continued) Author

Species

Findings

Davidovich et al. 2010 [4.45]

Tabularia fasciculata

Motile male gametes randomly search for stationary female gametes while spinning around their axis

Sato et al. 2011 [4.134]

Pseudostaurosira trainorii

Mate-finding amoeboid movement of male gametes to female mediated by unidentified pheromones

Davidovich et al. 2012 [4.46]

Tabularia tabulata, Tabularia fasciculata

Pseudopodia-like cytoplasmic projections mediates random ameboid movement in male gametes

Witkowski et al. 2012 [4.163]

Navicula sp.

Swarm-like (ring-formation) behavior on agar plates; swarming ceases under nitrate or phosphate starvation or hypo- or hypersaline conditions

Edgar et al. 2014 [4.55]

Tabularia fasciculata

Male gametes exhibit Brownian motion; Mate recognition of male to female gametes requires near contact

Podunay et al. 2014 [4.118]

Ulnaria ulna

Pseudopodia-like projection of male to encounter female gametes

Daglio et al. 2016 [4.44]

Halamphora luciae

Phosphate deficiency increased EPS production and cell speed

Bondoc et al. 2016a [4.15]

Seminavis robusta, Navicula sp.

Simultaneous chemotactic and chemokinetic behavior of silicate-starved cells towards silicate gradients; the ability of cells to discriminate between resources (silicate and germanium)

Bondoc et al. 2016b [4.17]

Seminavis robusta

Simultaneous chemotactic and chemokinetic behavior of sexualized cells towards pheromone gradients of the opposite mating type; In silico comparison of silicate foraging and mate-finding behavior

Kaczmarska et al. 2017 [4.80]

Plagiogramma sp., Dimeregramma sp.

Pseudopodial motility of male gametes towards slightly motile female gametes

Bondoc et al. 2018 [4.18]

Seminavis robusta

Simultaneous chemotactic and chemokinetic behavior of phosphate-starved cells towards phosphate gradients; nitrogenstarved cells did not respond to nitrate or ammonium gradients (Continued)

86  Diatom Gliding Motility Table 4.1  Summary of advances in understanding diatom motility over the last 20 years (1999-2019). (Continued) Author

Species

Findings

Lembke et al. 2018 [4.93]

Seminavis robusta

Structure-activity relationships on diproline attraction; a core chemical structure is required, but slight structural modifications do not significantly affect attraction

Bondoc et al. 2019 [4.16]

Seminavis robusta

Cellular decision-making of cells across their life cycle towards multiple stimuli (silicate and pheromones); silicate attraction evident across all life stages; pheromone attraction is dependent on life stage and priming; priming abandoned by critical cells on the verge of extinction

Hu et al. 2020 [4.79]

Navicula arenaria var. rostellata

Novel circular/rotational run-reversal movement of silicate-starved cells towards silicate gradients

4.2 General Motility Models and Concepts Because of their very low Reynolds number (~10-5–10-4), there is significant viscous drag as diatoms glide on the sediment-water interface. As such, the average speed of a diatom during movement is fairly constant over the range of seconds, although over a shorter time scale, the movement may be generated by numerous bursts of sudden acceleration (see Sabuncu et al., this volume [4.130]). Even though resistance to movement is high (10-10 N), the cell’s energy expenditure to overcome this resistance is only ca. 0.1–1% from the predicted power generation of a cell [4.52]. If a typical diatom is migrating across the photic zone (~400 µm), it would spend only 0.03 pcal, which is a glucose expense of only 0.0001%. Hence, motility is a cost-beneficial process [4.52] [4.102]. The raphe (an elongated slit that runs along the central axis of a diatom’s biomineralized cell wall) is responsible for surface adhesion and motility (Figure 4.1c) [4.73] [4.126]. Within pennates, those cells that lack a raphe (i.e., araphids) are immotile, though some species can move slowly [4.117] [4.141]. On the other hand, raphid diatoms may possess a raphe slit either on one valve (monoraphid) or both valves (biraphid) and move via gliding, though several adnate or stalked species are sessile [4.135] [4.157]. Raphid pennates excrete EPS; sticky, mucilaginous compounds via their raphe (Figure 4.4a) [4.52] [4.53] [4.123] [4.139], that creates adhesive force between the cell and the substratum surface, which likely acts as the principal material that connects the internal cell protoplast with the exterior to allow gliding [4.26]. In this way, diatom adhesion and diatom motility are tightly coupled. Extrusion of EPS from the raphe can also vary between species, with some species releasing it via the whole raphe canal and others only from specific parts of the raphe [4.161]. EPS predominantly contains more proteins than sugars, although a recent study shows that the major component of extruded EPS is glycolipids [4.84] [4.91]. Production of extruded EPS is highly species-specific and not directly related to a

Motility of Biofilm-Forming Benthic Diatoms  87 (a)

(b)

Hydrophobic

Hydrophilic

mucilage trails

mucilage strands

15 µm

15 µm

Figure 4.4  (a) Scanning electron microscopy (SEM) images of EPS or mucilage trails and strands from the motile raphid diatom, Navicula sp. (scale bar: 5 µm). EPS trails were observed to be either straight or curved. (SEM images from Chen et al. [4.26]. (Reprinted under CC-BY license); (b) Adhesion of Nitzschia palea towards surfaces with different surface wettabilities (scale bar: 15 µm). Cells adhered more to hydrophobic surfaces after 4 hours. Insets are duplicate experiments. (Optical microscopy images from Laviale et al. Reprinted (adapted) with permission from Laviale et al. Copyright 2019 American Chemical Society [4.91]).

growth phase [4.5] [4.161], and different EPS types help to make the biomatrix conducive to attachment and growth of different types of external organisms such as bacteria [4.14]. These macromolecules are secreted in differing lengths depending on surface types, are highly adhesive and resistant to breaking, sustaining forces up to 60 nN, and elastic until 2.5 µm from the cell surface [4.5] [4.72]. However, some bacteria can produce possible lectin/agglutinin compounds that can bind these macromolecules, thereby significantly reducing both adhesion and motility [4.161]. The EPS secretion is also thought to enhance the sediments’ stability and cohesion, providing additional structures and a matrix through which the diatoms can move [4.149].

4.2.1 Adhesion Before cells can glide, they first need to contact and adhere to surfaces through the mucilage strands associated with their raphe, which provides traction for the movement. The chemical mechanism for mucilage adhesion is poorly understood. Such underwater adhesion requires water displacement at the surface interface with the mucilaginous material that must make numerous weak noncovalent adhesive bonds with the substratum (e.g., electrostatic or van der Waals interactions). Raphid diatoms are thought by some to have moderately hydrophobic surfaces, with many cells showing an adhesion preference to hydrophobic surfaces that have a higher contact angle and low surface energy (i.e., Craticula cuspidata, Stauroneis phoenicenteron, Nitzschia linnearis, Pinnularia viridis, Craspedostauros australis, Navicula perminuta, Phaeodactylum tricornutum, Seminavis robusta, Navicula jeffreyi, and Nitzschia palea) (Figure 4.4b) [4.60] [4.61] [4.77] [4.84] [4.91] [4.144] [4.151] that is possibly related to surface-associated glycolipids that form patchy adhesive clusters on the cell surface [4.91]. However, Amphora coffeaeformis and Navicula sp. have similar adhesion profiles for both hydrophobic and hydrophilic surfaces [4.5] [4.162], and adhesion tests on A. coffeaeformis have mixed results with regard to its surface preference [4.60] [4.161] [4.162]. As a whole, these studies are somewhat ambiguous, and the surface properties for both the cell and the substratum need to be further examined, both for pre-attachment conditions, as well as the surface changes and modifications made by the cell due to cell secretions.

88  Diatom Gliding Motility Atomic force microscopy analysis of the material extruding from the raphe slits indicates that at least some of the secreted mucilage material may be composed of 10-30 modular protein components with a rupture force of ca. 0.8 nN [4.49] [4.50]. More recently, however, there has been significant work on diatom mucilage adhesion [4.89] [4.122], in which the mucilage material has been analyzed by nuclear magnetic resonance (NMR) spectroscopy, chemical analysis, and proteomics, as well as immunolocalization, and found to be composed of about 70% carbohydrate and 30% protein. While numerous proteins were isolated from the mucilage, it is still unclear which of these proteins might be directly required or responsible for the adhesion. Under realistic physical conditions such as shear stress, the adhered cells can be dislodged and washed away, especially from hydrophilic surfaces [4.61] [4.76] [4.77]. However, resistance to shear stress is highly variable among species and even strains, with motility possibly being a factor [4.76] [4.77]. For example, four species with differing motility characteristics displayed different cell loss rates from the underside of a glass surface to which they had been previously attached [4.41]. Additionally, when cells are allowed to settle for an extended period, they adhere to surfaces better and can tolerate higher shear [4.76] [4.136]. Weakly adhered cells might be dislodged under stronger water flow due to the nature of the EPS produced, which is, in turn, dependent on surface wettability. When a substrate is hydrophilic, cells produce lower EPS [4.151] or undergo constant reversible attachment events [4.162], leading to low cohesion among the cells and substratum [4.61]. Overall, shear can enhance biofilm integrity by promoting EPS production and cell aggregation, especially in hydrophobic surfaces, or eliminating weakly adhered cells to maintain biofilm structural stability [4.61]. This promotion of additional EPS by water shear may explain why the loss of cells in a flume due to water flow displayed a loss rate that best followed a power curve rather than an exponential one [4.36], i.e., the longer the cells were attached, the harder they were to remove. This is in contrast to the loss of adhered cells from the underside of a coverslip, due to the force of gravity alone, where the loss follows an exponential curve, i.e., the probability of any particular cell losing contact is independent of time [4.41]. Interestingly, the exact relationship between adhesion and motility remains poorly understood. For example, A. coffeaeformis motility is reduced on hydrophobic surfaces [4.162]. However, another study on the same species showed that adhesion strength, which is a function of the surface’s hydrophobicity, was not related to its motility [4.77]. Adhesion profiles are also highly variable even within the same samples and can be dependent or independent of growth phases for different species [4.5] [4.161]. Perhaps one of the most direct observations on the direct link between adhesion and motility has come from observing the movement of diatoms on the underside of glass coverslips [4.41], where motile diatoms often stopped and lost contact with the coverslip at all but the trailing tip of the cell. In many cases, the cell could recover, pulling itself back up to the coverslip, and moving in the opposite direction (the previously attached tip becoming the new leading end of the cell). Such behavior would directly support the Edgar actin-myosin model of diatom movement [4.52–4.54], in which the adherent contact sites of mucilage strands are also directly connected to the forcegenerating mechanism within the cell. The adhesion of diatoms is also connected to environmental conditions. Upon mitotic cell divisions, the majority of the daughter cells stay within the vicinity of their parental cells, as environmental factors in an area that has already been successfully colonized often

Motility of Biofilm-Forming Benthic Diatoms  89 continue to be favorable for growth and subsequent reproduction [4.24]. This increased density of cells, secreting the mucilages used for adhesion and motility, eventually contribute to the significant formation of a biofilm that can further stabilize locally favorable environmental conditions. Cell aggregation mediated by the sticky nature of EPS can further enhance biofilm structure [4.24]. Both motility and adhesion can also be affected by temperature, as well as the presence of co-occurring cells and bacteria [4.21] [4.23] [4.36] [4.153] [4.161], with some bacteria inducing increased mucilage secretion. Overall, stronger adhesion is not a prerequisite for faster motility, suggesting that these processes are likely to be regulated independently of each other [4.36]. Motility clearly requires cellular adhesion that is strong enough to provide cell stability during motile force production, but weak enough to allow for the release of previous adhesion sites as a cell moves. Local biotic and abiotic conditions modulate this adhesion. Thus, the regulation of motility and associated adhesion is likely a complex interplay of various factors. Some of these factors that mediate adhesion properties of diatoms are likely small extracellular molecules that act as cell signaling molecules. One such stress response molecule, nitric oxide (NO), may mediate substrate selection by sensing its wettability (i.e., hydrophobicity/hydrophilicity) (Figure 4.3). NO production is reduced on hydrophilic surfaces, which subsequently hinders EPS secretion, allowing the cells to detach from the substrate [4.151]. Hence, the adhering-detaching patterns are more consistent on hydrophilic than hydrophobic surfaces [4.162]. Extracellular Ca2+ is also required for adhesion in both marine and freshwater diatoms, with the former requiring millimolar amounts [4.43]. Extracellular and intracellular Ca2+ may both be critical components of both adhesion and motility, although in C. cuspidata, the presence of measurable extracellular Ca2+ did not seem to be required for motility [4.34], suggesting motile regulation for freshwater diatoms may be mainly through intracellular Ca2+, or Ca2+ associated with the local EPS. In N. perminuta, a decreased Ca2+ influx reduced cell motility, and regulation of intracellular Ca2+ has been suggested to direct reversals of cell movement when generating a phobic response to light stimuli [4.104]. The oval benthic morphotype of P. tricornutum, which has similar adhesion capabilities as that of other pennates [4.144] and has low motility on Ca2+-deplete conditions, was recently used to uncover the Ca2+-dependent signaling in diatoms [4.71]. Diatoms were found to contain EukCatAs: single-domain, voltage-gated Na+/Ca2+ excitable channels that are functionally similar to the four-domain channels of most metazoans and regulate membrane excitability through a negative feedback mechanism with Ca2+. Knockout mutants are almost nonmotile and/or have decreased gliding speed, highlighting that EukCatAs control of Ca2+ influx is likely critical for motility [4.71].

4.2.2 Gliding Motility Raphid diatoms have long been known to possess significant motility in gliding over surfaces. The horizontal speed of raphid diatom gliding can range from ~1–29 µm s-1 [4.53], while their vertical speed is a magnitude lower [4.42], with cells having a mechanical limit for their highest maximum speed [4.152]. Mechanistically, many previous models have been proposed for the force generation in diatom movement (reviewed in this volume [4.67]). However, one of the most prominent hypotheses on gliding is the actin-based cytoskeletal system (for a review, see Edgar and Pickett-Heaps [4.54]). Briefly, cables of actin filaments underlying the raphe are functionally connected via transmembrane elements to

90  Diatom Gliding Motility mucilage strands adhered to the substratum. When force is applied to the transmembrane components via motor proteins acting along the actin cables, the force is transmitted to the mucilage, resulting in the cell’s movement in the opposite direction of the force [4.54]. Two parallel actin cables have been observed in numerous motile diatoms within the cytoplasm underlying the raphe, and Poulsen et al. [4.123] have provided evidence for the model using cytoskeletal inhibitors and further expanded the concept by including the motor protein responsible for force generation: myosin. Myosin is a motor protein known to be present in many forms of algae and plants [4.22] [4.87] [4.112], which can produce force within the range of diatom speeds and generate saltatory jerky movement at the microscale [4.125]. One of the fastest class of myosin motors, myosin XI, is thought to be an evolutionarily ancient form found in many algal types, and responsible for many types of organelle movement [4.112]. Recent work on myosin XI shows the association of this myosin type to a complex range of receptors and adapters that allow transport along actin cables in plants [4.87]. There is some evidence that at very small time scales the movement of diatoms occurs in saltatory steps with sudden bursts of acceleration, which can also be explained by myosin motors’ sudden stepwise displacements when placed under resistive force [4.31]. Under the Edgar model, the force is transduced from the motor protein/actin cable complex via the mucilage strands secreted through the raphe. This model is also supported by observations that small ink particles or fluorescent beads attached to the raphe area aggregate in very focused locations and move at the same average rate of the diatoms [4.48] [4.78] [4.167]. Thus, this model requires motility-specific EPS secreted through the raphe that are independent of other types of mucilage secretions. Such differentiation of EPS types is not surprising since diatoms are known to have numerous types of mucilage secretions specialized by location, function, and excretion [4.1] [4.75] [4.164] [4.165]. Expulsion of rapidly hydrated mucilage from regulated positions along the raphe has also been postulated as a mechanism to generate the force for motility [4.67]. While this model might explain the rapid, jerky accelerations seen in diatoms, we are more inclined to support the actin-myosin model, as the rapid expulsion model generates more difficulties to explain: 1) directional changes that occur along the raphe observed in single aggregates of ink, sediment, or fluorescent particles; 2) direction changes and recovery seen in cells attached at single sites at the trailing end of cells (e.g., cells on the underside of surfaces as described above [4.41]); and 3) the relative rapid sensitivity of the cells to light stimuli at the two ends in the regulation of direction. Nonetheless, it is critical to note that no light or EM observations have definitively shown any transmembrane connections connecting the mucilage strands excreted from the raphe with the underlying actin cables. Such connections would be needed to transduce the force from actin to the adhered mucilage strands and move the cell. Another recent and somewhat more controversial model for raphid motility has recently been presented by Wang et al. [4.159], who suggested that circular actin-containing structures located at the polar ends of Navicula sp. are the source of actin-filled stalks or “pseudopods,” which protrude into the EPS that is adhering to the substratum. They present detailed observations of mucilage trails exhibiting regular periodic pits in the trail, suggesting that they are regulated by pressing and releasing the EPS projections into the mucilage.  The coordinated control of these projections into the EPS results in the directionally biased movement, with force for motility generated by the positive pressure developed from

Motility of Biofilm-Forming Benthic Diatoms  91 the adhesion force [4.26] [4.159]. Cells rotate at a high speed by firmly adhering to the substrate at one site while generating sufficient torque for the movement [4.159] by extensions at a secondary site.  We remain unconvinced of this model for several reasons: 1) It seems unlikely that actin-based protrusions can extend and retract through the raphe at any frequency due to the strong resistance by the curved and narrow nature of the raphe fissure, 2) there are no microscopic observations of any actin extensions extending perpendicularly out from the valve face, 3) similar observations in other cells have not supported the actin immunofluorescence seen in this work, and 4) it is unclear from their model what the source is for generating the force of actin extension and retraction, and how the two circular organelles would be regulated to give the light and chemical-based directional movements seen in diatoms. The secretion and production of diatom mucilage have been associated with several types of motility and several types of path curvatures [4.41] [4.54], and several types of motile behaviors are associated with the frustule mucilage [4.11]. The mucilage released through the frustule as part of the motility process might also be used not only for motility but also as part of the complex used for initial diatom attachment to surfaces [4.160]. The requirement of raphe-based mucilage secretion in the production of movement is also supported by the paths taken by moving cells. Most pennate diatoms have a fairly straight path of movement, corresponding to the raphe’s shape, with symmetric cells tending to have smaller angular velocity compared to asymmetric ones [4.154]. Some symmetric cells have significant path curvature (e.g., Pinnularia spp. [4.41] [4.54]), although the shape of the raphe itself within these cells may be less linear. The mechanisms by which cells generate nonlinear paths of movement over surfaces are ambiguous, but they are likely related to both the raphe curvature and the mechanism of mucilage interaction between the substratum and cellular structures as the force is generated (see, e.g., this volume Harbich [4.69]). The unusual pennate raphid, Bacillaria paxillifer, is a colonial diatom in which cells can glide across each other in a bidirectional rhythmic and coordinated manner. The mechanism underlying the coordination and control of the rhythmic gliding is unclear, however, it is thought to be an adaptation of the standard diatom gliding motion requiring actin and myosin, with additional components that allow adjacent cells to remain attached. Previous work has suggested that adjacent cells are connected through electron-dense or protein-rich regions observed between the actin filaments and the plasma membrane, which conceivably help regulate the bidirectional gliding [4.166]. A model designed to help explore this motility has been published elsewhere in this volume [4.3]. It should be noted that some araphid diatoms (lacking a raphe system) can also display motility. For example, Ardissonea can display some movement [4.117], and the araphid diatom Licmophora hyalina can also glide towards its broader polar head in an almost straight direction while excreting mucilaginous material from the basal polar end [4.27]. Motile ball-like colonies that rotate forward are formed when multiple cells join their basal ends [4.4]. This type of araphid motility is likely similar in terms of mucilage secretion to the specialized pores observed in desmids [4.83] [4.113].

4.2.3 Motility and Environmental Responsiveness Regardless of the shape of the path, diatoms can alter their direction in response to external stimuli to accomplish productive movement. Two distinct movement patterns, taxis

92  Diatom Gliding Motility and kinesis, will be discussed to describe how cells position themselves in response to the external signals. Generally, both behaviors are used by motile organisms to accumulate or disperse from the stimulus source. Taxis is based on spatial sensing, wherein a cell would make comparisons of signals present at different locations, while kinesis depends on signal comparisons detected by the cell over time, leading to more indirect guiding of cells through regulation of motility characteristics such as speed or turning frequency [4.4] [4.51] [4.100]. Besides bacteria [4.7] [4.63], diatoms are the only other known single-celled microorganism that can simultaneously utilize taxis and kinesis to respond to environmental signals [4.15] [4.17] [4.18] [4.103]. Interestingly, both bacteria and diatoms exhibit motility in which their position is in part determined by the regulation of directional run/reversal frequency (see Figure 4.5a), and this specific movement might be linked to the use of both sensing mechanisms [4.15] [4.17] [4.18]. Understanding individual behavioral patterns requires appropriate time intervals for cell tracking and modifying spatial spaces for the cell to move [4.108] [4.109]. Light regulated biasing of diatom directional changes has been an area of active research (e.g., see review in [4.40]) and suggests that diatoms can use directional bias to both move into physiologically appropriate areas of light and avoid areas of light that are too high [4.33–4.36]. Light is also thought to help regulate movement in diurnal exposure/refuge cycles of diatoms within the sediment and help the development of the resulting biofilm [4.42]. The movement path also appears to be important in these processes, as Pinnularia sp., which exhibits a circular path, shows a significantly reduced ability to explore and extend its presence into new environments through movement alone [4.35]. Nonetheless, P. viridis did show directional change responses to light at some light conditions [4.37]. Even the colonial Bacillaria species described above exhibits coordinated sliding movement that seems photosensitive and able to generate diel cycles of extension and retraction of the colony [4.81] that may contribute to increased ecological success and survival, similar to other diatom migration patterns. Similarly, circular run-reversal gliding was observed in Nitzschia communis and Navicula arenaria var. rostellata (Figure 4.5b) exposed to isotropic and varying concentrations of Si(OH)4 (Si(OH)4 (noted as dSi in the remainder of the article) [4.68] [4.79]. The cells form arc-like runs, followed by a brief pause, and then reversal runs in the opposite direction

(a)

(b)

Figure 4.5  (a) A representative cell trajectory of Seminavis robusta showing the run-reverse motility of cells from high cell density experiments. (Cell track replotted from data from Bondoc et al. [4.15]). (b) The circular run-reverse gliding technique observed in Nitzschia communis under isotropic environmental conditions. Cells form arc-like runs with constant speed, reverse, and continue the arc-like runs in the opposite clock direction. Filled squares signifies the starting point of the cells. (Figure from Gutiérrez-Medina et al. [4.68]. Reprinted with permission.)

Motility of Biofilm-Forming Benthic Diatoms  93 (i.e., counterclockwise to clockwise or vice-versa) with each run characterized as having a constant speed and arc diameter [4.68] [4.79]. Similar circular-like trajectories consistent over long periods were also observed in a Nitzschia sp., although reversal was not found. Such movements indicate that a gliding diatom may make long-range path corrections as they move, consistent with previous observations [4.68] [4.79] [4.110]. This motion is also evident in both low- and high-density Seminavis robusta cells exposed to either isotropic or nutrient-gradient conditions (KGBN, unpublished data). The circular run-reversal gliding is reminiscent of bacterial tumbling but with arc-like paths instead of straight and reversal arcs with only a slight change of orientation, leading to limited diffusion (i.e., spread) of the cells [4.68]. While the rate of reversals can be influenced by nutrient gradients, such as in the case of N. arenaria, a constant circular run-reverse strategy with consistent extrusion of EPS in isotropic conditions may pave the way for the initiation of biofilm formation and provide a conditioning film for other diatoms or algae [4.68].

4.3 Light-Directed Vertical Migration A large amount of research on benthic biofilms concentrates on the vertical migration of pennates as a response to light and its importance at the community level (for review, see Cohn [4.33]). On an individual scale, it has been shown that raphids can respond both tactically and kinetically to specific light properties leading to accumulation or dispersal [4.40]. This simultaneous use of taxis and kinesis can only be exclusively seen with true benthic diatoms, as the tychoplanktonic Cylindrotheca closterium only responds to light kinetically [4.103]. The rhythmic diel behavior of diatoms in migrating to the surface and then retreating to refuge beneath the surface has been shown to enhance not just the survival of motile cells against burial and erosion but also their chances to accumulate carbon reserves through photosynthesis [4.70] [4.127]. Benthic diatoms are continually challenged since only 1% of light can attenuate through 0.1–13 mm of the sediment, making the sediment photic zone shallow (Figure 4.3) [98]. However, benthic pennate diatoms are highly sensitive to light and are thought to possess receptors on both the trailing and leading tips of the cell, enabling them to spatially sense changes in irradiance level and wavelength in a species-specific manner [4.35] [4.38]. Light properties can also regulate the cell’s speed, with the highest cell speed observed at an intensity and wavelength specific to the cell [4.103] [4.150]. Interestingly, previously photoacclimated cells can actively select their position across a light gradient where they can have the highest photosynthetic efficiency without compromising overexposure [4.57]. Initial evidence suggests that upward vertical migration (Figure 4.3) could be an endogenous negative geotactic response coupled with positive chemotaxis, with the former prevailing as cells migrate upward to the sediment surface a few hours before sunrise [4.32] [4.62]. An intense peak in cell density is observed after sunrise with species-specific rhythms at preferred irradiances. Increased ambient light prompts smaller cells to migrate downward, while larger cells migrate upward [4.91] [4.96] [4.127] [4.131] [4.156]. This species-specific behavior promotes micro-migration and niche-segregation [4.37] [4.86] [4.156], which leads to the daily cycling of cells in the upper story of a benthic mat [4.156] and constant productivity on the biofilm surface [4.86] [4.156]. The small-scale differences in optimal wavelengths further suggest that benthic diatoms reduce competition

94  Diatom Gliding Motility by mediating their behavioral patterns, therefore ensuring the productive success of the whole system. This is further supported by the finding that some diatom species can modulate the behavioral response to light in ways that are dependent on both the individual species as well as the presence of other neighboring species, further supporting an interspecies interplay in a benthic biofilm with resulting niche partitioning [4.35]. However, lightmediated behavior is also regulated by circadian rhythms, as cells exposed to continuous darkness still exhibit migratory patterns [4.91] [4.127]. Other exogenous factors, such as tidal cycles, seasons (day-night length), light attenuation to sediment, and temperature can also affect the diel migration [4.96] [4.128].

4.4 Stimuli-Directed Movement In a relatively viscous environment, the transport of chemical stimuli is only through molecular diffusion [4.20] [4.51] [4.146], and factors affecting bulk water transport, such as water turbulence, can enhance the delivery of stimuli to the cells [4.115]. At the same time, cell motility allows the cells to expand their effective range beyond diffusion limits, enhancing their residence in areas of beneficial nutrients and decreasing their exposure to toxins or other damaging conditions. Pennate diatoms are highly sensitive to their dynamic environment and can respond to chemical changes within a matter of seconds [4.58]. As benthic biofilms have temporally and spatially distributed signals, rapid stimuli perception and active motility could be beneficial behavioral adaptations in exploiting environmental stimuli. With the emergence of single-cell imaging techniques, there has been a significant increase in behavioral studies directed towards understanding the cellular responses to dissolved nutrients [4.15-4.18] [4.44] [4.163] and mating partners/pheromones [4.17] [4.27] [4.45] [4.46] [4.55] [4.80 [4.93] [4.118] [4.134].

4.4.1 Nutrient Foraging In the benthos, chemical gradients, particularly nutrients, are affected by photosynthetic and respiratory processes (Figure 4.3). During tidal emersion and night periods, remineralization of the mineral-based nutrients, silicate (dSi) and orthophosphate ( PO34− , noted as dP) is enhanced due to increased respiratory processes [4.25] [4.88] [4.94] [4.133] [4.140]. Meanwhile, different forms of dissolved nitrogen (denoted by dN) dominate the nutrient content, depending on the prevailing nitrification and denitrification processes and the influx from the water column [4.88] [4.133]. These processes lead to different depthdependent concentrations of nutrients across the vertical space of the biofilm. Due to denitrification, nitrate ( NO3− ) and nitrite ( NO−2 ) have decreasing concentration with depth, whereas ammonium ( NH+4 ), dP, and dSi exhibit increasing concentration with depth [4.95] [4.97] [4.99] (Figure 4.3). Benthic motile pennates are highly sensitive to some of these specific nutrients, as cells depleted with dSi and dP increase their EPS production and speed [4.15] [4.18] [4.44]. Specific motility behavior such as swarming also ceases under nitrate or dP starvation for soft agar plate-grown cells [4.163]. As raphid diatoms predominantly exist at the sediment-water interface, the research performed using liquid cultures containing available surface material for cell attachment (e.g., beads) is likely more ecologically

Motility of Biofilm-Forming Benthic Diatoms  95 relevant than research investigating cells existing on a completely solid surface (e.g., Petri dishes, well plates). In recent years, chemoattraction assays mimicking hotspots gradients were devised by Bondoc et al. [4.15] using neutral aluminum oxide beads that can adsorb dissolved nutrients and release nutrient gradients upon release into a medium. This technique was modified by Hu et al. using polystyrene beads [4.79]. The resulting formed gradient mimics nutrient concentrations that were previously observed in biofilms [4.99]. Dissolved silicate (dSi) is a unique requirement for diatoms, as they utilize it to form their biomineralized cell wall [4.74]. Starved cells of S. robusta and N. arenaria can modify their behavior to forage for dSi optimally [4.15] [4.79]. High-density dSi-starved S. robusta exposed to dSi gradients accumulated around the source within 5 min of exposure, suggesting a high sensitivity for this resource (Figure 4.3). The movement patterns of dSi-foraging cells can be characterized as having: (i) increased cell speed in the immediate periphery of the source; (ii) directional persistence towards the stimuli; (iii) increased reversal behavior under an optimized dSi concentration; and (iv) continuous rotation on turning angles (i.e., circular movement) [4.15] [4.79]. As cells increase their speed around stimuli hotspots, they can display an increase in their nutrient uptake [4.19]. Overall, the rapid sensing mechanism by diatoms towards dSi gradients is simultaneous taxis and kinesis, previously observed on marine bacteria that display similar run-reverse motility [4.15]. The observed dSi-foraging is also substrate-specific since the chemically similar dissolved germanium (dGe) elicited a negative response, showing that cells can discriminate between these two resources [4.15] (Figure 4.6). The aforementioned behavioral characteristics were observed in both high- and lowcell density experiments, paving the way for a better understanding of how cells respond to a stimulus source. The use of low cell concentrations for studying behavior increases the opportunity to determine novel motility patterns that cannot be observed in higher density cultures where cells accumulate on the stimuli source, making careful observation of individual cell behavior arduous. For example, in low-density experiments, Hu et al. [4.79] observed the circular run-reverse gliding previously described by Gutiérrez-Medina under isotropic conditions [4.68]. However, when cells are exposed to dSi gradients, they can fine-tune their reversal behavior and turning angle, thereby maximizing their search area while foraging. Although differences in angular orientation for dSi and control beads were not observed by Bondoc et al. [4.15], a re-analysis of track data from these high cell concentration experiments showed that the cumulative frequencies over time of specific turning angles are indeed higher when dSi gradients are present (KGBN, unpublished data), concurrent with the observations on N. arenaria. Additionally, low-density experiments of S. robusta confirm the observed circular run-reverse motility on both isotropic and nutrient gradient conditions, a movement pattern not observed on high-density experiments (KGBN, unpublished data). Circular run-reverse motility, especially with low cell concentrations, might be a technique by raphid pennates to prepare the substrate for the colonization of other diatoms [4.68] through releasing a copious amount of EPS on a smaller surface area [4.47]. Modifying the movement pattern to increase diffusivity under the influence of dSi [4.79] enhances both the EPS deposited on the substrate as well as the possibility of establishing a biofilm on a dSi-replete environment. Meanwhile, well established biofilms with a high density of cells can further increase their diffusivity by switching to a run-reverse strategy.

96  Diatom Gliding Motility DPR

5 min dSi

SIPMitosis

dSi

Mitosis

SST 5 min

dSi

20 min

dGe

dP

dN

Figure 4.6  Motility of the model raphid pennate Seminavis robusta towards different stimuli across its life cycle. For every mitotic division, cells undergo size reduction until they reach a sexual size threshold (SST). Cells can either continue to undergo vegetative growth or sexual reproduction to reconstitute their size and escape death. Throughout vegetative growth, cells require nutrients. Gradients of dissolved silicate (dSi) and phosphate (dP) elicited starved cells to accumulate at point sources within 5 and 20 min, respectively. Dissolved germanium (dGe) did not elicit any attraction, pointing to substrate specificity response. Meanwhile, starved cells also did not respond to gradients of dissolved nitrate or ammonium (collectively called dN). Once cells reach SST, they release sex-inducing pheromones (SIPs) that control the production of diproline (DPR) on MT – cells. This pheromone is used by the MT+ cells as a chemical guide on locating MT– spatially and pair with it. Sexual reproduction requires trace amounts of dSi for the reconstitution of the silica frustule of the initial cells. Additionally, SIP priming is essential for cells that recently crossed SST. On the other hand, critically small-sized cells can bypass the priming process and be readily attracted to diproline. This self-priming could be a self-preserving strategy for the cells to avoid extinction. (Figure from Bondoc et al. [4.16]. Reprinted with CC-BY license.)

The dP-starved cells that were foraging employed a similar movement pattern but with a longer time for accumulation (~20 min vs. ~5 min for dSi), whereas the addition of nitrogen (NO3- or NH4) did not elicit any attraction even after prolonged starvation [4.18] (Figure 4.6). As dSi and dP remineralize in the benthos, they can be considered patchy resources [4.82], whereas dN can diffuse readily from the overlying water column and is thus readily available for uptake [4.97]. Therefore, foraging may be one of the primary ways cells generate downward migration towards a more stable environment containing slow-diffusing hotspots of the mineral-derived nutrients [4.18]. Nutrient foraging could be a dominant mechanism to seek out dSi- and dP-locations, as it was p ­ reviously observed that dSi attraction is present in all vegetative (i.e., diploid) stages of the cell [4.16]. A large population of cells undergoing active foraging can thereby have ecological consequences, such as controlling nutrient fluxes across the sediment-water interface, indirectly influencing phytoplankton abundance within the water column. This effect can particularly enhance the growth of non-siliceous organisms, causing effects such as harmful algal blooms [4.25] [4.90] [4.124] [4.140] [4.147].

Motility of Biofilm-Forming Benthic Diatoms  97

4.4.2 Pheromone-Based Mate-Finding Motility The stable deeper layers of the benthos also provide a protective environment for cell reproduction. Diatoms exhibit a diplontic life cycle described by a long asexual phase (months to years) with cell size reduction followed by a short burst of sexual events (hours to weeks) for size reconstitution [4.28] [4.56] [4.121] (for review see Poulíčková and Mann [4.121], and Chepurnov et al. [4.28]). While vegetative and sexual reproduction causes strong inhibition of cell movement, the active motility used in aiding the accumulation of nutrients and alignment with potential reproductive mates is crucial, as sufficient stored nutrients and proximity of mating cells are required before committing to either process [4.16] [4.65]. Dividing cells were exclusively observed only in the sediment’s anoxic zone, with their density correlating with depth [4.131], highlighting that mitotic division occurs primarily within this layer (Figure 4.3). On the other hand, sexual reproduction is a more complicated process as mating types need to encounter each other either through random motility or pheromone-guided attraction. Both araphids and raphids undergo sexual reproduction once they cross a certain sexual size threshold (SST), typically 30–40% of their original size, to restore their cell size and prevent cell death [4.28] [4.64]. Sexualization is triggered by either environmental cues (i.e., light, temperature, nutrient concentrations, or salinity) or pheromones from the sexually-induced partner [4.28] [4.56] [4.106]. For raphid diatoms, active motility is critical in pairing and subsequent mating of the distinct mating types, typically identified as male or female, or MT+ or MT-, respectively. Upon pairing, cells align, secrete a sheath of protective mucilage, and then undergo reduction divisions to produce gametes. The gametes then migrate and fuse, subsequently producing zygotes that undergo auxosporulation, enlarging in size and producing rings of protective silica girdle bands. Once the auxospore is complete, the cell develops a new frustule, completing the formation of a new initial large cell which often has significant differences in the pore and raphe structure [4.28] [4.39] [4.56] [4.101]. Once the initial cell frustule is complete, raphe-based cell gliding resumes, allowing the cell to break out of its surrounding protective sheath. In raphid diatoms, the primary copulation mechanism is not via the fusion of oogamous gametes in which one of the gametes is flagellated (i.e., gametogamy, as in many centric diatoms, e.g., [4.85]), but starts via the lateral pairing of gametangia or parental cells (i.e., gametangiogamy), which subsequently produce gametes that fuse to form a zygote [4.28]. Interestingly, gametangiogamy as a life history trait is believed to have evolved before the raphe system. As araphids release slow-moving gametes, they are dependent on circumstantial proximity of mates, as well as passive transport of water for mating. Active motility subsequently evolved, allowing cells to increase encounter rates in a crowded habitat, enhancing successful copulation and subsequent production of initial cells. This helped to drive the evolutionary radiation of diatoms across novel submerged habitats [4.7] [4.10] [4.56]. Araphids (e.g., Licmophora communis, Pseudostaurosira trainorii, Tabularia tabulata, T. fasciculata, Ulnaria ulna, Plagiogramma sp., Dimeregramma sp.) are consistent in their mate-finding strategy wherein spherical non-flagellated male gametes project one or multiple pseudopod-like extrusions to randomly search stationary or slow-moving females that are either tightly associated or near the gametangial thecae [4.27] [4.45] [4.46] [4.55] [4.80] [4.118] [4.134]. These protrusions are flexible and sticky and can extend and retract

98  Diatom Gliding Motility to “catch” females, albeit by chance [4.118] [4.134]. The thread-like protrusions require contact with a substratum or surface to initiate motility and move horizontally and vertically [4.46] [4.80]. Male gametes repeatedly extend and retract their thread-like projections as an aid to (i) spin around on their axis with or without displacement, maintaining their spherical shape, and (ii) undergo amoeboid movement characterized by the gamete changing from spherical to an elongated ellipsoid. The latter is the movement technique that enables the male gamete to travel longer distances when a compatible mating type is in the vicinity [4.45] [4.46] [4.134]. Briefly, when a spherical gamete retracts its thread, it elongates and produces a blob-like structure that is simultaneously absorbed by the cell as it spins around its axis [4.134]. Finer projections with an autonomous movement were also observed in the gamete surface, but its use in motility or mate-finding is unknown [4.134]. Except for P. trainorii, male gametes have no directed movement towards females, even in proximity [4.27] [4.45] [4.46] [4.55] [4.80] [4.118]. Thus, random motility or Brownian motion is the optimum strategy by most araphids to increase encounter rates for successful mating [4.55], and this behavior is consistent with all araphids, including the basal araphids Plagiogramma sp. and Dimeregramma sp. [4.80]. However, some initial studies on P. trainorii also hint at the role of pheromones to orient motile gametes towards their partner, and to induce gametogenesis [4.134]. Raphe-bearing pennates were observed to utilize pheromones to increase mating efficiency [4.17] [4.66] [4.105] (Figure 4.3). The raphid diatom S. robusta has been used as a model in understanding mating dynamics, particularly pheromone cascade events (Figure 4.6). Unknown sex-inducing pheromones (SIPs) serve as priming signals for sexualization and trigger cell cycle arrest on both mating types. They also induce the production of the attracting pheromone diproline (i.e., l‑proline-derived diketopiperazine) in MT– (i.e., females). SIPs are used by MT+ (i.e., males) for detecting MT– presence, and diproline is used for searching and guiding purposes [4.17] [4.66] [4.105]. Diproline is the only diatom pheromone characterized to date, resulting in numerous studies to attempt and understand pheromone-based behaviors. Structure-activity studies show that diproline attraction is restricted to diketopiperazines with a standard central chemical structure, and slight structural modifications do not significantly modify their activity. Hence, the pheromones used in attraction may not be very specific, and SIPs might play a role in orchestrating successful mating through a more cell or species-specific pheromone cascade [4.93]. Pheromone production is both density-dependent and light-activated [4.66], implying that these processes are coupled with vertical migration, making mate searching highly efficient in a complex benthic environment. Additionally, biofilm-associated bacteria can either enhance or diminish released diproline concentrations [4.29] [4.30]. By using beads that adsorb diproline that can release pheromone upon water contact, Gillard et al. [4.66] has shown that this release seems to mimic the pheromone release of females (MT– ), as behavioral observations showed that MT+ cells repeatedly attempt to pair with the bead. Significant changes in speed and directionality increased upon pheromone encounters, signifying that the attraction is simultaneously chemokinetic and chemotactic [4.17]. The calculated threshold concentration of diproline, or the lowest concentration eliciting attraction and searching is 19 pM diproline [4.19], similar to concentrations reported for planktonic flagellated brown algal gametes [4.119]. Thus, pheromone-induced motility highlights a sensitive and effective guiding mechanism to increase search efficiency, encounters, and successful pairing in a dense microhabitat [4.19].

Motility of Biofilm-Forming Benthic Diatoms  99

4.4.3 Prioritization Between Co-Occurring Stimuli Both nutrients and mating partners/pheromones can co-occur, but their distribution can be spatially and temporally patchy. Hence, motility also provides a profitable strategy for diatoms to effectively position themselves within regions of numerous mutually beneficial stimuli, such as dissolved nutrients and pheromones from prospective mating partners. Understanding how cells prioritize between multiple signals is crucial in determining the overall behaviors of diatoms. The raphid diatom S. robusta is an exemplary system for understanding the interplay between motility and multiple stimuli sensing, as well as prioritization mechanisms in microeukaryotes [4.16] (Figure 4.6). In silico comparisons of dSi acquisition vs. diproline attraction on S. robusta showed that the latter significantly enhanced behavioral searching parameters [4.17]. Combining experimental and modeling approaches, Bondoc et al. [4.16] described the decisionmaking behavior (nutrient foraging vs. mating) of S. robusta across its life stages. Large cells are above the SST and not competent for sexual reproduction, so nutrient foraging is a top priority for these cells. Meanwhile, cells that recently crossed the SST can choose between nutrient foraging or mating, depending on their physiological need. In this case, starved cells are only attracted to dSi, while sexually-induced cells respond solely to diproline. However, critically small-sized cells bypass the induction of SIPs and are readily attracted to diproline beads. This breach in the pheromone cascade highlights the cell’s priority to mate to reconstitute their size and escape extinction. However, trace amounts of dSi are needed by the cell for pheromone attraction to occur, since this nutrient is required to build the silicified frustule of the new initial cell (Figure 4.6). Overall, the ability of benthic diatoms to undergo decision-making between different chemical stimuli highlights the capacity of unicellular cells to select their most suitable environment actively [4.16]. When sexually induced, the planktonic diatom Pseudo-nitzschia multistriata downregulates the genetic expression of nutrient transporters and cyclins, and upregulates secondary messenger cyclic nucleotides (cGMP/ cAMP), which are responsible for pheromone perception, indicating crosstalk between the regulatory pathways responsible for nutrient sensing and mate finding [4.9] [4.16].

4.5 Conclusion Benthic diatoms are significant players in the aquatic environment as they can substantially affect the wider carbon and silica cycles. These biofilm-forming cells thrive in a complex and dynamic microhabitat through active motility, and understanding the mechanics of their behavior and movement could give insights on how they might thrive in such an environment. The ability of pennates to choose a suitable substrate, adhere to it, form a biofilm, and actively seek out specific locations based on beneficial signals demonstrates a model for understanding how unicellular behavior results from cell-level decision-making and prioritization. Moreover, their rapid response to gradients of environmental and chemical signals highlights sensitive adaptive behavior that could explain their rapid diversification across novel habitats. This behavior can also lead to spatial and temporal structuring of biofilms such as patchiness and enhanced ecological viability via micro-niche partitioning. For the last 20 years, advances in singlecell technologies have helped to understand diatom behavior and motility. However,

100  Diatom Gliding Motility bridging microscale research across a continuum of physically and biologically-relevant scales is a daunting task. Population- and global-scale studies have shown that diatoms can not only promote sediment stabilization but can drive proper large-scale ecosystem functioning through their modification of biogeochemical cycles. Utilizing these large-scale studies, along with accomplishing more microscale research that focuses on single-cell behavior, can yield a more thorough understanding of the significance of diatoms—both on a microscale patch of a biofilm as well as in the broader ecosystems they inhabit.

References [4.1] Abdullahi, A.S., Underwood, G.J.C., Gretz, M.R., Extracellular matrix assembly in diatoms (Bacillariophycae). V. Environmental effects on polysaccharide synthesis in the model diatom, Phaedactylum Tricornutum. J. Phycol., 42, 363–378, 2006. [4.2] Admiraal, W., Peletier, H., Brouwer, T., The seasonal succession patterns of diatom species on an intertidal mudflat - an experimental analysis. Oikos, 42, 30–40, 1984. [4.3] Alicea, B., Gordon, R., Harbich, T., Singh, A., Varma, V., Mehan, P., Singh, U., Towards a digital diatom: Image processing and deep learning analysis of Bacillaria paradoxa dynamic morphology, in: Diatom Gliding Motility [DIGM, Volume 2 in the series: Diatoms: Biology & Applications, series editors: Richard Gordon & Joseph Seckbach], S.A. Cohn, K.M. Manoylov, R. Gordon (Eds.), Wiley-Scrivener, Beverly, MA, USA, 2020, This volume. [4.4] Amsler, C.D. and Iken, K.B., Chemotaxis and chemokinesis in marine algae and bacteria, in: Marine Chemical Ecology, J.B. McClintock and B.J. Baker (Eds.), pp. 413–430, CRC Press, Boca Raton, FL, 2001. [4.5] Arce, F.T., Avci, R., Beech, I.B., Cooksey, K.E., Wigglesworth-Cooksey, B., A live bioprobe for studying diatom-surface interactions. Biophys. J., 87, 4284–4297, 2004. [4.6] Azovsky, A., Chertoprood, M., Kucheruk, N., Rybnikov, P., Sapozhnikov, F., Fractal properties of spatial distribution of intertidal benthic communities. Mar. Biol., 136, 581–590, 2000. [4.7] Barbara, G.M. and Mitchell, J.G., Marine bacterial organisation around point-like sources of amino acids. FEMS Microbiol. Ecol., 43, 99–109, 2003. [4.8] Barnett, A., Meleder, V., Blommaert, L., Lepetit, B., Gaudin, P., Vyverman, W., Sabbe, K., Dupuy, C., Lavaud, J., Growth form defines physiological photoprotective capacity in intertidal benthic diatoms. ISME J., 9, 32–45, 2015. [4.9] Basu, S., Patil, S., Mapleson, D., Russo, M.T., Vitale, L., Fevola, C., Maumus, F., Casotti, R., Mock, T., Caccamo, M., Montresor, M., Sanges, R., Ferrante, M.I., Finding a partner in the ocean: Molecular and evolutionary bases of the response to sexual cues in a planktonic diatom. New Phytol., 215, 140–157, 2017. [4.10] Bell, W.J., Searching Behaviour: The Behavioural Ecology of Finding Resources, Chapman and Hall, London, England, 1990. [4.11] Bertrand, J., Mouvements des diatomées. 1- L’équilibre dynamique chez Rhoicosphenia Abbreviata. Cryptogam. Algol., 12, 11–29, 1991. [4.12] Blommaert, L., Huysman, M.J.J., Vyverman, W., Lavaud, J., Sabbe, K., Contrasting NPQ dynamics and xanthophyll cycling in a motile and a non-motile intertidal benthic diatom. Limnol. Oceanogr., 62, 1466–1479, 2017. [4.13] Blommaert, L., Lauvaud, J., Vyverman, W., Sabbe, K., Behavioural versus physiological photoprotection in epipelic and epipsammic benthic diatoms. Eur. J. Phycol., 53, 146–155, 2018. [4.14] Bohórquez, J., McGenity, T.J., Papaspyrou, S., García-Robledo, E., Corzo, A., Underwood, G.J.C., Different types of diatom-derived extracellular polymeric substances drive changes

Motility of Biofilm-Forming Benthic Diatoms  101 in heterotrophic bacterial communities from intertidal sediments. Front. Microbiol., 8, Article 245, 2017. [4.15] Bondoc, K.G.V., Heuschele, J., Gillard, J., Vyverman, W., Pohnert, G., Selective silicatedirected motility in diatoms. Nat. Commun., 7, 10540, 2016. [4.16] Bondoc, K.G.V., Lembke, C., Lang, S., Germerodt, S., Schuster, S., Vyverman, W., Pohnert, G., Decision-making of the benthic diatom Seminavis Robusta searching for inorganic nutrients and pheromones. ISME J., 13, 537–546, 2019. [4.17] Bondoc, K.G.V., Lembke, C., Vyverman, W., Pohnert, G., Searching for a mate: pheromonedirected movement of the benthic diatom Seminavis robusta. Microb. Ecol., 72, 287–294, 2016. [4.18] Bondoc, K.G.V., Lembke, C., Vyverman, W., Pohnert, G., Selective chemoattraction of the benthic diatom Seminavis robusta to phosphate but not nitrate contributes to biofilm structuring. Microbiol. Open, 8, e00694, 2018. [4.19] Bondoc, K.G.V., Directed Motility of Benthic Diatoms. Dissertation, Friedrich-SchillerUniversität Jena, 2017, https://www.db-thueringen.de/receive/dbt_mods_00033653. [4.20] Borchardt, M., Algal Ecology: Freshwater Benthic Ecosystems, Academic Press, San Diego, 1996. [4.21] Bruckner, C.G., Rehm, C., Grossart, H.-P., Kroth, P.G., Growth and release of extracellular organic compounds by benthic diatoms depend on interactions with bacteria. Environ. Microbiol., 13, 1052–1063, 2011. [4.22] Buchnik, L., Abu-Abied, M., Sadot, E., Role of plant myosins in motile organelles: Is a direct interaction required? J. Integr. Plant Biol., 57, 23–30, 2015. [4.23] Buhmann, M.T., Schulze, B., Förderer, A., Schleheck, D., Kroth, P.G., Bacteria may induce the secretion of mucin-like proteins by the diatom Phaeodactylum tricornutum. J. Phycol., 52, 463–74, 2016. [4.24] Cao, S., Wang, J., Li, D., Chen, D., Ecological and social modeling for migration and adhesion pattern of a benthic diatom. Ecol. Modell., 250, 269–278, 2013. [4.25] Carlton, R.G. and Wetzel, R.G., Phosphorus flux from lake sediments: Effect of epipelic algal oxygen production. Limnol. Oceanogr., 33, 562–570, 1988. [4.26] Chen, L., Weng, D., Du, C., Wang, J., Cao, S., Contribution of frustules and mucilage trails to the mobility of diatom Navicula sp. Sci. Rep., 9, 1–12, 2019. [4.27] Chepurnov, V.A. and Mann, D.G., Auxosporulation of Licmophora communis (Bacillariophyta) and a review of mating systems and sexual reproduction in araphid pennate diatoms. Phycol. Res., 52, 1–12, 2004. [4.28] Chepurnov, V.A., Mann, D.G., Sabbe, K., Vyverman, W., Experimental studies on sexual reproduction in diatoms. Int. Rev. Cytol., 237, 91–154, 2004. [4.29] Cirri, E., De Decker, S., Bilcke, G., Werner, M., Osuna-Cruz, C., DeVeylder, L., Vandepoele, K., Werz, O., Vyverman, W., Pohnert, G., Associated bacteria affect sexual reproduction by altering gene expression and metabolic processes in a biofilm inhabiting diatom. Front. Microbiol., 10, 1790, 2019. [4.30] Cirri, E., Vyverman, W., Pohnert, G., Biofilm interactions—bacteria modulate sexual reproduction success of the diatom Seminavis robusta. FEMS Microbiol. Ecol., 94, 2018. [4.31] Clemen, A.E.-M., Vilfan, M., Jaud, J., Zhang, J., Bärmann, M., Rief, M., Force-dependent stepping kinetics of myosin-V. Biophys. J., 88, 4402–4410, 2005. [4.32] Coelho, H., Vieira, S., Serôdio, J., Endogenous versus environmental control of vertical migration by intertidal benthic microalgae. Eur. J. Phycol., 46, 271–281, 2011. [4.33] Cohn, S.A., Photo-stimulated effects on diatom motility, in: Photomovement. Comprehensive Series in Photosciences 1, D.P. Häder and M.L. Lebert (Eds.), pp. 375–401, Elsevier, Amsterdam, 2001.

102  Diatom Gliding Motility [4.34] Cohn, S.A. and Disparti, N.C., Environmental factors influencing diatom cell motility. J. Phycol., 30, 818–828, 1994. [4.35] Cohn, S.A., Dunbar, S., Ragland, R., Schulze, J., Suchar, A., Weiss, J., Wolske, A., Analysis of light quality and assemblage composition on diatom motility and accumulation rate. Diatom Res., 31, 173–184, 2016. [4.36] Cohn, S.A., Farrell, J.F., Munro, J.D., Ragland, R.L., Weitzell Jr, R.E., Wibisono, B.L., The effect of temperature and mixed species composition on diatom motility and adhesion. Diatom Res., 18, 225–243, 2003. [4.37] Cohn, S.A., Halpin, D., Hawley, N., Ismail, A., Kaplan, Z., Kordes, T., Kuhn, J., Macke, W., Marhaver, K., Ness, B., ..., Zapata, Y., Comparative analysis of light-stimulated motility responses in three diatom species. Diatom Res., 30, 213–225, 2015. [4.38] Cohn, S.A., Spurck, T.P., Pickett-Heaps, J.D., High energy irradiation at the leading tip of moving diatoms causes a rapid change of cell direction. Diatom Res., 14, 193–206, 1999. [4.39] Cohn, S.A., Spurck, T.P., Pickett-Heaps, J.D., Edgar, L.A., Perizonium and initial valve formation in the diatom Navicula cuspidata (Bacillariophyceae). J. Phycol., 25, 15–26, 1989. [4.40] Cohn, S.A., Warnick, L., Timmerman, B., Photophobic responses of diatoms - Motility and inter-species modulation, in: Diatom Gliding Motility [Volume 2 in the series: Diatoms: Biology & Applications, series editors: Richard Gordon & Joseph Seckbach], S.A. Cohn, K.M. Manoylov, R. Gordon (Eds.), Wiley-Scrivener, Beverly, MA, USA, 2020, This volume. [4.41] Cohn, S.A. and Weitzell, Jr., R.E., Ecological considerations of diatom cell motility I. Characterization of motility and adhesion in four diatom species. J. Phycol., 32, 928–939, 1996. [4.42] Consalvey, M., Paterson, D.M., Underwood, G.J.C., The ups and downs of life in a benthic biofilm: Migration of benthic diatoms. Diatom Res., 19, 181–202, 2004. [4.43] Cooksey, K.E., Requirement for calcium in adhesion of a fouling diatom to glass. Appl. Environ. Microbiol., 41, 1378–1382, 1981. [4.44] Daglio, Y., Maidana, N.I., Matulewicz, M.C., Rodriguez, M.C., Changes in motility and induction of enzymatic activity by nitrogen and phosphate deficiency in benthic Halamphora luciae (Bacillariophyceae) from Argentina. Phycologia, 55, 493–505, 2016. [4.45] Davidovich, N.A., Kaczmarska, I., Ehrman, J.M., Heterothallic and homothallic sexual reproduction in Tabularia fasciculata (Bacillariophyta). Fottea, 10, 251–266, 2010. [4.46] Davidovich, N.A., Kaczmarska, I., Karpov, S.A., Davidovich, O.I., MacGillivary, M.L., Mather, L., Mechanism of male gamete motility in araphid pennate diatoms from the genus Tabularia (Bacillariophyta). Protist, 163, 480–494, 2012. [4.47] Decho, A.W., Microbial biofilms in intertidal systems: An overview. Cont. Shelf Res., 20, 1257–1273, 2000. [4.48] Drum, R.W. and Hopkins, J., Diatom locomotion: An explanation. Protoplasma, 62, 1–33, 1966. [4.49] Dugdale, T.M., Dagastine, R., Chiovitti, A., Mulvaney, P., Wetherbee, R., Single adhesive nanofibers from a live diatom have the signature fingerprint of modular proteins. Biophys. J., 89, 4252–4260, 2005. [4.50] Dugdale, T.M., Dagastine, R., Chiovitti, A., Wetherbee, R., Diatom adhesive mucilage contains distinct supramolecular assemblies of a single modular protein. Biophys. J., 90, 2987– 2993, 2006. [4.51] Dusenbery, D., Living at Microscale: The Unexpected Physics of Being Small, Harvard University Press, Cambridge, MA, 2011. [4.52] Edgar, L.A., Diatom locomotion: A consideration of movement in a highly viscous situation. Br. Phycol. J., 17, 243–251, 1982.

Motility of Biofilm-Forming Benthic Diatoms  103 [4.53] Edgar, L.A. and Pickett-Heaps, J., The mechanism of diatom locomotion. I. An ultrastructural study of the motility apparatus. Proc. R. Soc London, Ser. B: Biol. Sci., 218, 331–343, 1983. [4.54] Edgar, L.A. and Pickett-Heaps, J., Diatom locomotion, in: Progress in Phycological Research, vol. 3, F. Round and D. Chapman (Eds.), pp. 47–88, Biopress, Bristol, 1984. [4.55] Edgar, R., Drolet, D., Ehrman, J.M., Kaczmarska, I., Motile male gametes of the araphid diatom Tabularia fasciculata search randomly for mates. PloS One, 9, e101767, 2014. [4.56] Edlund, M.B. and Stoermer, E.F., Ecological, evolutionary, and systematic significance of diatom life histories. J. Phycol., 33, 897–918, 1997. [4.57] Ezequiel, J., Laviale, M., Frankenbach, S., Cartaxana, P., Serôdio, J., Photoacclimation state determines the photobehaviour of motile microalgae: the case of a benthic diatom. J. Exp. Mar. Biol. Ecol., 468, 11–20, 2015. [4.58] Falciatore, A., d’Alcalà, M.R., Croot, P., Bowler, C., Perception of environmental signals by a marine diatom. Science, 288, 2363–2366, 2000. [4.59] Fenchel, T., The ecology of marine microbenthos. IV. Structure and function of the benthic ecosystem, its chemical and physical factors and the microfauna communities with special reference to the ciliated Protozoa. Ophelia, 6, 1–182, 1969. [4.60] Finlay, J., Callow, M., Ista, L., Lopez, G., Callow, J., Adhesion strength of settled spores of the green alga Enteromorpha and the diatom Amphora: The influence of surface wettability. Integr. Comp. Biol., 42, 1116–1122, 2002. [4.61] Finlay, J.A., Schultz, M.P., Cone, G., Callow, M.E., Callow, J.A., A novel biofilm channel for evaluating the adhesion of diatoms to non-biocidal coatings. Biofouling, 29, 401–411, 2013. [4.62] Frankenbach, S., Pais, C., Martinez, M., Laviale, M., Ezequiel, J., Serôdio, J., Evidence for gravitactic behaviour in benthic diatoms. Eur. J. Phycol., 49, 429–435, 2014. [4.63] Garren, M., Son, K., Raina, J.B., Rusconi, R., Menolascina, F., Shapiro, O.H., Tout, J., Bourne, D.G., Seymour, J.R., Stocker, R., A bacterial pathogen uses dimethyl-sulfoniopropionate as a cue to target heat-stressed corals. ISME J., 8, 999–1007, 2014. [4.64] Geitler, L., Der Formwechsel der pennaten Diatomeen. Arch. Protistenkd., 78, 1–226, 1932. [4.65] Gillard, J., Devos, V., Huysman, M.J., DeVeylder, L., D’Hondt, S., Martens, C., Vanormelingen, P., Vannerum, K., Sabbe, K., Chepurnov, V.A., Inze, D., Vuylsteke, M., Vyverman, W., Physiological and transcriptomic evidence for a close coupling between chloroplast ontogeny and cell cycle progression in the pennate diatom Seminavis robusta. Plant Physiol., 148, 1394–1411, 2008. [4.66] Gillard, J., Frenkel, J., Devos, V., Sabbe, K., Paul, C., Rempt, M., Inze, D., Pohnert, G., Vuylsteke, M., Vyverman, W., Metabolomics enables the structure elucidation of a diatom sex pheromone. Angew. Chem. Int. Ed. Engl., 52, 854–857, 2013. [4.67] Gordon, R., The whimsical history of proposed motors for diatom motility, in: Diatom Gliding Motility [DIGM, Volume 2 in the series: Diatoms: Biology & Applications, series editors: Richard Gordon & Joseph Seckbach], S.A. Cohn, K.M. Manoylov, R. Gordon (Eds.), Wiley-Scrivener, Beverly, MA, USA, 2020, This volume. [4.68] Gutiérrez-Medina, B., Guerra, A.J., Maldonado, A.I.P., Rubio, Y.C., Meza, J.V.G., Circular random motion in diatom gliding under isotropic conditions. Phys. Biol., 11, 066006, 2014. [4.69] Harbich, T., Some observations of movements of pennate diatoms in cultures and their possible interpretation, in: Diatom Gliding Motility [DIGM, Volume 2 in the series: Diatoms: Biology & Applications, series editors: Richard Gordon & Joseph Seckbach], S.A. Cohn, K.M. Manoylov, R. Gordon (Eds.), Wiley-Scrivener, Beverly, MA, USA, 2020, This volume. [4.70] Harper, M., Movement and migration of diatoms on sand grains. Br. Phycol. J., 4, 97–103, 1969.

104  Diatom Gliding Motility [4.71] Helliwell, K.E., Chrachri, A., Koester, J.A., Wharam, S., Verret, F., Taylor, A.R., Wheeler, G.L., Brownlee, C., Alternative mechanisms for fast Na+/Ca2+ signaling in eukaryotes via a novel class of single-domain voltage-gated channels. Curr. Biol., 29, 1503–1511. e1506, 2019. [4.72] Higgins, M.J., Crawford, S.A., Mulvaney, P., Wetherbee, R., Characterization of the adhesive mucilages secreted by live diatom cells using atomic force microscopy. Protist, 153, 25–38, 2002. [4.73] Hildebrand, M. and Lerch, S.J., Diatom silica biomineralization: Parallel development of approaches and understanding. Semin. Cell Dev. Biol., 46, Elsevier, 27–35, 2015. [4.74] Hildebrand, M., Lerch, S.J.L., Shrestha, R.P., Understanding diatom cell wall silicification— Moving forward. Front. Mar. Sci., 5, 1–19, 2018. [4.75] Hoagland, K.D., Rosowski, J.R., Gretz, M.R., Roemer, S.C., Diatom extracellular polymeric substances: Function, fine structure, chemistry, and physiology. J. Phycol., 29, 537–566, 1993. [4.76] Hodson, O.M., Monty, J.P., Molino, P.J., Wetherbee, R., Novel whole cell adhesion assays of three isolates of the fouling diatom Amphora coffeaeformis reveal diverse responses to surfaces of different wettability. Biofouling, 28, 381–393, 2012. [4.77] Holland, R., Dugdale, T., Wetherbee, R., Brennan, A., Finlay, J., Callow, J., Callow, M.E., Adhesion and motility of fouling diatoms on a silicone elastomer. Biofouling, 20, 323–329, 2004. [4.78] Hopkins, J.T. and Drum, R.W., Diatom motility: An explanation and a problem. Br. Phycol. Bull., 3, 63–67, 1966. [4.79] Hu, W.-S., Huang, M., Zhang, H.-P., Zhang, F., Vyverman, W., Liu, Q.-X., Movement behavioral plasticity of benthic diatoms driven by optimal foraging. bioRxiv, 2020, 682153. [4.80] Kaczmarska, I., Gray Jr, B.S., Ehrman, J.M., Thaler, M., Sexual reproduction in plagiogrammacean diatoms: First insights into the early pennates. PloS One, 12, e0181413, 2017. [4.81] Kapinga, M.R.M. and Gordon, R., Cell motility rhythms in Bacillaria paxillifer. Diatom Res., 7, 221–225, 1992. [4.82] Karl, D.M., Microbially mediated transformations of phosphorus in the sea: New views of an old cycle. Ann. Rev. Mar. Sci., 6, 279–337, 2014. [4.83] Kiemle, S.N., Domozych, D.S., Gretz, M.R., The extracellular polymeric substances of desmids (Conjugatophyceae, Streptophyta): chemistry, structural analyses and implications in wetland biofilms. Phycologia, 46, 617–627, 2007. [4.84] Klein, G.L., Pierre, G., Bellon-Fontaine, M.-N., Zhao, J.-M., Breret, M., Maugard, T., Graber, M., Marine diatom Navicula jeffreyi from biochemical composition and physico-chemical surface properties to understanding the first step of benthic biofilm formation. J. Adhes. Sci. Technol., 28, 1739–1753, 2014. [4.85] Koester, J.A., Brawley, S.H., Karp-Boss, L., Mann, D.G., Sexual reproduction in the marine centric diatom Ditylum brightwellii (Bacillariophyta). Eur. J. Phycol., 42, 351–366, 2007. [4.86] Kromkamp, J., Barranguet, C., Peene, J., Determination of microphytobenthos PSII quantum efficiency and photosynthetic activity by means of variable chlorophyll fluorescence. Mar. Ecol. Prog. Ser., 162, 45–55, 1998. [4.87] Kurth, E.G., Peremyslov, V.V., Turner, H.L., Makarova, K.S., Iranzo, J., Mekhedov, S.L., Koonin, E.V., Dolja, V.V., Myosin-driven transport network in plants. Proc. Natl. Acad. Sci. (USA), 114, E1385–E1394, 2017. [4.88] Kuwae, T., Kibe, E., Nakamura, Y., Effect of emersion and immersion on the porewater nutrient dynamics of an intertidal sandflat in Tokyo Bay. Estuarine Coastal Shelf Sci., 57, 929–940, 2003. [4.89] Lachnit, M., Buhmann, M.T., Klemm, J., Kroger, N., Poulsen, N., Identification of proteins in the adhesive trails of the diatom Amphora coffeaeformis. Philos. Trans. R. Soc.-B-Biol. Sci., 374, 20190196, 2019.

Motility of Biofilm-Forming Benthic Diatoms  105 [4.90] Laruelle, G.G., Regnier, P., Ragueneau, O., Kempa, M., Moriceau, B., Longphuirt, S.N., Leynaert, A., Thouzeau, G., Chauvaud, L., Benthic–pelagic coupling and the seasonal silica cycle in the Bay of Brest (France): New insights from a coupled physical–biological model. Mar. Ecol. Prog. Ser., 385, 15–32, 2009. [4.91] Laviale, M., Beaussart, A., Allen, J., Quilès, F., El-Kirat-Chatel, S., Probing the adhesion of the common freshwater diatom Nitzschia palea at nanoscale. ACS Appl. Mater. Interfaces, 11, 48574–48582, 2019. [4.92] Laviale, M., Frankenbach, S., Serôdio, J., The importance of being fast: Comparative kinetics of vertical migration and non-photochemical quenching of benthic diatoms under light stress. Mar. Biol., 163, 10, 2016. [4.93] Lembke, C., Stettin, D., Speck, F., Ueberschaar, N., De Decker, S., Vyverman, W., Pohnert, G., Attraction pheromone of the benthic diatom Seminavis robusta: studies on structureactivity relationships. J. Chem. Ecol., 44, 354–363, 2018. [4.94] Leynaert, A., Longphuirt, S.N., An, S., Lim, J.-H., Claquin, P., Grall, J., Kwon, B.O., Koh, C.H., Tidal variability in benthic silicic acid fluxes and microphytobenthos uptake in intertidal sediment. Estuarine Coastal Shelf Sci., 95, 59–66, 2011. [4.95] Leynaert, A., Longphuirt, S.N., Claquin, P., Chauvaud, L., Ragueneau, O., No limit? The multiphasic uptake of silicic acid by benthic diatoms. Limnol. Oceanogr., 54, 571–576, 2009. [4.96] Longphuirt, S.N., Leynaert, A., Guarini, J.-M., Chauvaud, L., Claquin, P., Herlory, O., Amice, E., Huonnic, P., Ragueneau, O., Discovery of microphytobenthos migration in the subtidal zone. Mar. Ecol. Prog. Ser., 328, 143–154, 2006. [4.97] Longphuirt, S.N., Lim, J.H., Leynaert, A., Claquin, P., Choy, E.J., Kang, C.K., An, S., Dissolved inorganic nitrogen uptake by intertidal microphytobenthos: nutrient concentrations, light availability and migration. Mar. Ecol. Prog. Ser., 379, 33–44, 2009. [4.98] MacIntyre, H.L., Geider, R.J., Miller, D.C., Microphytobenthos: the ecological role of the “secret garden” of unvegetated, shallow-water marine habitats. I. Distribution, abundance and primary production. Estuaries, 19, 186–201, 1996. [4.99] Magni, P. and Montani, S., Seasonal patterns of pore-water nutrients, benthic chlorophyll a and sedimentary AVS in a macrobenthos-rich tidal flat. Hydrobiologia, 571, 297–311, 2006. [4.100] Maier, I., Gamete orientation and induction of gametogenesis by pheromones in algae and plants. Plant Cell Environ., 16, 891–907, 1993. [4.101] Mann, D.G., Sexual reproduction and systematics of Navicula protracta. Diatom Res., 3, 227–236, 1988. [4.102] Marques da Silva, J., Duarte, B., Utkin, A.B., Travelling Expenses: The energy cost of diel vertical migrations of epipelic microphytobenthos. Front. Mar. Sci., 7, 433, 2020. [4.103] McLachlan, D.H., Brownlee, C., Taylor, A.R., Geider, R.J., Underwood, G.J.C., Light-induced motile responses of the estuarine benthic diatoms Navicula perminuta and Cylindrotheca closterium (Bacillariophyceae). J. Phycol., 45, 592–599, 2009. [4.104] McLachlan, D.H., Underwood, G.J., Taylor, A.R., Brownlee, C., Calcium release from intracellular stores is necessary for the photophobic response in the benthic diatom Navicula perminuta (Bacillariophyceae). J. Phycol., 48, 675–681, 2012. [4.105] Moeys, S., Frenkel, J., Lembke, C., Gillard, J.T.F., Devos, V., Van den Berge, K., Bouillon, B., Huysman, M.J.J., De Decker, S., Scharf, J., Bones, A., Brembu, T., Winge, P., Sabbe, K., Vuylsteke, M., Clement, L., De Veylder, L., Pohnert, G., Vyverman, W., A sex-inducing pheromone triggers cell cycle arrest and mate attraction in the diatom Seminavis robusta. Sci. Rep., 6, 19252, 2016. [4.106] Moore, E.R., Bullington, B.S., Weisberg, A.J., Jiang, Y., Chang, J., Halsey, K., Morphological and transcriptomic evidence for ammonium induction of sexual reproduction in Thalassiosira pseudonana and other centric diatoms. PloS One, 12, e0181098, 2017.

106  Diatom Gliding Motility [4.107] Morales, E., Staurosira construens var. venter. Web Page on morphology. https://diatoms. org/species/staurosira_construens_var._venter. Accessed Sept. 14 2020, 2010. [4.108] Murase, A., Kubota, Y., Hirayama, S., Kumashiro, Y., Okano, T., Mayama, S., Umemura, K., Two-dimensional trajectory analysis of the diatom Navicula sp. using a micro chamber. J. Microbiol. Methods, 87, 316–319, 2011. [4.109] Murase, A., Kubota, Y., Hori, S., Hirayama, S., Mayama, S., Umemura, K., Importance of observation interval in two-dimensional video analysis of individual diatom cells. Eur. Biophys. J., 41, 545–550, 2012. [4.110] Murguía, J., Rosu, H.C., Jimenez, A., Gutiérrez-Medina, B., García-Meza, J., The Hurst exponents of Nitzschia sp. diatom trajectories observed by light microscopy. Physica A: Stat. Mech. Appl., 417, 176–184, 2015. [4.111] Nakov, T., Beaulieu, J.M., Alverson, A.J., Insights into global planktonic diatom diversity: The importance of comparisons between phylogenetically equivalent units that account for time. ISME J., 12, 2807–2810, 2018. [4.112] Nebenführ, A. and Dixit, R., Kinesins and Myosins: Molecular motors that coordinate cellular functions in plants. Annu. Rev. Plant Biol., 69, 329–361, 2018. [4.113] Oertel, A., Aichinger, N., Hochreiter, R., Thalhamer, J., Lütz-Meindl, U., Analysis of mucilage secretion and excretion in Micrasterias (Chlorophyta) by means of immunoelectron microscopy and digital time lapse video microscopy. J. Phycol., 40, 711–720, 2004. [4.114] Parsons, T. and Harrison, P., Nutrient cycling in marine ecosystems, in: Physiological Plant Ecology IV, O. Lange, P. Nobel, C. Osmond, H. Ziegler (Eds.), pp. 85–115, Springer, Berlin, 1983. [4.115] Pasciak, W.J. and Gavis, J., Transport limitation of nutrient uptake in phytoplankton. Limnol. Oceanogr., 19, 881–898, 1974. [4.116] Passy, S.I., Diatom ecological guilds display distinct and predictable behavior along nutrient and disturbance gradients in running waters. Aquat. Bot., 86, 171–178, 2007. [4.117] Pickett-Heaps, J., Hill, D.R.A., Blaze, K.L., Active gliding motility in an araphid marine diatom, Ardissonea (formerly Synedra) Crystallina. J. Phycol., 27, 718–725, 1991. [4.118] Podunay, Y.A., Davidovich, O., Davidovich, N., Mating system and two types of gametogenesis in the fresh water diatom Ulnaria ulna (Bacillariophyta). Algologia, 24, 3–19, 2014. [4.119] Pohnert, G. and Boland, W., The oxylipin chemistry of attraction and defense in brown algae and diatoms. Nat. Prod. Rep., 19, 108–122, 2002. [4.120] Potapova, M., Navicula radiosa, 2011, Web Page on morphology. Accessed Sept. 14 2020. https://diatoms.org/species/navicula_radiosa. [4.121] Poulíčková, A. and Mann, D.G., Diatom sexual reproduction and life cycles, in: Diatoms: Fundamentals and Applications, J. Sekbach and R. Gordon (Eds.), pp. 245–272, WileyScrivener, Beverly, MA, USA, 2019. [4.122] Poulsen, N., Kröger, N., Harrington, M.J., Brunner, E., Paasch, S., Buhmann, M.T., Isolation and biochemical characterization of underwater adhesives from diatoms. Biofouling, 30, 513–523, 2014. [4.123] Poulsen, N.C., Spector, I., Spurck, T.P., Schultz, T.F., Wetherbee, R., Diatom gliding is the result of an actin-myosin motility system. Cell Motil. Cytoskeleton, 44, 23–33, 1999. [4.124] Pringle, C.M., Nutrient spatial heterogeneity: Effects on community structure, physiognomy, and diversity of stream algae. Ecology, 71, 905–920, 1990. [4.125] Rief, M., Rock, R.S., Mehta, A.D., Mooseker, M.S., Cheney, R.E., Spudich, J.A., Myosin-V stepping kinetics: A molecular model for processivity. Proc. Natl. Acad. Sci. (USA), 97, 9482–9486, 2000.

Motility of Biofilm-Forming Benthic Diatoms  107 [4.126] Round, F.E., Crawford, R.M., Mann, D.G., Diatoms: Biology and Morphology of the Genera, Cambridge University Press, Cambridge, United Kingdom, 1990. [4.127] Round, F. and Happey, C., Persistent, vertical-migration rhythms in benthic microflora: Part IV. a diurnal rhythm of the epipelic diatom association in non-tidal flowing water. Br. Phycol. Bull., 2, 463–471, 1965. [4.128] Round, F. and Palmer, J., Persistent, vertical-migration rhythms in benthic microflora.: II. Field and laboratory studies on diatoms from the banks of the River Avon. J. Mar. Biol. Assoc. U. K., 46, 191–214, 1966. [4.129] Sabater, S., Timoner, X., Borrego, C., Acuña, V., Stream biofilm responses to flow intermittency: from cells to ecosystems. Front. Environ. Sci., 4, 14, 2016. [4.130] Sabuncu, A.C., Gordon, R., Richer, E., Manoylov, K.M., Beskok, A., The kinematics of explosively jerky diatom motility: A natural example of active nanofluidics, in: Diatom Gliding Motility [Volume 2 in the series: Diatoms: Biology & Applications, series editors: Richard Gordon & Joseph Seckbach], S.A. Cohn, K.M. Manoylov, R. Gordon (Eds.), WileyScrivener, Beverly, MA, USA, 2020, This volume. [4.131] Saburova, M.A. and Polikarpov, I.G., Diatom activity within soft sediments: behavioural and physiological processes. Mar. Ecol. Prog. Ser., 251, 115–126, 2003. [4.132] Saburova, M.A., Polikarpov, I.G., Burkovsky, I.V., Spatial structure of an intertidal sandflat microphytobenthic community as related to different spatial scales. Mar. Ecol. Prog. Ser., 129, 229–239, 1995. [4.133] Sakamaki, T., Nishimura, O., Sudo, R., Tidal time-scale variation in nutrient flux across the sediment–water interface of an estuarine tidal flat. Estuarine Coastal Shelf Sci., 67, 653–663, 2006. [4.134] Sato, S., Beakes, G., Idei, M., Nagumo, T., Mann, D.G., Novel sex cells and evidence for sex pheromones in diatoms. PloS One, 6, e26923, 2011. [4.135] Sato, S. and Medlin, L.K., Motility of non-raphid diatoms. Diatom Res., 21, 473–477, 2006. [4.136] Schultz, M.P., Finlay, J.A., Callow, M.E., Callow, J.A., A turbulent channel flow apparatus for the determination of the adhesion strength of microfouling organisms. Biofouling, 15, 243–251, 2000. [4.137] Serôdio, J., Marques da Silva, J., Catarino, F., Nondestructive tracing of migratory rhythms of intertidal benthic microalgae using in vivo chlorophyll A fluorescence. J. Phycol., 33, 542–553, 1997. [4.138] Seymour, J. and Stocker, R., Resource patch formation and exploitation throughout the marine microbial food web. Am. Nat., 173, E15–E29, 2009. [4.139] Shniukova, Е.I. and Zolotareva, E.K., Diatom exopolysaccharides: A review. Int. J. Algae, 17, 50–67, 2015. [4.140] Sigmon, D.E. and Cahoon, L.B., Comparative effects of benthic microalgae and phytoplankton on dissolved silica fluxes. Aquat. Microb. Ecol., 13, 275–284, 1997. [4.141] Sims, P.A., Mann, D.G., Medlin, L.K., Evolution of the diatoms: insights from fossil, biological and molecular data. Phycologia, 45, 361–402, 2006. [4.142] Son, K., Brumley, D.R., Stocker, R., Live from under the lens: Exploring microbial motility with dynamic imaging and microfluidics. Nat. Rev.: Microbiol., 13, 761–765, 2015. [4.143] Spaulding, S.A., Bishop, I.W., Edlund, M.B., Lee, S., Furey, P., Jovanovska, E., Potapova, M., Diatoms of North America, 2020, Website. https://diatoms.org, Accessed April 5, 2021. [4.144] Stanley, M.S. and Callow, J.A., Whole cell adhesion strength of morphotypes and isolates of Phaeodactylum tricornutum (Bacillariophyceae). Eur. J. Phycol., 42, 191–197, 2007. [4.145] Stocker, R., The 100 µm length scale in the microbial ocean. Aquat. Microb. Ecol., 76, 189– 194, 2015.

108  Diatom Gliding Motility [4.146] Stocker, R. and Seymour, J.R., Ecology and physics of bacterial chemotaxis in the ocean. Microbiol. Mol. Biol. Rev., 76, 792–812, 2012. [4.147] Sundbäck, K., Enoksson, V., Graneli, W., Pettersson, K., Influence of sublittoral microphytobenthos on the oxygen and nutrient flux between sediment and water: A laboratory continuousflow study. Mar. Ecol. Prog. Ser., 74, 263–279, 1991. [4.148] Sundbäck, K. and Granéli, W., Influence of microphytobenthos on the nutrient flux between sediment and water: a laboratory study. Mar. Ecol. Prog. Ser., 43, 63–69, 1988. [4.149] Sutherland, T.F., Grant, J., Amos, C.L., The effect of carbohydrate production by the diatom Nitzschia curvilineata on the erodibility of sediment. Limnol. Oceanogr., 43, 65–72, 1998. [4.150] Taira, H., Kondo, S., Kumashiro, Y., Mayama, S., Umemura, K., Differences in dynamic behavior of single diatom cells caused by changing wavelengths. Micron, 108, 1–5, 2018. [4.151] Thompson, S.E.M., Taylor, A.R., Brownlee, C., Callow, M.E., Callow, J.A., The role of nitric oxide in diatom adhesion in relation to substratum properties. J. Phycol., 44, 967–976, 2008. [4.152] Umemura, K., Haneda, T., Tanabe, M., Suzuki, A., Kumashiro, Y., Itoga, K., Okano, T., Mayama, S., Semi-circular microgrooves to observe active movements of individual Navicula pavillardii cells. J. Microbiol. Methods, 92, 349–354, 2013. [4.153] Umemura, K., Miyabayashi, T., Taira, H., Suzuki, A., Kumashiro, Y., Okano, T., Mayama, S., Use of a microchamber for analysis of thermal variation of the gliding phenomenon of single Navicula pavillardii cells. Eur. Biophys. J., 44, 113–119, 2015. [4.154] Umemura, K., Sadoya, Y., Nagao, K., Oikawa, R., Hanada, Y., Sugioka, K., Mayama, S., Single cell analysis using a glass microchamber for studying movement fluctuations of Navicula pavillardii and Seminavis robusta diatom cells. Micron, 77, 41–43, 2015. [4.155] Underwood, G.J.C. and Kromkamp, J., Primary production by phytoplankton and microphytobenthos in estuaries. Adv. Ecol. Res., 29, 93–153, 1999. [4.156]  Underwood, G., Perkins, R., Consalvey, M., Hanlon, A., Oxborough, K., Baker, N., Paterson, D., Patterns in microphytobenthic primary productivity: Species-specific variation in migratory rhythms and photosynthetic efficiency in mixed-species biofilms. Limnol. Oceanogr., 50, 755–767, 2005. [4.157] Underwood, G.J. and Paterson, D.M., The importance of extracellular carbohydrate production by marine epipelic diatoms. Adv. Bot. Res., 40, 183–240, 2003. [4.158] Van Colen, C., Underwood, G.J., Serôdio, J., Paterson, D.M., Ecology of intertidal microbial biofilms: Mechanisms, patterns and future research needs. J. Sea Res., 92, 2–5, 2014. [4.159] Wang, J.D., Cao, S., Du, C., Chen, D.R., Underwater locomotion strategy by a benthic pennate diatom Navicula sp. Protoplasma, 250, 1203–1212, 2013. [4.160] Wetherbee, R., Lind, J.L., Burke, J., Quatrano, R.S., The first kiss: Establishment and control of initial adhesion by raphid diatoms. J. Phycol., 34, 9–15, 1998. [4.161] Wigglesworth-Cooksey, B. and Cooksey, K.E., Use of fluorophore-conjugated lectins to study cell-cell interactions in model marine biofilms. Appl. Environ. Microbiol., 71, 428–435, 2005. [4.162] Wigglesworth-Cooksey, B., van der Mei, H., Busscher, H., Cooksey, K., The influence of surface chemistry on the control of cellular behavior: studies with a marine diatom and a wettability gradient. Colloids Surf. B: Biointerfaces, 15, 71–80, 1999. [4.163]Witkowski, A., Brehm, U., Palinska, K.A., Rhiel, E., Swarm-like migratory behaviour in the laboratory of a pennate diatom isolated from North Sea sediments. Diatom Res., 27, 95–100, 2012. [4.164]Wustman, B., Gretz, M.R., Hoagland, K.D., Extracellular Matrix Assembly in Diatoms (Bacillariophyceae): I. A model of adhesives based on chemical characterization and localization of polysaccharides from the marine diatom Achnanthes longipes and other diatoms. Plant Physiol., 113, 1119–1069, 1997.

Motility of Biofilm-Forming Benthic Diatoms  109 [4.165]  Wustman, B.A., Lind, J., Wetherbee, R., Gretz, M.R., Matrix assembly in diatoms (Bacillariophyceae): III. Organization of fucoglucuronogalactans within the adhesive stalks of Achnanthes longipes. Plant Physiol., 116, 1431–1441, 1998. [4.166] Yamaoka, N., Suetomo, Y., Yoshihisa, T., Sonobe, S., Motion analysis and ultrastructural study of a colonial diatom, Bacillaria paxillifer. J. Electron Microsc., 65, 211–221, 2016. [4.167] Zelner, D.J., Analysis of Directed Extracellular Fluorescent Bead Movements over the Surface of Diatoms, DePaul University Master’s Thesis, 2005, DePaul Special Collections Call# 579.85 Z51a2005, Chicago IL, USA.

5 Photophobic Responses of Diatoms – Motility and Inter-Species Modulation Stanley A. Cohn*, Lee Warnick and Blake Timmerman Department of Biological Sciences, DePaul University, Chicago, Illinois, USA

Abstract

Many benthic diatoms have long been known to be photosensitive, altering the direction of their movement in response to changes in ambient light conditions detected at the tips of the cells. This review outlines the current understanding of some of the basic photoresponses of such cells and the light conditions responsible for triggering positive photophobic (out-of-light) and negative photophobic (into-light) responses of diatoms. Our research has substantially come from work on four species of diatoms (Craticula cuspidata (Kützing) D.G. Mann, Stauroneis phoenicenteron (Nitzsch) Ehrenberg, Nitzschia linearis (Agardh) W. Smith, and Pinnularia viridis (Nitzsch) Ehrenberg), all isolated from the same pond. By selectively irradiating tips of cells with short bursts of light at various irradiance levels and wavelengths, as well as measuring the rate of cell accumulation into small spots of light, we determined the relative wavelengths and intensities of light that elicited light-based motile responses in each species. In addition, analysis of cellular responses suggests that each species has its own characteristic responses with regards to into-light movement, out-of-light movement, and habituation of response. These studies suggest that C. cuspidata is a rapidly moving cell, with lower persistent substratum adhesion, that is most sensitive to blue light, moving into moderate level blue light, and out of high- and low-level light. In contrast, S. phoenicenteron is a strongly adhering cell sensitive to moving into moderate to low-level red light. The P. viridis cells are cells with moderate adhesion, relatively insensitive to changes in light regimes, and a circular path of movement that minimizes large distance accumulation into light spots. In addition, these responses can be modulated when the cells are in multi-species assemblages in the presence of one or more of the other diatom species. These differences in light-based motile responses of the three species suggest that diatoms not only have generally adaptive behavioral mechanisms, but also likely have the mechanisms to undergo niche partitioning and separation of species within local environments that might lower competition and increase ecological success. Keywords:  Diatom motility, photophobic response, diatom ecology

*Corresponding author: [email protected] Lee Warnick: [email protected] Blake Timmerman: [email protected] Stanley Cohn, Kalina Manoylov and Richard Gordon (eds.) Diatom Gliding Motility, (111–134) © 2021 Scrivener Publishing LLC

111

112  Diatom Gliding Motility

5.1 Introduction Diatom cells are contained within hardened silica cell walls creating severe restrictions on any direct cell/substratum contact. Nonetheless, pennate benthic diatoms are often highly motile cells, displaying active locomotion over surfaces in their local environments. This active movement appears to require secretions of mucilaginous material from the cell through specialized slits (raphes) in their cell wall, a motile behavior that has been observed and studied for over 125 years [5.11] [5.21] [5.23] [5.27] [5.36] [5.39] [5.48] [5.55] [5.56]. The gliding movement of these motile diatoms requires contact with an underlying surface and shows characteristic smooth paths of movement that generally align with the direction and orientation of the raphe [5.4] [5.5] [5.18] [5.22]. As typical for many photosynthetic motile algae, diatom movement is regulated, with light-based motile responses that allow the cells to bias their movement to inhabit favorable and avoid unfavorable light conditions [5.27] [5.29] [5.51] [5.54]. This light-sensitive behavior appears to be due to a light response that causes cells to positively or negatively bias their change of direction at light/dark boundaries, or when there is a strong light differential between the two ends of the cells. Because the orientation of movement in most motile diatoms is somewhat fixed along the axis defined by the raphe, this type of phototaxis has historically been characterized as a photophobic response to light [5.49] [5.50], either a positive photophobic response (moving the cell out of the light) or a negative photophobic response (moving the cell into the light). While the specific photoreceptors responsible for regulating the movement of diatoms remain unknown, the localization of diatom photoresponses at the tips of the cell, and not necessarily in the area of the chloroplasts [5.10] [5.11] [5.15] [5.16] [5.27] [5.58-5.62], along with the potential need for intracellular Ca2+ [5.45], suggest that the photoreception requires one of the several types of membrane-based photoreceptors [5.8] [5.19] [5.25] [5.31] that are unrelated to those used in photosynthesis. These photoresponses are thought to be responsible for the well-known observation of some benthic diatoms to use light to modify their vertical movements within sediments [5.2] [5.28] [5.37], or their relative accumulation in localized areas of light [5.44] in order to increase their ability to gain light energy and maintain their ecological success. This paper will summarize research on diatom photophobic behavior, concentrating on the work from our lab involved in investigating species-specific and inter-specific responses, to better understand the nature of their movement and potential ecological ramifications.

5.2 Types of Observed Photoresponses Our lab has investigated several types of light-based motile responses, listed below, in order to characterize the spectral and irradiance sensitivity of the cells in each of these types of behaviors.

5.2.1 Light Spot Accumulation One of the first types of observed photoresponsive diatom behaviors to be critically examined was described by Wenderoth [5.58-5.62], analyzing the ability of diatom movement to be responsive to small patches of light exposed on the cell that caused the cells to move

Photophobic Responses of Diatoms  113 towards and accumulate in areas of light. Two underlying aspects of this behavior of cells to move into appropriate areas of light—the directional behavior biasing individual cells reaching light/dark boundaries, and the resulting bulk accumulation of cells into light spots—have been further measured and analyzed in our lab. Light/Dark Boundaries. In order to observe the photobehavior of motile cells arriving at light/dark boundaries, a standard upright microscope with a 100W tungsten light was fitted with a small field aperture and narrow band light filters, allowing for irradiation of cells with 400-700 nm light, in 50 nm increments [5.16]. Single cells could be placed within a small light spot (ca. 200 µm) of moderate irradiance (ca. 50 µmol/m2s), allowed to move, and then analyzed for their behavior when the cells arrived at the light/dark boundary. This technique allowed us to adjust the wavelength of the filter used, and directly compare the rate of direction change at light/dark boundaries with the basal rate of cell direction change for cells under similar light conditions but without such light boundaries. In this way we could determine both the basal rate of direction change for different species under particular light conditions, as well as the spectral responsiveness for each species at light/ dark boundaries. Each of the four species tested had characteristic wavelengths at which the cells would most actively change direction and reverse movement to re-enter the light spot [5.16]. The C. cuspidata cells had the strongest light preference for moving into blue-green light (ca. 60% of cells changing direction at the boundary) and a weaker response to move out of red light. The N. linearis cells also had a strong blue-green preference, but no out-of-light red response. In contrast, the S. phoenicenteron cells had a strong preference for moving into red light (ca. 30% net positive response) and a weak out-of-light blue response. The P. viridis also showed some blue preference and red avoidance, but with a weaker response than the other cells. These species also showed two general trends in terms of basal rate of direction changes under exposure to the different light frequencies (i.e., no distinct boundary). The C. cuspidata, P. viridis, and N. linearis cells all showed a general decrease in basal rate of direction change under blue-green light, and increased basal rate of direction change under red light. In contrast, the S. phoenicenteron cells showed a decreased rate of basal rate of direction in red light. In both groups of cells, it appeared that cells changed direction less under those general light conditions when the whole cell was exposed to the light and would most rapidly change direction when only one of the ends of the cell was exposed to the light. While the four species each showed characteristic average cell speeds, there was no significant wavelength- or irradiance-dependent change in cell speed [5.9] [5.16]. It thus seems likely that rather than cell speed, the light exposures primarily affect the degree to which cells are biased or stimulated to change their direction of movement. That is, light quality affects the cells’ directional bias and propensity for spending time in a particular type of light, but not the actual rate at which they move into the light. Cell Accumulation into Light Areas. To determine the rate at which cells would accumulate into defined areas of light, small holes of light (ca. 3-7 mm diameter) were made in the bottom of otherwise light-restricted Petri plates. Samples of cells from culture were inoculated into liquid diatom medium within these Petri plates. These small culture plates were then placed onto a stereomicroscope fitted with a fiber optic light source and broadband

114  Diatom Gliding Motility filters, and subsequently exposed to various intensities and wavelengths to determine the rate of net accumulation of cells into the light spots [5.12]. As with the direction change response, accumulation of cells into the light spot was dependent on both light intensity and wavelength [5.12]. Similar to the responses seen with the boundary response assay, light accumulation into spots showed that C. cuspidata had strong positive accumulation response with blue light, but little accumulation response with red light. In contrast, S. phoenicenteron had a moderate accumulation response in low-level red light along with a strong accumulation response into moderatelevel blue light. The P. viridis had little accumulation into any of the light spots despite active cell movement, likely due to their constrained circular path of cell movement [5.16] [5.22] resulting in their poor ability to move long distances to search out new surfaces for light. The rates by which the cells accumulate into the light spots correlates well with a mathematical model of cell accumulation based on assigning into-light and out-of-light rate constants for cell movement at the light/dark boundary [5.9]. Our measurements of the accumulation rates of cells into the light spots match the model in terms of the net cell density increases being inversely proportional to the radius of the spot, and the rate of net cell increase into the spot (cells/min) being directly proportional to the perimeter (and thus radius) of the spot. This supports the concept that diatom accumulation is based on the frequency of direction change decisions made by the cells as they encounter boundaries of light with different wavelength or irradiance, in which the two ends of a cell each detect different light conditions. Perhaps most important to the understanding of localized diatom distribution, the spot accumulation studies showed that species distributions in an area of light can be modulated by the quality of the light exposure (see, e.g., fig. 5.4 in [5.12]). By placing a mixed assemblage of C. cuspidata and S. phoenicenteron cells in the same culture, and then exposing the culture to different light wavelengths, the relative abundance of the two species present in the light spot could be changed dramatically. If the mixed assemblage was exposed to a light spot with lower levels (10-25 µmol/m2s) of blue light, the relative abundance of C. cuspidata cells in the spot increased with little change in S. phoenicenteron abundance, causing a 2-3fold increase in the C. cuspidata:S. phoenicenteron ratio. Subsequently shifting that same culture into the same irradiance level red light resulted in a similarly asymmetric increase in S. phoenicenteron abundance, greatly lowering the C. cuspidata:S. phoenicenteron ratio. These accumulation experiments thus provide direct evidence that the quality of light can directly affect the relative abundance and distribution of diatom populations that are present within a localized area of light exposure.

5.2.2 High-Intensity Light Responses Early work on diatom photophobic responses, particularly by Nultsch and Wenderoth, suggested that the triggers regulating these light responses were localized at the tips of the cells, and very much responsive to levels of light irradiance [5.49] [5.50] [5.58-5.60]. In addition to the into-light photophobic response of the diatoms under low to moderate light conditions described above, it is now clear that motile diatoms also show a dramatic behavior of rapidly changing direction to avoid remaining in the presence of high-intensity light that could potentialy induce photodamage [5.10] [5.14] [5.15]. Our lab attempted to

Photophobic Responses of Diatoms  115 determine some of the characteristic parameters influencing this photo-induced direction reversal response, such as the total energy exposure needed for the response, the temporal persistence of the response, any habituation of the response, and whether there were species-specific differences of the response. Localization of High-Irradiance Response. Localization experiments in C. cuspidata confirmed that the light sensitivity for the reversal response appears to be localized at cell tips [5.15]. Irradiation of moving cells with 1s high irradiance blue light (ca. 470 nm, >104/ m2s) at the tips of the leading end of cells caused 100% of treated cells to change direction within ca. 30 s. Treatment of cells further toward the middle of the cell, away from the tip, or on the edge of the cell (where the chloroplast was located) caused significantly lower response rates (ca. 10% and 25% cells responding respectively). These rates were insignificantly different than rates of cell reversal in control cells that were not irradiated, and also suggested that cellular responses were not due to irradiation of the chloroplast photosystems. Evidence also suggested that both tips of the cell are simultaneously sensitive to light [5.15]. High irradiance exposure of the whole cell (thus irradiating both tips), or of the leading tip directly after a prior exposure of the trailing tip of the cell, caused about a third of the cells to undergo multiple rapid direction changes, i.e., oscillating back and forth with no clear directionality. Moreover, the exposure at the trailing end seems responsive in repressing the direction change of cells, even when no irradiation of the leading tip occurs. While about 15% of cells changed direction on their own in unirradiated control cells, no cells changed direction within 30 s after the trailing tip of the cell was irradiated [5.15]. If the trailing end was repeatedly irradiated with high irradiance exposures [5.14], cells could be induced to continue moving in the same direction for well over 10 minutes (cells that are not irradiated typically change direction within 1-4 minutes depending on cell type). Thus, high irradiance exposure at leading ends of cells stimulated a reversal in cell direction, while high irradiance exposure at trailing ends repressed such a change in direction, generating directional persistence, while irradiation in the middle of the cells generates little effect. Stimulation of both ends at the same time causes a variable effect with cells often stopped, or oscillated rapidly, changing direction every few seconds. Energy and Exposure Threshold. Investigations into the relationship between the light exposures and the cell direction change responses demonstrated that for very high irradiance, the cells require a minimum exposure of about 20-30 ms to initiate a cell reversal response [5.10]. Moreover, the response seems to require an energy threshold for total light exposure (i.e., total energy accumulated during the exposure). When cells were exposed with reducing irradiance levels, a 10-fold or 100-fold reduction in irradiance resulted in a respective 10-fold or 100-fold increase in the exposure time required to induce a direction change. These early estimates suggested that any light receptor mechanism requires a light pulse of a minimum total energy exposure of ca. 600 J m-2 to trigger the direction change (corresponding to an actual energy exposure on the cell of ca. 60 nJ). This seems to be coupled to the requirement of a minimum irradiance of > 100 µmol m-2 s-1 in order to trigger the out-of-light response. Lower irradiances seem to trigger the into-light responses, even if exposed for longer periods. Having this lower light level for effective movement into the

116  Diatom Gliding Motility light would correlate well with other measurements of 10-50 µmol m-2 s-1 for the light levels used for optimal photosynthesis [5.46]. An estimate of the rate of energy transfer by a putative light receptor was determined by flashing the tip of the cell with two 15 ms pulses of high-irradiance light separated by increasing amounts of time. Below 0.5 ms spacing the two pulses produced a strong stimulation of direction change (> 60% of cells changing direction), similar to a single 30 ms pulse, while spacing the light pulses apart at > 5 ms had no stimulation of direction change above background level. Irradiation of cells with 100 ms total exposure separated into multiple pulses also showed a strong reduction of response when the space between pulses was greater than 0.3 ms. The 30 ms stimulus threshold for high irradiance effect is comparable to the time required for high irradiance blue light to generate second messenger production using the photoreceptors in Euglena [5.63]. Likewise, the ca. 1 ms range for the loss of membrane signaling is typical for the response half-life for many types of membrane photoreceptors [5.32] [5.42]. The overall threshold requirement for a 30 ms pulse of ca. 5 × 105 µmol m-2s-1 blue light irradiance (or ca. 15 mmol m-2 photon exposure) is well within the range of light requirements seen to induce responses in other algae [5.20], and similar to the rapid kinetics seen in some diatom Na+/Ca2+ membrane channels correlated with the photophobic response [5.31]. Species Characterizations of Into-light and Out-Of-Light Responses. Each of the three large species normally cultured in our lab was analyzed for their response to single pulse irradiations at the tip of the cell at different irradiance levels. By irradiating at either the leading or the trailing end we could determine whether the irradiation induced an intolight or out-of-light response. If an irradiation at the trailing end caused the cell to change direction quickly it would be a sign of an into-light response (suggesting light conditions favorable for the cell), while a rapid change of direction after a leading-end irradiation would indicate an out-of-light response. By the addition of neutral density filters or narrow band wavelength filters into the light path we could regulate the light irradiance and wavelength for the light pulses and determine the threshold for these into-light and out-of-light responses for the two ends for each of the three species at different wavelengths. The three species were tested using 1 s exposures at irradiance levels over the range of ca. 105 to 103 µmol m-2s-1 at each of 470 nm (blue), 540 nm (green), and 650 nm (red) wavelengths. Such tests [5.14] showed that as with previous studies, each of the three species (S. phoenicenteron, C. cuspidata, P. viridis) had distinct light sensitivities. At the highest levels of irradiance, all three species showed strong out-of-light responses at blue, green and red light. The high irradiance pulses given at the leading end of the cell generated relatively rapid changes in cell direction (20-50 s), while irradiations at the trailing end tended to cause a slight repression of direction change compared to control (unirradiated) cells, with cells usually changing direction after 150-200 s. However, other than these responses at the highest irradiance level of blue light, each species displayed characteristic differences. For C. cuspidata cells, the strong out-of-light response for leading-end exposures to blue light (ca. 450 nm) steadily decreased as the blue irradiance level was reduced, so that cells were responding at near control (unirradiated) levels when exposed to pulses at 1% of maximal irradiance. In contrast, average into-light direction change responses at the trailing end remained negligible at high blue light irradiance (> 130 s for direction change), with responsiveness increasing dramatically at the intermediate levels of light (ca. 40 s for

Photophobic Responses of Diatoms  117 response time), then again showed increasing response times at diminishing light levels. In contrast to the blue light responses, the C. cuspidata cells showed little response to green light (ca. 100-120 s response times) across a wide range of irradiance levels, either at the leading or trailing ends. The red light irradiation response of C. cuspidata cells at the leading end was similar to that of blue light, starting out with a significant response time (ca. 40 s) at the highest irradiance, diminishing as the irradiance levels were lowered to that of no significant response over control at the lowest light levels tested. The exposures at the trailing end of C. cuspidata showed negligible into-light response in red light (ca. 120-160 s response times). For P. viridis cells, there was a moderate out-of-light response for leading-end exposures to blue light that decreased as irradiance decreased (from ca. 50 s to 150 s), with little intolight response at the trailing end except a small into-light blue response at dim light levels. The P. viridis cells showed little light sensitivity at all to green or red light. The S. phoenicenteron also showed a decreasing out-of-light response to blue light as the irradiance decreased (from ca. 50 s to 150 s), with little into-light response with blue irradiances (ca. 150-200 s). The S. phoenicenteron also displayed no response to green light irradiations at all but the highest irradiance. However, as with the light boundary and light spot accumulation tests, S. phoenicenteron was the only species we have observed to show red light directional sensitivity, displaying an increasing into-light response at low to dim red light levels (response times diminishing from ca. 250 s to ca. 100 s). Thus, each of the three species tested displayed a distinct and characteristic response to the light exposures, with responses that correlated strongly with our earlier light boundary and light spot responses—C. craticula is strongly responsive to blue light, and S. phoenicenteron is responsive to red light. There is little green light sensitivity in any of the species, which is reasonable considering there is little selective pressure to move into areas of green light, the least well captured wavelengths for photosynthesis. Cellular Memory of Light Responses. The degree to which a light pulse represses or modifies any subsequent responses (i.e., the effective physiological memory of a previous light pulse) could be measured by giving a leading-end or trailing-end irradiation to a cell, followed by a subsequent leading-end test irradiation. Such experiments [5.14] showed that a high-irradiance exposure at the trailing end of a cell caused a strong repression of a subsequent leading-end response for a period of time that was characteristic for each species. The C. cuspidata cells were repressed for 5-10 s at all wavelengths tested, P. viridis were repressed for 10-30 s for blue and green light (red light had no effect on P. viridis), and S. phoenicenteron were repressed for up to 30 s. Such experiments indicate that high irradiance pulses at the trailing end reinforce the out-of-light response, keeping the cell moving away from high-irradiance light. The trailing-end irradiation results were further confirmed by testing the ability of sequential trailing-end irradiations to repress any direction changes. When C. cuspidata cells were given repeated trailing end pulses of blue light spaced 15 s apart, cells could be driven to maintain their cell direction for well over 10 minutes [5.14]. The amount of time the cell sustained its direction decreased as the interval time between pulses increased, dropping significantly once the pulses were 60-90 s apart. The S. phoenicenteron cells treated in a similar way also showed reduced repression when the pulses were spaced greater than 30-60 s apart, showing this same effect for irradiations with both red and blue light. The P. viridis cells also

118  Diatom Gliding Motility showed strong repression by trailing-end irradiations, although this repression appeared to be longer lasting, showing repression even when the pulses were over 120s apart. Similarly, by irradiating cells at their leading end, letting them change directions, and then testing the response of a second irradiation at the new leading end, we could determine the degree and duration of leading-end repression of light effects. Such experiments [5.10] showed that the response time of a second leading-end exposure was 60-70% greater if given within 20 sec of the first exposure. The second response time returned to control cell levels by 60 s after the first irradiation. Taken together, these experiments suggest that the high light irradiations generate cellular effects that repress subsequent changes to the cell direction for at least 15 sec and up to 1 min. This refractory period likely gives greater effectiveness to driving the cell in the properly stimulated direction as cells search out new areas with optimal light levels. In this way the diatoms would avoid getting stuck in an oscillating state, remaining in an area with less effective light. Habituation of Light Responses. The degree to which cells can acclimate, or habituate, to these light-induced responses can also be determined. To measure this effect, C. cuspidata and S. phoenicenteron cells were irradiated at their leading ends, allowed to change direction, then re-irradiated again at the new leading end [5.14]. The response time could then be measured as a function of the number of times the cell had been irradiated. The C. cuspidata and S. phoenicenteron showed distinct and characteristic responses. The C. cuspidata cells showed strong blue light repression after the first irradiation, with the second and third irradiations having about twice the direction change response times of the initial response. This diminished somewhat to close to initial levels after about ten irradiations. In contrast, the red light irradiations of C. cuspidata showed an increasing repression with number of exposures, with again the response time doubling by the tenth exposure, and then diminishing somewhat for more than 12 irradiations. The S. phoenicenteron cells also showed a slowly increasing repression as a function of blue light irradiations, peaking at around six to seven irradiations, and then diminishing back to control response times after about 10 irradiations. However, the red light irradiations showed great variability, and no significant change in repression as a function of number of irradiations.

5.3 Inter-Species Effects of Light Responses The adhesion of diatoms to surfaces, as measured by the rate of diatom loss from surfaces due to gravity or water flow, has been shown to be modified by the presence of other diatom species. For example, the rate of P. viridis cell loss from the underside of a coverslip was significantly increased when there were either C. cuspidata or S. phoenicenteron also present in the assemblage [5.13]. Similarly, C. cuspidata cells had increased rates of cell loss in the presence of P. viridis. Interestingly, the S. phoenicenteron loss rate was unaffected by the presence of P. viridis. There is considerable evidence in pennate diatoms for a direct connection between the adhesion from raphe secreted mucilage and resulting motility (e.g., [5.33]), suggesting that the adhesion changes resulting from differing diatom composition in the assemblage may affect motile behavioral responses as well.

Photophobic Responses of Diatoms  119

5.3.1 Inter-Species Effects on High Irradiance Direction Change Response To test the effect of photoresponse changes due to differing assemblage composition, the response time for P. viridis, C. cuspidata, and S. phoenicenteron to change direction upon a high irradiance 1 s blue light pulse at the leading tip was measured in the presence and absence of other species [5.12]. While most multi-species interactions showed no significant changes in motile response times, S. phoenicenteron showed significantly increased direction change response times (i.e., reduced responsiveness) when in the presence of C. cuspidata, P. viridis, or both. This multi-species response was concentration-dependent, with the response times for S. phonicenteron increasing as the C. cuspidata:S. phoenicenteron ratio increased [5.12]. Subsequent experiments have shown that this effect is also time-dependent, with C. cuspidata requiring 20-30 min of co-incubation time with S. phoenicenteron to get maximal repression of Stauroneis light response (Figure 5.1). The effect has also been confirmed for S. phoenicenteron in the presence of P. viridis. When S. phoenicenteron are placed directly in the presence of P. viridis there is a significant increase in the S. phoenicenteron response times to the 1 s blue light irradiations (Figure 5.2). In order to determine if the effects are due to some type of small rapidly diffusing molecule secreted by the cells, two types of experiments were performed. In the first approach, S. phoenicenteron were tested for light-stimulated response time in the presence of culture medium in which either S. phoenicenteron alone, C. cuspidata alone, or a C. cuspidata/ S. phoenicenteron mix were grown for approximately a week. The medium was obtained from the growing cell culture, then centrifuged briefly in a microcentrifuge to remove any 70

Response Tims (s)

60 50 40 30 20 10 0

Staur Alone

0-10

10-20 20-30 30-40 Incubation Time (min)

40-50

50-60

Figure 5.1  The effect of incubation time on Stauroneis response times in a mixed culture of Craticula cuspidata and Stauroneis phoenicenteron. This graph displays the average direction change response times for Stauroneis phoenicenteron cells in the presence of ca. 9:1 ratio of live C. cuspidata: S. phoenicenteron cells. S. phoenicenteron cells were isolated and washed from culture and incubated together with C. cuspidata cells (C/S) in a 9:1 C. cuspidata:S. phoenicenteron ratio. Cells were then irradiated at their leading end with high irradiance (ca. 104 µmol/m2s) 1s pulses of blue (470 nm) light, and observed to determine the time until they changed direction (response time). Response times significantly increased almost 2-fold from the initial incubation interval (0-10 min) within 20-30 min (30±2 µm/s and 57±7 µm/s respectively, P=0.003). Graphs represent the mean values of response times ± 1 SE. For comparison, unirradiated Stauroneis cell response times were 155±11 µm/s. Error bars represent ± 1 SE.

120  Diatom Gliding Motility 160

Avg. Response Time (s)

140 120 100 80 60 40 20 0 Control Stauroneis

Isolated Stauroneis

Stauroneis in presence of C. cuspidata

Stauroneis in presence of P. viridis

Treatment Group

Figure 5.2  The effect of Craticula cuspidata or Pinnularia viridis presence on Stauroneis phoenicenteron response times. This graph displays the average direction change response times for S. phoenicenteron cells alone on slide chamber (Control). On a mixed slide chamber in a group by themselves (Isolated) or in a slide chamber in the presence of a high ratio of live C. cuspidata or P. viridis cells to S. phoenicenteron cells. Cells were then irradiated at their leading end with high irradiance 1s pulses of blue light, and observed to determine the time until they changed direction (response time). The presence of either C. cuspidata or P. viridis caused significant increases (P < 0.03) in cell response time from either control or isolated S. phoenicenteron. Error bars represent ± 1 SE.

contaminating cells or cell debris. While there was a very slight increase in response time in the non-Stauroneis media, none of these media caused any significant change from the control S. phoenicenteron response time (Figure 5.3, P=0.41). In addition, we tested cells using a cell chamber in which a population of S. phoenicenteron alone was on one side of the chamber, and a mix of S. phoenicenteron with either C. cuspidata or P. viridis on the other side, both immersed in the same medium. In other words, it was a chamber in which both Stauroneis cell populations (either alone or in the presence of other cells) were bathed in the same cell medium. In these cases, the S. phoenicenteron in the mixed populations showed significantly increased cell response times relative to the Stauroneis in the isolated single-species group (e.g., Figure 5.4, P 50 citings of the term were first recorded in the published literature (Web of Science). Different researchers place varied emphasis on the component populations of the MPB that includes both eukaryotes (benthic algae and protists) and prokaryotes (cyanobacteria, photosynthetic bacteria). Motile diatoms (epipelon) often dominate the microphytobenthos (Figure 6.1), particularly in muddy sediments [6.55]. In these habitats, the diatoms and the extracellular polymers they produce can lead to the formation of substantial, visible biofilms at the sediment-water interface. These biofilm matrices can modulate the flux of nutrients across the boundary [6.111] [6.21], influencing biogeochemical cycling and resource availability to the water column, an important functional role. Although individual diatom cells are usually small and inconspicuous, often less than 80 µm, the collective effects of their assemblages can be substantial. Their unusual mechanism of locomotion leads to the accumulation of EPS at the sediment surface, and while this may be advantageous to the diatoms [6.33] [6.119], it is becoming clear that these organic molecules may have much wider and sometimes surprising impacts, as will be described later in this chapter. While the differences between planktonic and benthic diatoms are considerable, benthic diatoms are often resuspended, which contributes directly to the planktonic community in nearshore waters. These resuspended cells support important suspension-feeding shellfish [6.74] and coastal fisheries [6.60]. More recently it has been shown that some wading shorebirds also exploit MPB [6.102], having structures in their bills to sift biofilm from the sediment surface [6.40]. There is a further category of lightly silicified diatoms that have adopted a “mixed” lifestyle of alternating benthic and suspended phases and these facultative phytoplankton are termed “Tychoplankton” [6.4]. Thus, benthic diatoms vary in lifestyle between fully sessile and immobile forms (e.g., stalked and adpressed) and highly motile MPB, some even forming tubes within which they can move rapidly (Figure 6.1).

6.3 The Ecological Importance of Locomotion Effective locomotion was an important evolutionary development among epipelic diatoms. Movement allows cell in a sediment matrix to optimize their local conditions of light and nutrients as well as avoid potential toxins and pathogens. The ability to move also allowed the evolution of complex patterns of assemblage migration towards and away from the sediment water/air interface, as conditions demand [6.25]. This response seems highly controlled [6.22] [6.39] [6.93] and acts at an assemblage level, with many different species behaving broadly in harmony or in response to very similar cues. As with many aspects of diatom research, the detailed mechanisms of these behaviors and the cues that drive them are still not fully understood. In the context of the transient biofilms that microphytobenthos form, even relatively small changes in vertical position can be extremely important. The most relevant variable for the MPB is probably photon flux density (light) which is rapidly attenuated with depth, particularly in cohesive sediments [6.55] [6.16]. In addition, the spectrum available in a particular layer with depth from the sediment surface, is dependent on the absorption and scattering of light in the layer above. Light can be too bright, too dim or of poor spectral quality and all of these features change on a micron scale within the sediment bed. High light intensity can reduce the photosynthetic capacity of MPB [6.106] but complete

138  Diatom Gliding Motility

Figure 6.1  Low-temperature scanning electron micrographs of diatom life styles (clockwise from top left): Stalked, tube forming, adpressed, epipelic. (Images: Irvine Davidson, University of St Andrews.)

photoinhibition is rare in MPB [6.124]. This may be due to the evolution of “micromigration” in diatoms, which provides an effective behavioral adaptation to avoid damaging light levels [6.93]. This locomotory mechanism of dealing with light compliments physiological adaptations, such as non-photochemical quenching [6.101], serving to protect the cells from the photo-damage that would otherwise take place [6.118]. Light can also influence microphytobenthic community composition and the migratory patterns of the different taxa will regulate the overall community level photosynthetic capacity [6.12]. Motility is therefore an important capability that influences the productivity and evolutionary success of microphytobenthic biofilms. This example outlines the importance of position in the sediment, but what is the capacity of diatoms to respond to this? The absolute distances involved are relatively small, and in cohesive sediments, light may become fully attenuated within 2 millimeters of the surface and conditions may also become anoxic [6.55]. Therefore diatoms do not have to move far to alter their physicochemical environment and since velocities of up to 35 µm.s-1 have been recorded for diatom locomotion [6.38], this would mean cells would be capable of moving across the 2 mm sediment photic zone in less than 1 minute. Change in diatom biomass at the sediment surface, both as an increase or reduction [6.26] [6.85] [6.116], have often been reported, confirming that a change in diatom distribution can be rapid, and within minutes. In terms of body length, the measured speed of diatom locomotion would be equivalent to a human running a hundred meters in 20 seconds, about half the speed of Usain Bolt, the past Olympic sprint champion. However, while this sounds impressive it’s not quite correct. It is important to distinguish between movement over surfaces measured in the laboratory, usually on glass slides, and movement within and over natural sediments and among the grains. The velocities reported under these conditions are a magnitude lower than those reported above [6.46]. In addition, speed

Diatom Biofilms and Ecosystem Engineering  139

Figure 6.2  Low temperature electron micrograph of the pathway of an epipelic diatom moving thorough fine silt (Bar marker = 10 um). (Image: Irvine Davidson, University of St Andrews.)

may not be as critical as power, since diatom cells have to force a pathway through sediment grains (Figure 6.2) and can often be observed moving large grains aside in sandy substrata and easily observable in sediment-rich cultures on glass slides (pers. obs.). Temperature may also differentially influence the locomotive ability of different species [6.23] so that diatom speed is highly variable, responding to ambient conditions and external stressors [6.24] [6.28], both increasing and declining in response to stimulus and sometimes also showing “hormesis” [6.14], an increase in activity (in this case, locomotive speed) caused by a normally toxic compound in low concentration. The importance of locomotion in depositional environments is therefore clear but subtleties in cell positioning and local context such as light, temperature and sediment properties are still being revealed [6.56] [6.101] [6.45].

6.4 Ecosystem Engineering and Functions 6.4.1 Ecosystem Engineering Individual organisms or populations that strongly affect an ecosystem, through direct modification of their environment, and subsequently influence the behavior and potential evolutionary fitness of other species, are often referred to as “keystone species” [6.81] and their activity referred to as “ecosystem engineering” [6.57]. In the early literature, examples of keystone species or ecosystem engineers were usually large charismatic species such as beavers, otters, etc., but while an individual microbial cell may not have great impact, the collective and combined activity of many microbes certainly can [6.10]. It is therefore sensible to recognize that an assemblage of microbial cells can indeed act as ecosystem engineers, often achieving significant biogeochemical transformations such as nitrogen-fixation, primary production and physical effects as in the case of sediment stabilization by MPB [6.88].

140  Diatom Gliding Motility MPB are not unique in this regard, as a similar ecosystem engineering function has been attributed to microbial assemblages, dominated by cyanobacteria, such as those forming the biological soil crusts (BSC) found on coastal dunes [6.104].

6.4.2 Ecosystem Functioning The terminology around environmental science is continually evolving with the need for more precise, descriptive terms for new and emerging concepts but also, apparently, with an innate human love of jargon. The definition of ecosystem function or functioning is often given as “processes occurring in the ecosystem” but there are many definitions available [6.90], often reflecting the interest of the definer (Table 6.1). The growing emphasis on ecosystem function and services in marine management emphasizes the importance of organisms (biodiversity) in the functionality of natural and impacted ecosystems. The link between biodiversity and ecosystem function is based on the theory that biodiversity supports the critical transformational processes that occur in an ecosystem [6.134]. The impetus for this research is that unless “healthy” ecosystems are maintained then the damage will be experienced globally and at all social, economic and societal levels. This paradigm often rested on the loss of charismatic systems or species (e.g., coral reefs, mangroves) but every ecosystem is a dynamic complex of components and it may be the loss of a less recognized system or pathway that leads to the decline. Diatoms and other microphytobenthic organisms are vital for our marine ecosystems [6.50], however, they are often disregarded as ecosystem engineers or even as functionally important organisms. Nonetheless, the carbon fixation performed by diatoms and other microphytobenthic organisms, supports

Table 6.1  Varied definitions of ecosystem function. Ecosystem function defined: • The collective intraspecific and interspecific interactions of the biota, such as primary and secondary production and mutualistic relationship (W 1). • The interactions between organisms and the physical environment, such as nutrient cycling, soil development, water budgeting, and flammability (W 1). • The physical, chemical, and biological processes or attributes that contribute to the selfmaintenance of the ecosystem; in other words, what the ecosystem does. Some example of ecosystem functions are wildlife habitat, carbon cycling, or trapping nutrients (W 2). • The characteristics exchanges within an ecosystem are called ecosystem functions and in addition to energy and nutrient exchanges, involve decomposition and production of biomass (W 3). • The biophysical processes that take place within an ecosystem. These can be characterized apart from any human context (e.g., fish and waterfowl habitat, cycling carbon, trapping nutrient). The level of function depends on the capacity of the ecosystem (onsite features) and certain aspects of its landscape context (e.g., connectedness to other natural/human features, accessibility to birds, fish) (W 4). Key: W1: Ecosystem function: http://www.biology-online.org/dictionary/Ecosystem_function. W2: Ecosystem valuation: http//Ecosystemvaluation.org/glossary.htm. W3: Ecosystem function and services: http://www. sustainablescale.org/ConceptualFramework/ UnderstandingScale/BasicConcepts/ EcosystemFunctions Services.aspx. W4: Ecosystem valuation http://www.ecosystemvaluation.org/Indicators/economvalind.htm.

Diatom Biofilms and Ecosystem Engineering  141 the secretion of extracellular polymeric substances (EPS). EPS secretion and motility are closely linked to a variety of functional roles. For example, biofilms that are formed by motile diatoms are an important labile carbon source for bacteria [6.114] and metazoan food webs. The secretion of EPS, the formation of a biofilm matrix and diatom migration in surface sediments can also mediate large-scale processes such as sediment transport and the related geomorphology of coastal sediments [6.82] [6.17]. Biofilms also contribute to carbon and nitrogen recycling, and can modulate the flux of nutrients into the water column, and retention within the sediment bed [6.27] [6.76] [6.78]. Motile diatoms are therefore critical components of ecological networks and mediate biogeochemical cycles, sediment stability, water quality and ecosystem productivity [6.50]. Diatom assemblages have also been used to assess environmental conditions since each species has a range of tolerance that if properly understood can be used to interpret the current or past environmental context [6.121]. More recently, Virta et al. [6.123] have linked the productivity of Baltic coastal habitats to the functionality and diversity of benthic diatom assemblages, suggesting that if changing conditions were to affect diatom biodiversity, perhaps as a consequence of climate change, this would have serious implications for the ecosystem. Drylie et al. [6.36] also noted that the MPB contribution to benthic primary production may also help build resilience against changes in the water column such as increasing turbidity. However, there is no single easy explanation for understanding the complex linkages among components of ecological systems.

6.5 Microphytobenthos as Ecosystem Engineers 6.5.1 Sediment Stabilization The engineering activity of biota stabilizing sediments, although not initially recognized in this way, was recorded as early as the 1970s [6.75] and later by Manzenrieder [6.69]. The latter author noted that erosion measurements made in the field on natural sediments did not obey the expected relationships predicted by physical principles or correlate with previously described erosion relationships such as the Shields Curve [6.108] or the Yalin parameter [6.131]. An early experiment, probably not to be repeated, was conducted by de Boer [6.9] when the authors “poisoned” a large area of sandflats in the Netherlands to remove the effect of the natural biology, including MPB. This resulted in a rapid change in sediment transport and resultant bed morphology. It is now well-documented that EPS secreted by cyanobacteria and diatoms can modify the erosional properties of the sediment by increasing the cohesion between sediment grains. While this is particularly effective in fine-grained sediments, EPS can influence bedform development and sediment erosion thresholds in different sediments [6.82] [6.49] [6.117]. These effects are recognized as a form of ecosystem engineering (Table 6.2) and benthic diatoms are excellent examples of individually small but collectively effective ecosystem engineers [6.83]. The potential influence of diatoms, biofilm formation and EPS accumulation in natural systems is now receiving increasing research attention. This is partly due to a strong interdisciplinary drive to better understand the biogenic mediation of sediment behavior in order to improve the modeling of sediment transport under natural conditions [6.44] [6.19] [6.47] [6.3] [6.67]. A recent investigation carried out by Parsons et al. [6.82] supported the

142  Diatom Gliding Motility Table 6.2  The role of epipelic diatoms in ecosystem functions, ecosystem engineering effects and the ecosystem services that these contribute towards. Ecosystem functions

Ecosystem engineering effects

Ecosystem services

Role of locomotion

Carbon fixation

Oxygen penetration depth

Carbon sequestration, Primary productivity

Community migration optimizes productivity

Alteration of biochemical environment

Nutrient re-mineralization

Biofilm formation intercepts nutrients at the sedimentwater interface

The formation of dense patches of biofilm increase grazer abundance

Supporting higher trophic levels

Vertically structured, diverse biofilms can support greater diversity of grazers

Photosynthesis O2 production Biomass production Community metabolism Secondary productivity

Production of food and other raw materials

Makes spatially constrained nutritious food resource for marine food webs Rich in essential nutrients such as fatty acids

Organic carbon production and release

Sediment stability

Sediment formation

Alteration of biochemical environment

Shoreline protection Habitat formation & diversity

Motile diatoms reduce sediment erosion via the secretion of EPS and the formation of biofilms Creation of new habitat

Resilience Nutrient processing Water purification

Interception, storing and cycling of nutrients (N, P and Si)

Habitat formation Nutrient cycling Water purification

Biofilms formation regulates benthic-pelagic exchange of nutrients Biofilms trap pollutants and sediment from the water column Interception and competition for nutrients modulates the availability to other organisms Large motile diatoms store more nutrients for longer periods of time The resuspension of motile diatoms transports nutrients to adjacent habitat

Diatom Biofilms and Ecosystem Engineering  143 early work of de Boer [6.9] whereby the addition of a commercial polymer (used as a proxy for natural EPS) evoked a considerable change in the evolution of bed morphology across a variety of sediments. This study also demonstrated that the effects were dependent on EPS concentrations in the sediment. While most experiments in this field largely concentrate on the effects of EPS on erosion and bedload transport, it should be recognized that biological stabilization affects all stages of the erosion, transport, deposition and consolidation (ETDC) cycle of sediment dynamics. The interactions between biological and sedimentary processes can also lead to spatially self-organized patterns of diatom distribution that influence large-scale geomorphological features [6.3]. There also seems to be a feedback between geomorphological features, such that the stabilization potential of motile diatoms can be influenced by the topology [6.7] with diatom assemblages favoring, or replicating faster, on some parts of the bed over others, leading to a patchy distribution [6.54]. More recently, Malarkey et al. [6.67], Parsons et al. [6.82], Baas et al. [6.3] and Hope et al. [6.49] have confirmed the wider influence of EPS on bed geomorphology and structure across different sedimentary habitats. Given this recent research, it is not too much of a surprise that complex feedback might be expected. As diatoms secrete EPS, the more substantial biofilm matrix can help trap fine particles within sandy sediments [6.43], reducing the grain size [6.31] and modifying the local habitat structure and the environment for benthic fauna. The consequences of EPS “trapping” particles and creating coatings on other larger particles has also become as subject of interest for sedimentary and petroleum geologists as this can affect the quality of petroleum reservoirs. For example, the suggestion by Wooldridge et al. [6.130] is that biofilm-forming organisms, with diatoms identified as critical drivers, maintain a clay layer around larger particles, enhancing porosity and leading to more commercially attractive high-porosity sandstone deposits. This relatively fine-scale ecosystem engineering, the creation of sand grain coating [6.129] [6.130] is of considerable economic importance and again demonstrates that apparently a small change can have much greater consequences. Thus, biofilm formation around a grain may have an eventual geological impact but effects can also be more immediate. Van de Koppel et al. [6.59] considered the transition between sandy and muddy habitats to be an example of alternate stable states influenced by the successional stage of the MPB biofilm.

6.5.2

Beyond the Benthos

It is clear that the polymeric material secreted by MPB can have a dramatic effect on the stability and structure of the local environment. However, can this effect extend beyond the benthic habitat where EPS is secreted by motile diatoms? EPS on the sediment surface can be dissolved by incoming tides and becomes resuspended into the water column in dissolved forms so this is plausible. Furthermore, resuspended sediment can be coated with EPS, which can play an important role in the formation of clay-coated sand grains [6.130] and influence water column processes such as flocculation [6.68]. Recently, the influence of EPS on the structure and behavior of marine “oil snow” resulting from oil spills has been investigated [6.110] and may have significant influence on the distribution and behavior of spilled oil. The influence of organic coatings on cohesive sediment rheology [6.107] is now also under investigation and for the first time, Craig et al. [6.30], have demonstrated the influence of EPS on the behavior of gravity-driven density flows. If this work is confirmed in situ, this would be a very important advance in the understanding of these density flows that are critical for

144  Diatom Gliding Motility transporting sediment and organic carbon to the deep ocean. Furthermore, this information would help to predict the effect of density flows and the consequence of underwater landslips. EPS can therefore significantly modify the behavior of particles throughout nearshore environments and into the deeper ocean, having a more distributed range of engineering effects than formally recognized, many of which are not yet fully understood.

6.5.3 Diatom Architects The ability of diatoms to lay down highly regular species-specific microstructures using silica has attracted the attention of nanotechnologists [6.11]. While the formation of intricate silicate frustules may not initially seem a form of ecosystem engineering, the consequences of this ability is considerable, influencing cell fitness and the cycling of nutrients and minerals on a global scale [6.132]. Aspects of interest in diatom nanoarchitecture include drug delivery, optical properties and mechanical strength [6.135]. For example, planktonic diatoms seem more able to cope with shear than some competitors (e.g., dinoflagellate) and may actually be stimulated by turbulence in the water column [6.73]. The mechanisms that increase diatom fitness under turbulence are not well-known and further research is required. This resistance and even stimulation by surface shear is likely to be a consequence of the nature of the silica cell wall and its microfluidic properties. However, it has taken rapid advances in technology to recognize some of these effects. The importance of surface structure has been shown in planktonic cells but there may be advantages in terms of selective surface transport and dynamics that also increase the fitness of benthic forms since both the planktonic and benthic species share aspects of cell nanoarchitecture [6.72].

6.5.4 Working with Others: Combined Effects Benthic diatoms produce significant amounts of EPS, but so do other organisms, including bacteria, cyanobacteria and benthic fauna. Other organisms, especially macrobenthos, process sediments creating tubes and burrows, and generally have been considered to restrict biofilm growth and biogenic stabilization [6.13]. However, the reality may be more complex than this. Several studies have reported stabilization by other benthic organisms, but this effect is often due to the stimulation of the MPB rather than the other species directly affecting the stability [6.1] [6.84]. The complexity of ecosystem engineering interactions has been noted before [6.35], and bioturbation by macrofauna may reduce or increase MPB biomass and productivity in both intertidal [6.37] [6.20] and subtidal systems. An increase in MPB growth due to fertilization was demonstrated in mesocosm experiments by Androuin et al. [6.1], who were examining the impact of the invasive slipper limpet, Crepidula fornicata. In their impressively large field study, Donadi et al. [6.35] also noted the context dependency of the interactions of macrofauna and MPB but perhaps underplayed the stabilizing role of MPB as ecosystem engineers themselves. This situation is similar to that reported by Passarelli et al. [6.84], where two or more different ecosystem engineers (MPB and tube-forming polychaetes) both led to stabilization of the substratum and may also have increased each other’s fitness by reducing the risk of bed erosion. These authors termed this as “cooperative ecosystem engineering.” The effects of these context-dependent variations are unsurprising in complex ecological systems; however, we are not yet at a stage of knowledge where they could be easily modeled. Laboratory mesocosm, while informative, are still

Diatom Biofilms and Ecosystem Engineering  145 a simplified version of the natural variation arising from external dynamics and interaction with biotic components such as biogenic reef formation and enhanced sediment deposition among diverse assemblages.

6.5.5 The Dynamic of EPS Generally, MPB stabilization is related to polymer secretion but in many cases we simply do not know the short- and long-term fate of EPS produced by different groups of organisms. Even if motile diatoms are not the solitary source of stabilizing EPS, there is a wealth of evidence supporting the importance of benthic diatoms for sediment stability [6.115] [6.49]. More recent technological advances may perhaps resolve the uncertainty surrounding these contributions and effects. For example, the ability to image sediment particles and flocs in high resolution is increasing our understanding of the wider ecosystem engineering potential of EPS [6.127] [6.133]. This is, at present, painstaking work but is providing much needed, detailed information on the structure and distribution of materials within flocs. The use of molecular approaches such as next generation sequencing is also starting to reveal the functional attributes of target organisms and their potential role in EPS production and sediment stability [6.2] [6.110].

6.5.6 Nutrient Turnover and Biogeochemistry The presence of EPS and related biofilms affect more than the physical properties of soft sediment ecosystems. Jones et al. [6.57] defined ecosystem engineers as those organisms that “directly or indirectly modulate the availability of resources (other than themselves) to other species.” Hence, the mediation of the supply of inorganic nutrients by MPB to other organisms in the system (trophic interactions aside) constitutes an important ecosystem function due to their influence on resources and on biogeochemical gradients in the sediment. Motile diatoms are particularly important in mediating biogeochemical cycles, including the cycling of carbon, oxygen, nitrogen, and silica [6.63] [6.27] [6.112] [6.125]. Photosynthesis and migration of diatoms at the sediment-water interface together with their non-trophic interactions with benthic macro- and meiofauna oxygenate surface sediments [6.62] [6.103], providing a suitable habitat for other organisms. This oxygenation can substantially alter biogeochemical gradients within the bed, facilitate coupled ncycling processes [6.62] and is paramount in the recovery from disturbances such as hypoxia [6.61]. These processes are increasingly important for ecosystem resilience to other anthropogenic stressors. For example, under low nitrogen conditions, MPB growth is reduced but production of EPS is high due to overflow metabolism. Increasing nitrogen loads allows MPB growth and a more balanced metabolism and a relative decrease in EPS production [6.27] [6.96]. Since EPS production facilitates diatom migration, these environmental changes could alter the MPB community, with fewer taxa that exhibit behavioral (movement) responses to high light conditions, and more taxa that are physiologically adapted for less varied light conditions (less motile). This could alter other ecosystem functions, including sediment stability, and increase the likelihood of erosion. Silica is a key element for benthic diatoms. Investigations in the Severn Estuary, UK, have revealed a small but important role for motile benthic diatoms in the temporal transformation of silica between dissolved and biogenic forms [6.120]. More recently, benthic

146  Diatom Gliding Motility biogenic silica was linked to the productivity of motile diatoms as well as the growth and development of biofilms that are formed during migration [6.125]. However, these authors concluded that the exchange of silica between different compartments was regulated by local hydrodynamic conditions. The role of MPB in mediating nutrient cycling in coastal sediments enhances their position as ecosystem engineers. In clear water systems, the MPB can assimilate the majority of dissolved inorganic nitrogen inputs [6.48], out-competing bacteria [6.97], but this role can be altered as turbidity increases [6.92]. This may result in a feedback whereby MPB help promote clear waters which in turn supports enhanced MPB growth and further sediment stabilization, and if this is disturbed the system may revert to another alternate state, as proposed by van de Koppel et al. [6.59]. Nonetheless, nitrogen uptake and retention often exceeds permanent nitrogen removal via processes such as denitrification [6.112], particularly in low nutrient systems, and diatom migration can redistribute nitrogen deeper in the sediment [6.109] [6.65]. Although the scale of this redistribution may seem insignificant (on the millimeter scale), this is an important mechanism by which nitrogen is provided to bacteria and other organisms that would otherwise have no access to this resource. This results in a close dependency between diatoms and bacteria [6.58]. Other processes of nutrient retention, recycling and removal are also influenced by diatom migration. For example, EPS secreted during diel migrations provides a high-quality labile carbon source to n-cycling bacteria [6.114] and the formation of biofilms can alter sediment permeability, modifying porewater pressure gradients [6.52] and geomorphological features as discussed above. This in turn, influences the benthic-­ pelagic exchange of nutrients.

6.6 Niche Construction and Epipelic Diatoms Ecosystem engineering is a recently recognized concept [6.57] [6.98], although ecosystem engineering itself has a very long history. As soon as life emerged on the surface of the Earth, these newly evolved organisms began to change their environment. At first, these changes would be localized and relatively insignificant but as cells proliferated these effects began to accumulate, processing material, creating fluxes, metabolizing new products and excreting waste. A successful lineage of these first organisms, most of which probably became extinct, produced the “last universal common ancestor” (LUCA), which according to monophyletic theory, led to the evolution of all subsequent life [6.113]. This gradual transformation of material, changing the environment and habitat conditions for other organisms, is a clear case of ecosystem engineering and took place well before the more well-known “agronomic revolution” of bioturbation [6.15]. This supports the Boogert et al. [6.10] argument that it is not only the most commonly cited charismatic species (e.g., beavers, sea otters, elephants, etc.) that are effective ecosystem engineers but that the cumulative effect of smaller organisms should also be considered as ecosystem engineering. So when does a cell become an engineer? At one extreme, it may be considered that any organism that changes energy flux or material flows within an environment is an ecosystem engineer, so essentially everything is probably an engineer to some extent. At the other extreme, some consider that only organisms shown to have a very significant and measurable effect on the dynamics of the system might be characterized as ecosystem engineers. Drawing a line between these extremes is difficult, highly subjective and also contextually dependent. Life stage, ambient

Diatom Biofilms and Ecosystem Engineering  147 environment and resources might all affect engineering capacity at any particular moment. However, accepting the argument that ecosystem engineering can be delivered by microbial assemblages it follows that ecosystem engineering will have emerged early in the evolutionary calendar. One of the first remnants of evolving life on Earth are layered patterns, fossilized in rock, that represent ancient biofilms. These structures, known as “stromatolites,” date back nearly 3.5 billion years; but there are other fine structures and possible biological signatures [6.77] that are considered an indication of the presence of living entities, pushing back the origin of life to 3.7 or even 4.2 billion years ago [6.34]. An early representation of ecosystem engineering [6.86] was the stabilization of sediment and the establishment of gradients, promoting niche separation and advancing evolution as early organisms adapted to differing conditions. The potential of modern bacteria to evolve rapidly is wellknown and niche separation is an effective driver of evolutionary change [6.99] but we know less about the adaptability of ancient lineages. Interestingly, modern equivalents of ancient stromatolic assemblages exist today but the modern versions are probably quite different in composition from the ancient forms [6.95]. Specifically modern stromatolitic assemblages (in Australia, Israel, South Africa, the UK and the Bahamas) are usually “contaminated” by modern photosynthetic organisms (including diatoms and foraminifera) augmenting the cyanobacteria that are often considered the major components of more ancient stromatolitic assemblages [6.126] [6.87]. There is evidence that the functional ability of stromatolite assemblages to bind sediments has been enhanced by their photosynthetic members such as diatoms, at least for intertidal systems such as the Bahamian shallow subtidal [6.87] and peritidal South African [6.126] stromatolites. For example, Bahamian stromatolitic assemblages kept in the light became more stable than similar assemblages kept in darkness [6.87]. This was because the photosynthetic microbes (in this case diatoms and cyanobacteria) in the illuminated assemblage produce and secrete extracellular polymer substances (EPS), which have a role in preserving the structure of the depositional systems formed by the microbes and supporting stability, a step on the way to the formation of stromatolites. Now, very few areas of the planet are truly abiotic, with even habitats considered to be “extreme” colonized and impacted by life, and the effects of biology on the planet has been pronounced. For example, the accumulation of atmospheric oxygen, generated through photosynthesis began about 3.5 billion years ago [6.8]. The effects of photosynthesis could be taken as another example of ecosystem engineering [6.42], whereby the action of an organism changes the nature of the habitat [6.57] and has consequences for other organisms using it. With time and as more oxygen was introduced into the atmosphere, lifeforms had to evolve to withstand increasing levels of atmospheric oxygen or become restricted to habitats where oxygen did not penetrate, for example, anaerobic sediments. This is an impressive example of global change, but the precise role of ecosystem engineering in the way that life has evolved and how it should be considered in evolutionary theory is a matter of surprisingly strong contention. Some theorists consider that it is sufficient to include the actions of organisms under standard evolutionary theory but within an “extended phenotype” approach, describing the action of the organisms as part and parcel of their existence and needing no special consideration. Protagonists of the emerging niche construction theory (NCT [6.80]) view the evolutionary process slightly differently and propose that the action of organisms that change the environment for themselves and other species

148  Diatom Gliding Motility (ecosystem engineering) should be recognized as a separate and important force for evolutionary change. This is where the term “niche construction” arises, the proposal that the activity of these organisms can create or significantly alter the “niche space,” even creating new niches (bower bird nest, coral reef) that are available for other organisms. NCT theorists suggest that this ecosystem engineering should be recognized as an evolutionary pressure (niche construction) in its own right [6.105]. The activity of benthic diatoms as an example of ecosystem engineering is highly relevant. As previously outlined (Table 6.1), the protection of sediment from erosion is an important property of diatom biofilms [6.83] and one of their most recognized ecosystem functions and services [6.50] [6.89]. For this to have an evolutionary aspect, we would expect that increasing the stability of these sediments would lead to an increase in the fitness of the population(s) responsible and also require other local conspecifics to adapt and potentially benefit from the enhanced stability of the system. This type of niche construction has been suggested for the freshwater flatworm, Polycelis tenuis, through its secretion of polymer layers on the sediment surface, influencing the development of the meiofaunal community through a trophic mechanism [6.128]. In terms of diatom EPS, there are a number of potential advantages to sediment stabilization, which might lead to increased evolutionary success (Figure 6.3). There may also be a further aspect of temporal change in the development of ecosystem engineering, with Erwin [6.42] suggesting that the extent of ecosystem engineering has changed with geological time, stating that modern ecosystems display considerably more ecosystems engineering than did the Palaeozoic. Epipelic diatoms are an example of assemblages that have become effective ecosystem engineers, and it is likely that this capacity has Pressure

Stabilising biofilm

Regular erosion

Turbidity

Reduced

Increased

Surface light

Increased

Reduced

Bed gradients

More stable

Less stable

Remote colonisation

Reduced

Increased

Food resource

Bed grazers

Filter feeders Surface biofilm

ure face Edge of fract

Sub-surface cells and sediment

Figure 6.3  Top: Change in environmental driver (pressure) dependent on the presence of a “stabilizing biofilm” or under “regular erosion.” Bottom: Low-temperature scanning electron micrograph of MPB biofilm structure. (Image: Irvine Davidson, University of St Andrews.)

Diatom Biofilms and Ecosystem Engineering  149 helped to support their global success and, critical in terms of the niche construction theory, influenced the evolutionary trajectory of other conspecifics [6.79] [6.80]. An example is given by Passarelli et al. [6.84]. The authors noted that a recognized engineering species may indeed enhance the fitness of another population, but the growth and success of that population also enhanced the fitness of the original ecosystem engineer. These populations enhance the activity of one another, creating a synergy between the organisms that essentially modifies the environment in a way that benefited both species, Passarelli et al. [6.84], and thus far this has not been considered as part of the niche construction debate. The understanding that different functional attributes of individual species (traits) complement or interfere with each other is not new [6.41], and has been investigated in the biodiversity-­ ecosystem function debate many times—but in the context of ecosystem engineering this is less well-recognized.

6.7 Conclusion The ecological range of benthic diatoms is greater than might at first be imagined because of the capacity of diatoms to adapt to photosynthesis at low light levels (photoacclimation). Benthic diatom assemblages exist in areas that might be considered extreme and inhospitable such as under sea ice [6.2] [6.64], tidal areas during winter, epizoan spp. associated with the coat of diving marine mammals and turtles [6.94] and as parasites [6.5]. Mainly, however, epipelic diatoms are associated with muddy intertidal and shallow subtidal sediments where a range of species often coexist, many with a cosmopolitan distribution. While modern oceanic sediments cover 70% of the planet’s surface, the scope for epipelic diatom colonization is reduced by the need for light. However, the region of light penetration is still extensive and epipelic diatoms have been found to be at a depth of more than 150 meters, allowing colonization of deposited sediments in coastal regions and in shallow shelf waters. The occurrence of MPB on depositional intertidal deposits is wellknown, but as the depth increases our knowledge declines. In a recent paper, Pinckney [6.91] suggested that considering the light penetration and water depth, 80-90% of the South Atlantic Bight of the southeastern United States continental shelf was suitable for MPB colonization. This analysis probably also applies to many shelf areas where there is less information on the nature of the bed and the potential for microphytobenthic growth. What seems certain is that the importance of MPB is globally underestimated, as suggested by Pinckney, including in the Gulf of California [6.100]. It is certainly the case that MPB are not always noticed and can be easily overlooked leading to the “Secret Garden” terminology of MacIntyre [6.66] despite their central roles in multiple ecosystem functions and the delivery of ecosystem services [6.50]. The ideas of niche construction may provide a further reason for recognizing the importance of MPB, suggesting that the engineering activity of one organism may affect the fitness to survive of other species. This implies a rather passive role of the organisms affected by the ecosystem engineer, adapting to and sometime benefiting from the change in conditions. However, ecological interactions are complex, just as understanding the dynamics of evolution are [6.105], and it would not be surprising if the interaction between the engineering, and the affected populations, were more nuanced.

150  Diatom Gliding Motility

Acknowledgments DMP was partly supported by the Marine Alliance for Science and Technology for Scotland (MASTS), funded by the Scottish Funding Council (grant reference HR09011), NERC project BLUE-COAST (N E/N 0 160 0 9/) and the Templeton Foundation ID 60501.

References [6.1] Androuin, T., Polerecky, L., Decottignies, P., Dubois, S.F., Dupuy, C., Hubas, C., Jesus, B., Le Gall, E., Marzloff, M.P., Carlier, A., Subtidal Microphytobenthos: A Secret Garden Stimulated by the Engineer Species Crepidula Fornicata. Front. Mar. Sci., 5, 1–12, 2018, https://doi. org/10.3389/fmars.2018.00475. [6.2] Aslam, S.N., Strauss, J., Thomas, D.N., Mock, T., Underwood, G.J.C., Identifying Metabolic Pathways for Production of Extracellular Polymeric Substances by the Diatom Fragilariopsis Cylindrus Inhabiting Sea Ice. ISME J., Springer US, 12, 1237–1251, 2018, https://doi. org/10.1038/s41396-017-0039-z. [6.3] Baas, J.H., Baker, M.L., Malarkey, J., Bass, S.J., Manning, A.J., Hope, J.A., Peakall, J., Lichtman, I.D., Ye, L., Davies, A.G., Parsons, D.R., Paterson, D.M., Thorne, P.D., Integrating Field and Laboratory Approaches for Ripple Development in Mixed Sand–Clay–EPS. Sedimentology, 66, 2749–2768, 2019, https://doi.org/10.1111/sed.12611. [6.4] Barnett, A., Méléder, V., Blommaert, L., Lepetit, B., Gaudin, P., Vyverman, W., Sabbe, K., Dupuy, C., Lavaud, J., Growth Form Defines Physiological Photoprotective Capacity in Intertidal Benthic Diatoms. ISME J., 9, 32–45, 2015, https://doi.org/10.1038/ismej.2014.105. [6.5] Bavestrello, G., Arillo, A., Calcinai, B., Cattaneo-Vietti, R., Cerrano, C., Gaino, E., Penna, A., Sarà, M., Parasitic Diatoms inside Antarctic Sponges. Biol. Bull., 198, 29–33, 2000, https:// doi.org/10.2307/1542801. [6.6] Beninger, P.G., Cuadrado, D., Van De Koppel, J., Sedimentary and Biological Patterns on Mudflats, in: Mudflat Ecology, P.G. Beninger, (Ed.), pp. 185–211, Springer Nature Switzerland AG, 2018, https://doi.org/10.1007/978-3-319-99194-8. [6.7] Blanchard, G.F., Paterson, D.M., Stal, L.J., Richard, P., Galois, R., Huet, V., Kelly, J., Honeywill, C., De Brouwer, J., Dyer, K., Christie, M., Seguignes, M., The Effect of Geomorphological Structures on Potential Biostabilisation by Microphytobenthos on Intertidal Mud flats. Cont. Shelf Res., 20, 1243–1256, 2000, https://doi.org/10.1016/ S0278-4343(00)00021-2. [6.8] Blankenship, R.E., Early Evolution of Photosynthesis. Plant Physiol., 154, 434–438, 2010, https://doi.org/10.1104/pp.110.161687. [6.9] de Boer, P.L., Mechanical Effects of Micro-Organisms on Intertidal Bedform Migration. Sedimentology, 28, 129–132, 1981, https://doi.org/10.1016/b978-0-240-80620-4.50028-8. [6.10] Boogert, N.J., Paterson, D.M., Laland, K.N., The Implications of Niche Construction and Ecosystem Engineering for Conservation Biology. BioScience, 56, 570, 2006, https://doi. org/10.1641/0006-3568(2006)56[570:TIONCA]2.0.CO;2. [6.11] Bradbury, J., Nature’s Nanotechnologists: Unveiling the Secrets of Diatoms. PloS Biol., 2, 1512–1515, 2004, https://doi.org/10.1371/journal.pbio.0020306. [6.12] Brotas, V., Serôdio, J., Risgaard-Petersen, N., Dalsgaard, T., Serôdio, J., Ottosen, L., Ottosen, L., In Situ Measurements of Photosynthetic Activity and Respiration of Intertidal Benthic Microalgal Communities Undergoing Vertical Migration. Ophelia, 57, 13–26, 2003, https:// doi.org/10.1080/00785236.2003.10409502.

Diatom Biofilms and Ecosystem Engineering  151 [6.13] Brustolin, M.C., Thomas, M.C., Mafra, L.L., da Cunha Lana, P., Bioturbation by the Sand Dollar Encope Emarginata (Echinoidea, Mellitidae) Changes the Composition and Size Structure of Microphytobenthic Assemblages. Hydrobiologia, Springer Int. Publishing, 779, 183–192, 2016, https://doi.org/10.1007/s10750-016-2815-6. [6.14] Calabrese, E.J. and Baldwin, L.A., Defining Hormesis. Hum. Exp. Toxicol., 21, 91–97, 2002, https://doi.org/10.1191/0960327102ht217oa. [6.15] Canfield, D.E. and Farquhar, J., Animal Evolution, Bioturbation, and the Sulfate Concentration of the Oceans. Proc. Natl. Acad. Sci. U. S. A., 106, 8123–8127, 2009, https:// doi.org/10.1073/pnas.0902037106. [6.16] Cartaxana, P., Cruz, S., Gameiro, C., Kühl, M., Regulation of Intertidal Microphytobenthos Photosynthesis over a Diel Emersion Period Is Strongly Affected by Diatom Migration Patterns. Front. Microbiol., 7, 1–11, 2016, https://doi.org/10.3389/fmicb.2016.00872. [6.17] Chapman, M.G., Tolhurst, T.J., Murphy, R.J., Underwood, A.J., Complex and Inconsistent Patterns of Variation in Benthos, Micro-Algae and Sediment over Multiple Spatial Scales. Mar. Ecol. Prog. Ser., 398, 33–47, 2010, https://doi.org/10.3354/meps08328. [6.18] Chavez, F.P., Messié, M., Pennington, J.T., Marine Primary Production in Relation to Climate Variability and Change. Annu. Rev. Mar. Sci., 3, 227–260, 2011, https://doi.org/10.1146/ annurev.marine.010908.163917. [6.19] Chen, X., Zhang, C., Paterson, D.M., Townend, I.H., Jin, C., Zhou, Z., Gong, Z., Feng, Q., The Effect of Cyclic Variation of Shear Stress on Non-Cohesive Sediment Stabilization by Microbial Biofilms: The Role of ‘Biofilm Precursors.’. Earth Surf. Processes Landforms, 44, 1471–1481, 2019, https://doi.org/10.1002/esp.4573. [6.20] Chennu, A., Volkenborn, N., De Beer, D., Wethey, D.S., Woodin, S.A., Polerecky, L., Effects of Bioadvection by Arenicola Marina on Microphytobenthos in Permeable Sediments. PloS One, 10, 1–16, 2015, https://doi.org/10.1371/journal.pone.0134236. [6.21] Christianen, M.J.A., Middelburg, J.J., Holthuijsen, S.J., Jouta, J., Compton, T.J., van der Heide, T., Piersma, T., Sinninghe Damsté, J.S., van der Veer, H.W., Schouten, S., Olff, H., Benthic Primary Producers Are Key to Sustain the Wadden Sea Food Web: Stable Carbon Isotope Analysis at Landscape Scale. Ecology, 98, 1498–1512, 2017, https://doi.org/10.1002/ ecy.1837. [6.22] Coelho, H., Vieira, S., Serôdio, J., Endogenous versus Environmental Control of Vertical Migration by Intertidal Benthic Microalgae. Eur. J. Phycol., 46, 271–281, 2011, https://doi. org/10.1080/09670262.2011.598242. [6.23] Cohn, S.A., Farrell, J.F., Munro, J.D., Ragland, R.L., Weitzell, R.E., Wibisono, B.L., The Effect of Temperature and Mixed Species Composition on Diatom Motility and Adhesion. Diatom Res., 18, 225–243, 2003, https://doi.org/10.1080/0269249X.2003.9705589. [6.24] Cohn, S.A. and McGuire, J.R., Using Diatom Motility as an Indicator of Environmental Stress: Effects of Toxic Sediment Elutriates. Diatom Res., 15, 19–29, 2000, https://doi.org/10. 1080/0269249X.2000.9705484. [6.25] Consalvey, M., Jesus, B., Perkins, R.G., Brotas, V., Underwood, G.J.C., Paterson, D.M., Monitoring Migration and Measuring Biomass in Benthic Biofilms: The Effects of Dark/ Far-Red Adaptation and Vertical Migration on Fluorescence Measurements. Photosynth. Res., 81, 91–101, 2004, https://doi.org/10.1023/B:PRES.0000028397.86495.b5. [6.26] Consalvey, M., Paterson, D.M., Underwood, G.J.C., The Ups and Downs of Life in a Benthic Biofilm: Migration of Benthic Diatoms. Diatom Res., 19, 181–202, 2004, https://doi.org/10.1 080/0269249X.2004.9705870. [6.27] Cook, P.L.M., Veuger, B., Böer, S., Middelburg, J.J., Effect of Nutrient Availability on Carbon and Nitrogen Incorporation and Flows through Benthic Algae and Bacteria in Near-Shore Sandy Sediment. Aquat. Microb. Ecol., 49, 165–180, 2007, https://doi.org/10.3354/ame01142.

152  Diatom Gliding Motility [6.28] Coquillé, N., Jan, G., Moreira, A., Morin, S., Use of Diatom Motility Features as Endpoints of Metolachlor Toxicity. Aquat. Toxicol., Elsevier B.V., 158, 202–210, 2015, https://doi. org/10.1016/j.aquatox.2014.11.021. [6.29] Costanza, R., de Groot, R., Sutton, P., van der Ploeg, S., Anderson, S.J., Kubiszewski, I., Farber, S., Turner, R.K., Changes in the Global Value of Ecosystem Services. Global Environ. Change, 26, 152–158, 2014, https://doi.org/10.1016/j.gloenvcha.2014.04.002. [6.30] Craig, M.J., Baas, J.H., Amos, K.J., Strachan, L.J., Manning, A.J., Paterson, D.M., Hope, J.A., Nodder, S.D., Baker, M.L., Biomediation of Sediment Gravity Flow Dynamics. Geology, 48, 1–26, 2019, https://doi.org/doi.org/10.1130/G46837.1. [6.31] Dade, W.B., Davis, J.D., Nichols, P.D., Nowell, A.R.M., Thistle, D., Trexler, M.B., White, D.C., Effects of Bacterial Exopolymer Adhesion on the Entrainment of Sand. Geomicrobiol. J., 8, 1–16, 1990, https://doi.org/10.1080/01490459009377874. [6.32] Darley, W.M., Wimpee, B.B., Ohlman, C.T., Heterotrophic and Photoheterotrophic Utilization of Lactate by the Diatom, Cylindrotheca Fusiformis. Br. Phycological J., 16, 423– 428, 1981, https://doi.org/10.1080/00071618100650481. [6.33] Decho, A.W., Microbial Exopolymer Secretions in Ocean Environments: Their Role (s) in Food Webs and Marine Processes. Oceanogr. Mar. Biol., CRC Press, 28, 73–153, 1990. [6.34] Dodd, M.S., Papineau, D., Grenne, T., Slack, J.F., Rittner, M., Pirajno, F., O’Neil, J., Little, C.T.S., Evidence for Early Life in Earth’s Oldest Hydrothermal Vent Precipitates. Nature, 543, 60–64, 2017, https://doi.org/10.1038/nature21377. [6.35] Donadi, S., Westra, J., Weerman, E.J., van der Heide, T., van der Zee, E.M., van de Koppel, J., Olff, H., Piersma, T., van der Veer, H.W., Eriksson, B.K., Non-Trophic Interactions Control Benthic Producers on Intertidal Flats. Ecosystems, 16, 1325–1335, 2013, https://doi. org/10.1007/s10021-013-9686-8. [6.36] Drylie, T.P., Lohrer, A.M., Needham, H.R., Bulmer, R.H., Pilditch, C.A., Benthic Primary Production in Emerged Intertidal Habitats Provides Resilience to High Water Column Turbidity. J. Sea Res., Elsevier, 142, 101–112, 2018, https://doi.org/10.1016/J. SEARES.2018.09.015. [6.37] Echappé, C., Gernez, P., Méléder, V., Jesus, B., Cognie, B., Decottignies, P., Sabbe, K., Barillé, L., Satellite Remote Sensing Reveals a Positive Impact of Living Oyster Reefs on Microalgal Biofilm Development. Biogeosciences, 15, 905–918, 2018, https://doi.org/10.5194/ bg-15-905-2018. [6.38] Edgar, L.A. and Pickett-Heaps, J.D., Diatom Locomotion, in: Phycological Research, F.E. Round, and D.J. Chapman, (Eds.), pp. 47–88, Biopress, Bristol, 1984. [6.39] Edgar, L.A., Diatom Locomotion: A Consideration of Movement in a Highly Viscous Situation. Br. Phycological J., 17, 243–251, 1982, https://doi.org/10.1080/00071618200650261. [6.40] Elner, R.W., Beninger, P.G., Jackson, D.L., Potter, T.M., Evidence of a New Feeding Mode in Western Sandpiper (Calidris Mauri) and Dunlin (Calidris Alpina) Based on Bill and Tongue Morphology and Ultrastructure. Mar. Biol., 146, 1223–1234, 2005, https://doi.org/10.1007/ s00227-004-1521-5. [6.41] Emmerson, M.C., Solan, M., Emes, C., Paterson, D.M., Raffaelli, D., Consistent Patterns and the Idiosyncratic Effects of Biodiversity in Marine Ecosystems. Nature, 411, 73–77, 2001, https://doi.org/10.1038/35075055. [6.42] Erwin, D.H., Macroevolution of Ecosystem Engineering, Niche Construction and Diversity. Trends Ecol. Evol., 23, 304–310, 2008, https://doi.org/10.1016/j.tree.2008.01.013. [6.43] Garwood, J.C., Hill, P.S., MacIntyre, H.L., Law, B.A., Grain sizes Retained by Diatom Biofilms during Erosion Linked to Sediment Texture, Continental Shelf Research, 104, 37–44, 2015.

Diatom Biofilms and Ecosystem Engineering  153 [6.44] Grabowski, R.C., Droppo, I.G., Wharton, G., Erodibility of Cohesive Sediment: The Importance of Sediment Properties. Earth-Sci. Rev., Elsevier B.V., 105, 101–120, 2011, https://doi.org/10.1016/j.earscirev.2011.01.008. [6.45] Haro, S., Brodersen, K.E., Bohórquez, J., Papaspyrou, S., Corzo, A., Kühl, M., Radiative Energy Budgets in a Microbial Mat Under Different Irradiance and Tidal Conditions. Microb. Ecol., 77, 852–865, 2019, https://doi.org/10.1007/s00248-019-01350-6. [6.46] Hay, S.I., Maitland, T.C., Paterson, D.M., The Speed of Diatom Migration through Natural and Artificial Substrata. Diatom Res., 8, 371–384, 1993, https://doi.org/10.1080/02692 49x.1993.9705268. [6.47] Le Hir, P., Monbet, Y., Orvain, F., Sediment Erodability in Sediment Transport Modelling: Can We Account for Biota Effects? Cont. Shelf Res., 27, 1116–1142, 2007, https://doi. org/10.1016/j.csr.2005.11.016. [6.48] Hochard, S., Pinazo, C., Grenz, C., Evans, J.L.B., Pringault, O., Impact of Microphytobenthos on the Sediment Biogeochemical Cycles: A Modeling Approach. Ecol. Modell., 221, 1687– 1701, 2010, https://doi.org/10.1016/j.ecolmodel.2010.04.002. [6.49] Hope, J.A., Malarkey, J., Baas, J.H., Peakall, J., Parsons, D.R., Manning, A.J., Bass, S.J., Lichtman, I.D., Thorne, P.D., Ye, L., Paterson, D.M., Interactions between Sediment Microbial Ecology and Physical Dynamics Drive Heterogeneity in Contextually Similar Depositional Systems. Limnol. Oceanogr., 65, 2403–2419, 2020, https://doi.org/10.1002/lno.11461. [6.50] Hope, J.A., Paterson, D.M., Thrush, S.F., The Role of Microphytobenthos in Soft-Sediment Ecological Networks and Their Contribution to the Delivery of Multiple Ecosystem Services. J. Ecol., 108, 815–830. 2020, https://doi.org/10.1111/1365-2745.13322. [6.51] Hubas, C., Passarelli, C., Paterson, D.M., Microphytobenthic Biofilms: Composition and Interactions, in: Mudflat Ecology, P.G. Beninger, (Ed.), pp. 63–90, Springer Nature Switzerland AG, 2018, https://doi.org/10.1007/978-3-319-99194-8. [6.52] Huettel, M., Berg, P., Kostka, J.E., Benthic Exchange and Biogeochemical Cycling in Permeable Sediments. Annu. Rev. Mar. Sci., 6, 23–51, 2014, https://doi.org/10.1146/ annurev-marine-051413-012706. [6.53] Ishii, K.I. and Kamikawa, R., Growth Characterization of Non-Photosynthetic Diatoms, Nitzschia Spp., Inhabiting Estuarine Mangrove Forests of Ishigaki Island, Japan. Plankton Benthos Res., 12, 164–170, 2017, https://doi.org/10.3800/pbr.12.164. [6.54] Jesus, B., Brotas, V., Marani, M., Paterson, D.M., Spatial Dynamics of Microphytobenthos Determined by PAM Fluorescence. Estuar. Coast. Shelf Sci., 65, 30–42, 2005, https://doi. org/10.1016/j.ecss.2005.05.005. [6.55] Jesus, B., Brotas, V., Ribeiro, L., Mendes, C.R., Cartaxana, P., Paterson, D.M., Adaptations of Microphytobenthos Assemblages to Sediment Type and Tidal Position. Cont. Shelf Res., 29, 1624–1634, 2009, https://doi.org/10.1016/j.csr.2009.05.006. [6.56] Jesus, B., Mendes, C.R., Brotas, V., Paterson, D.M., Effect of Sediment Type on Microphytobenthos Vertical Distribution: Modelling the Productive Biomass and Improving Ground Truth Measurements. J. Exp. Mar. Biol. Ecol., 332, 60–74, 2006, https:// doi.org/10.1016/j.jembe.2005.11.005. [6.57] Jones, C.G., Lawton, J.H., Shachak, M., Organisms as Ecosystem Engineers. Oikos, 69, 373– 386, 1994. [6.58] Koedooder, C., Stock, W., Willems, A., Mangelinckx, S., De Troch, M., Vyverman, W., Sabbe, K., Diatom-Bacteria Interactions Modulate the Composition and Productivity of Benthic Diatom Biofilms. Front. Microbiol., 10, 1–11, 2019, https://doi.org/10.3389/fmicb. 2019.01255.

154  Diatom Gliding Motility [6.59] Van De Koppel, J., Herman, P.M.J., Thoolen, P., Heip, C.H.R., Do Alternate Stable States Occur in Natural Ecosystems? Evidence from a Tidal Flat. Ecology, 82, 3449–3461, 2001, https://doi.org/10.1890/0012-9658(2001)082[3449:DASSOI]2.0.CO;2. [6.60] Kritzer, J.P., DeLucia, M.B., Greene, E., Shumway, C., Topolski, M.F., Thomas-Blate, J., Chiarella, L.A., Davy, K.B., Smith, K., The Importance of Benthic Habitats for Coastal Fisheries. BioScience, 66, 274–284, 2016, https://doi.org/10.1093/biosci/biw014. [6.61] Larson, F. and Sundbäck, K., Role of Microphytobenthos in Recovery of Functions in a Shallow-Water Sediment System after Hypoxic Events. Mar. Ecol. Prog. Ser., 357, 1–16, 2008, https://doi.org/10.3354/meps07426. [6.62] Laverock, B., Gilbert, J.A., Tait, K., Osborn, A.M., Widdicombe, S., Bioturbation: Impact on the Marine Nitrogen Cycle. Biochem. Soc Trans., 39, 315–320, 2011, https://doi.org/10.1042/ BST0390315. [6.63] Leynaert, A., Longphuirt, S.N., An, S., Lim, J.H., Claquin, P., Grall, J., Kwon, B.O., Koh, C.H., Tidal Variability in Benthic Silicic Acid Fluxes and Microphytobenthos Uptake in Intertidal Sediment. Estuar. Coast. Shelf Sci., Elsevier Ltd, 95, 59–66, 2011, https://doi.org/10.1016/j. ecss.2011.08.005. [6.64] Lohrer, A.M., Cummings, V.J., Thrush, S.F., Altered Sea Ice Thickness and Permanence Affects Benthic Ecosystem Functioning in Coastal Antarctica. Ecosystems, 16, 224–236, 2013, https://doi.org/10.1007/s10021-012-9610-7. [6.65] Longphuirt, S.N., Lim, J.H., Leynaert, A., Claquin, P., Choy, E.J., Kang, C.K., An, S., Dissolved Inorganic Nitrogen Uptake by Intertidal Microphytobenthos: Nutrient Concentrations, Light Availability and Migration. Mar. Ecol. Prog. Ser., 379, 33–44, 2009, https://doi.org/10.3354/ meps07852. [6.66] MacIntyre, H.L., Geider, R.J., Miller, D.C., Microphytobenthos: The Ecological Role of the “Secret Garden” of Unvegetated, Shallow-Water Marine Habitats. I. Distribution, Abundance and Primary Production. Estuaries Coasts, 19, 186–201, 1996, https://doi. org/10.2307/1352224. [6.67] Malarkey, J., Baas, J.H., Hope, J.A., Aspden, R.J., Parsons, D.R., Peakall, J., Paterson, D.M., Schindler, R.J., Ye, L., Lichtman, I.D., Bass, S.J., Davies, A.G., Manning, A.J., Thorne, P.D., The Pervasive Role of Biological Cohesion in Bedform Development. Nat. Commun., Nat. Publishing Group, 6, 6257, 2015, https://doi.org/10.1038/ncomms7257. [6.68] Manning, A.J., Baugh, J.V., Spearman, J.R., Whitehouse, R.J.S., Flocculation Settling Characteristics of Mud: Sand Mixtures. Ocean Dyn., 60, 237–253, 2010, https://doi. org/10.1007/s10236-009-0251-0. [6.69] Manzenrieder, H., Retardation of Initial Erosion under Biological Effects in Sandy Tidal Flats. Australasian Conference on Coastal and Ocean Engineering, Institution of Engineers, Australia, 1985. [6.70] Medlin, L.K., Mini Review: Diatom Species as Seen through a Molecular Window. Rev. Bras. Bot., Springer Int. Publishing, 41, 457–469, 2018, https://doi.org/10.1007/s40415-018-0444-1. [6.71] Miller, D.C., Geider, R.J., MacIntyre, H.L., Microphytobenthos: The Ecological Role of the “Secret Garden” of Unvegetated, Shallow-Water Marine Habitats. II. Role in Sediment Stability and Shallow-Water Food Webs. Estuaries Coasts, 19, 202–212, 1996, https://doi. org/10.2307/1352224. [6.72] Mitchell, J.G., Whense Is the Diversity of Diatom Frustules Derived?, in: Diatom Nanotechnology: Progress and Emerging Applications, vol. 44, D. Losic, (Ed.), pp. 1–13, Royal Society of Chemistry, Cambridge, 2017, https://doi.org/10.1039/9781788010160-00001. [6.73] Mitchell, J.G., Seuront, L., Doubell, M.J., Losic, D., Voelcker, N.H., Seymour, J., Lal, R., The Role of Diatom Nanostructures in Biasing Diffusion to Improve Uptake in a Patchy Nutrient Environment. PloS One, 8, 5, 2013, https://doi.org/10.1371/journal.pone.0059548.

Diatom Biofilms and Ecosystem Engineering  155 [6.74] Morioka, H., Kasai, A., Miyake, Y., Kitagawa, T., Kimura, S., Food Composition for Blue Mussels (Mytilus Edulis) in the Menai Strait, UK, Based on Physical and Biochemical Analyses. J. Shellfish Res., 36, 659–668, 2017, https://doi.org/10.2983/35.036.0315. [6.75] Neumann, A.C., Gebelein, C.D., Scoffin, T.P., The Composition, Structure and Erodibility of Subtidal Mats Abaco, Bahamas. J. Sediment. Res., 40, 274–297, 1970, https://doi.org/https:// doi.org/10.1306/74D71F2D-2B21-11D7-8648000102C1865D. [6.76] Nielsen, S.L., Risgaard-petersen, N., Banta, G.T., Nitrogen Retention in Coastal Marine Sediments — a Field Study of the Relative Importance of Biological and Physical Removal in a Danish Estuary. Estuaries Coasts, 40, 1276–1287, 2017, https://doi.org/10.1007/ s12237-017-0216-3. [6.77] Noffke, N. and Paterson, D., Microbial Interactions with Physical Sediment Dynamics, and Their Significance for the Interpretation of Earth’s Biological History. Geobiology, 6, 1–4, 2008, https://doi.org/10.1111/j.1472-4669.2007.00132.x. [6.78] Oakes, J.M. and Eyre, B.D., Transformation and Fate of Microphytobenthos Carbon in Subtropical, Intertidal Sediments: Potential for Long-Term Carbon Retention Revealed 13C-Labeling. Biogeosciences, 11, 1927–1940, 2014, https://doi.org/10.5194/bg-111927-2014. [6.79] Odling-smee, F.J., Laland, K.N., Feldman, M.W., Niche Construction. Am. Nat., 147, 641– 648, 1996, https://doi.org/10.1086/285870. [6.80] Odling-smee, F., Laland, K.N., Feldman, M.W., Niche Construction: The Neglected Process in Evolution (MPB-37), Princeton University press, New Jersey, 2013. [6.81] Paine, R.T., A Conversation on Refining the Concept of Keystone Species. Conserv. Biol., 9, 962–964, 2011. [6.82] Parsons, D.R., Schindler, R.J., Hope, J.A., Malarkey, J., Baas, J.H., Peakall, J., Manning, A.J., Ye, L., Simmons, S., Paterson, D.M., Aspden, R.J., Bass, S.J., Davies, A.G., Lichtman, I.D., Thorne, P.D., The Role of Biophysical Cohesion on Subaqueous Bed Form Size. Geophys. Res. Lett., 43, 1566–1573, 2016, https://doi.org/10.1002/2016GL067667. [6.83] Passarelli, C., Hubas, C., Paterson, D.M., Mudflat Ecosystem Engineers and Services, in: Mudflat Ecology. Aquatic Ecology Series, vol. 7, P.G. Beninger, (Ed.), pp. 243–269, Springer, Cham, 2018. [6.84] Passarelli, C., Olivier, F., Paterson, D.M., Meziane, T., Hubas, C., Organisms as Cooperative Ecosystem Engineers in Intertidal Flats. J. Sea Res., Elsevier B.V., 92, 92–101, 2014, https:// doi.org/10.1016/j.seares.2013.07.010. [6.85] Paterson, D.M., Wiltshire, K.H., Miles, A., Blackburn, J., Davidson, I., Yates, M.G., McGrorty,  S., Eastwood, J.A., Microbiological Mediation of Spectral Reflectance from Intertidal Cohesive Sediments. Limnol. Oceanogr., 43, 1207–1221, 1998, https://doi. org/10.4319/lo.1998.43.6.1207. [6.86] Paterson, D.M., Aspden, R.J., Reid, R.P., Biodynamics of Modern Marine Stromatolites, in: Microbial Mats, J. Seckbach, and A. Oren, (Eds.), pp. 223–235, Springer, Dordrecht, 2010, https://doi.org/10.1007/978-90-481-3799-2. [6.87] Paterson, D.M., Aspden, R.J., Visscher, P.T., Consalvey, M., Andres, M.S., Decho, A.W., Stolz,  J., Reid, R.P., Light-Dependant Biostabilisation of Sediments by Stromatolite Assemblages. PloS One, 3, e3176, 2008, https://doi.org/10.1371/journal.pone.0003176. [6.88] Paterson, D.M., Hope, J.A., Kenworthy, J., Biles, C.L., Gerbersdorf, S.U., Form, Function and Physics: The Ecology of Biogenic Stabilisation. J. Soils Sediments, 18, 3044–3054, 2018, https://doi.org/10.1007/s11368-018-2005-4. [6.89] Paterson, D.M., Hope, J.A., Kenworthy, J., Biles, C.L., Gerbersdorf, S.U., Form, Function and Physics: The Ecology of Biogenic Stabilisation. J. Soils Sediments, 18, 3044–3054, 2018, https://doi.org/10.1007/s11368-018-2005-4.

156  Diatom Gliding Motility [6.90] Paterson, D.M., Defew, E.C., Jabour, J., Ecosystem Function and Co Evolution of Terminology in Marine Science and Management, in: Marine Biodiversity and Ecosystem Functioning: Frameworks, Methodologies and IItegration, M. Solan, R.J. Aspden, D.M. Paterson, (Eds.), Oxford University Press, Oxford, UK, 2012. [6.91] Pinckney, J.L., A Mini-Review of the Contribution of Benthic Microalgae to the Ecology of the Continental Shelf in the South Atlantic Bight. Estuaries Coasts, 41, 2070–2078, 2018, https://doi.org/10.1007/s12237-018-0401-z. [6.92] Pratt, D.R., Pilditch, C.A., Lohrer, A.M., Thrush, S.F., The Effects of Short-Term Increases in Turbidity on Sandflat Microphytobenthic Productivity and Nutrient Fluxes. J. Sea Res., 92, 170–177, 2014, https://doi.org/10.1016/j.seares.2013.07.009. [6.93] Prins, A., Deleris, P., Hubas, C., Jesus, B., Effect of Light Intensity and Light Quality on Diatom Behavioral and Physiological Photoprotection. Front. Mar. Sci., Front. Media 2020, 7, 203, 2020, https://doi.org/10.3389/fmars.2020.00203. [6.94] Riaux-Gobin, C., Witkowski, A., Kociolek, J.P., Ector, L., Chevallier, D., Compère, P., New Epizoic Diatom (Bacillariophyta) Species from Sea Turtles in the Eastern Caribbean and South Pacific. Diatom Res., 32, 109–125, 2017, https://doi.org/10.1080/0269249X.2017.1299042. [6.95] Riding, R., The Nature of Stromatolites: 3,500 Million Years of History and a Century of Research, in: Advances in Stromatolite Geobiology, Lecture Notes in Earth Sciences, J. Reitner, N.-V. Quéric, G. Arp, (Eds.), Springer-Verlag, Berlin, 2011, https://doi.org/10.1007/ 978-3-642-10415-2_3. [6.96] Riekenberg, P.M., Oakes, J.M., Eyre, B.D., Short-Term Fate of Intertidal Microphytobenthos Carbon under Enhanced Nutrient Availability: A13C Pulse-Chase Experiment. Biogeosciences, 15, 2873–2889, 2018, https://doi.org/10.5194/bg-15-2873-2018. [6.97] Riekenberg, P., Oakes, J.M., Eyre, B., Uptake of Dissolved Organic and Inorganic Nitrogen in Microalgae-Dominated Sediment : Comparing Dark and Light in Situ and Ex Situ Additions of 15N. Mar. Ecol. Prog. Ser., 571, 29–42, 2017, https://doi.org/10.3354/meps12127. [6.98] Romero, G.Q., Gonçalves-Souza, T., Vieira, C., Koricheva, J., Ecosystem Engineering Effects on Species Diversity across Ecosystems: A Meta-Analysis. Biol. Rev., 90, 877–890, 2015, https://doi.org/10.1111/brv.12138. [6.99] San Roman, M. and Wagner, A., An Enormous Potential for Niche Construction through Bacterial Cross-Feeding in a Homogeneous Environment. PloS Computat. Biol., 14, 1–29, 2018, https://doi.org/10.1371/journal.pcbi.1006340. [6.100] Santema, M. and Huettel, M., Dynamics of Microphytobenthos Photosynthetic Activity along a Depth Transect in the Sandy Northeastern Gulf of Mexico Shelf. Estuar. Coast. Shelf Sci., Elsevier, 212, 273–285, 2018, https://doi.org/10.1016/j.ecss.2018.07.016. [6.101] Savelli, R., Dupuy, C., Barillé, L., Lerouxel, A., Guizien, K., Philippe, A., Bocher, P., Polsenaere, P., Le Fouest, V., On Biotic and Abiotic Drivers of the Microphytobenthos Seasonal Cycle in a Temperate Intertidal Mudflat: A Modelling Study. Biogeosciences, 15, 7243–7271, 2018, https://doi.org/10.5194/bg-15-7243-2018. [6.102] Schnurr, P.J., Drever, M.C., Kling, H.J., Elner, R.W., Arts, M.T., Seasonal Changes in Fatty Acid Composition of Estuarine Intertidal Biofilm: Implications for Western Sandpiper Migration. Estuar. Coast. Shelf Sci., Elsevier Ltd., 224, 94–107, 2019, https://doi.org/10.1016/j. ecss.2019.04.047. [6.103] Schratzberger, M. and Ingels, J., Meiofauna Matters: The Roles of Meiofauna in Benthic Ecosystems. J. Exp. Mar Biol. Ecol., Elsevier B.V., 502, 12–25, 2018, https://doi.org/10.1016/j. jembe.2017.01.007. [6.104] Schulz, K., Mikhailyuk, T., Dreßler, M., Leinweber, P., Karsten, U., Biological Soil Crusts from Coastal Dunes at the Baltic Sea: Cyanobacterial and Algal Biodiversity and Related Soil Properties. Microb. Ecol., 71, 178–193, 2016, https://doi.org/10.1007/s00248-015-0691-7.

Diatom Biofilms and Ecosystem Engineering  157 [6.105] Scott-Phillips, T.C., Laland, K.N., Shuker, D.M., Dickins, T.E., West, S.A., The Niche Construction Perspective: A Critical Appraisal. Evolution, 68, 1231–1243, 2014, https://doi. org/10.1111/evo.12332. [6.106] Serôdio, J., Vieira, S., Cruz, S., Photosynthetic Activity, Photoprotection and Photoinhibition in Intertidal Microphytobenthos as Studied in Situ Using Variable Chlorophyll Fluorescence. Cont. Shelf Res., 28, 1363–1375, 2008, https://doi.org/10.1016/j.csr.2008.03.019. [6.107] Shakeel, A., Kirichek, A., Chassagne, C., Is Density Enough to Predict the Rheology of Natural Sediments? Geo-Mar. Lett., 39, 427–434, 2019, https://doi.org/10.1007/ s00367-019-00601-2. [6.108] Shields, A.F., Application of Similarity Principals and Turbulence Research to Bed-Load Movement, in: Mitteilungen der Preussischen Versuchsanstalt fur Wasserbau und Schiffbau, Berlin, vol. 26, pp. 5–24, 1936. [6.109] Stief, P., Kamp, A., de Beer, D., Role of Diatoms in the Spatial-Temporal Distribution of Intracellular Nitrate in Intertidal Sediment. PloS One, 8, 1–15, 2013, https://doi.org/10.1371/ journal.pone.0073257. [6.110] Suja, L.D., Chen, X., Summer, S., Paterson, D.M., Gutierrez, T., Chemical Dispersant Enhances Microbial Exopolymer (EPS) Production and Formation of Marine Oil/ Dispersant Snow in Surface Waters of the Subarctic Northeast Atlantic. Front. Microbiol., 10, 553, 2019, https://doi.org/10.3389/fmicb.2019.00553. [6.111] Sundbäck, K., Alsterberg, C., Larson, F., Effects of Multiple Stressors on Marine Shallow-Water Sediments: Response of Microalgae and Meiofauna to Nutrient-Toxicant Exposure. J. Exp. Mar. Biol. Ecol., Elsevier B.V., 388, 39–50, 2010, https://doi.org/10.1016/j.jembe.2010.03.007. [6.112] Sundbäck, K., Miles, A., Linares, F., Nitrogen Dynamics in Nontidal Littoral Sediments: Role of Microphytobenthos and Denitrification. Estuaries Coasts, 29, 1196–1211, 2006. [6.113] Theobald, D.L., A Formal Test of the Theory of Universal Common Ancestry. Nature, Nat. Publishing Group, 465, 219–222, 2010, https://doi.org/10.1038/nature09014. [6.114] Tobias, C., Giblin, A., Mcclelland, J., Tucker, J., Peterson, B., Sediment DIN Fluxes and Preferential Recycling of Benthic Microalgal Nitrogen in a Shallow Macrotidal Estuary. Mar. Ecol. Prog. Ser., 257, 25–36, 2003, https://doi.org/10.3354/meps257025. [6.115]  Tolhurst, T.J., Black, K.S., Paterson, D.M., Muddy Sediment Erosion: Insights from Field Studies. J. Hydraul. Eng., 135, 73–87, 2009, https://doi.org/10.1061/ (ASCE)0733-9429(2009)135:2(73. [6.116]  Tolhurst, T.J., Jesus, B., Brotas, V., Paterson, D.M., Diatom Migration and Sediment Armouring - An Example from the Tagus Estuary, Portugal. Hydrobiologia, 503, 183–193, 2003, https://doi.org/10.1023/B:HYDR.0000008474.33782.8d. [6.117]  Tolhurst, T.J., Gust, G., Paterson, D.M., The Influence of an Extracellular Polymeric Substance (EPS) on Cohesive Sediment Stability, in: Fine Sediment Dynamics in the Marine Environment, J.C. Winterwerp, and C. Kranenburg, (Eds.), pp. 409–425, Elsevier, Amsterdam, The Netherlands, 2002. [6.118] Underwood, G.J.C., Perkins, R.G., Consalvey, M.C., Hanlon, A.R.M., Oxborough, K., Baker, N.R., Paterson, D.M., Patterns in Microphytobenthic Primary Productivity: SpeciesSpecific Variation in Migratory Rhythms and Photosynthetic Efficiency in Mixed-Species Biofilms. Limnol. Oceanogr., 50, 755–767, 2005, https://doi.org/10.4319/lo.2005.50.3.0755. [6.119] Underwood, G.J.C. and Paterson, D.M., The Importance of Extracellular Carbohydrate Productionby Marine Epipelic Diatoms. Adv. Bot. Res., 40, 183–240, 2003, https://doi. org/10.1016/S0065-2296(05)40005-1. [6.120] Underwood, G.J.C., Microphytobenthos and Phytoplankton in the Severn Estuary, UK: Present Situation and Possible Consequences of a Tidal Energy Barrage. Mar. Pollut. Bull., Elsevier Ltd, 61, 83–91, 2010, https://doi.org/10.1016/j.marpolbul.2009.12.015.

158  Diatom Gliding Motility [6.121] Vilar, A.G., Donders, T., Cvetkoska, A., Wagner-Cremer, F., Seasonality Modulates the Predictive Skills of Diatom Based Salinity Transfer Functions. PloS One, 13, 1–19, 2018, https://doi.org/10.1371/journal.pone.0199343. [6.122] Villanova, V., Fortunato, A.E., Singh, D., Bo, D.D., Conte, M., Obata, T., Jouhet, J., Fernie, A.R., Marechal, E., Falciatore, A., Pagliardini, J., Le Monnier, A., Poolman, M., Curien, G., Petroutsos, D., Finazzi, G., Investigating Mixotrophic Metabolism in the Model Diatom Phaeodactylum Tricornutum. Philos. Trans. R. Soc B: Biol. Sci., 5, 372, 1728, 2017, https:// doi.org/10.1098/rstb.2016.0404. [6.123] Virta, L., Gammal, J., Järnström, M., Bernard, G., Soininen, J., Norkko, J., Norkko, A., The Diversity of Benthic Diatoms Affects Ecosystem Productivity in Heterogeneous Coastal Environments. Ecology, 100, 1–11, 2019, https://doi.org/10.1002/ecy.2765. [6.124] Welsby, H.J., Hendry, K.R., Perkins, R.G., The Role of Benthic Biofilm Production in the Mediation of Silicon Cycling in the Severn Estuary, UK. Estuar. Coast. Shelf Sci., 176, 124– 134, 2016, https://doi.org/10.1016/j.ecss.2016.04.008. [6.125] Welsby, H.J., Hendry, K.R., Perkins, R.G., The Role of Benthic Biofilm Production in the Mediation of Silicon Cycling in the Severn Estuary, UK. Estuar. Coast. Shelf Sci., Elsevier Ltd, 176, 124–134, 2016, https://doi.org/10.1016/j.ecss.2016.04.008. [6.126] Weston, R.L.A., Perissinotto, R., Rishworth, G.M., Steyn, P.P., Benthic Microalgal Variability Associated with Peritidal Stromatolite Microhabitats along the South African Coast. Aquat. Microb. Ecol., 82, 253–264, 2019, https://doi.org/10.3354/ame01895. [6.127] Wheatland, J.A.T., Bushby, A.J., Spencer, K.L., Quantifying the Structure and Composition of Flocculated Suspended Particulate Matter Using Focused Ion Beam Nanotomography. Environ. Sci. Technol., 51, 8917–8925, 2017, https://doi.org/10.1021/acs.est.7b00770. [6.128] Wilden, B., Majdi, N., Kuhlicke, U., Neu, T.R., Traunspurger, W., Flatworm Mucus as the Base of a Food Web. BMC Ecol., BioMed Central, 19, 1–9, 2019, https://doi.org/10.1186/ s12898-019-0231-2. [6.129] Wooldridge, L.J., Worden, R.H., Griffiths, J., Thompson, A., Chung, P., Biofilm Origin of Clay-Coated Sand Grains. Geology, 45, 875–878, 2017, https://doi.org/10.1130/G39161.1. [6.130] Wooldridge, L.J., Worden, R.H., Griffiths, J., Utley, J.E.P., Thompson, A., The Origin of Clay-Coated Sand Grains and Sediment Heterogeneity in Tidal Flats. Sediment. Geol., 373, 191–209, 2018, https://doi.org/10.1016/j.sedgeo.2018.06.004. [6.131] Yalin, M.S., Mechanics of Sediment Transport, Pergamon Press, Oxford, UK, 1972. [6.132] Yool, A. and Tyrrell, T., Role of Diatoms in Regulating the Ocean’s Silicon Cycle. Global Biogeochem. Cycles, 17n (4), 1103, 2003, n/a-n/a, https://doi.org/10.1029/2002gb002018. [6.133] Zhang, N., Thompson, C.E.L., Townend, I.H., Rankin, K.E., Paterson, D.M., Manning, A.J., Nondestructive 3D Imaging and Quantification of Hydrated Biofilm-Sediment Aggregates Using x-Ray Microcomputed Tomography. Environ. Sci. Technol., 52, 13306–13313, 2018, https://doi.org/10.1021/acs.est.8b03997. [6.134]  Solan, M., Aspden, R.J., Paterson, D.M. (Eds.), Marine Biodiversity and Ecosystem Functioning: Frameworks, Methodologies, and Integration, Oxford University Press, Oxford, UK, 2012, https://doi.org/10.1093/acprof:oso/9780199642250.001.0001. [6.135] Losic, D. (Ed.), Nanoscience & Diatom Nanotechnology: Progress and Emerging Applications, vol. 44, Royal Society of Chemistry, Cambridge, UK, 2017.

7 Diatom Motility: Mechanisms, Control and Adaptive Value João Serôdio

*

Department of Biology and CESAM – Centre for Environmental and Marine Studies, University of Aveiro, Aveiro, Portugal

Abstract

Cell motility is one of the most conspicuous traits of pennate diatoms. They use it to respond to a wide range of environmental factors, actively exploiting spatial heterogeneities and, under endogenous control, to anticipate changing environmental conditions. In benthic communities dominated by raphid diatoms, rhythms in cell motility determine the patterns of primary productivity and the associated production of extracellular polymeric substances, promoting the biostabilization of sediments. Directed motility is believed to have conferred a critical adaptive advantage to raphid pennate species, explaining their evolutionary success and their fast diversification, resulting in the most recent and the most diverse group of diatoms. However, directed motility is not restricted to the pennates. Planktonic centric species undergo vertical migrations across large distances in the upper layers of the water column, causing vertical fluxes of nutrients and facilitating primary productivity in vast regions of the ocean. This work reviews the current knowledge on the forms of motility in diatoms, their underlying mechanisms, and the factors that control cell motility under natural conditions, considering the possible adaptive advantages and ecological consequences. Keywords:  Buoyancy, centric diatoms, motility, microphytobenthos, pennate diatoms, phytoplankton, raphe, vertical migration

7.1 Introduction Diatoms are classified into two main morphological categories, the centrics and the pennates. These two groups differ markedly in key cytological, biological, and ecological attributes, including cell wall symmetry, chloroplast number and morphology, and sexual reproduction [7.4] [7.106]. Within the pennates, a distinction is made between the araphids and the raphids, groups that differ mostly by the presence of a raphe, a longitudinal thin and long slit through the surface of the valve, and a determinant for the ability of directed motility [7.80]. The contrasting differences between centrics and pennates are also reflected in the types of cell motility that evolved in the two groups. These forms of cell motility differ Email: [email protected]

*

Stanley Cohn, Kalina Manoylov and Richard Gordon (eds.) Diatom Gliding Motility, (159–184) © 2021 Scrivener Publishing LLC

159

160  Diatom Gliding Motility radically regarding aspects like the interaction with the physical medium, the underlying propulsion mechanisms, the velocities attained, and the distances covered. Centric diatoms are typically planktonic and use motility to migrate vertically in the water column, between regions of different nutrient content and covering distances in the order of meters. Pennate species, on the other hand, often inhabit sedimentary environments or reside within epilithic biofilms, where they use motility to exploit environmental micro-heterogeneity on much shorter spatial scales.

7.2 Forms and Mechanisms of Motility in Diatoms 7.2.1 Motility in Centric Diatoms Several species of large planktonic centric diatoms inhabiting oligotrophic oceanic waters migrate vertically in the water column, between the well-lit surficial layers and the deep nutrient pools [7.133]. This migratory behavior is a central trait of the life history of these species, occurring consistently across their ranges of distribution [7.134]. Vertical migration comprises a strategy to exploit the light and nutrient heterogeneity within the water column. Cells sink below the nutricline to uptake nutrients, primarily nitrogen but likely other macro- and micronutrients as well, and ascend to the photic zone to carry out photosynthesis [7.132] [7.134]. The best-studied case of migratory non-flagellate phytoplankton is of the large diatom species within the genus Rhizosolenia sp., which migrate as either free-living cells or within macroscopic mats (up to 30 cm in diameter), composed of aggregations of cells from various taxa [7.135]. Other diatom species have also been documented to exhibit a vertical migration behavior, such as species within the genera Skeletonema [7.40], Phaeodactylum [7.39], Leptocylindricus, Thalassiosira, Eucampia, and Chaetoceros [7.91]. A range of experimental evidence has inferred that vertical movement of centric diatoms is associated with nutrient uptake, including differences in chemical composition, and in cellular nitrate content and nitrate reductase activity, between ascending and sinking cells [7.117] [7.132]. Migrating diatom cells can cover vertical distances of up to 50−100 m, on a multiple-day time scale, and Rhizosolenia mats have been found at depths of several hundred meters [7.134]. Vertical speeds can be high enough to allow individual cells to cover the entire euphotic zone in less than one day [7.132]. Mean descent rates vary between 0.25 m d-1 [7.39] and more than 2.2 m d-1 [7.91], with maximum values reaching above 7.4 m d-1 [7.118]. Rates of ascent can vary from 0.26 m d-1 [7.39] to 6.9 m d-1 [7.134]. However, studies carried out in experimentally controlled conditions have shown large intra-­population ­variability, denoting a large degree of phenotypic plasticity in the capacity for vertical migration [7.40]. For upward migration to occur, the sinking caused by the heavy silica cell wall must be counteracted by positive buoyancy. Centric planktonic diatoms are able to control buoyancy by regulating the ionic solute content of their large vacuoles [7.101] [7.135]. Nitrate, ammonium, phosphate, amino acids (e.g., proline) and sugars (e.g., mannose) are exchanged between the cell and the surrounding medium [7.40]. Nitrate has a particularly relevant role in the control of vertical migration. Nitrate uptake increases buoyancy, as shown for Skeletonema and Rhizosolenia mats, while cells with low internal nitrate pools are negatively buoyant [7.40] [7.131]. Positive buoyancy is also favored by factors such as low cellular

Mechanisms, Control and Adaptive Value  161 carbohydrate:nitrogen ratios [7.117] [7.135], the formation of aggregates retaining exuded lipids and oxygen bubbles produced by photosynthesis [7.40], and large cell sizes, having high surface:volume ratios [7.62] [7.130].

7.2.2 Motility in Pennate Raphid Diatoms The most common form of motility in pennate diatoms is termed “gliding” [7.38]. It consists in directed movement, typically along the direction parallel to the longitudinal axis of the cell, when the cell is in close contact with hard surfaces. The movement involves the extrusion of mucilaginous extracellular polymeric substances (EPS) through the raphe, an elongated slit located in the cell wall. The excretion of adhesive EPS results in the transient attachment of the cell to the substratum while moving, which is a requirement for diatom gliding as well as for other forms of motility and adhesion to substrates, like the ones involving pseudopods or stalks [7.77] [7.100]. Gliding is the most well-studied type of movement in diatoms, as this behavior is easily observed under the microscope and has been documented in detail since the first microscopists [7.67]. It is also the most ecologically relevant type of movement in pennate diatoms, being used for exploiting heterogeneous microenvironments, and affecting key processes such as nutrient uptake, sexual reproduction and photosynthesis within benthic diatom communities, dominated by these diatom forms [7.26] [7.110] [7.116]. Cell movement through gliding is endogenous and rhythmic, through which the cells move back and forth, reversing direction after a certain time (autonomous reversal rhythm); [7.138] [7.140]. Directed motility, for example towards a stimulus, is achieved by varying the time between the reversal of direction, causing forward progression when the movement in the direction of the stimulus lasts longer than away from it [7.3] [7.19]. In the absence of a vectorial stimulus, cells tend move randomly [7.71]. Gliding speeds up to 25 μm s-1 have been reported [7.20], but typical values are much lower, ranging within 2.5−4.7 μm s-1 for horizontal movement and within 0.17−0.28 μm s-1, for cells moving vertically [7.26]. Movement in pennate diatoms is not restricted to gliding, and a wide array of movement modalities have been reported [7.3] [7.22] [7.61] [7.97]. These alternative forms of movement include cell rolling, rocking or pirouetting, and may play a role in complementing gliding within heterogeneous substrata, like fine sediments. The mechanism behind diatom gliding has been an object of interest since the earliest microscopy observations [7.20] and several hypotheses have been put forward over the years [7.22]. Today, the model proposed by Edgar and Pickett-Heaps [7.38] is widely accepted as providing a good general description of the complex processes on which diatom gliding is based [7.26] [7.77] [7.100]. This model is based on an actin-myosin mechanism, involving the secretion of EPS through the raphe, which can be summarized as follows [7.38] [7.77] [7.100]. EPS are secreted by the merging of cytoplasmatic vesicles with the plasma membrane and the release of their content into the extracellular space. The mucilage forms strands, which attach to the substrate surface, causing adhesion between the cell and substrate. The other end of the strands is connected through the plasma membrane, via a transmembrane connector involving myosin, to a pair of bundles of actin filaments, located immediately beneath the membrane, and running the length of the raphe. Cell gliding results from the movement of the strands along actin cables parallel to the raphe, connected to the transmembrane components, causing the displacement of the cell in the

162  Diatom Gliding Motility opposite direction. As the cell moves, the EPS strands detach from the cell membrane as they approach the tip of the raphe, resulting in the deposition of a trail of mucilage. The connection between actin filaments and substratum surface was later hypothesized to be supported by an Adhesion Complex (AC; [7.141]), composed of proteins and myosin molecules extending from the actinic filaments through transmembrane connections to the EPS strands. The mechanical force required for cell movement is generated by an actin-myosin molecular motor associated with ATPase activity. This model is supported by experimental evidence which includes the requirement of calcium ions to generate motive force [7.72], and the motility inhibiting effects of an ATPase inhibitor and of the actin-destabilizing drug Lantruculin [7.100]. Despite the fact that several aspects of the AC model remain unproven [7.77], this model explains some of the main features of diatom gliding behavior. These include the requirements for mucilage excretion and for the contact between the cell and the substratum, the ability for sudden reversal of gliding direction, and the similar speeds in both directions of movement [7.3] [7.20]. The model is also compatible with the observation that adhesion and the generation of motile force are controlled independently [7.22]. Pennate diatoms inhabiting non-sedimentary benthic habitats can also move horizontally (e.g., between different sites in a freshwater stream) by detaching from the substrate and drifting in the current [7.96]. Diatom drifts may show marked diel variation, peaking at mid-day, and are caused by both active and passive release from substrate. Passive mechanisms include the entrainment of cells into the water column by oxygen bubbles produced by photosynthetic activity and advection currents due to the increase in water temperature. Active release is attributed to increases in cell buoyancy associated with cell division [7.96]. This type of motility seems to confer adaptive advantages to the diatom colonizing freshwater epilithic habitats, as it allows cells to avoid adverse environmental conditions, such as darkness [7.13] or pollutants [7.78], also by facilitating the recolonization and recovery of disturbed areas [7.78] [7.96].

7.2.3 Motility in Other Substrate-Associated Diatoms Motility in centric diatoms occurs mostly through regulation of buoyancy in the water column. However, some centric species show motility when in contact with surfaces. Species of the genera Odontella [7.98], Actinocyclus [7.73], and Toxarium [7.65] were reported to display a weak form of movement associated with the secretion of mucilage through the labiate processes, pores formed by simple slits in the cell wall. While in Odontella the movements are not directed, with the cells remaining fixed while undergoing small oscillations, in Actinocyclus and Toxarium the cell movement is directional, resulting in speeds up to 4 µm s-1 [7.65]. The occurrence of motility in centric diatoms, despite seeming unimportant and restricted to a small number of species, appears to support the contention that the raphe may have evolved from the labiate process [7.38]. Within the pennates, motility is often considered limited to the raphid forms. However, slow movements have been documented for the araphid diatoms, such as Ardissonea crystallina [7.97] or Licmophora hyalina [7.111]. As in the centrics species mentioned above, the movement of these araphid cells is thought to be promoted by the secretion of mucilage, in this case through pores at the tips of the cell.

Mechanisms, Control and Adaptive Value  163

7.2.4 Vertical Migration in Diatom-Dominated Microphytobenthos Nowhere in the natural world is the motility of diatoms more easily observable and more ecologically relevant than within the microphytobenthos inhabiting intertidal and shallow subtidal sedimentary habitats. These communities of phototrophic microalgae and cyanobacteria known for forming dense and highly productive biofilms, are typically dominated by motile pennate diatoms [7.2] [7.70] [7.125]. Particularly in the case of intertidal communities, diatoms use their motility to migrate vertically in the upper layers of the sediment, in synchronization with day-night and tidal cycles [7.18] [7.86]. The phenomenon of vertical migration is the most evident macroscopic manifestation of diatom motility. It is detectable to the naked eye by a dramatic change in sediment coloration, associated with the synchronized and rhythmic movement of massive numbers of cells to and from the sediment surface [7.43]. Vertical migration has a central role in many key aspects of the biology and ecology of microphytobenthic diatom species. It controls photosynthetic activity, affecting photoacclimation and patterns of carbon assimilation [7.116], determines nutrient uptake and sexual reproduction [7.110], and is associated with the operation of an endogenous biological clock [7.86]. It also has important ecological implications, through the enhancement of productivity and the production of EPS [7.122]. The migratory behavior of diatoms in sediments is considered a critical trait and a key adaptation to the sedimentary environment and has attracted considerable research interest in the last decades. Consequently, motility responses like geotaxis or its endogenous control, have been almost exclusively studied in intact microphytobenthos assemblages, detected and quantified through vertical migration. Vertical migration is mostly readily observed in fine sediments, where the microphytobenthos is dominated by diatoms classified as “epipelic” [7.2] [7.104]. The “epipelon” is formed by biraphid motile cells, most belonging to the genera Navicula, Nitzschia, Gyrosigma, Pleurosigma and Diploneis, that move freely through gliding between the sediment particles. On more coarse sediments, diatoms belong to the “epipsammon,” a group comprised predominantly of non-motile or partially motile forms, including species of the genera Achnanthes, Opephora, Cocconeis, or Fragillaria, but also small-celled species within the motile genera of Nitzschia, Navicula or Amphora. Like the epipelic species, the motile epipsammic species are pennate raphids while the epipsammic non-motile are pennate araphids [7.5]. Epipsammic forms live attached to sediment particles by means of mucilage stalks or pads [7.59], and generally remain in hollows on the surface of grains [7.51]. Although they may show motility, migrating rhythmically like their epipelic counterparts [7.51], their migratory ability is generally much less pronounced [7.103]. The most commonly observed pattern of vertical migration in microphytobenthos comprises the rhythmic and synchronized movement of cells upward towards the surface of the sediment at the beginning of daytime periods of low tide, followed by the downward migration in anticipation of tidal flood or night (e.g., [7.42] [7.50] [7.86] [7.92] [7.115]). These migratory events only take place during daylight hours, but a bimodal pattern, in which cells emerge twice during the same day, may be observed during long summer days at high intertidal sites, when two periods of low tide exposure occur [7.18] [7.109]. Similar vertical migration rhythms have been found in non-tidal systems, the main difference being that the rhythm, not influenced by the tides, is essentially circadian [7.44] [7.52] [7.61] [7.108]. This general large-scale pattern of migration may be complemented by a more

164  Diatom Gliding Motility subtle “micro-migration” [7.66], consisting of the replacement of surface cells by others migrating from deeper layers [7.94] [7.127].

7.3 Controlling Factors of Diatom Motility Directed motility allows diatoms to respond behaviorally to a wide array of abiotic and biotic factors and cues, such as light, gravity, temperature, salinity, disturbance, desiccation, nutrients, or pheromones. Most experimental evidence comes from manipulative studies designed to isolate the effects of specific factors. Experimental studies focusing on the influence of external factors on the motility of planktonic centric species migrating via regulated buoyancy are scarce, likely due to difficulties in simulating in the laboratory the vertical heterogeneity of the oceanic water column (e.g., [7.39] [7.40] [7.102]). As such, most of the available data discussed in this work regards the motility of pennate diatoms and directed movement through gliding.

7.3.1 Motility Responses to Vectorial Stimuli 7.3.1.1 Light Intensity Responses to intensity and spectral composition of light are amongst the most important for diatoms, as they not only directly determine photosynthetic rates and growth, but also photodamage and possibly cell death. The effects of light on diatom motility are probably also the better-studied ones. The motility response of pennate diatoms to changes in light intensity is well documented, consisting in the avoidance of both darkness and low light and of very high light intensities, and the preference of intermediate irradiance levels. This general pattern has been found, and described in detail, not only for diatom-dominated sedimentary biofilms [7.18] [7.33] [7.90] [7.94] [7.112] [7.113], but also for isolated cells grown in unialgal cultures [7.19] [7.23] [7.34] [7.41] [7.71]. In biofilms, this behavior results in downward migration under low and high incident light levels, and in upward migration, and consequent accumulation of large numbers of cells at the surface, under intermediate irradiances. The relationship between the accumulation of cells at biofilm surface and ambient light intensity is well defined, and can be characterized by a “biomass vs. light” response curve [7.113]. These curves allow quantification of the main features of the migratory photoresponse: a biphasic pattern, with low cell densities under low irradiances, a steep increase in cell accumulation as light increases until reaching a well-defined peak, and a subsequent and more gradual decrease under higher irradiances. Maximum photoaccumulation generally occurs in irradiances from 70 to 250 µmol m-2 s-1 [7.34] [7.41] [7.69] [7.71] [7.113], but accumulation maxima have been reported at irradiances as low as 50 and as high as 500 µmol m-2 s-1 [7.33] [7.112]. Interestingly, similar optimum irradiance levels have been determined for biofilms on sediments and for isolated cells [7.71]. However, the light responses, and in particular the light levels which induce maximum photoaccumulation, are species-dependent [7.33] [7.71], and are affected by the photoacclimation state and susceptibility of cell to photoinhibition [7.41]. Manipulative experiments carried out on natural sedimentary assemblages have shown that light is a critical factor determining the cell’s migratory response, overriding other

Mechanisms, Control and Adaptive Value  165 environmental stimuli, like submersion in water, or circadian rhythms. Exposure to light upon incoming tidal inundation or at sunset causes diatoms that otherwise would migrate downwards to remain at the sediment surface [7.18] [7.51] [7.71] [7.75]. The response patterns observed at the biofilm level to changes in light intensity result from the combination of several types of light-regulated motility reactions by individual cells. These are comprised of light-induced changes in cell speed (photokinesis) and direction of cell movement, either towards or away from a light source (positive and negative phototaxis, respectively). Some authors further distinguish two types of phototaxis in diatoms: photo-topotaxis, when the cell’s light-oriented movement involves the perception of the direction of the incident light, and photo-phobotaxis, in the case of a reversal of the direction of movement away from the source, induced by a sudden change in light intensity [7.81] [7.139]. The photobehavior of pennate raphid diatoms, such as the movements toward preferred light conditions, and avoidance of less favorable conditions (darkness, excessive light intensities), is thought to result primarily from the modulation of their autonomous reversal rhythm by varying the time between direction changes at light-dark gradients. When moving in a favorable direction, the reversal of cell direction is delayed; if moving into less preferred conditions, the time between direction changes is shortened [7.19] [7.24] [7.71]. However, the regulation of movement is more complex, as negative photokinesis (reduction in cell speed with increasing irradiance) also plays a role, at least in promoting the avoidance of low light levels (but not escaping excessive ones), and in some species the photoresponses do not require phototaxis [7.71]. Light-induced changes in the direction of cell movement are triggered by illumination of photoresponsive regions located at the tips of the cells. Unlike the central region of the cell, the cell poles are sensitive to light intensity, the induced changes in movement direction requiring only brief (millisecond time-scale) exposure to light [7.24]. Changes in movement direction occur as the two tips respond to light separately and differently depending on their position relative to the direction of movement (leading, trailing) [7.24]. For example, exposure of the leading pole to high irradiance causes cells to quickly reverse direction, while exposure of both the leading and the trailing poles to the same light conditions does not induce any change in direction. This results in cells maintaining their direction when moving away from high light, causing a photophobic, out-of-light, response [7.23] [7.24]. The mechanisms underlying light perception and its transduction to motility are unclear but are thought to involve the regulation of mucilage secretion or the interaction between mucilage strands and the underlying actin-myosin system [7.21].

7.3.1.2 Light Spectrum The photobehavior of pennate diatoms is controlled not only by the intensity of incident light but also by its spectrum. As with the incident light intensity, the photodetection systems located at the cell’s tips are also differentially responsive to different light wavelengths [7.23] [7.81] [7.82] [7.140]. Evidence from a variety of species and benthic assemblages has shown that blue and red light are those that cause stronger motility responses. Blue light was shown to induce phototaxis in a wide range of pennate raphid species, such as Craticula cuspidate [7.24] [7.25], Navicula perminuta [7.71], Nitzschia communis [7.81], and Surirella gemma [7.56]. Blue

166  Diatom Gliding Motility light was seen to trigger both positive and negative phototaxis (depending on the intensity applied), indicating that the same type of photoreceptors is involved in the two responses [7.24] [7.71]. Also, for diatom-dominated sediments, blue light is the most effective in inducing the upward migration and accumulation of cells at the surface [7.140]. Although phototaxis can also be induced by light of a wide range of wavelengths in the UV-blue region, from 335 nm to 555 nm [7.71] [7.81], the wavelengths that most effectively induce movement vary between 430 nm [7.71] and 500 nm [7.24]. Red light also promotes phototaxis, although to a lesser extent than blue light, as confirmed by the accumulation of diatoms at the surface of sediments when exposed to red light [7.140]. Red light has, however, as opposed to blue light, a strong effect in diatom photokinesis. When compared to other wavelengths, red light tends to increase cell speed, contributing to the dispersion of cells, with maximum responses occurring for 686 nm [7.71]. Light of other wavelengths have typically much smaller effects on diatom motility, although some species respond to exposure to light in the 500–550 nm range [7.71] [7.81]. When trying to unveil the photoreceptors involved in these spectrally-dependent responses, the question arises of distinguishing the possible roles of specific receptors from the actions of photosynthetic pigments. This is difficult because both motility responses and pigment absorption show maxima in the blue and red regions of the visible spectrum. The hypothesis that pigments are the main photoreceptors involved in diatom motility responses was long dismissed, at least regarding the phototaxis of some species, since red light (absorption maxima for chlorophyll a) may not induce phototactic responses [7.81]. However, genes for a variety of photoreceptors, including blue-absorbing cryptochromes and phototropines, and red-absorbing phytochromes, have been identified in the genome of diatoms [7.14] [7.31]. As key components of the light detection systems in other organisms, it has been suggested that these molecules may have an alternative or complementary role in regulating photoresponses of pennate diatoms as well [7.23] [7.140].

7.3.1.3 UV Radiation Pennate diatoms also have the ability to detect and respond behaviorally to ultraviolet radiation (UVR). Early tests carried out on sedimentary microphytobenthos dominated by the species Gyrosigma balticum indicated that these diatoms respond to UVB radiation (280–315 nm, applied simultaneously with visible light) through negative phototaxis, migrating away from the surface [7.120] [7.126]. It was later found that benthic diatoms in natural assemblages were capable of detecting specifically UVB radiation. They responded promptly through vertical migration to UVB radiation, independently of UVA (315–400 nm) or visible light [7.136].

7.3.1.4 Gravity Motility responses to gravity have been postulated since the earlier works on microphytobenthic diatoms [7.18] [7.56] [7.57] [7.75] [7.85] [7.86] [7.107]. Despite being often mentioned, experimental evidence supporting diatom gravitaxis was limited [7.56] [7.58]. Only recently gravitaxis was demonstrated experimentally in sedimentary microphytobenthos [7.45]. By distinguishing gravitactic from surface-oriented cell movements, it could be confirmed that pennate benthic diatoms have the capacity to sense and use gravity to orient

Mechanisms, Control and Adaptive Value  167 themselves relative to the surface in the absence of directional physical or chemical cues. Both negative and positive gravitaxis were shown through the record of upward and downward migration in the absence of other significant stimuli [7.45]. Gravitactic responses can explain the vertical migratory behavior observed under constant conditions, when no other stimuli appear to influence the cells, such as the upward movement in anticipation of sunrise (negative geotaxis; [7.44]); or, conversely, the downward migration under constant light, anticipating high tide or sunset (positive geotaxis; [7.18]).

7.3.1.5 Chemical Gradients There is mounting evidence that pennate diatom species are capable of detecting sources or gradients of selected substances and of actively responding behaviorally, resulting in a directional movement towards the source of these chemicals and in the accumulation of cells in areas of higher concentrations. This phenomenon was first quantified in the species Amphora coffeaeformis, which showed positive chemotaxis along gradients of glucose and other sugars [7.28]. Similar chemical-oriented behaviors were later documented in other diatom species, regarding silica [7.11] and sex pheromones [7.12]. Directed movement is caused not only by chemotactic responses (directed movement along a gradient of increasing chemical concentration) but also by chemokinetic ones (changes in speed or turning frequency) [7.12]. Although relatively small, the number of species and isolates for which these behaviors have been observed suggests that chemical-oriented motility might be a widespread type of response in pennate diatoms [7.11].

7.3.2 Motility Responses to Non-Vectorial Stimuli 7.3.2.1 Temperature Temperature has a marked effect on diatom gliding motility. Although responses vary from species to species, a general pattern has emerged from experiments carried out with isolated species grown in culture, as well as with natural assemblages. Low temperatures (2−6 °C) cause a decrease in cell motility, observable as a decrease in the speed of isolated cells [7.22] [7.55] or in the vertical migration of natural assemblages in sediments [7.35] [7.109]. Maximum motility was observed under temperatures that vary between 15−17.5 °C for the species Navicula cyprinus, Pleurosigma angulatum, P. balticum, Nitzschia closterium, Stauroneis salina and Tropidoneis vitrae [7.55], 17−25 °C for Navicula pavillardii [7.123], and 30−35 °C for Craticula cuspidata, Stauroneis phoenicenteron, Nitzschia linearis, and Pinnularia viridis [7.22] [7.33]. In natural diatom assemblages, the highest rates of cell accumulation at the sediment surface were observed at temperatures ranging between 12−20 °C [109]. Increases in temperature above these optimum values typically induce a sharp decline in cell motility. The temperature response curve is clearly asymmetric, with the decrease in cell speed relative to the increase of temperature above optimum levels being much steeper than responses observed under lower temperatures [7.22]. Exposure to temperatures above 40 °C caused an almost complete elimination of directed cell movement in four diatom species. Interestingly, this effect is not permanent since cells retain motility when returned to lower temperatures [7.22]. The effects of temperature on diatom motility

168  Diatom Gliding Motility appears to be related to the rate of mucilage secretion and to changes in the viscosity of the cytoplasm in the raphe, and not to adhesion to the substrate [7.22] [7.121].

7.3.2.2 Salinity Salinity has a strong direct effect on the motility of pennate diatoms from intertidal marine and estuarine habitats. Extreme low and high salinities (5 and 60, respectively) cause a decrease in cell motility, as detected by a decrease in cell accumulation at the surface of sediments when compared to the behavior displayed under salinity 35 [7.112]. Results on isolated cells generally confirmed these observations, but further showed that extreme salinities cause a change in the type of movement performed by the diatoms. Low salinities tend to induce an alteration from gliding to other motility modalities, while high salinities cause the cessation of motility altogether [7.3].

7.3.2.3 pH Diatom motility is affected by a range of chemical substances. pH, as it is strongly and rapidly affected by photosynthetic activity and other processes taking place in the sedimentary environment, appears as particularly important. Tests carried out on cultures of pennate diatoms of the genera Craticula and Nitzschia have shown that gliding speeds are highest when pH is between 5−8, with maximum speeds occurring between 6–7 [7.20]. Field observations showed that changes in pH in the range 7.5−8.3 do not induce a significant migratory response [7.55]. pH values lower and higher than this optimum interval caused a marked decrease in cell motility, with almost complete cessation of cell movement for pH below 4 and above 10 [7.20]. Responses to changes in pH were also studied from gassing sedimentary assemblages with CO2 [7.140]. Flushing with CO2, causing a significant drop in pH from 7 to 4−5, induced the downward migration of motile diatoms and a decrease in cell numbers at the surface [7.140]. This migratory response to the acidification of the top layers of the sediment was observed even under light exposure, overriding the positive phototaxis expected under these conditions.

7.3.2.4 Calcium Calcium has long been known to be positively involved in raphid diatom motility. Early studies have shown that the reduction of Ca2+ ions in the external medium induced a decrease in cell motility [7.27] and cell adhesion [7.29]. The role of calcium on diatom motility was further confirmed by the inhibiting effects of calcium channel inhibitors and of calcium chelators on cell movement [7.20]. The observed strong effects of calcium on diatom motility are consistent with the accepted model involving actin and myosin, in which Ca2+ ions are required to generate motive force [7.72]. However, the mechanisms involved seem to vary with the type of behavioral response. Earlier indications that inhibitors of Ca2+ influx prevented motility [7.27] were confirmed by more recent results showing that the reducing influx of extracellular Ca2+ into intercellular stores, and not the release of Ca2+ from these stores, decreased cell speed [7.72]. However, inhibition of Ca2+ influx had no effect on blue light-induced negative phototactic responses, which seems to depend on the maintenance of large intracellular pools [7.72].

Mechanisms, Control and Adaptive Value  169 Calcium is also thought to be involved in detection of environmental stimuli. Based on the observed effects of light exposure on the persistence of cell direction reversal responses, calcium was hypothesized to act as a messenger, mediating light responses [7.24]. Recent evidence points to a role of cytosolic Ca2+ in motility responses to stress and to various environmental stimuli [7.72].

7.3.2.5 Other Factors In sedimentary habitats, pennate raphid diatoms respond behaviorally to a range of factors through vertical migration. These factors include those addressed above, but also others, acting concurrently, affecting the physicochemical conditions of the microenvironment at the surface layers of the sediment. One important factor is the sediment water content. For intertidal sites, during low tide, direct exposure to sunlight and wind causes the fast evaporation of water and an intense de-watering of the top layers of the sediment. Consequent changes in the water content of these layers induces a downward migration into deeper and wetter layers, resulting in a fast and marked decrease in surface diatom numbers [7.17]. However, downward migration also occurs as a response to sediment rewetting after prolonged air exposure, interpreted as being perceived by the cells as a signal of the incoming tidal flood [7.15] [7.57]. The sensitivity of diatom cells to rewetting, namely through capillary water, was shown to increase over the low tide period in anticipation of the tidal flood [7.57]. Freshwater, in the form of rain during low tide, also affects the migratory behavior of benthic diatoms through the acidification of surficial layers, as rainwater can be acid (pH 4−5) and low pH can induce motility responses (see above). Water may also trigger downward migration due to physical disturbance alone, associated with heavy rain during low tide [7.57] or wave action during submersion [7.63].

7.3.2.6 Inhibitors of Diatom Motility A number of chemicals are known to inhibit diatoms’ gliding movements. The most well studied are the Latrunculins (Lat) A and B, which sequester monomeric actinic and dissolve raphe-associated actin filaments, and 2,3-butanedione monoxime (BDM), which inhibits the actin-myosin interaction by inhibiting myosin ATPase activity [7.100]. Other substances, including the sugar D-mannose [7.27], the drugs against microtubules podophyllotoxin and vinblastine, and the drugs against actin filaments cytochalasins D and E [7.137], also have a strong effect on diatom motility, although inhibiting it only partially [7.100]. Lat A, Lat B, and BDM have been used as tools for studying the nature of the gliding mechanism in diatoms [7.100]. Lat A has been adopted to experimentally inhibit diatom motility in natural assemblages, often to study the photoprotective role of vertical migration [7.16] [7.32] [7.47] [7.49] [7.93] [7.114] [7.128].

7.3.3 Species-Specific Responses and Interspecies Interactions Many motility responses are strongly species-specific, an aspect that may be overlooked in studies based only on the bulk response of benthic diatom assemblages. Many cases of species-dependent behavioral responses have been reported. Species-specific responses are known to include cell motility [7.8] [7.25] [7.33] [7.34], responsiveness to light intensity

170  Diatom Gliding Motility [7.19] [7.23] [7.25] [7.33] [7.81] [7.112] and light spectrum [7.25] [7.71] [7.81], temperature [7.33], adhesion ability [7.22] [7.25], mucilage secretion [7.7] [7.21], and production of allelopathic substances affecting cell motility [7.129]. Marked differences in the behavior of individual species are also evident from studies on the vertical migration of undisturbed sedimentary biofilms, showing differences in timing of cells surfacing and permanence at the surface [7.66] [7.84] [7.88] [7.94] [7.109] [7.127]. Particularly interesting is the fact that motility responses are not only species-specific but are modulated by the presence of cells of a second diatom species. Such interspecies interactions have been demonstrated regarding the light sensitivity, cell speed, and cell adhesion [7.22] [7.23]. For example, the speed of cells of Pinnularia viridis was reduced in the presence of cells of Stauroneis phoenicenteron, but not in the presence of cells of Craticula cuspidata [7.22].

7.3.4 Endogenous Control of Motility One of the most fascinating aspects of raphid diatom motility is the fact that, when expressed as vertical migration in sedimentary environments, the phenomenon is partially endogenously controlled, exhibiting a self-sustained rhythm synchronized with environmental day-night and tidal cycles (for an extensive review, see [7.26]). Endogenous rhythms in vertical migratory behavior have been known since early studies on microphytobenthos [7.15] [7.42] [7.48], detected by the persistence of vertical movements in samples kept under constant conditions in the absence of external stimuli, either under low light or in darkness. Since then, endogenously controlled vertical migration rhythms have been demonstrated in a multitude of studies, most for natural assemblages in intertidal habitats [7.18] [7.45] [7.50] [7.52] [7.75] [7.88] [7.109] [7.112] [7.115] but also for freshwater, non-tidal environments [7.44] [7.107]. Persistent rhythmicity was also found for the motility and cell speed of isolated diatom cells [7.50]. In most cases, the vertical migratory rhythm follows a circatidal pattern, with diatom cells retaining the timing of upward migration corresponding to the diurnal low tide of the day of collection (e.g., [7.86] [7.115]. Other studies reported circatidal rhythms but following the daily delay of the tide [7.50] or a simple circadiurnal rhythm, with diatom cells remaining at the surface during the whole photoperiod, irrespective of the phase of the tidal cycle [7.105].

7.3.5 A Model of Diatom Vertical Migration Behavior in Sediments Under natural conditions, the motility behavior of pennate diatoms results from the combination of endogenous control and responses to external stimuli. The relative role of endogenous versus environmental forcing, the latter associated to the day-night and tidal cycles, has been studied by comparing the patterns of vertical migration in undisturbed microphytobenthos samples kept under constant conditions and under ambient light conditions [7.18] [7.50] [7.57] [7.108] [7.109]. Several models have been proposed to explain the general patterns of benthic diatom vertical migration, involving different forms of coupling between rhythms in cell motility and in the sign and strength of phototaxis and geotaxis [7.18] [7.57] [7.109]. From this and more recent evidence, an integrated, general model can be proposed, distinguishing three phases during the daytime exposure period (Figure 7.1):

Mechanisms, Control and Adaptive Value  171 Low light

Surface diatom biomass (r.u.)

100

3

80

4 60 40 2 20 0 10:00

1

Dark 12:00

14:00

16:00

18:00

Time of day Geotaxis (−) Geotaxis (+) Phototaxis (+, −)

Figure 7.1  Proposed model of the control of vertical migration by sediment-inhabiting benthic pennate diatoms, as responding to main directional environmental stimulus, light and gravity. The figure illustrates the variation with the time of day of the diatom biomass at the sediment surface on samples kept in the dark (closed circles) and exposed to constant low light (150 µmol m-2 s-1) during the subjective low tide period (open circles) (for more details, see [7.18]). Example of a day when the low tide takes place during the middle of the day. The gray horizontal plots represent the strength (bar thickness) of photo- and geotaxis, of negative and positive signal, along the day. (1) Upward migration starts before the beginning of the light period, driven by negative geotaxis. (2) Negative geotaxis ceases roughly at the time expected for start of the low tide light period: if no light is available at surface, the diatoms stop migrating upwards and the incipient biofilms start to disaggregate, due to random cell movement or weak positive geotaxis. (3) During light exposure, cell movements are controlled mainly by phototaxis, either positive (under low light intensities) or negative (under high light intensities). In the particular case of the data in the figure, positive phototaxis dominates, as samples were exposed to low intensity. (4) Anticipating the end of the light period, geotaxis becomes dominant over phototaxis, as cells begin to migrate downwards without any changes in incident illumination. Vertical gray areas represent periods of darkness. Vertical white area represents the period of light exposure (150 µmol m-2 s-1) of the light exposed samples.

(i) Upward migration and surfacing at the start of the light period. Earlier models proposed positive phototaxis to explain the upward migration of cells early in the morning or at the beginning of daytime low tide. During this period, the positive phototaxis would overcome a positive geotaxis, which would increase in strength throughout the day, explaining the downward migration at the end of the light period [7.57] [7.108] [7.109]. However, the widespread observation that the upward migration anticipates the start of the light period and occurs independently of light exposure led to the thinking that negative geotaxis could play a role [7.18] [7.44] [7.107] [7.108] [7.115]. More recent results on the influence of geotaxis on benthic diatom motility

172  Diatom Gliding Motility support the idea that upward migration starts before first light, driven by a strong endogenous negative geotaxis [7.45]. The accumulation of cells at the surface and the formation of a biofilm at the start of the low tide exposure is likely a two-phase process. The first phase is an upward migration pulse that starts hours before the start of the light period and is fully endogenously controlled; this is followed by a second phase, when geotaxis ceases and the movement becomes driven by environmental factors, mainly by a strong positive phototaxis, resulting in a fast accumulation rate of cells at the surface, but only if illuminated [7.18]; (ii) Regulation of surface biomass during daytime exposure. Once in the vicinity of the surface, the movement of diatom cells becomes mainly regulated by exogenous stimuli, particularly irradiance, overriding the endogenous control observed under constant conditions, and allowing the cells to respond quickly to changing environmental conditions. Experimental evidence for this environmental forcing is abundant, especially regarding phototaxis and the responses to light intensity (see above, Sections 7.3.1.1–7.3.1.3); (iii) Downward migration at the end of the light period. When approaching the end of the low tide light period, due to either tidal flood or sunset, the biofilm begins to spontaneously disaggregate due to downward migration of large numbers of cells. The consistent observation that this behavior occurs in constant conditions, both under illumination and in the dark, points to the action being under endogenous control, likely a positive geotaxis [7.18] [7.107]. Of importance is also the variation of the motility behavior throughout the fortnightly spring-neap tidal cycle. The “rephasing” of the migratory cycle when the light exposure during low tide begins to be truncated by sunset, with a new phase appearing early in the morning, was noticed in earlier studies [7.50] [7.86]. Later experiments have shown that the relative effect of the endogenous component of the migratory behavior varies periodically along the spring-neap tidal cycle, revealing an entrainment associated with the variation of the timing and duration of low tide exposure periods during the fortnightly tidal cycle. The exposure to increasingly longer light periods as the time of day of low tide progresses during the spring-neap cycle reinforces the endogenous behavior, reaching a maximum response during the days when low tide occurs in the middle of the day. When tidal flood begins to take place after sunset, the interruption of the low tide light period is memorized, resulting in such that during the next day the downward migration anticipates the approach of night [7.18].

7.4 Adaptive Value and Consequences of Motility 7.4.1 Planktonic Centrics For planktonic centric diatoms, the capacity to move vertically in the water column has been thought to represent an adaptation to living in stratified waters, typical of warm, oligotrophic regions [7.62]. Vertical migration confers these cells the capability of obtaining

Mechanisms, Control and Adaptive Value  173 nutrients from sub-euphotic layers and for photosynthetizing in well-lit conditions, and for avoiding direct competition with much more abundant, non-migratory smaller phytoplankton, overcoming the disadvantages of their larger size [7.117] [7.134]. The vertical migration of large centric diatoms represents a form of “nutrient mining,” through which substantial amounts of nutrients are transported upwards across the nutricline, replenishing the otherwise nutrient-depleted photic layer and contributing to new production [7.117]. On the other hand, carbon fixed through photosynthesis near the surface is transported downwards and respired at subphotic layers. These processes have a large impact on the vertical fluxes of nitrogen, and presumably on other nutrients like phosphorous. It has been estimated that vertical migration contributes to more than one quarter of the surface nitrate pool [7.117]. The vertical migration of centric diatoms is widespread, and has major biogeochemical consequences, calling for the reassessment of the role of motility in marine phytoplankton and the reevaluation of the predicted implications of global warming on changes in phytoplankton diversity [7.62] [7.134].

7.4.2 Benthic Pennates Raphe-associated motility is a recent trait in diatom evolution, having appeared during the Palaeocene epoch, ca. 30 million years ago [7.4]. Directed motility is thought to have conferred important adaptive advantages, allowing raphids to respond efficiently to environmental gradients and to actively exploit habitat heterogeneity regarding resource distribution (light, nutrients, carbon) [7.23] [7.26] [7.80]. It has been hypothesized that motility has enabled raphid diatoms to colonize new niches, including the sedimentary microenvironment, and, by enhancing sexual reproduction, to be the primary driver of the rapid and large diversification that has made this group the most diverse of present-day diatoms [7.64] [7.80]. In the benthic sedimentary environment, diatom motility is mostly expressed through vertical migration. Vertical migratory movements, directed by light conditions, gravity, chemical gradients or pheromones (e.g., the pheromone L-proline-derived diketopiperazine, involved in the motility of cells of the species Seminavis robusta [7.12]), confer obvious benefits for the cells and appear particularly significant in terms of exploiting environmental heterogeneity. Because of the comparable scales of the spatial variability of resources like light or nutrients, and the sizes of diatom cells, raphids are able to cover significant ranges of the environmental gradients found among the sedimentary microhabitats [7.11] [7.23] [7.113]. Several additional advantages have been speculated to be associated with diatom vertical migration. They are often considered individually in the literature, but it is likely that many are linked to some degree: (i) Enhanced nutrient uptake. Vertical migration, in combination with the capacity to forage for nutrients using chemotaxis [7.11], allows movement between the well-lit surface layers where nutrients and carbon may be depleted during the day, and the deeper nutrient-rich regions at night (e.g., [7.6] [7.50] [7.109]); (ii) Facilitated sexual reproduction. Vertical movements enable the cells to reach deep nutrient-rich layers where reproduction is favored [7.110], also

174  Diatom Gliding Motility increasing the likelihood of contact between potentially mating cells, and possibly contributing to faster rates of adaptive divergence [7.80]; (iii) Reduced exposure to grazers. The ability to migrate downwards away from the surface may reduce the predation pressure from animals grazing at the surface of the sediment [7.60] [7.83]; (iv) Reduced impacts of the instability of the intertidal sedimentary environment. Downward vertical migration allows diatom cells to escape from resuspension into the water column due to the constant action of waves, currents, tides, wind, sediment deposition and bioturbation (e.g., [7.42] [7.54] [7.57] [7.63]). Vertical migration, guided by gravitaxis, may also attenuate the effects of the removal of cells from the surface due to resuspensiondeposition events and burial by permitting buried cells to return to the surface and act as an “inoculum” for new populations [7.30] [7.36]. A related effect is the reduction in the rates of passive dispersal, causing a greater spatial isolation between local populations, and increased rates of speciation [7.80]; (v) Anticipation of environmental periodicity. Because their vertical movement is (partially) under control of an internal biological clock, sediment-­ inhabiting raphid diatoms display endogenous vertical migratory rhythms. These rhythmic movements are synchronized with the day-night and tidal cycles, allowing the cells to anticipate main periodic events such as sunrise or sunset, and tidal ebb or flood [7.18] [7.26] [7.109]; (vi) Benefits from EPS production. The production of large amounts of EPS associated with vertical migration has known advantages for biofilm-forming diatom cells, namely protection against desiccation and high salinities [7.1] [7.119], and as a mechanism to deal with photosynthetic overflow [7.124]. Particularly discussed in the literature is the possible adaptive advantage conferred by the complex motility response to light, characterized by the combination of positive phototaxis under low light and negative phototaxis under high light, resulting in the definition of an intensity optimum, under which maximum cell accumulation is observed. The similar dimensions of many diatom cells and of the sediment photic zone, and the possibility to cover, through vertical migration, a large range of light levels in a short time, led earlier researchers to consider that diatoms could use motility to position themselves towards optimal light conditions [7.2] [7.105]. In what became known as the “behavioral photoprotection” hypothesis, diatoms would use vertical motility as a form of photoregulation, optimizing light exposure by simultaneously avoiding low light regions, where photosynthesis is light-limited, and escaping regions of photodamaging light levels, minimizing excessive excitation pressures to photosystem II (e.g., [7.2] [7.113] [7.136]). The ubiquity and regularity of these photobehavioral patterns made it widely accepted that light-induced vertical migration confers important adaptive advantages in the variable and extreme light environment of the intertidal sedimentary habitat (e.g., [7.20] [7.23] [7.26] [7.95] [7.121]). The photoregulatory role of vertical migration is well supported by experimental evidence: (i) diatom movements and positioning in the vertical light gradient are regulated by light intensity, including photodamaging UV radiation [7.23] [7.79] [7.113] [7.136]; (ii) diatom-dominated biofilms treated with the motility inhibitor Lat A suffer a decrease in photosynthetic activity under high light [7.93] [7.114]; (iii) light levels preferred by

Mechanisms, Control and Adaptive Value  175 diatoms depend on their photophysiological (photoacclimation) state and susceptibility to photodamage [7.41]; (iv) light-induced vertical migration is as fast as the activation of physiological photoprotection mechanisms [7.9] [7.47] [7.69]; (v) a trade-off seems to exist between motility-based and physiological photoprotective mechanisms, with motile species relying less on physiological photoprotection than their non-motile counterparts [7.5] [7.46]. Importantly, there seems to exist a mutual dependency between photobehavior and photophysiological needs, and a complex interdependency among the regulation of the two types of mechanisms [7.41] [7.69] [7.142]. On one hand, the motility responses are conditioned by the previous light history, which determined the cell’s photophysiological state and light preferences [7.41]. On the other, the inverse may also happen: photobehavior, namely by systematically avoiding high light exposure, may modulate the light regime experienced by the cells and thus influence their photoacclimation state [7.46] [7.69]. The relative role of motility- and physiological-based photoregulation is likely defined by the relative cost and benefits provided by each process. Vertical migration has been speculated to be energetically cheaper, more flexible, and faster than physiological photoacclimation [7.116]. Energetic costs of diatom movements are hard to ascertain, although indications exist that they may be relatively minor considering the overall cellular metabolism [7.37] [7.53]. However, the flexibility and promptness provided by cell motility certainly seems highly advantageous relative to the slowly reversible changes in photoacclimation state.

7.4.3 Ecological Consequences of Vertical Migration 7.4.3.1 Motility-Enhanced Productivity By conferring important adaptive benefits, the motility of pennate diatoms is thought to favor the productivity of microphytobenthic communities. Vertical migration is considered a key factor in increasing the fitness of benthic diatoms, helping them to cope with the highly variable and extreme conditions of the intertidal habitat, ultimately contributing to the high rates of primary productivity typical of these areas [7.68] [7.114]. This motility-­enhanced productivity may be due, besides to the above-discussed benefits of active exploitation of environmental heterogeneity, to improved resource partitioning between different species, minimizing direct interspecific competition [7.23] [7.71]. One form of resource partitioning would be the one resulting from the asynchronous vertical migration of cells of different species (“micro-migration” [7.66]): cells of different species migrate to the surface at different times of the day, resulting in a continuous replacement of the cells in the photic zone. Such a pattern was observed on natural microphytobenthos assemblages, with cells of the species Stauroneis amphioxus dominant during early morning hours being replaced by cells of Pleurosigma angulatum at midday [7.94]. This behavior would confer benefits to the cells, when considered individually, but a far greater advantage to biofilm-level productivity [7.127]. Another important ecological implication of diatom motility is the periodic short-term variability in time scales ranging from minutes to weeks, in primary productivity rates [7.26] [7.99] [7.116]. It is well known that the variability of photosynthesis over tidal emersion periods is strongly determined by migration patterns, as surface microphytobenthos

176  Diatom Gliding Motility biomass is a major controlling factor of the community-level photosynthesis on short-term time scales (e.g., [7.36] [7.99] [7.116]).

7.4.3.2 Carbon Cycling and Sediment Biostabilization The large amounts of carbon-rich mucilages excreted during diatom vertical migration cycles represent a major source of organic carbon, which is rapidly used to fuel the growth of heterotrophic bacteria [7.10] [7.74] [7.76]. A well-studied effect of produced EPS associated with diatom movements is the promotion of sediment stabilization in estuarine benthic habitats. The EPS accumulated in the upper layers of the sediment over periods of low tide increases critical shear stress and thus resistance to erosion, promoting habitat stability [7.77] [7.89] [7.143]. Due to this role in sediment biostabilization, benthic pennate diatoms are thus increasingly recognized as crucial “ecosystem engineers” in tidal environments [7.87].

Acknowledgments The author thanks FCT/MCTES for the financial support to CESAM (UIDP/ 50017/2020+UIDB/50017/2020), through national funds. I thank William Schmidt for revising the manuscript, and one anonymous reviewer for valuable comments.

References [7.1] Abdullahi, A.S., Underwood, G.J.C., Gretz, M.R., Extracellular matrix assembly in diatoms (Bacillariophyceae). V. Environmental effects on polysaccharide synthesis in the model diatom, Phaeodactylum tricornutum. J. Phycol., 42, 2, 363–378, 2006. [7.2] Admiraal, W., The ecology of estuarine sediment-inhabiting diatoms, in: Progress in Phycological Research, vol. 3, pp. 269–322, 1984. [7.3] Apoya-Horton, M.D., Yin, L., Underwood, G.J.C., Gretz, M.R., Movement modalities and responses to environmental changes of the mudflat diatom Cylindrotheca closterium (Bacillariophyceae). J. Phycol., 42, 2, 379–390, 2006. [7.4] Armbrust, E.V., The life of diatoms in the world’s oceans. Nature, 459, 7244, 185–192, 2009. [7.5] Barnett, A., Méléder, V., Blommaert, L., Lepetit, B., Gaudin, P., Vyverman, W., Lavaud, J., Growth form defines physiological photoprotective capacity in intertidal benthic diatoms. ISME J., 9, 1, 32–45, 2015. [7.6] Barranguet, C., Kromkamp, J., Peene, J., Factors controlling variations in primary production and photosynthetic characteristics of intertidal microphytobenthos. Mar. Ecol. Prog. Ser., 173, 117–126, 1998. [7.7] Bellinger, B., Abdullahi, A., Gretz, M., Underwood, G., Biofilm polymers: relationship between carbohydrate biopolymers from estuarine mudflats and unialgal cultures of benthic diatoms. Aquat. Microb. Ecol., 38, 2, 169–180, 2005. [7.8] Bertrand, J., La vitesse de deplacement des diatomees. Diatom Res., 5, 2, 223–239, 1990. [7.9] Blommaert, L., Lavaud, J., Vyverman, W., Sabbe, K., Behavioural versus physiological photoprotection in epipelic and epipsammic benthic diatoms. Eur. J. Phycol., 53, 2, 146–155, 2018.

Mechanisms, Control and Adaptive Value  177 [7.10] Bohórquez, J., McGenity, T.J., Papaspyrou, S., García-Robledo, E., Corzo, A., Underwood, G.J.C., Different types of diatom-derived extracellular polymeric substances drive changes in heterotrophic bacterial communities from intertidal sediments. Front. Microbiol., 8, 245, 2017. [7.11] Bondoc, K.G.V., Heuschele, J., Gillard, J., Vyverman, W., Pohnert, G., Selective silicate-­ directed motility in diatoms. Nat. Commun., 7, 10540, 2016. [7.12] Bondoc, K.G.V., Lembke, C., Vyverman, W., Pohnert, G., Searching for a mate: Pheromonedirected movement of the benthic diatom Seminavis robusta. Microb. Ecol., 72, 2, 287–294, 2016. [7.13] Bothwell, M.L., Suzuki, K.E., Bolin, M.K., Hardy, F.J., Evidence for dark avoidance by phototrophic periphytic diatoms in lotic systems. J. Phycol., 25, 1, 85–94, 1989. [7.14] Bowler, C., Vardi, A., Allen, A.E., Oceanographic and biogeochemical insights from diatom genomes. Annu. Rev. Mar. Sci., 2, 1, 333–365, 2010. [7.15] Callame, B. and Debyser, J., Observations sur les mouvements des diatomees a la surface des sediments marins de la zone intercotidale. Vie Milieu, 5, 242–249, 1954. [7.16] Cartaxana, P. and Serôdio, J., Inhibiting diatom motility: a new tool for the study of the photophysiology of intertidal microphytobenthic biofilms. Limnol. Oceanogr.: Methods, 6, 9, 466–476, 2008. [7.17] Coelho, H., Vieira, S., Serôdio, J., Effects of desiccation on the photosynthetic activity of intertidal microphytobenthos biofilms as studied by optical methods. J. Exp. Mar. Biol. Ecol., 381, 2, 98–104, 2009. [7.18] Coelho, H., Vieira, S., Serôdio, J., Endogenous versus environmental control of vertical migration by intertidal benthic microalgae. Eur. J. Phycol., 46, 3, 271–281, 2011. [7.19] Cohn, S.A., Bahena, M., Davis, J.T., Ragland, R.L., Rauschenberg, C.D., Smith, B.J., Characterisation of the diatom photophobic response to high irradiance. Diatom Res., 19, 2, 167–179, 2004. [7.20] Cohn, S.A. and Disparti, N.C., Environmental factors influencing diatom cell motility. J. Phycol., 30, 5, 818–828, 1994. [7.21] Cohn, S.A., Dunbar, S., Ragland, R., Schulze, J., Suchar, A., Weiss, J., Wolske, A., Analysis of light quality and assemblage composition on diatom motility and accumulation rate. Diatom Res., 31, 3, 173–184, 2016. [7.22] Cohn, S.A., Farrell, J.F., Munro, J.D., Ragland, R.L., Weitzell, R.E., Wibisono, B.L., The effect of temperature and mixed species composition on diatom motility and adhesion. Diatom Res., 18, 2, 225–243, 2003. [7.23] Cohn, S.A., Halpin, D., Hawley, N., Ismail, A., Kaplan, Z., Kordes, T., Zapata, Y., Comparative analysis of light-stimulated motility responses in three diatom species. Diatom Res., 30, 3, 213–225, 2015. [7.24] Cohn, S.A., Spurck, T.P., Pickett-Heaps, J.D., High energy irradiation at the leading tip of moving diatoms causes a rapid change of cell direction. Diatom Res., 14, 2, 193–206, 1999. [7.25] Cohn, S.A. and Weitzell, R.E., Ecological considerations of diatom cell motility. I. Characterization of motility and adhesion in four diatom species. J. Phycol., 32, 6, 928–939, 1996. [7.26] Consalvey, M., Paterson, D.M., Underwood, G.J.C., The ups and downs of life in a benthic biofilm: Migration of benthic diatoms. Diatom Res., 19, 2, 181–202, 2004. [7.27] Cooksey, B. and Cooksey, K.E., Calcium is necessary for motility in the diatom Amphora coffeaeformis. Plant Physiol., 65, 1, 129–131, 1980. [7.28] Cooksey, B. and Cooksey, K.E., Chemical signal-response in diatoms of the genus Amphora. J. Cell Sci., 91, 4, 523–529, 1988.

178  Diatom Gliding Motility [7.29] Cooksey, K.E., Requirement for calcium in adhesion of a fouling diatom to glass. Appl. Environ. Microbiol., 41, 6, 1378–1382, 1981. [7.30] Delgado, M., de Jonge, V.N., Peletier, H., Jonge, V.N., Peletier, H., Experiments on resuspension of natural microphytobenthos populations. Mar. Biol., 108, 2, 321–328, 1991. [7.31] Depauw, F.A., Rogato, A., D’Alcalá, M.R., Falciatore, A., Ribera d’Alcalá, M., Falciatore, A., Exploring the molecular basis of responses to light in marine diatoms. J. Exp. Bot., 63, 4, 1575–1591, 2012. [7.32] Du, G., Yan, H., Liu, C., Mao, Y., Behavioral and physiological photoresponses to light intensity by intertidal microphytobenthos. Chin. J. Oceanol. Limnol., 36, 2, 293–304, 2018. [7.33] Du, G.Y., Li, W.T., Li, H., Chung, I.K., Migratory responses of benthic diatoms to light and temperature monitored by chlorophyll fluorescence. J. Plant Biol., 55, 2, 159–164, 2012. [7.34] Du, G.Y., Oak, J.-H.H., Li, H., Chung, I.-K.K., Effect of light and sediment grain size on the vertical migration of benthic diatoms. Algae, 25, 3, 133–140, 2010. [7.35] Du, G.Y., Son, M., An, S., Chung, I.K., Ying, G., Kyo, I., Temporal variation in the vertical distribution of microphytobenthos in intertidal flats of the Nakdong River estuary, Korea. Estuar. Coast. Shelf Sci., 86, 1, 62–70, 2010. [7.36] Easley, J.T., Hymel, S.N., Plante, C.J., Temporal patterns of benthic microalgal migration on a semi-protected beach. Estuar. Coast. Shelf Sci., 64, 2–3, 486–496, 2005. [7.37] Edgar, L.A., Diatom locomotion: A consideration of movement in a highly viscous situation. Br. Phycol. J., 17, 3, 243–251, 1982. [7.38] Edgar, L.A. and Pickett-Heaps, J.D., Diatom locomotion, in: Progress in Phycological Research, vol. 3, pp. 47–88, 1984. [7.39] Erga, S.R., Lie, G.C., Aarø, L.H., Aursland, K., Olseng, C.D., Frette, Ø., Hamre, B., Fine scale vertical displacement of Phaeodactylum tricornutum (Bacillariophyceae) in stratified waters: Influence of halocline and day length on buoyancy control. J. Exp. Mar. Biol. Ecol., 384, 1–2, 7–17, 2010. [7.40] Erga, S.R., Lie, G.C., Aarø, L.H., Frette, Ø., Hamre, B., Migratory behaviour of Skeletonema grethae (Bacillariophyceae) in stratified waters. Diatom Res., 30, 1, 13–25, 2015. [7.41] Ezequiel, J., Laviale, M., Frankenbach, S., Cartaxana, P., Serôdio, J., Photoacclimation state determines the photobehaviour of motile microalgae: The case of a benthic diatom. J. Exp. Mar. Biol. Ecol., 468, 11–20, 2015. [7.42] Fauré-Fremiet, E., The tidal rhythm of the diatom Hantzchia amphioxys. Biol. Bull., 100, 3, 173–177, 1951. [7.43] Fauvel, P. and Bohn, G., Le rhythme des marees chez les diatomees littorales. C. R. Séances Soc Biol., 62, 121–123, 1907. [7.44] Fischer, H., Gröning, C., Köster, C., Groning, C., Koster, C., Vertical migration rhythm in freshwater diatoms. Hydrobiologia, 56, 3, 259–263, 1977. [7.45] Frankenbach, S., Pais, C., Martinez, M., Laviale, M., Ezequiel, J., Serôdio, J., Evidence for gravitactic behaviour in benthic diatoms. Eur. J. Phycol., 49, 4, 429–435, 2014. [7.46] Frankenbach, S., Schmidt, W., Frommlet, J., Serôdio, J., Photoinactivation, repair and the motility-physiology trade-off in microphytobenthos. Mar. Ecol. Prog. Ser., 601, 41–57, 2018. [7.47] Frankenbach, S. and Serôdio, J., One pulse, one light curve: Fast characterization of the light response of microphytobenthos biofilms using chlorophyll fluorescence. Limnol. Oceanogr.: Methods, 15, 6, 554–566, 2017. [7.48] Ganapati, P.N., Lakshmanu Rao, M.V., Subba Rao, D.V., Tidal rhythms of some diatoms and dinoflagellates inhabiting the intertidal sands of the Visakhapatnam beach. Curr. Sci., 11, 450–451, 1959.

Mechanisms, Control and Adaptive Value  179 [7.49] Goessling, J., Frankenbach, S., Ribeiro, L., Serôdio, J., Kühl, M., Modulation of the light field related to valve optical properties of raphid diatoms: implications for niche differentiation in the microphytobenthos. Mar. Ecol. Prog. Ser., 588, 29–42, 2018. [7.50] Happey-Wood, C.M. and Jones, P., Rhythms of vertical migration and motility in intertidal benthic diatoms with particular reference to Pleurosigma angulatum. Diatom Res., 3, 1, 83–93, 1988. [7.51] Harper, M.A., Movement and migration of diatoms on sand grains. Br. Phycol. J., 4, 1, 97–103, 1969. [7.52] Harper, M.A., Migration rhythm of the benthic diatom Pinnularia viridis on pond silt. N. Z. J. Mar. Freshwater Res., 10, 2, 381–384, 1976. [7.53] Hay, S.I., Maitland, T.C., Paterson, D.M., The speed of diatom migration through natural and artificial substrata. Diatom Res., 8, 2, 371–384, 1993. [7.54] Heckman, C.W., The development of vertical migration patterns in the sediments of estuaries as a strategy for algae to resist drift with tidal currents. Int. Rev. Gesamten Hydrobiol. Hydrogr., 70, 1, 151–164, 1985. [7.55] Hopkins, J.T., A study of the diatoms of the Ouse Estuary, Sussex. I. The movement of the mud-flat diatoms in response to some chemical and physical changes. J. Mar. Biol. Assoc. U. K., 43, 3, 653–663, 1963. [7.56] Hopkins, J.T., The behaviour of the diatom Surirella gemma in response to some tidal and and diurnal stimuli. Br. Phycol. Bull., 2, 6, 513–514, 1965. [7.57] Hopkins, J.T., The role of water in the behaviour of an estuarine mud-flat diatom. J. Mar. Biol. Assoc. U. K., 46, 3, 617–626, 1966. [7.58] Hopkins, J.T. and Drum, R.W., Diatom motility: An explanation and a problem. Br. Phycol. Bull., 3, 1, 63–67, 1966. [7.59] Jewson, D.H., Lowry, S.F., Bowen, R., Co-existence and survival of diatoms on sand grains. Eur. J. Phycol., 41, 2, 131–146, 2006. [7.60] Joint, I., Gee, J., Warwick, R., Determination of fine-scale vertical distribution of microbes and meiofauna in an intertidal sediment. Marine Biol., 72, 2, 157–164, 1982. [7.61] Jönsson, B., Sundbäck, K., Nilson, C., An upright life-form of an epipelic motile diatom: on the behaviour of Gyrosigma balticum. Eur. J. Phycol., 29, 1, 11–15, 1994. [7.62] Kemp, A.E.S. and Villareal, T.A., High diatom production and export in stratified waters - A potential negative feedback to global warming. Prog. Oceanogr., 119, 4–23, 2013. [7.63] Kingston, M.B., Wave effects on the vertical migration of two benthic microalgae: Hantzschia virgata var. intermedia and Euglena proxima. Estuaries, 22, 1, 81, 1999. [7.64] Kooistra, H., Gersonde, R., Medlin, L., Mann, D., The origin and evolution of the diatoms: their adaptation to a planktonic existence, in: Evolution of Primary Producers in the Sea, P.G. Falkowski and A. Knoll (Eds.), pp. 207–249, Elsevier Academic Press, Amsterdam, 2007. [7.65] Kooistra, W.H.C.F., De Stefano, M., Mann, D.G., Salma, N., Medlin, L.K., Phylogenetic position of Toxarium, a pennate-like lineage within centric diatoms (Bacillariophyceae). J. Phycol., 39, 1, 185–197, 2003. [7.66] Kromkamp, J., Barranguet, C., Peene, J., Determination of microphytobenthos PSII quantum yield efficiency and photosynthetic activity by means of variable chlorophyll fluorescence. Mar. Ecol. Prog. Ser., 162, 45–55, 1998. [7.67] Lauterborn, R., Untersuchungen über Bau, Kernteilung und Bewegung der Diatomeen, Leipzig: W. Englemann. 165 pp., 1896. [7.68] Laviale, M., Barnett, A., Ezequiel, J., Lepetit, B., Frankenbach, S., Méléder, V., Lavaud, J., Response of intertidal benthic microalgal biofilms to a coupled light-temperature stress: evidence for latitudinal adaptation along the Atlantic coast of Southern Europe. Environ. Microbiol., 17, 10, 3662–3677, 2015.

180  Diatom Gliding Motility [7.69] Laviale, M., Frankenbach, S., Serôdio, J., The importance of being fast: comparative kinetics of vertical migration and non-photochemical quenching of benthic diatoms under light stress. Mar. Biol., 163, 1, 1–12, 2016. [7.70] MacIntyre, H.L., Geider, R.J., Miller, D.C., Microphytobenthos: The ecological role of the “secret garden” of unvegetated, shalow-water marine habitats. I. Distribution, abundance and primary production. Estuaries, 19, 2A, 186–201, 1996. [7.71] McLachlan, D.H., Brownlee, C., Taylor, A.R., Geider, R.J., Underwood, G.J.C., Light-induced motile responses of the estuarine benthic diatoms Navicula perminuta and Cylindrotheca closterium (Bacillariophyceae). J. Phycol., 45, 3, 592–599, 2009. [7.72] McLachlan, D.H., Underwood, G.J.C., Taylor, A.R., Brownlee, C., Calcium release from intracellular stores is necessary for the photophobic response the benthic diatom Navicula perminuta (Bacillariophyceae). J. Phycol., 48, 3, 675–681, 2012. [7.73] Medlin, L.K., Crawford, R.M., Andersen, R.A., Histochemical and ultrastructural evidence for the function of the labiate process in the movement of centric diatoms. Br. Phycol. J., 21, 3, 297–301, 1986. [7.74] Middelburg, J.J.B.C., Barranguet, C., Boschker, H.T.S., Herman, P.M.J., Moens, T., Heip, C.H.R., The fate of intertidal microphytobenthos carbon: An in situ 13C-labeling study. Limnol. Oceanogr., 45, 6, 1224–1234, 2000. [7.75] Mitbavkar, S. and Anil, A.C.C., Vertical migratory rhythms of benthic diatoms in a tropical intertidal sand flat: influence of irradiance and tides. Mar. Biol., 145, 1, 9–20, 2004. [7.76] Miyatake, T., Moerdijk-Poortvliet, T.C.W., Stal, L.J., Boschker, H.T.S., Tracing carbon flow from microphytobenthos to major bacterial groups in an intertidal marine sediment by using an in situ 13C pulse-chase method. Limnol. Oceanogr., 59, 4, 1275–1287, 2014. [7.77] Molino, P.J. and Wetherbee, R., The biology of biofouling diatoms and their role in the development of microbial slimes. Biofouling, 24, 5, 365–379, 2008. [7.78] Morin, S., Lambert, A.S., Artigas, J., Coquery, M., Pesce, S., Diatom immigration drives biofilm recovery after chronic copper exposure. Freshwater Biol., 57, 8, 1658–1666, 2012. [7.79] Mouget, J.L., Perkins, R., Consalvey, M., Lefebvre, S., Migration or photoacclimation to prevent high irradiance and UV-B damage in marine microphytobenthic communities. Aquat. Microb. Ecol., 52, 3, 223–232, 2008. [7.80] Nakov, T., Beaulieu, J.M., Alverson, A.J., Accelerated diversification is related to life history and locomotion in a hyperdiverse lineage of microbial eukaryotes (Diatoms, Bacillariophyta). New Phytol., 219, 1, 462–473, 2018. [7.81] Nultsch, W., Phototactic and photokinetic action spectra of the diatom Nitzschia communis. Photochem. Photobiol., 14, 705–712, 1971. [7.82] Nultsch, W. and Häder, D.-P., Photomovement in motile microorganisms - II. Photochem. Photobiol., 47, 6, 837–869, 1988. [7.83] Orvain, F. and Sauriau, P.G., Environmental and behavioural factors affecting activity in the intertidal gastropod Hydrobia ulvae. J. Exp. Mar. Biol. Ecol., 272, 2, 191–216, 2002. [7.84] Oxborough, K., Hanlon, A.R.M., Underwood, G.J.C., Baker, N.R., In vivo estimation of the photosystem II photochemical efficiency of individual microphytobenthic cells using highresolution imaging of chlorophyll a fluorescence. Limnol. Oceanogr., 45, 6, 1420–1425, 2000. [7.85] Palmer, J.D. and Round, F.E., Persistent, vertical-migration rhythms in benthic microflora. I. The effect of light and temperature on the rhythmic behaviour of Euglena obtusa. J. Mar. Biol. Assoc. U. K., 45, 3, 567–582, 1965. [7.86] Palmer, J.D. and Round, F.E., Persistent, vertical-migration rhythms in benthic microflora. VI. The tidal and diurnal nature of the rhythm in the diatom Hantzschia virgata. Biol. Bull., 132, 1, 44–55, 1967.

Mechanisms, Control and Adaptive Value  181 [7.87] Passarelli, C., Olivier, F., Paterson, D.M., Meziane, T., Hubas, C., Organisms as cooperative ecosystem engineers in intertidal flats. J. Sea Res., 92, 8, 92–101, 2014. [7.88] Paterson, D.M., The migratory behaviour of diatom assemblages in a laboratory tidal micro-ecosystem examined by low temperature scanning electron microscopy. Diatom Res., 1, 2, 227–239, 1986. [7.89] Paterson, D.M., Short-term changes in the erodibility of intertidal cohesive sediments related to the migratory behavior of epipelic diatoms. Limnol. Oceanogr., 34, 1, 223–234, 1989. [7.90] Paterson, D.M., Wiltshire, K.H., Miles, A., Blackburn, J., Davidson, I., Yates, M.G., Eastwood, J.A., Microbiological mediation of spectral reflectance from intertidal cohesive sediments. Limnol. Oceanogr., 43, 6, 1207–1221, 1998. [7.91] Peperzak, L., Colijn, F., Koeman, R., Gieskes, W.W.C., Joordens, J.C.A., Phytoplankton sinking rates in the Rhine region of freshwater influence. J. Plankton Res., 25, 4, 365–383, 2003. [7.92] Perkins, E.J.J., The Diurnal rhythm of the littoral diatoms of the River Eden Estuary, Fife. J. Ecol., 48, 2, 725–728, 1960. [7.93] Perkins, R.G., Lavaud, J., Serôdio, J., Mouget, J.L., Cartaxana, P., Rosa, P., Jesus, B.M., Vertical cell movement is a primary response of intertidal benthic biofilms to increasing light dose. Mar. Ecol. Prog. Ser., 416, 93–103, 2010. [7.94] Perkins, R.G., Oxborough, K., Hanlon, A.R.M., Underwood, G.J.C., Baker, N.R., Can chlorophyll fluorescence be used to estimate the rate of photosynthetic electron transport within microphytobenthic biofilms? Mar. Ecol. Prog. Ser., 228, 47–56, 2002. [7.95] Perkins, R.G., Underwood, G.J.C., Brotas, V., Snow, G.C., Jesus, B., Ribeiro, L., Responses of microphytobenthos to light: primary production and carbohydrate allocation over an emersion period. Mar. Ecol. Prog. Ser., 223, 101–112, 2001. [7.96] Peterson, C.G., Mechanisms of Lotic Microalgal Colonization Following Space-Clearing Disturbances Acting at Different Spatial Scales. Oikos, 77, 3, 417–435, 1996. [7.97] Pickett-Heaps, J.D., Hill, D., Blaze, K., Active gliding motility in an araphid marine diatom, Ardissonea (formely Synedra) crystallina. J. Phycol., 27, 6, 718–725, 1991. [7.98] Pickett-Heaps, J.D., Hill, D.R.A., Wetherbee, R., Cellular movement in the centric diatom Odontella sinenis. J. Phycol., 22, 3, 334–339, 1986. [7.99] Pinckney, J. and Zingmark, R.G., Effects of tidal stage and sun angles on intertidal benthic microalgal productivity. Mar. Ecol. Prog. Ser., 76, 1, 81–89, 1991. [7.100] Poulsen, N.C., Spector, I., Spurck, T.P., Schultz, T.F., Wetherbee, R., Diatom gliding is the result of an actin-myosin motility system. Cell Motil. Cytoskeleton, 44, 1, 23–33, 1999. [7.101] Raven, J.A. and Waite, A.M., The evolution of silicification in diatoms: Inescapable sinking and sinking as escape? New Phytol., 162, 1, 45–61, 2004. [7.102] Richardson, T.L., Ciotti, A.M., Cullen, J.J., Villareal, T.A., Were, H.P., Physiological and optical properties of Rhizosolenia formosa (Bacillariophyceae) in the context of open-ocean vertical migration. J. Phycol., 32, 5, 741–757, 1996. [7.103] Round, F.E., The epipsammon, a relatively unknown freshwater algal association. Br. Phycol. Bull., 2, 6, 456–462, 1965. [7.104] Round, F.E., Benthic marine diatoms, in: Oceanography and Marine Biology: An Annual Review, vol. 9, pp. 83–139, 1971. [7.105] Round, F.E., Occurence and rhythmic behaviour of Tropidoneis lepidoptera in the epipelon of Barnstable Harbor, Massachusetts, USA. Mar. Biol., 54, 3, 215–217, 1979. [7.106] Round, F.E., Crawford, R.M., Mann, D., The Diatoms - Biology & Morphology of the genera, Cambridge University Press, Cambridge, 1990.

182  Diatom Gliding Motility [7.107] Round, F.E. and Eaton, J.W., Persistent, vertical-migration rhythms in benthic microflora. III. The rhythm of epipelic algae in a freshwater pond. J. Ecol., 54, 3, 609–615, 1996. [7.108] Round, F.E. and Happey, C.M., Persistent, vertical-migration rhytms in benthic microflora. IV. A diurnal rhythm of the epipelic diatom association in non-tidal flowing water. Br. Phycol. Bull., 2, 6, 463–471, 1965. [7.109] Round, F.E. and Palmer, J.D., Persistent, vertical-migration rhythms in benthic microflora. II. Field and laboratory studies on diatoms from the banks of the River Avon. J. Mar. Biol. Assoc. UK, 46, 1, 191–214, 1966. [7.110] Saburova, M.A. and Polikarpov, I.G., Diatom activity within soft sediments: behavioural and physiological processes. Mar. Ecol. Prog. Ser., 251, 115–126, 2003. [7.111] Sato, S. and Medlin, L.K., Motility of non-raphid diatoms. Diatom Res., 21, 2, 473–477, 2006. [7.112] Sauer, J., Wenderoth, K., Maier, U.G., Rhiel, E., Effects of salinity, light and time on the vertical migration of diatom assemblages. Diatom Res., 17, 1, 189–203, 2002. [7.113] Serôdio, J., Coelho, H., Vieira, S., Cruz, S., Microphytobenthos vertical migratory photoresponse as characterised by light-response curves of surface biomass. Estuar. Coast. Shelf Sci., 68, 3–4, 547–556, 2006. [7.114] Serôdio, J., Ezequiel, J., Barnett, A., Mouget, J., Méléder, V., Laviale, M., Lavaud, J., Efficiency of photoprotection in microphytobenthos: role of vertical migration and the xanthophyll cycle against photoinhibition. Aquat. Microb. Ecol., 67, 2, 161–175, 2012. [7.115] Serôdio, J., Silva, J.M., Catarino, F., Nondestructive tracing of migratory rhythms of intertidal benthic microalgae using in vivo chlorophyll a fluorescence. J. Phycol., 33, 3, 542–553, 1997. [7.116] Serôdio, J., Da Silva, J.M., Catarino, F., Use of in vivo chlorophyll a fluorescence to quantify short-term variations in the productive biomass of intertidal microphytobenthos. Mar. Ecol. Prog. Ser., 218, 45–61, 2001. [7.117] Singler, H.R. and Villareal, T.A., Nitrogen inputs into the euphotic zone by vertically migrating Rhizosolenia mats. J. Plankton Res., 27, 6, 545–556, 2005. [7.118] Smayda, T.J. and Boleyn, B.J., Experimental observations on the flotation of marine diatoms. II. Skeletonema costatum and Rhizosolenia setigera. Limnol. Oceanogr., 11, 1, 18–34, 1966. [7.119] Steele, D.J., Franklin, D.J., Underwood, G.J.C., Protection of cells from salinity stress by extracellular polymeric substances in diatom biofilms. Biofouling, 30, 8, 987–998, 2014. [7.120] Sundbäck, K., Odmark, S., Wulff, A., Nilsson, C., Wängberg, S.-Å., Effects of enhanced UVB radiation on a marine benthic diatom mat. Mar. Biol., 128, 1, 171–179, 1997. [7.121] Svensson, F., Norberg, J., Snoeijs, P., Diatom cell size, coloniality and motility: trade-offs between temperature, salinity and nutrient supply with climate change. PloS One, 9, 10, e109993, 2014. [7.122] Taylor, I.S. and Paterson, D.M., Microspatial variation in carbohydrate concentrations with depth in the upper millimetres of intertidal cohesive sediments. Estuar. Coast. Shelf Sci., 46, 3, 359–370, 1998. [7.123] Umemura, K., Miyabayashi, T., Taira, H., Use of a microchamber for analysis of thermal variation of the gliding phenomenon of single Navicula pavillardii cells. Eur. J. Biophys., 44, 3, 113–119, 2015. [7.124] Underwood, G., Fietz, S., Papadimitriou, S., Thomas, D., Dieckmann, G., Distribution and composition of dissolved extracellular polymeric substances (EPS) in Antarctic sea ice. Mar. Ecol. Prog. Ser., 404, 1–19, 2010. [7.125] Underwood, G.J.C. and Kromkamp, J., Primary production by phytoplankton and microphytobenthos in estuaries. Adv. Ecol. Res., 29, 93–153, 1999.

Mechanisms, Control and Adaptive Value  183 [7.126] Underwood, G.J.C., Nilsson, C., Sundbäck, K., Wulff, A., Short-term effects of UVB radiation on chlorophyll fluorescence, biomass, pigments, and carbohydrate fractions in a benthic diatom mat. J. Phycol., 35, 4, 656–666, 1999. [7.127] Underwood, G.J.C., Perkins, R.G., Consalvey, M.C., Hanlon, A.R.M., Oxborough, K., Baker, N.R., Paterson, D.M., Patterns in microphytobenthic primary productivity: Species-specific variation in migratory rhythms and photosynthetic efficiency in mixed-species biofilms. Limnol. Oceanogr., 50, 3, 755–767, 2005. [7.128] Utkin, A.B., Vieira, S., Marques da Silva, J., Lavrov, A., Leite, E., Cartaxana, P., Compact low-cost detector for in vivo assessment of microphytobenthos using laser induced fluorescence. Opt. Spectrosc., 114, 3, 471–477, 2013. [7.129] Vanelslander, B., Paul, C., Grueneberg, J., Prince, E.K., Gillard, J., Sabbe, K., Vyverman, W., Daily bursts of biogenic cyanogen bromide (BrCN) control biofilm formation around a marine benthic diatom. Proc. Natl. Acad. Sci., 109, 7, 2412–2417, 2012. [7.130] Villareal, T.A., Positive buoyancy in the oceanic diatom Rhizosolenia debyana H. Peragallo. Deep Sea Res. Part A. Oceanogr. Res. Pap., 35, 6, 1037–1045, 1988. [7.131] Villareal, T.A. and Lipschultz, F., Internal nitrate concentrations in single cells of large phytoplankton from the Sargasso Sea. J. Phycol., 31, 5, 689–696, 1995. [7.132]  Villareal, T.A., McKay, R.M.L., Al-Rshaidat, M.M.D., Boyanapalli, R., Sherrell, R.M., Compositional and fluorescence characteristics of the giant diatom Ethmodiscus along a 3000 km transect (28°N) in the central North Pacific gyre. Deep-Sea Res. Part I: Oceanogr. Res. Pap., 54, 8, 1273–1288, 2007. [7.133] Villareal, T.A., Pilskaln, C., Brzezinski, M., Lipschultz, F., Dennett, M., Gardner, G.B., Upward transport of oceanic nitrate by migrating diatom mats. Nature, 397, 6718, 423–425, 1999. [7.134] Villareal, T.A., Pilskaln, C.H., Montoya, J.P., Dennett, M., Upward nitrate transport by phytoplankton in oceanic waters: balancing nutrient budgets in oligotrophic seas. PeerJ, 2, e302, 2014. [7.135] Villareal, T.A., Woods, S., Moore, J.K., Culver-Rymsza, K., Vertical migration of Rhizosolenia mats and their significance to NO3- fluxes in the North Pacific gyre. J. Plankton Res., 18, 7, 1103–1121, 1996. [7.136] Waring, J., Baker, N.R., Underwood, G.J.C., Responses of estuarine intertidal microphytobenthic algal assemblages to enhanced ultraviolet B radiation. Global Change Biol., 13, 7, 1398–1413, 2007. [7.137] Webster, D.R., Cooksey, K.E., Rubin, R.W., An investigation of the involvement of cytoskeletal structures and secretion in gliding motility of the marine diatom, Amphora coffeaeformis. Cell Motil., 5, 2, 103–122, 1985. [7.138] Wenderoth, K., Action spectrum of step-down reaction in diatoms, IWF (Göttingen), Film. Germany, 1984. [7.139] Wenderoth, K., Marquardt, J., Rhiel, E., The big trail: Many migrate at the expense of a few. Diatom Res., 19, 1, 115–122, 2004. [7.140] Wenderoth, K. and Rhiel, E., Influence of light quality and gassing on the vertical migration of diatoms inhabiting the Wadden Sea. Helgol. Mar. Res., 58, 3, 211–215, 2004. [7.141] Wetherbee, R., Lind, J.L., Burke, J., Quatrano, R.S., The first kiss: Establishment and control of initial adhesion by raphid diatoms. J. Phycol., 34, 1, 9–15, 1998. [7.142] Wilde, A. and Mullineaux, C.W., Light-controlled motility in prokaryotes and the problem of directional light perception. FEMS Microbiol. Rev., 41, 6, 900–922, 2017. [7.143] Yallop, M.L., De Winder, B., Paterson, D.M., Stal, L.J., Comparative structure, primary production and biogenic stabilization of cohesive and non-cohesive marine sediments inhabited by microphytobenthos. Estuar. Coast. Shelf Sci., 39, 6, 565–582, 1994.

8 Motility in the Diatom Genus Eunotia Ehrenb. Paula C. Furey

*

Department of Biology, St. Catherine University, St. Paul, Minnesota, USA

Abstract

Motility in diatoms, largely focused on motile diatoms with complex raphe systems, remains understudied for diatoms with short raphe systems, like those in the genus Eunotia. Examination of motility in this weakly or slightly motile genus may provide unique insight into motility in diatoms overall, especially for raphid diatoms. Coverage of historical and current accounts of motility in a handful of Eunotia species reveals a variety of movement types that allow cells to move forward, pivot, and reorient to ventral-girdle side down where raphe ends can connect with the substratum. Variations in overall valve morphology, especially with respect to raphe characteristics and the presence of rimoportulae, likely drive patterns in and flexibility of movement types in this genus. Motility in Eunotia may expand their scope for resource tracking and habitat selection. Weak motility in Eunotia may suggest that this trait is in either an early or a degenerate stage of evolution, so further focus on eunotioid diatoms might broaden our understanding of motility in diatoms more generally. Keywords:  Ecology, Eunotia, eunotioid, morphology, motility, rimoportula, raphid diatom, substratum-associated microbiota

8.1 Introduction In the context of diatom motility, the genus Eunotia Ehrenb. is of particular interest due to the wide variation in growth form, overall valve morphology (Figures 8.1–8.11, 8.13–8.14, 8.16, 8.18, 8.20), location and shape of the raphe (Figures 8.1–8.11, 8.14, 8.16–8.20), and the number and location of rimoportulae (labiate processes; tubular structures through the valves of some diatoms [8.21]) (Figures 8.15, 8.21, 8.22). Features that typically characterize the genus Eunotia include, in part, asymmetry about the apical axis, uniseriate striae, helictoglossae (terminal nodules; internal, distal termination of the raphe [8.21]) located on the mantle, the presence of one or more rimoportula, and the presence of a short raphe system [8.66]. In contrast with most other raphe-bearing diatoms, the raphe in Eunotia typically lies to the side of the valve with the raphe slit first starting on the valve mantle then curving to varying degrees onto the valve face [8.66] [8.78]. These characteristics suggest Eunotia (and other taxa in the Eunotiaceae) form a basal lineage within the raphid diatoms * Email: [email protected] Dr. Paula C. Furey: http://drpcfurey.com/, https://www.researchgate.net/profile/P_Furey, https://scholar.google. com/citations?user=fK27GYwAAAAJ&hl=en&oi=sra

Stanley Cohn, Kalina Manoylov and Richard Gordon (eds.) Diatom Gliding Motility, (185–210) © 2021 Scrivener Publishing LLC

185

186  Diatom Gliding Motility 1

2

3

4

5

6

7

9 8

10

11

12

13

s

E 10 µm

Figures 8.1–8.13  LM images of Eunotia taxa in valve view (Figures 8.1–8.8) and girdle view, ventral-side up (Figures 8.9–8.11) to show variation in morphology and raphe shape and location (arrow in some images). (Figure 8.1 [8.28]) E. bilunaris (Ehrenb.) Souza – raphe recurved almost 180°. (Figure 8.2) E. serra Ehrenb. – raphe on valve face with slight curve toward apex. (Figure 8.3 [8.28]) E. areniverma Furey, Lowe et Johansen – raphe follows margin of apex from ventral to dorsal margin, and up onto dorsal mantle. (Figure 8.4 [8.28]). E. pectinalis var. ventricosa (Ehrenb.) Grunow (E. pectinalis var. ventralis (Ehrenb.) Hust. = synonym). (Figure 8.5) E. bigibba Kütz. – raphe with slight recurve. (Figure 8.6) E. incisa Smith ex. Gregory – raphe (not visible) only present on valve mantle (see Figure 8.17). (Figure 8.7) E. muscicola Krasske – raphe (not visible) on the valve face with slight recurve (see Figure 8.20). (Figure 8.8) Close up of valve apex and raphe of E. bilunaris in Figure 8.1. (Figure 8.9) E. bigibba and (Figure 8.10) E. tetraodon Ehrenb. – ventral-girdle view. Depth of valve only permits part of raphe to be in focus (on one plane) at a time. (Figure 8.11) Unknown valve in ventral-girdle view. Raphe almost all in focus. (Figure 8.12) Epiphytic cells of E. bilunaris (E) standing up on end from the mucilaginous sheath(s) of the cyanobacterium Hapalosiphon Nägeli ex Bornet et Flahault (Image credit R.L. Lowe). (Figure 8.13) SEM image of Eunotia on a bryophyte. For all images except for Figure 8.8: black scale bar = 10 μm. Figure 8.8: white scale bar = 5 μm. (Figures 8.1, 8.3, 8.4 originally published in Furey et al. [8.28] www.schweizerbart.de/journals/bibl_diatom).

Motility in the Diatom Genus Eunotia  187 14

10 µm

15

r h

R 16

2 µm

17

10 µm

R 18

R

5 µm rp

1 µm

19 R

20 R R

5 µm

21

22 r

r

0.5 µm

h

h

2 µm

Figures 8.14–8.22  SEM images of Eunotia taxa to show variation in morphology, along with external and internal raphe (R) shape and location on the valve face and valve mantle, helictoglossa (h), shape, location, and internal expression of the rimoportula (r), and external expression of the rimoportula pore (rp). (Figure 8.14 [8.28]) E. areniverma – raphe follows margin of apex from ventral to dorsal margin, and up onto dorsal mantle. (Figure 8.15 [8.28]) E. areniverma – internal view of apex. Rimoportula located mid apex. (Figure 8.16 [8.28]) E. pectinalis var. ventricosa – external view showing path of raphe from mantle onto valve face with slight recurve. (Figure 8.17 [8.28]) E. incisca – raphe located completely on valve mantle. External expression of rimoportula. (Figure 8.18) E. bigibba – curve of raphe onto valve face with slight recurve. (Figure 8.19) E. serra – raphe on valve face with slight curve toward apex. (Figure 8.20) E. muscicola – curve of raphe onto valve face with a slight recurve and (Figure 8.21) internal view of valve apex with rimoportula located close to the helictoglossa. (Figure 8.22) E. serra – internal view of valve apex with rimoportula located closer to dorsal margin. Scale bars as shown. (Figures 8.14–8.17 originally published in Furey et al. [8.28] www.schweizerbart. de/journals/bibl_diatom).

188  Diatom Gliding Motility [8.38] [8.53] [8.66], though their position could be derived if a more complex raphe system became reduced (discussed by Kociolek [8.44] and Siver and Wolfe [8.69]). Movement in eunotioid diatoms with their short raphe system, typically described as slightly (< 2 µm/sec [8.31]) or weakly motile (2 to 4 µm/sec [8.31]) (see eunotioid taxa in DONA [8.21] – Furey [8.26]), contrasts with that of more motile forms like naviculoid, nitzschioid, or surirelioid diatoms with more extensive raphe systems, typically described as moderately to highly motile. Examination of motility in Eunotia species may provide unique insight into motility in diatoms, especially for diatoms with more complex raphe systems. A search for the terms “motile” or “move” in florae focused on Eunotia (e.g., [8.20] [8.28] [8.48] [8.51]) and in >40 manuscripts with descriptions of Eunotia taxa new to science revealed little to no mention of motility. This chapter focuses on motility in the diatom genus Eunotia, but does not cover cellular or biomechanical details around the mechanisms of movement, as other chapters in this book discuss these aspects at length. • The first section covers accounts of motility in Eunotia, starting with historical reports. • The second section examines valve characteristics (valve morphology in girdle and valve view, raphe shape and location, and internal valve structure) in the context of movement to provide groundwork for future research. • The third section highlights additional research gaps through a discussion of the role of motility in the ecology of this genus in the context of growth form and habitat. • The fourth section briefly considers motility in Eunotia in the context of the evolution of diatoms. • The chapter concludes with a brief summary and suggestions for future research.

8.2 Accounts of Movement in Eunotia Earlier studies of diatom motility focused on taxa with extensive raphe systems like those present in naviculoid and nitzschioid diatoms (e.g., [8.31] [8.37] [8.41]), while motility in diatom taxa with short ­­­­raphe systems, like those in the genus Eunotia, remains understudied. Despite this, historical and recent literature identify a number of movement types for a handful of Eunotia taxa, including some that form chains (Table 8.1). Here I provide an account of movement types reported for various Eunotia taxa. Movement in Eunotia was first studied in two species: E. major (Smith) Rabenh. [8.60] and E. bilunaris (Ehrenb.) Souza ([8.80] – reported as E. lunaris (Ehrenb.) Grunow; see Souza and Moreira-Filho [8.71]; Figure 8.1). These early observations of forward movement in Eunotia, described as “jerky” relative to the smooth gliding mentioned for Pinnularia Ehrenb. and Nitzschia Hassall [8.60], suggest the location and shape of the short raphe system on both the valve face and valve mantle drive patterns in motility despite the limited knowledge of internal and external ultrastructure at the time. Hustedt [8.40] also alluded to challenges to movement presented by a short raphe system in taxa like Eunotia, because only part of the raphe extends to the valve surface.

Motility in the Diatom Genus Eunotia  189 Table 8.1  Summary of movement types observed in various Eunotia taxa. Movement type

Comments on movements

Taxon (citation)

Forward movement – Apical displacement (girdle view)

Series of crawl or jerk-like movements Frustule situated girdle view, ventral side down (dorsal side in view). The two forward raphe branches are in contact with the substratum, while the other end slightly raises up (Figure 8.23). Movement forward in oblique direction in line with direction of raphe branches on opposite ends of opposite valves (e.g., Figure 8.24, arrow B to C with raphe on C end active, then along A to D, with the raphe on the D end active). Can result in oscillation laterally (i.e., up to ± 15°) as diatom movement alternates between raphe branches (Figure 8.25; [8.6])

E. arcus – [8.6] E. bidentula – [8.5] E. bilunaris – [8.5] [8.80] E. major – [8.60] E. pectinalis – [8.5] [8.6] E. praerupta – [8.5] [8.6] E. sp. – [8.30]

Forward movement – Apical displacement (valve view)

Frustule situated valve view

E. bilunaris – [8.6]

Lateral movement (in the direction of the pervalvar axis)

Frustule situated girdle view, ventral side down (dorsal-girdle in view). Rather than the two forward raphe branches of both valves directing movement, raphe branches of one valve contribute to the lateral movement.

E. sp. – [8.30]

Vertical polar pivot – returns cell upright (ventral girdle down)

A vertical, polar pivoting of up to 180° (Figure 8.26) Frustule starts situated in girdle view, dorsal side down (ventral-girdle in view) Frustule rises slowly on apices, rotates through 180° to land in girdle view, ventral side down. Valve apices of opposite valves remain in contact with slide (e.g., Figure 8.24, A-B or C-D ends).

E. major – [8.60] E. pectinalis – [8.5] [8.6]

Valve to girdle pivot

Cells in valve view turn 90° around the longitudinal axis. Two points remained in direct contact with the substratum. Can return a cell to girdle view, ventral side down (with the dorsal side up). Rocking back and forth required before the cells can stand on end.

E. bilunaris – [8.80]

(Continued)

190  Diatom Gliding Motility Table 8.1  Summary of movement types observed in various Eunotia taxa. (Continued) Movement type

Comments on movements

Taxon (citation)

Horizontal polar pivot – Valve

Cell in valve view balances on one end and swings the other end in an arc in the horizontal plane. Cells can follow the pivot in the horizontal plane with a rapid swing back to the point of origin (oscillation back and forth) (Zauer 1950).

E. bilunaris – [8.5] [8.80]

Horizontal polar pivot – Girdle

Cell in girdle view, ventral side down, balances on two apices on the same end of two valves and rotates up to 360° or more (Figure 8.27)

E. bilunaris – [8.5] E. bidentula – [8.5] E. pectinalis – [8.5] [8.6] E. praerupta – [8.5] E. sp. – [8.30]

Transapical rotation (along longitudinal axis)

The longitudinal length of the cell acts like the hinge For example, ­the dorsal-girdle side down returns to the ventral-girdle side down passing through the valve face, or from valve view to girdle view.

E. bilunaris – [8.5] E. bidentula – [8.5] E. major – [8.60] E. pectinalis –­[8.5] [8.6] E. praerupta – [8.5]

Movement around an object

Cells turn onto an edge such that the ventral side of the frustule gains contact with the object and moves around it. Just as either end of the cell successfully makes it past the object, the object is ‘kicked’ back.

E. major – [8.60]

Palmer [8.60] presents some of the earliest observations of movement for Eunotia. A sample rich in a chain-forming species of Eunotia, E. major, cleaned of debris and cultured in a bottle kept in semi-dark conditions, provided material for up to a month for both experimentation and general observation. Observations generally occurred at 200x magnification, with motile cells individually tracked for more than a half hour and occasionally up to two and half hours. As longer chains of frustules (up to 10 cm long) in the culture progressively shortened over time, isolated cells and short chains of cells (2 to 6 cells) exhibited “slow and erratic” movement. Further observation of cells from material dropped onto a slide captured three different types of movement: forward movement (Figures 8.23–8.25), a hinged arc or vertical polar pivot to reorient a cell via the apex back to the ventral-girdle side down (Figure 8.26), and rotation along the longitudinal length of the cell (Table 8.1). Palmer [8.60] also described how cells moved around an object. Forward movement – apical displacement (Figures 8.23–8.24): To move forward, cells of E. major tilted slightly so the anterior valve ends where the raphes are located remained in contact with the slide, and the posterior ends slightly raised (Figure 8.23). One corner would

Motility in the Diatom Genus Eunotia  191 advance forward a little; then the other corner would do the same (e.g., Figure 8.24, movement oriented in the angles created by the B to C arrow, then A to D direction, where the raphe on the C or D valve end becomes active). A “disjointed” or “jerky” movement resulted, in contrast with the smooth movement of other diatoms. It is not clear if both valve apices remain in contact with the surface during these jerk-like movements. Valves observed consistently only moved in one direction, in contrast with other diatoms that readily moved back and forth, such as Nitzschia and Pinnularia. Vertical polar pivot – to return a cell in girdle view from dorsal- to ventral-side down (Figure 8.26): Cells of E. major dropped onto a slide often landed in girdle view, but dorsal side down. To return to a ventral-side down position, cells typically rose onto an end, rotated 180°, and then fell back to the slide. The end of the cell acted like a hinge for the cell to pivot around.

Transapical rotation (along the longitudinal axis) ­– to return a cell in girdle view from dorsal- to ventral-side down: Occasionally the longitudinal length of the cell acted like the hinge such that the flip occurred at right angles to those described above in the vertical polar pivot (e.g., if the valve in Figure 8.26a rotated from side A and C, or side B and D as shown in Figure 8.24).



Movement around objects: Though objects slowed or paused the movement of E. major, the cells eventually turned onto an edge such that the ventral side of the frustule made contact with the object and moved around it. Just as the other end of the cell successfully made it past the object, the object was “kicked” back.

Similarly, Zauer [8.80] observed movement of cells of E. bilunaris in samples collected in a forest reserve near Leningrad (nowadays St. Petersburg, Russia). Cells, observed over the course of several hours (details on magnification not provided), only began movement after dark with a 40-watt bulb placed near the microscope. The periodic addition of water to the slides kept materials moist. Cells, originally found as isolated individuals or as cells arranged radially in groups with the valve up (like “wheels with bent spokes”), became active, especially as the fans of diatoms disassembled. Generally, cells over 50 μm in length participated in these movements. Cells exhibited three types of movement (Table 8.1). Forward movement – apical displacement (Figures 8.23–8.24): Similar to observations by Palmer [8.60], forward motion of E. bilunaris frustules required the cell to balance on its end, and resulted in “jerky” or “jolty” movements forward (e.g., Figure 8.24, movement oriented in the angles created by the B to C arrow, then the A to D direction, where the raphe on valve end C or D becomes active). Disruption of the balance of the cell caused motion to stop.

Valve to girdle pivot: Cell in valve view rotates 90° around the longitudinal axis to return to girdle view ventral side down (with the dorsal side up). Cell would end up on its ends, with two points in direct contact with the substratum.

192  Diatom Gliding Motility Zauer [8.80] described the shift in position as unstable with several attempts of rocking back and forth required before the cell could stand on its end.

Horizontal polar pivot: Zauer [8.80] described how cells of E. bilunaris, balanced on one end, took the frustule ends that were up off the surface and swung them in an arc of up to 150° through the horizontal plane. The cells followed this “jerk” in the horizontal plane with a rapid swing back to the point of origin.

Since these early studies, several newer technologies such as scanning electron microscopy (SEM), high-resolution microscopy, and image-capture and video technologies coupled with more complex software programs, facilitate the study and capture of movement in more detail. Descriptions of similar and additional movements, including schematics, are described in Bertrand [8.4] [8.5] [8.6] for E. bilunaris and other Eunotia species (E. arcus Ehrenb., E. bidentula W. Smith, E. pectinalis (Kütz.) Rabenh. and E. praerupta Ehrenb.), and are video documented by Harbich [8.30] (summarized in Table 8.1). Bertrand [8.6] provides a schematic diagram for forward movement to show how as the cell of E. pectinalis tips up on its forward valve apices it moves forward along the diagonal orientation of the forward raphe branch in action (e.g., Figure 8.24 raphe line B to C represents diagonal line of direction where the raphe on the C end becomes active). Then it oscillates in the other direction to move the cell forward along the contrasting oblique orientation of the forward raphe branch on the other valve apex (e.g., Figure 8.25, arrow represents diagonal line of direction). This forward movement shares similarities with the description provided by Palmer [8.60] and Zauer [8.80]. A recent description of movements in a chain-forming Eunotia accompanied by a set of short videos by Harbich [8.30] video documents this forward movement and other movement patterns. Bertrand [8.6] indicated that E. bilunaris also could move forward on its valve face, though only in short distances of 10 to 30 μm (Table 8.1). In addition to moving forward, cells can move sideways. Harbich [8.30] video captured lateral movement (Table 8.1), in the direction of the pervalvar axis, where the cell seems to “float” on the slide, even though the cell remains in contact with the slide. These and other movements primarily occurred in individual cells over those still in filaments [8.6] [8.30] [8.60]. Different types of pivots reported for different taxa (Table 8.1) can change the orientation of the cell; for example, to move a cell up onto its end (Figure 8.12), to reorient a cell to ­ventral-girdle side down so the raphe connects with the substratum, or to change direction. To return cells to ventral-girdle side down like in E. major [8.60], Bertrand [8.5] [8.6] indicated the 180° vertical polar pivot (Figure 8.26) for E. pectinalis, and transapical movements for E. bilunaris, E. bidentula, E. pectinalis, and E. praerupta [8.5]. Eunotia bilunaris in particular seems to exhibit some additional types of pivots from the valve face, such as the horizontal polar pivot [8.6] or the valve to girdle pivot [8.80]. The horizontal polar pivot with the ventral-valve side down (Figure 8.27), which would help a cell change direction, appears common in Eunotia taxa observed to date. Harbich [8.30] captured this motion in video. These horizontal polar pivots occur relatively fast with rotations of 360° or more in a few seconds (Figure 8.27b; [8.6] [8.30]). Interactions with the slide and coverslip can result in additional variations of the movements, where, for example, cells can move along the coverslip rather than the slide, or cells

Motility in the Diatom Genus Eunotia  193 23 a

b

24

25 a C

A

b

B

D

26 a

b

c 27 a

b A

C

B

D

Figures 8.23–8.27  Schematic representation of some of the movement types for Eunotia. (Figure 8.23) Schematic of forward movement – apical displacement where cells tilt slightly so the anterior ends of the valves remain in contact with the surface and the posterior ends become slightly raised (**schematic modeled after Palmer [8.60] plate vi. fig. 2, and Bertrand [8.6] fig. 1**). (Figure 8.24) Valve in girdle view. (Each raphe branch labeled after Bertrand [8.6] (**see also Harbich [8.30]**). Black arrow following the line of the raphes on B to C apices represents diagonal line of direction, where the raphe on the C end becomes active. (Figure 8.25) Black horizontal arrow represents diagonal line of direction. Bidirectional arrow shows transition between raphes involved in forward motion (**schematic modeled after Bertrand [8.6], fig. 1**). (Figure 8.26) Schematic of a vertical polar pivot which can return a cell in girdle view, dorsal-side down (a) to ventral-side down (b,c) where a cell can then continue forward movement (c) (**schematic modeled after Palmer [8.60], plate vi. fig. 2, and Bertrand ([8.6], fig. 5]**). (Figure 8.27) Schematic of a horizontal, polar pivot (a,b) to show direction of raphe activity (straight arrows, A and B). Note the cell is depicted ventral side up so the raphe branches are visible (rather than dorsal side up as the movement occurs). Curved arrows show direction of rotation. Dot represents the pivot point. (**Modified after images from Harbich [8.30]). See additional schematics in Bertand [8.5].

194  Diatom Gliding Motility can use the coverslip as an additional pivot point [8.5] [8.6]. However, some of these movements may be an artefact of the viewing method [8.32]. Use of the coverslip for movement may not occur in all Eunotia taxa [8.5]. On the other hand, these observed movements may occur where a substratum exhibits a similar structural layout; for example, if two surfaces are positioned in close proximity relative to each other. Overall, Eunotia bilunaris and E. pectinalis seem to show higher diversity of movements relative to the other Eunotia taxa. Despite limited knowledge of the external and internal ultrastructural features of Eunotia at the time of the first studies, the movements by E. major and E. bilunaris described hinted at the importance of cell structure. The next section connects these cell movements with diatom structure and proposes avenues for further investigation.

8.3 Motility in the Context of Valve Structure The movement patterns described in historical and more recent studies demonstrate how Eunotia uses its short raphe system to move forward, change direction, rotate, and flip upright in ways that make this genus quite motile! The varied morphology of Eunotia suggests that movement patterns across species and overall forms could be quite variable. Within a species, the placement of the raphe seems to remain consistent [8.54] [8.55]. Thus, for taxa that exhibit movement, cells across the size diminution series, beyond the initial valve, likely all would be capable of movement, though the variation in valve morphology across the size range could influence the nature, speed, or energetic cost of the movements. The cleaning of diatoms prior to examination of live material (or minimal time spent observing live material) limits understanding of motility in diatoms [8.1], especially for less motile taxa like Eunotia. Despite this, some patterns emerge to connect valve structure with movement. Here I consider motility in the context of aspects of valve morphology, particularly with respect to characteristics of the raphe system.

8.3.1 Motility and Morphological Characteristics in Girdle View Eunotia cells observed under the microscope typically present in girdle view (personal observation). This orientation aligns with the optimal orientation for movement as it places all four raphe branches in contact with the surface (Figures 8.9–8.11, 8.13). Depending on how concave the ventral side of the valve is, more or less of the raphe branches can be in contact with the surface at any one time (Figures 8.9–8.11). This may explain why observations of forward movement describe how the cell tips up more or less on one end (Figure 8.23). Additionally, cells with more concave morphologies might experience less friction during movement than might otherwise be created if the entire ventral surface remained in contact with the surface. The schematic diagram of forward movement with apical displacement in Bertrand ([8.6], fig. 1) shows how the diagonal orientation of the forward raphe branches can result in a ~15° horizontal oscillation as the cell transitions between each active raphe branch in contact with the slide (Figure 8.25). These descriptions of oblique movement that align with the raphe angle (example Figure 8.25a) clarify the origin of the “jerky” motion described by both Palmer [8.60] and Zauer [8.80], and digitally documented by Harbich [8.30]. It also may explain why motility of cells in the samples observed by Palmer [8.60] and by Harbich [8.30] primarily occurred in individual cells over

Motility in the Diatom Genus Eunotia  195 those still in chains, especially long chains. Harbich [8.30] suggests that if all four raphe branches remained in contact with the substratum, then both raphes on each diagonal line could contribute to the forward motion. The alignment of the movement path in Eunotia with the shape of the raphe is similar to other raphid taxa where movement patterns follow the curve of the raphe [8.5] [8.22] [8.66]. Other combinations of active raphe branches result in different movements. For example, action by raphe branches located on the same valve can explain lateral movement in the pervalvar axis (Harbich [8.30]; here Figure 8.24 – raphe branches A and C, or B and D). Similarly, active raphe branches on the same end of the cell by both valves can result in the horizontal polar pivot if one valve “works in apical direction, the other in distal direction” to create the motion ([8.30]; also [8.5]; here Figure 8.27, raphes A and B). Eunotia valves in girdle view show a consistent angular layout of the four raphe branches, where the proximal raphe end is inset farther relative to the longitudinal valve margin compared to that of the distal part of the raphe (closest to the apex) which typically then extends onto the valve face (e.g., Figures 8.9–8.11, 8.16). In an examination of four Eunotia taxa over their life cycles, Mayama [8.56] found remarkably little difference between the maximum and minimum depth of the epitheca across cell size within a taxon. The number of areolae in the region between the juncture of valve face and mantle and the middle of raphe branch on the mantle, and the number of areolae between the middle of raphe branch on the mantle and the mantle margin also stayed consistent. The stability of these characteristics relative to the raphe may relate to the mechanics of motility, such as applies to forward motion. The similarity in layout, which can make identification of different Eunotia species in girdle view challenging at times, also suggests movements such as the forward movement, lateral movement, and the horizontal polar pivot should be possible across many species in this genus. Subtle variations in the length, angle, or curve of the raphe on the ventral girdle might influence technical aspects of movement, such as the degree of choppiness or “jerkiness” as the cell shifts between the raphe branches driving a forward motion or rate of movement (e.g., Figures 8.24–8.25). The general shortness of the raphe branches, combined with the jerky, oblique movements in Eunotia relative to other diatom groups results in slow forward movement. Bertrand [8.4] found E. pectinalis moved at a rate of 1.95 ± 0.31 μm/s. Harbich [8.30] found the typical forward speed of a Eunotia taxon ranged from 1.1 to 2.3 µm/s. This speed is slow relative to diatoms with more complex raphe systems such as Nitzschia and Craticula Grunow (range of 2 to 18–20 µm/s) or Pinnularia (avg. ­­4 to 6 µm/s; range 1 to 12 µm/s) [8.17]. In Eunotia, the length of the raphe present on the girdle relative to valve length varies across the size spectrum within a species­where the raphe length does not change proportionally as valve length increases (e.g., see SEM images of E. macroglossa Furey, Lowe et Johansen [8.27] or E. pectinalis var. ventralis (Ehrenb.) Hust. [8.28]) and between species, i.e., in a longer narrower taxon like E. bilunaris versus a shorter taxon like E. muscicola Krasske (see SEM images in Costa et al. [8.20]). In an analysis of over 110 raphid diatoms, Bertrand [8.5] suggests that ranges in speed are more characteristic of species and that within a species the max speed remains constant regardless of valve length. Between different Eunotia species, there may be patterns in speed, especially for forward movement, based on overall frustule morphology (such as degree of curve to ventral valve) and characteristics of the raphe on the ventral girdle. In Eunotia, the frustule is often wider in girdle view than in valve view ([8.56]; e.g., here: ratio of valve width to girdle width: Figure 8.16: 5.9 to 8.7 based on min. max. averages:

196  Diatom Gliding Motility n = 16 for valve view, n=5 for girdle view; Figure 8.18: 0.8, n = 15 for valve view n=5 for girdle view). The number of siliceous girdle bands present, which varies by species, increases the width in girdle view ([8.56]; e.g., Furey et al. [8.28], plate 20 – fig. 6, plate 27 – fig. 8; Costa et al. [8.20], plate 26 – fig. 2, plate 36 – fig. 7, plate 72 – fig. 5). The number of girdle bands increases over the course of the cell cycle to make the girdle view widest just prior to cell division [8.66]. In amphoroid shaped Eunotia species the angle of deflection between the valves faces becomes more exaggerated over the cell cycle due to the wider dorsal cingulum compared to the ventral cingulum (e.g., E. arcuoides Foged [8.2], E. relicta Ferrari, Wetzel et Ector [8.24]). Changes in shape and mass added by the girdle bands likely complicate motility, especially for pivot movements that require the cell to arc 180 degrees. For example, in the valve to girdle pivot described by Zauer [8.80], the movement required multiple attempts, in part due to the instability of the transition. Bertrand [8.6] described how the vertical, polar pivot by E. pectinalis must occur in the vertical plane because of the cell resting on the large surface area of the girdle. The role of exopolymeric substances that provide different types of adhesion can help stabilize the cell and provide transition points so the cell can overcome some of the inertia created by the mass of the cell [8.6] [8.66]. Given the potential connection between movement and characteristics of the raphe on the ventral girdle and the nature of the girdle bands, future descriptions and documentation of Eunotia taxa should include both light microscope and scanning electron microscope images of the ventral-girdle view of cells. Just as the morphological characteristics of the girdle view of cells shape motility, so do the morphological characteristics of the valve view of cells.

8.3.2 Motility and Morphological Characteristics in Valve View Variations in the location, shape, and degree to which the raphe extends up onto the valve surface, which differ substantially across Eunotia species (e.g., Figures 8.1–8.8, 8.14, 8.16– 8.20), likely influence the more complex movements in Eunotia, such as the ability of a cell to execute a vertical polar pivot, horizontal polar pivot in valve view, or transapical rotation. It is not surprising that now with the ability to examine valve structure with scanning electron microscopes the descriptions of movement, like the horizontal polar pivot in valve view in E. bilunaris, follow the raphe shape where the fissure on the valve face strongly recurves almost to 180° (Figures 8.1, 8.8). The presence of a longer raphe branch on the valve face may allow for the short forward movement via the valve face (versus the girdle) in E. bilunaris relative to other Eunotia taxa observed to date [8.6]. Other taxa with a similar strongly re-curved raphe likely are also capable of similar pivot and forward movements via the valve face. For example, these taxa include members of the E. bilunaris complex, or other long, narrow taxa with a similar raphe pattern, such as in E. eurycephala (Grunow) NörpelSchempp et Lange-Beralot, E. flexuosa (Bréb. ex Kütz.) Kütz., E. gallica Lange-Bertalot et Witkowski, E. latitaenia Kobayasi, Ando, et Nagumo, E. macedonia Lange-Bert., Pavlov et Levkov, E. panda Veselá et Johansen, or E. spatulata Veselá et Johansen (see SEM images in Lange-Bertalot et al. [51], Pavlov and Levkov [8.63], Veselá and Johansen [8.76], and Costa et al. [8.20]). A more pronounced arc or longer forward movement might be possible in taxa like E. weisingii Lange-Bertalot where the recurve of the raphe extends farther along the valve face (see SEM images in Lange-Bertalot et al. [8.51]). Similarly, a less pronounced arc could occur in taxa like E. cantonatii Lange-Bertalot et Tagliaventi, E. fureyae LangeBertalot, or E. mucophila (Lange-Bertalot et Nörpel-Schempp) Lange-Bertalot where the

Motility in the Diatom Genus Eunotia  197 re-curve of the raphe is less than 180° and the raphe length on the valve face is shorter (see SEM images in Lange-Bertalot et al. [8.51], and Costa et al. [8.20]). In contrast, these movement types are likely not possible in taxa like E. incisa Smith ex Gregory (Figure 8.17) where the raphe remains on the valve mantle, in the cells of E. linearis (Carter) Vinšová, Kopalová et Van de Vijver where the raphe slit occurs only on the valve face [8.77], or in taxa like in E. diamantina Wetzel et Costa where the distal raphe endings almost remain on the valve mantle [8.20]. Another movement, the vertical polar pivot, seems to relate to the pattern of the raphe on the valve face. The vertical polar pivot reported for E. pectinalis [8.5] [8.6] and E. major [8.60], begins with the dorsal-girdle side down and ends with the ventral-girdle side down after a 180° rotation (Figure 8.26; [8.6]). This complex movement would seem a challenge because when the dorsal-girdle side is down, the raphe ends on the valve face lose contact with the substratum. Thus, to complete the movement, Bertrand [8.6] suggested that for E. pectinalis, substances emitted from the raphe branch on the valve face could extend and adhere to the substratum to create a traction point to start the pivot and then pull along the raphe to move the cell through the pivot. The hook of the raphe on the valve face then is key for successful execution of movement. Bertrand [8.6] described how cells that fell from the coverslip to the slide could land with more of the apex in contact with the slide to promote the development of the traction point. In the vertical polar pivot, a raphe branch with a curve on the valve face and proximity of the raphe relative to the apex edge and dorsal mantle might make the execution of this movement possible or more successful. The raphe of many Eunotia taxa curves onto the valve face to varying degrees and with varying distances from both the apex and the dorsal margin. Proximity of the raphe to the margin of the valve apex may allow the cell to make contact with the substratum in more situations or shorten the distance for any substances emitted to connect with the substratum. In taxa like E. major and E. areniverma Furey, Lowe, et Johansen (Figures 8.3, 8.14), the raphe branch extends onto the valve face and closely follows the margin of the valve apex up to and even onto the dorsal mantle. Where the raphe extends onto the valve mantle, it should be easier for a valve dorsal-girdle side down to connect with the substratum and the possibility of the occurrence and success of this movement should increase. Eunotia taxa with a similar overall valve outline to E. major tend to show this raphe pattern and would be good candidates to observe for this movement. These include, for example, E. areniverma, E. australomaior Van de Vijver, de Haan et Lange-Bertalot, E. didyma Hust. ex  Zimmermann, E. formica Ehrenb., E. formicina Lange-Bertalot, E. karenae Metzeltin et Lange-Bertalot, E. metamonodon Lange-Bertalot, or E. monodontiforma Lange-Bertalot et Nörpel-Schempp (see SEM images in Furey et al. [8.28], Lange-Bertalot et al. [8.51], and Costa et al. [8.20]). Similarly, this long extension of the raphe fissure along the valve apex is seen in other taxa with different overall valve outlines to E. major, such as in E. camelus Ehrenb., E. islandica Østrup, E. longicamelus Costa, Bicudo, et Wetzel, E. pseudincisa Kulikovskiy et Lange-Bertalot, and E. tecta Krasske (see SEM images in Lange-Bertalot et al. [8.51], Van de Vijver et al. [8.74], Kulikovskiy et al. [8.50], and Costa et al. [8.20]). Bertrand [8.7] suggests that pivot-based movements like the polar pivot and transapical rotation that return cells to their ventral girdle require either a point of fixation or resistance to stop movement of the frustule, such as by a drop of mucilage. A point of weak resistance provided by mucilage, likely the case for Eunotia, would allow the cell to pivot along the curve of the raphe [8.7]. Thus, the variations in curve morphology on the valve face of

198  Diatom Gliding Motility different Eunotia taxa could shape the nature of pivot or the ease with which the movement might begin or be carried out overall. An examination of the ability of Eunotia cells to maneuver or right themselves relative to the nature of the raphe position and curve could provide insight into Eunotia motility, as could an examination of the location of structures that produce extracellular substances, such as the rimoportula.

8.3.3 Motility and the Rimoportula The rimoportulae (labiate processes), structures thought to have been lost from other raphe-bearing diatoms, are present in the order Eunotiales [8.45]. Rimoportulae in Eunotia taxa typically occur as only one per valve, but can occur as two or more, or even none [8.23] [8.34] [8.48] [8.58] [8.66], or can be present on both apices, or vary within a species [8.10]. In Eunotia, the presence of a rimoportula, which can release polysaccharides and other carbon compounds, may facilitate movement [8.21] [8.66], as reported in two representatives of Centrales [8.57] [8.64]. However, some evidence suggests the pore is not a primary location for mucilage production in all diatoms [8.34] [8.36]. If, as Round et al. [8.66] speculate, the rimoportula is involved in the initial phases of attachment if it rapidly releases an unstructured adhesive, then the rimoportula could have a role, for example, in establishing a traction or adhesion point for motions such as the vertical polar pivot in Eunotia. The placement of the pores along the vertical axis varies in terms of their location along the apex, with some located more ventrally (e.g., Figure 8.21), and others centrally (e.g., Figure 8.15) or more dorsally (e.g., Figure 8.22) along the valve apex. In the case of taxa E. incisa, the prominent external expression of the rimoportula could indicate a unique role in motility due to the absence of the raphe branch on the valve face. Prominent external expression of the rimoportula occurs in some other taxa, like E. sudeticiformis Kociolek, You, Stepanek, Lowe et Wang [8.46]. Given these considerations, taxa with a rimoportula present at both apices or > one rimoportula per valve could potentially more easily carry out movements that originate from either apex or could conduct different movements that require more adhesion or a stronger anchor. The role of the rimoportula in motility in Eunotia remains unclear. Overall, a more detailed examination of movement in Eunotia taxa across a broader spectrum of frustule morphologies (i.e., long, skinny, short, broad, etc.), especially in the context of raphe shape, position, and length, will help connect the role of these features to movement patterns in both the valve and girdle positions in diatoms like Eunotia with short raphe systems. The potential and ease of carrying out different movements across Eunotia taxa could have implications for survival and competition for resources under different ecological scenarios.

8.4 Motility and Ecology of Eunotia Eunotia are common in ultraoligotrophic and oligotrophic waterways, often with a preference for acidic or dystrophic habitats [8.51] [8.61] [8.62]. Reported as free-floating, epiphytic, solitary or in filamentous colonies, Eunotia predominantly grow in association with the benthos (especially bryophytes; Figure 8.13) or metaphyton [8.12] [8.28] [8.66], but occasionally occur as plankton [8.44] [8.79]. At the scale of a diatom, motility, even if

Motility in the Diatom Genus Eunotia  199 limited or slow, allows cells to respond to resource needs (e.g., space, light, nutrients) and external pressures (e.g., sheer stress, grazing) in diverse ecological contexts [8.52]. The ability of Eunotia to move in a variety of ways could provide further flexibility for interacting with the dynamic nature of these environments. However, given that Eunotia is weakly or slightly motile rather than moderately or highly motile like other diatom groups, differences in the capacity to move or to carry out different types of movement may influence habitat preferences of taxa or may depend on the growth form. Additionally, environments that lead to or promote deformities in Eunotia cells, especially with respect to the location and shape of the rimoportula and the overall valve morphology (e.g., Furey et al. [8.27]) could disrupt patterns of motility. Eunotia growth forms, traditionally broadly categorized as solitary, attached by mucilaginous pads, or filamentous (in “band-like” colonies) [8.66], can also include branched chains [8.11], stalked forms, branched arborescent colonies [8.1], or stellate or zigzag colonies [8.79]. While Eunotia cells are often solitary, individual cells, for example, can detach from filaments to move to another location [8.30] [8.60]. Palmer [8.60] also observed movement of short chains of Eunotia. However, research that covers the taxonomy and ecology of Eunotia typically includes aspects of growth form and habitat but lacks details around movement. Examination of diatoms after cleaning limits understanding of the various characteristics associated with fresh material, like motility and growth form [8.1]. This makes it difficult to connect motility to aspects of diatom ecology. Here, habitat, growth form, and morphology, with particular consideration of raphe structure, inform a discussion on the ecological considerations of motility in Eunotia.

8.4.1 Substratum-Associated Environments In benthic and other substratum-associated environments, factors like the 3-dimensional nature of periphyton assemblages and substratum characteristics, like stability, complicate interactions with resources and exposure to physical disturbances.  Motility creates options for dealing with some of these complexities in the benthos, a response not available to non-motile taxa. Furthermore, variation in the types and speed of movement may be advantageous, especially where resources are scarce or patchy, as is common in the benthos. Response to Resources – Some raphid diatoms can sense and move in response to phosphate concentrations [8.9] or to a gradient of silica, a key resource for frustules [8.8]. Other diatoms can move toward or away from light of different quality and intensity to improve access to light for photosynthesis and reduce exposure to harmful light levels [8.13] [8.15] [8.16]. With less known for Eunotia, how species respond to various resource needs requires further study, particularly where movement is concerned. In Eunotia, the higher profile of stalked cells (e.g., E. amazonica Almeida et Wetzel [8.1]) and arborescent colonies (e.g., E. rabenhorstiana (Grunow) Hust. [8.1]) might confer a competitive advantage over lower profile algae. In contrast, motile taxa might be able to respond to resource needs in a more flexible manner. Like other diatom taxa, for example, motile Eunotia potentially could move toward and away from resources, or could selectively pivot a cell/s to stand on end and farther into the water column, as shown for E. bilunaris (see plate 11, fig. 82c of E. lunaris by Smith [8.70]; Zauer [8.80], Vanormelingen et al. [8.75, fig. 18]; Figure 8.12). In early

200  Diatom Gliding Motility experiments by Palmer [8.60], conditions of bright light (over dim light) promoted movement of E. major cells, though not in any specific direction on the slide. While many other raphid diatoms can change direction of movement in a back-forth orientation, Eunotia taxa would appear to need to pivot or swivel to change the direction of movement, which may take more time or require more lateral space. It is possible that movements like the horizontal polar pivot could stir water or the substratum to promote nutrient exchange through stagnant boundary layers, though there is no evidence for this to date. Some Eunotia taxa form chains that could extend cells up off the substratum or might alter the flow of water through the chain of cells to potentially reduce competition with more adnate taxa or aide in resource acquisition. In chain-forming Eunotia (e.g., E. intricans Metzeltin et Lange-Bertalot [8.1], E. leonardii Taylor et Cocquyt [8.73], E. major [8.60], E. meridiana Metzeltin et Lange-Bertalot [8.58]) movement applies in a couple of ways—one to cells that leave the chain, and two to reconnect cells in a broken chain. In a chain-forming Eunotia, Harbich [8.30] video-captured cells leaving both the end and middle of the chain. These released cells may facilitate dispersion. Where cells left the middle of the chain, the chain fragments on either side used lateral movement to re-form the filament, and thus maintained a longer chain. Lateral movement may be common in other chain-forming Eunotia taxa. Some raphid diatoms can sense factors like light, and reorient cell position (i.e., tip vs. middle) to align chloroplast-rich regions of the cell to the area of ideal light [8.14]. It is not clear if Eunotia cells can exhibit this level of finesse or sensitivity in their behavior in response to a resource. However, some Eunotia taxa can rotate and pivot in different ways, which might allow a cell to position itself relative to a resource if the cell can sense the resource gradient. Though centered on more motile diatom groups, the series of research questions around interactions between light and motility in diatoms proposed by Cohn [8.14] also apply to Eunotia. Different types of substrata that create variation in distances between surfaces could influence access to or response to resources where motility is concerned. For example, for Eunotia taxa living in association with bryophytes, different types of movement may confer advantages based on the arrangement of the leaves, e.g., more exposed nature of leafy liverworts versus whorled mosses with a variety of protective crevices and closer arrangement of surfaces (see SEM images in Knapp and Lowe [8.43]). Bertrand [8.6] described how a Eunotia cell hanging from a coverslip drops with gravity, falls back to girdle view, dorsal side down, then pivots to girdle view, ventral side down, where the cell could then move forward. While this movement occurred in an artificial setting, cells could potentially use this movement, for example, to navigate between close leaves of a whorled moss or in response to shifts in moss orientation with current or other forces. Similarly, the ability to move around an object (described in Palmer [8.60]) could be useful in crevices where materials might accumulate or on any surface where objects, such as other epiphytes or organic material, could impede access to resources. Substratum Stability – Many benthic environments contain moving particles like sand and silt that shift with current or other disturbances such as from wind or biota. At the scale of a diatom, particle movement can crush, bury, or dislodge cells. Thus, both adhesion and motility are essential for diatom survival as diatoms can either adhere to or move to protect themselves in microenvironments on a particle, or migrate to the surface.

Motility in the Diatom Genus Eunotia  201 Adhesion – Because the raphe plays such a key role in adhesion and movement [8.33] [8.37], diatoms with complex raphe systems typically dominate epipsammon (sandassociated floras) and epipelon (silt-associated floras). However, adhesion ability varies with species and by environment, where, for example, epipsammic diatoms move less than epipelic diatoms [8.31]. To help remain adhered to the substratum, epipsammic diatoms, like Gyrosigma Hassall and Nitzschia linearis (Agardh) W. Sm. can withstand shear stress forces >100 dyne/cm2 [8.33]. Other, non-epipsammic diatoms that can withstand 10 dyne/ cm2 are long and thin which provides a long raphe relative to the area of the diatom. For diatoms like Eunotia, with short raphe systems, lower adhesion capacity, and slower motility, these environments might be less hospitable. Indeed, Eunotia prefer slow moving/stagnant waters or substratum with protected microhabitats, such as bryophytes [8.28] [8.43] [8.61] [8.62], though cells might still need to move above depositing material to reach light. In epipsammic environments collected from urban streams in Brazil, the only Eunotia taxon included in the characterization of upstream, relatively less polluted, sites was E. bilunaris, a longer slender taxon, with a curved raphe on the valve face (Figures 8.1, 8.8) [8.3]. In this case, the long slender morphology, diversity of types of movements (Table 8.1), and raphe shape and location, likely aided in the tolerance of the less stable substratum. The short raphe system of Eunotia, which provides less adhesion capacity, also reduces speed of movement, and thus the ability to migrate. Migration – Motile taxa, typically biraphid forms, utilize their more complex raphe systems to carry out vertical and horizontal migration [8.18] [8.35] [8.67]. Without an extensive raphe system to promote quick movement, migration distances in sediments would be substantially restricted [8.35]. The slow rate of movement in Eunotia may limit growth to more stable environments. Debris (i.e., organic matter, sand, silt, etc.) can impede diatom movement. While some diatoms like Amphora ovalis (Kütz.) Kütz., readily exert strong forces to easily move particles [8.33], others, like Eunotia do not. For Eunotia to navigate particles, it may be easier to change direction. The ability for Eunotia cells to move around objects [8.60], even if relatively slow, might be of some benefit in the benthos. For example, movements that change direction, such as the horizontal polar pivot common among Eunotia taxa observed to date (Table 8.1), might help Eunotia cells navigate the terrain of the benthos. For Eunotia, where the raphe branches on the ventral-girdle side drive much of the forward and other movements, disturbance to cell orientation in these unstable environments could be problematic. The ventral-girdle orientation allows all four raphe branches to connect with the substratum to maximize movement potential. Thus, Eunotia taxa that can carry out movements like the transapical pivot or the vertical polar pivot that reorient cells to this position may be at an advantage over those that cannot (Table 8.1).

8.4.2 Planktonic Environments Less common than benthic forms of Eunotia, planktonic Eunotia [8.44] [8.68] [8.79] often produce narrow and weakly silicified valves [8.47] and form colonies [8.79] to remain in the photic zone. Thus, movement requirements for planktonic Eunotia likely remain minimal or pertain more to frustule-to-frustule contact, over movement relative to a substratum or object, or for reorientation of a cell to an upright position. Indeed, descriptions of

202  Diatom Gliding Motility planktonic taxa typically report a very short raphe system, such as the extremely short raphe located only on the valve mantle documented for E. zasuminensis (Cabejszekówna) Körner [8.79] or the short raphe slits of E. croatana Siver, Hamilton et Morales and E. pseudofragilaria Siver, Hamilton et Morales [8.68]. A number of planktonic Eunotia species described from the Brazilian Amazon form colonies with both zigzag and radial patterns where apices of the cells connect [8.79]. The short raphe present in these taxa either remains primarily on the valve mantle (e.g., E. tukanorum Wetzel et Bicudo; E. waimiriorum Wetzel) or turned up slightly onto the valve face (e.g., E. loboi Wetzel et Ector or E. gomesii Wetzel et Ector) [8.79]. The short raphe only minimally present on the valve face of E. enigmatica Costa et Wetzel also aligns with its planktonic habitat [8.19]. None of these descriptions of planktonic Eunotia taxa mentions movement, but the presence of the raphe indicates movement could be possible. Though movement likely contributes less to the ecology of planktonic Eunotia compared to benthic forms, the inclusion of observations on motility or lack of motility during different stages of planktonic taxa in future studies will further increase understanding of motility in this genus. For example, how or does movement impact colony formation or does movement play a role in reproduction or re-entrainment into the water column for cells that settle on the bottom of lentic habitats? Many questions remain around the connections between motility and ecology of Eunotia species. Given the extensive diversity of morphology, especially for the raphe, and the variety of movements observed to date in a small number of Eunotia species, future studies on motility in Eunotia species will provide insight not only into diatom ecology, but also into the evolution of diatoms.

8.5 Motility and Diatom Evolution Deeper knowledge of motility in eunotioid diatoms will contribute to discussions on the evolutionary relationships between diatom taxa. Hypotheses on the evolution of the raphe system in pennate diatoms focus, in part, on the position of groups with short raphe systems, like eunotioid diatoms. Similar to araphid diatoms, most Eunotiophycideae possess rimoportulae [8.66], while raphid diatoms do not [8.44]. One hypothesis proposes that Eunotia form a basal lineage to the raphid diatoms where the presence of rimoportulae represents an intermediate transition point between araphid and raphid diatoms [8.38] [8.39] [8.53] [8.66]. Another proposes a derived position for Eunotia where degeneration of more developed raphe systems leads to the reduced-raphe system (discussed by Kociolek [8.44] and Siver and Wolfe [8.69]). With the competing hypotheses around the position of Eunotia relative to other diatoms, unpacking connections between movement (including adhesion) and characteristics of the raphe and the production of extra polymeric substances exuded from various pores, like the rimoportula, may be informative. If Eunotia sit at a basal position relative to pennate diatoms, then patterns of movement types and mechanisms should emerge from an examination of motility in different Eunotia groups with similar raphe shapes and locations on the valve, or with rimoportula numbers and position. Across the Eunotia, the substantial variation in morphological characteristics provides a myriad of combinations of raphe and rimoportula traits for exploration, some of which should provide better motility than others do. Genetic explorations of the more and less motile Eunotia forms relative to other pennate diatoms and representatives of Centrales

Motility in the Diatom Genus Eunotia  203 (especially those with demonstrated motility) should also prove to be informative. Nakov et al. [8.59] discuss how motility promotes habitat diversification and colonization of new environments, and leads to increased sexual reproduction to improve genetic diversity. Motility in diatoms likely promoted an earlier transition from oogamy to anisogamy, to support outcrossing. Klebahn [8.42] also reported the use of motility for the purpose of conjugation of protoplasts into a single auxospore. Harper [8.32] further highlighted how, as Fritsch ([8.25], p. 622) suggested, movement would allow auxospore formation in individuals where genetic recombination would yield a genetic advantage over fission to produce connected daughter cells. Rather than meeting by chance, motility in Eunotia allows active pairing of cells in reproduction, such as reported for E. arcus [8.29] and E. bilunaris [8.75]. In contrast, if the raphe system in Eunotia resulted from loss over time, then variation in motility characteristics could be relevant to this transition. Reports of raphe loss are common in the evolution of diatoms (e.g., [8.49]). Any number of scenarios could alter motility criteria to lead to variation in raphe characteristics and movement. For example, diatoms that live in unstable benthic environments, like sediment-rich habitats, rely on their extensive raphe systems to move to keep from being crushed or buried, or on strong adhesion to remain in protective crevices of particles [8.65]. If taxa are located in other niches associated with the benthos but not directly in areas of high instability, then raphe properties designed for speed or connection to particles might be less critical. Shorter raphes or those designed for agility or maneuverability (movement flexibility in small spaces), such as pivoting, could be more useful, especially if less energy needs to be devoted to speed. On the one hand, this is less compelling, as many raphid diatoms exhibit a diverse suite of movements, such as medial pivots or transapical rotations [8.5] [8.32], appropriate to these habitats. Though not common in the Eunotia, taxa present in planktonic environments with their especially short raphes suggest a reduction of the raphe for taxa in these environments. Overall, there is a clear need for more research on the role motility in diatom evolution, with special consideration of eunotioid taxa.

8.6 Conclusion and Future Directions Here, key observations and findings about Eunotia motility are discussed to fuel curiosity, and questions are presented to inspire future explorations. Morphological Considerations: Overall, a more detailed examination of motility in Eunotia taxa across a broader spectrum of frustule morphologies (i.e., long, skinny, short, broad, etc.), especially in the context of raphe shape, position, and length will help connect the role of these features to movement patterns in both the valve and girdle positions. What movements, if any, are universal across Eunotia taxa? How do subtle variations in length, angle, or curve of the raphe on the ventral girdle influence technical aspects of movement or movement capacity in general? How does the shape of the raphe on the valve face (e.g., distance from apex margin, distance from the dorsal margin, degree of raphe present, extent of the terminal raphe fissure, degree and length of recurve, overall shape, etc.) confer a particular type of movement or lead to diversity of movement types? What role does the position, number, or presence on one or both apices of rimoportulae play in motility, including the

204  Diatom Gliding Motility use and production of exopolymeric substances used in motility (i.e., different types of pivots or establishing contact with the substratum)? What morphological characteristics promote adhesion used in different movements in Eunotia? Species Observations and Taxon Descriptions: Where possible, observations of Eunotia, including descriptions of new taxa, should include an examination of live material to increase  understanding around growth form and to capture movement if induced. This information will help determine consistencies and patterns in movement relative to morphology, raphe characteristics, growth form, and habitat. Light and scanning microscope examination across the size series should include valve and girdle views (especially ­ventral-girdle views) to capture raphe characteristics, along with internal and external expression of the rimoportula. To contribute to understanding of diatom motility overall, it would be particularly useful to capture video files of movement in various diatom taxa, and deposit them in an accessible location to build a library of diatom motility. Ecological Considerations: Observation of motility in live Eunotia under various ecological settings will help advance current approaches that use ecological functions and traits of algae to assess and characterize environments [8.72]. How do ecological preferences of Eunotia link to motility capacities of various taxa or morphotypes? Conversely, under different ecological settings, do differences in movement types (e.g., various pivots, rotations) between species or morphotypes provide competitive advantages for responses to resources? How does motility in Eunotia, including movement diversity, influence response patterns to various biological, chemical, and physical parameters? Phylogenetic Considerations: Continued explorations of the motility in Eunotia relative to other diatoms will contribute to current questions on the relationships between different diatom groups. Increased knowledge around motility in Eunotia, such as the types and sources of movement and exopolymeric substances, or genetic characterization of more Eunotia taxa with different motility characteristics, may provide evidence to help sort out current hypotheses or lead to the generation of new ones. Explorations of the broad spectrum of questions around motility in Eunotia discussed above will enrich knowledge of diatom motility. Furthermore, the diversity in morphology within and between Eunotia species brings opportunities for novel insight into diatom motility.

Acknowledgements Special thanks to Rex Lowe, Kalina Manoylov, and anonymous reviewers for constructive feedback that improved chapter content. Thank you to Bowling Green State University, Ohio, USA, and Macalester College, St. Paul, Minnesota, USA, for access to and use of their Scanning Electron Microscope facilities. A National Science Foundation Major Research Instrumentation Grant (MRI-1827514) supported collection of some of the light microscope images taken at St. Catherine University.

Motility in the Diatom Genus Eunotia  205

References [8.1] de Almeida, F.F., Santos-Silva, E.N., Ector, L., Wetzel, C.E., Eunotia amazonica sp. nov. (Bacillariophyta), a common stalk-forming species from the Rio Negro basin (Brazilian Amazon). Eur. J. Phycol., 53, 2, 166–179, 2018, https://doi.org/10.1080/09670262.2017.1402372. [8.2] Beals, J. and Potapova, M., Type material of the diatom Eunotia arcuoides Foged. Proc. Acad. Nat. Sci. Philadelphia, 162, 1, 25–32, 2013, https://doi.org/10.1635/053.162.0102. [8.3] Bere, T. and Tundisi, J.G., Epipsammic diatoms in streams influenced by urban pollution. Braz. J. Biol., 70, 921–930, 2010, http://dx.doi.org/10.1590/S1519-69842010000500002. [8.4] Bertrand, J., La vitesse de deplacement des diatomées. Diatom Res., 5, 2, 223–239, 1990, https://doi.org/10.1080/0269249X.1990.9705115. [8.5] Bertrand, J., Mouvements des diatomées. II - Synthèse des mouvements. Cryptogam.-Algol., 13, 1, 49–71, 1992. [8.6] Bertrand, J., Mouvements des diatomées. III - Le pivotement polaire vertical de Eunotia pectinalis (Kütz.) Rab. Essai de quantification des forces. Cryptogam.-Algol., 14, 157–172, 1993. [8.7] Bertrand, J., Mouvements des diatomées VIII: Synthèse et Hypothèse. Diatom Res., 23, 1, 19–29, 2008, https://doi.org/10.1080/0269249X.2008.9705734. [8.8] Bondoc, K.G.V., Heuschele, J., Gillard, J., Vyverman, W., Pohnert, G., Selective silicate-­ directed motility in diatoms. Nat. Commun., 7, 10540, 2016, https://doi.org/10.1038/ ncomms10540. [8.9] Bondoc, K.G.V., Lembke, C., Vyverman, C.W., Pohnert, G., Selective chemoattraction of the benthic diatom Seminavis robusta to phosphate but not to inorganic nitrogen sources contributes to biofilm structuring. Microbiol. Open, 8, e00694, 2018. [8.10] Brant, L.A. and Furey, P.C., Morphological variation in Eunotia serra, with a focus on the rimoportula. Diatom Res., 26, 2, 221–226, 2011, https://doi.org/10.1080/02692 49X.2011.601326. [8.11] Burliga, A.L., Torgan, L.C., Beaumord, A.C., Eunotia ariengae sp. nov., an epilithic diatom from Brazilian Amazon. Diatom Res., 22, 2, 247–253, 2007, https://doi.org/10.1080/02692 49X.2007.9705714. [8.12] Cantonati, M. and Lange-Bertalot, H., Diatom monitors of close-to-pristine, very-low alkalinity habitats: three new Eunotia species from springs in Nature Parks of the south-eastern Alps. J. Limnol., 70, 2, 209–221, 2011, https://doi.org/10.4081/jlimnol.2011.209. [8.13] Cartaxana, P. and Serodio, J., Inhibiting diatom motility: a new tool for the study of the photophysiology of intertidal microphytobenthic biofilms. Limnol. Oceanogr. Methods, 6, 466–476, 2008, https://doi.org/10.4319/lom.2008.6.466. [8.14] Cohn, S.A., Photo-stimulated effects on diatom motility, in: Photomovement, Comprehensive Series in Photosciences, vol. 1, D.-P. Häder and A.M. Breure (Eds.), pp. 375–401, Elsevier, Amsterdam, 2001, https://doi.org/10.1016/S1568-461X(01)80017-X. [8.15] Cohn, S.A., Dunbar, S., Ragland, R., Schulze, J., Suchar, A., Weiss, J., Wolske, A., Analysis of light quality and assemblage composition on diatom motility and accumulation rate. Diatom Res., 31, 3, 173–184, 2016, https://doi.org/10.1080/0269249X.2016.1193058. [8.16] Cohn, S.A., Halpin, D., Hawley, N., Ismail, A., Kaplan, Z., Kordes, T., Kuhn, J., Macke, Zapata, Y., Comparative analysis of light-stimulated motility responses in three diatom species. Diatom Res., 30, 3, 213–225, 2015, https://doi.org/10.1080/0269249X.2015.1058295. [8.17] Cohn, S.A. and Weitzell Jr., R.E., Ecological considerations of diatom cell motility. I. Characterization of motility and adhesion in four diatom species. J. Phycol., 32, 6, 928–939, 1996, https://doi.org/10.1111/j.0022-3646.1996.00928.x.

206  Diatom Gliding Motility [8.18] Consalvey, M., Paterson, D.M., Underwood, G.J., The ups and downs of life in a benthic biofilm: migration of benthic diatoms. Diatom Res., 19, 2, 181–202, 2004, https://doi.org/10. 1080/0269249X.2004.9705870. [8.19] Costa, L., Wetzel, C., Ector, L., Williams, D., Bicudo, C., Eunotia enigmatica sp. nov., a new planktonic diatom from Brazil and the transfer of Fragilaria braunii Hustedt to the genus Peronia (Bacillariophyceae). Fottea, 17, 103–113, 2017a, https://doi.org/10.5507/fot. 2016.023. [8.20] Costa, L.F., Wetzel, C.E., Lange-Bertalot, H., Ector, L., Bicudo, D.C., Taxonomy and ecology of Eunotia species (Bacillariophyta) in southeastern Brazilian reservoirs. Biblioth. Diatomol., 64, 1–302, 2017b, 108 pls. [8.21] DONA (Diatoms of North America), Eunotioid pages https://diatoms.org/morphology/ eunotioid from website: https://diatoms.org. [8.22] Edgar, L.A. and Pickett-Heaps, J.D., Diatom locomotion, in: Progress in Phycological Research, F.E. Round and D.J. Chapman (Eds.), pp. 47–88, Biopress Ltd., Bristol, 1984. [8.23] Edlund, M.B. and Brant, L.A., Eunotia charliereimeri, a new Eunotia species (Bacillariophyceae) with amphoroid frustule symmetry. Proc. Acad. Nat. Sci. Philadelphia, 160, 47–56, 2010, https://doi.org/10.1635/053.160.0106. [8.24] Ferrari, F., Wetzel, C.E., Ector, L., Bicudo, D.C., Bicudo, C.D.M., A new uncommon epilithic Eunotia (Bacillariophyceae, Eunotiaceae) from the Chapada Diamantina region, Northeast Brazil. Phytotaxa, 164, 3, 161–174, 2014, https://doi.org/10.11646/phytotaxa.164.3.1. [8.25] Fritsch, F.E., The Structure and Reproduction of the Algae, vol. I, pp. 564–651, Cambridge University Press, Cambridge, 1935. [8.26] Furey, P., Eunotia, in: Diatoms of North America, 2010, https://diatoms.org/genera/eunotia. [8.27] Furey, P.C., Lowe, R.L., Johansen, J.R., Teratology in Eunotia taxa in the Great Smoky Mountains National Park and description of Eunotia macroglossa sp. nov. Diatom Res., 24, 2, 273–290, 2009, https://doi.org/10.1080/0269249X.2009.9705802. [8.28] Furey, P.C., Lowe, R.L., Johansen, J.R., Eunotia Ehrenberg (Bacillariophyta) of the Great Smoky Mountains National Park, USA. Biblioth. Diatomol., 56, 1–134, 2011. [8.29] Geitler, L., Kopulation und Formwechsel von Eunotia arcus. Österr. Bot. Z., 98, 292–337, 1951, https://doi.org/10.1007/BF01290407. [8.30] Harbich, T., Examples of complex paths, in: Observations of Diatoms, 2018, https://www. diatoms.de/en/example-of-complex-paths. [8.31] Harper, M.A., Movement and migration of diatoms on sand grains. Br. Phycol. J., 4, 1, 97–103, 1969, https://doi.org/10.1080/00071616900650081. [8.32] Harper, M.A., Movements, in: The Biology of Diatoms, D. Werner (Ed.), pp. 214–249, Blackwell, Oxford, 1977. [8.33] Harper, M.A. and Harper, J.F., Measurements of diatom adhesion and their relationship with movement. Br. Phycol. Bull., 3, 2, 195–207, 1967, https://doi.org/10.1080/00071616700650051. [8.34] Hasle, G.R., The ‘mucilage pore’ of pennate diatoms. Beih. Nova Hedwigia, 45, 167–194, 1973. [8.35] Hay, S.I., Maitland, T.C., Paterson, D.M., The speed of diatom migration through natural and artificial substrata. Diatom Res., 8, 2, 371–384, 1993, https://doi.org/10.1080/02692 49X.1993.9705268. [8.36] Hoagland, K.D., Rosowski, J.R., Gretz, M.R., Roemer, S.C., Diatom extracellular polymeric substances: function, fine structure, chemistry, and physiology. J. Phycol., 29, 537–566, 1993, https://doi.org/10.1111/j.0022-3646.1993.00537.x. [8.37] Hopkins J.T., Drum, R.W., Diatom motility: An explanation and a problem. Br. Phycol. Bull., 3, 1, 63–67, 1966, https://doi.org/10.1080/00071616600650081. [8.38] Hustedt, F., Raphe und Gallerporten der Eunotiodeae. Ber. Dtsch. Bot. Ges., 44, 142–150, 1926.

Motility in the Diatom Genus Eunotia  207 [8.39] Hustedt, F., Neue und wenig bekannte Diatomeen. III. Phylogenetische variationen bei den raphidioiden Diatomeen. Ber. Dtsch. Bot. Ges., 65, 133–144, 1952. [8.40] Hustedt, F., Kieselalgen (Diatomeen), 3rd Edition, W. Verlagsbuchhandlung, (Ed.), Keller and Co., Franckh’sche, Stuttgart, 1965. [8.41] Jackson, D.D., The movements of diatoms and other microscopic plants. J. R. Microsc. Soc, 25, 554–557, 1905, https://doi.org/10.1111/j.1365-2818.1905.tb00221.x. [8.42] Klebahn, H., Beiträge zur kenntnis der auxosporenbildung. I. Rhopalodia gibba. Jahrb. Wiss. Bot., 29, 595–654, 1896. [8.43] Knapp, J. and Lowe, R.L., Spatial distribution of epiphytic diatoms on lotic bryophytes. Southeast. Nat., 8, 305–316, 2009, https://doi.org/10.1656/058.008.0209. [8.44] Kociolek, J.P., Valve ultrastructure of some Eunotiaceae (Bacillariophyceae), with comments on the evolution of the raphe system. Proc. Calif. Acad. Sci., 52, 11–21, 2000. [8.45] Kociolek, J.P., Spaulding, S.A., Lowe, R.L., Bacillariophyceae: The Raphid Diatoms, in: Freshwater Algae of North America, J. Wehr, R. Sheath, J.P. Kociolek (Eds.), pp. 707–770, Academic Press, New York, 2015, https://doi.org/10.1016/b978-0-12-385876-4.00016-5. [8.46] Kociolek, J.P., You, Q.-M., Stepanek, J., Lowe, R.L., Wang, Q.-X., A new Eunotia (Bacillariophyta: Eunotiales) species from Karst formations of southern China. Phytotaxa, 265, 285–293, 2016, https://doi.org/10.11646/phytotaxa.265.3.10. [8.47] Kooistra, W., Forlani, G., De Stefano, M., Adaptations of araphid pennate diatoms to a planktonic existence. Mar. Ecol., 30, 1–15, 2009, https://doi.org/10.1111/j.1439-0485.2008.00262.x. [8.48] Krammer, K. and Lange-Bertalot, H., Bacillariophyceae. 3. Teil: Centrales, Fragilariaceae, Eunotiaceae, in: Süsswasserflora von Mitteleuropa, vol. 2(3), H. Ettl, J. Gerloff, H. Heynig, D. Mollenhauer (Eds.), pp. 1–576, Gustav Fisher Verlag, Stuttgart, Germany, 1991. [8.49] Kulikovskiy, M.S., Andreeva, S.A., Gusev, E.S., Kuznetsova, I.V., Annenkova, N.V., Molecular phylogeny of monoraphid diatoms and raphe significance in evolution and taxonomy. Biol. Bull., 43, 5, 398–407, 2016. [8.50] Kulikovskiy, M., Lange-Bertalot, H., Witkowski, A., Khursevich, G.K., Kociolek, J.P., New species of Eunotia (Bacillariophyta) from Lake Baikal with comments on morphology and biogeography of the genus. Phycologia, 54, 3, 248–260, 2015, https://doi.org/10.2216/14-98.1. [8.51] Lange-Bertalot, H., Bak, M., Witkowski, A., Tagliaventi, N., Eunotia and some related genera, in: Diatoms of Europe Vol. 6: Diatoms of the European Inland Waters and Comparable Habitats, Lange-Bertalot (Ed.), ARG Ganter Verlag, Königstein, Germany, 2011. [8.52] Lowe, R.L., The importance of scale in understanding the natural history of diatom communities, in: The Diatom World, J. Seckbach and P. Kociolek (Eds.), pp. 293–311, Springer, Dordrecht, 2011. [8.53] Mann, D.G., An ontogenetic approach to diatom systematics, in: Proceedings of the 7th International Diatom Symposium, O. Koeltz, Koenigstein, pp. 113–144, 1984. [8.54] Mayama, S. and Kobayasi, H., Observations of Eunotia arcus Ehrenb., type species of the genus Eunotia (Bacillariophyceae). Jpn. J. Phycol., 39, 131– 141, 1991. [8.55] Mayama, S., Morphology of Eunotia multiplastidica sp. nov. (Bacillariophyceae) examined throughout the life cycle. Korean J. Phycol., 7, 1, 45–54, 1992. [8.56] Mayama, S., Valuable taxonomic characters in the valve mantle and girdle of some Eunotia species, in: Lange-Bertalot-Festschrift, R. Jahn, J.P. Kociolek, A. Witkowski, P. Compere (Eds.), pp. 119–130, A.R.G., Gantner Verlag, Ruggell Königstein, Germany, 2001. [8.57] Medlin, L.K., Crawford, R.M., Andersen, R.A., Histochemical and ultrastructural evidence for the function of the labiate process in the movement of centric diatoms. Br. Phycol. J., 21, 3, 297–301, 1986, https://doi.org/10.1080/00071618600650351. [8.58] Metzeltin, D. and Lange-Bertalot, H., Tropical Diatoms of South America I. Iconogr. Diatomol., 5, 1–695, 1998.

208  Diatom Gliding Motility [8.59] Nakov, T., Beaulieu, J.M., Alverson, A.J., Accelerated diversification is related to life history and locomotion in a hyperdiverse lineage of microbial eukaryotes (Diatoms, Bacillariophyta). New Phytol., 219, 1, 462–473, 2018, https://doi.org/10.1111/nph.15137. [8.60] Palmer, T.C., Observations on errant frustules of Eunotia major. Proc. Acad. Nat. Sci. Philadelphia, 50, 110–119, 1898. [8.61] Patrick, R., Ecology of freshwater diatoms-diatom communities, in: The Biology of Diatoms, D. Werner (Ed.), pp. 284–332, Univ. of Cal., Press, Berkeley, 1977. [8.62] Patrick, R. and Reimer, C.W., The Diatoms of the United States, in: Volume 1. Monographs, vol. 13, pp. 1–688, Academy of the Natural Sciences, Philadelphia, PA, 1966. [8.63] Pavlov, A. and Levkov, Z., Diversity and distribution of taxa in the genus Eunotia Ehrenberg (Bacillariophyta) in Macedonia. Phytotaxa, 86, 1–117, 2013, https://doi.org/10.11646/ phytotaxa.86.1.1. [8.64] Pickett-Heaps, J.D., Hill, D.R., Wetherbee, R., Cellular movement in the centric diatom Odontella sinenis. J. Phycol., 22, 334–339, 1986, https://doi.org/10.1111/j.1529-8817.1986. tb00032.x. [8.65] Round, F.E., The epipsammon; a relatively unknown freshwater algal association. Br. Phycol. Bull., 2, 6, 456–462, 1965, https://doi.org/10.1080/00071616500650071. [8.66] Round, F.E., Crawford, R.M., Mann, D.G. (Eds.), Diatoms: biology and morphology of the genera, Cambridge University Press, Cambridge, UK, 1990. [8.67] Round, F.E. and Happey, C.M., Persistent, vertical-migration rhythms in benthic microflora. Br. Phycol. Bull., 2, 6, 463–471, 1965, https://doi.org/10.1080/00071616500650081. [8.68] Siver, P.A., Hamilton, P.B., Morales, E.A., Two new planktic species of Eunotia (Bacillariophyceae) from freshwater waterbodies in North Carolina, U.S.A. Algol. Stud., 119, 1–16, 2006, https://doi.org/10.1127/1864-1318/2006/0119-0001. [8.69] Siver, P.A. and Wolfe, A.P., Eunotia spp.(Bacillariophyceae) from Middle Eocene lake sediments and comments on the origin of the diatom raphe. Botany, 85, 1, 83–90, 2007, https:// doi.org/10.1139/b06-143. [8.70] Smith, W., A. synopsis of the British Diatomaceae: with remarks on their structure, functions and distribution; and instructions for collecting and preserving specimens, in: British Diatomaceae, vol. 1(1), 1853, https://doi.org/10.5962/bhl.title.10706. [8.71] Souza, M.G.M. and de Moreira-Filho, H., Diatoms (Bacillariophyceae) of two aquatic macrophyte banks from Lagoa Bonita, Distrito Federal, Brazil, I: Thalassiosiraceae and Eunotiaceae. Bull. Natl. Belg., 67, 1/4, 259–278, 1999, https://doi.org/10.2307/3668431. [8.72] Tapolczai, K., Bouchez, A., Stenger-Kovács, C., Padisák, J., Rimet, F., Trait-based ecological classifications for benthic algae: review and perspectives. Hydrobiologia, 776, 1, 1–17, 2016, https://doi.org/10.1007/s10750-016-2736-4. [8.73] Taylor, J.C., Cocquyt, C., Mayama, S., New and interesting Eunotia (Bacillariophyta) taxa from the Democratic Republic of the Congo, tropical central Africa. Plant Ecol. Evol., 149, 3, 291–307, 2016, https://doi.org/10.5091/plecevo.2016.1219. [8.74] Van de Vijver, B., De Haan, M., Lange-Bertalot, H., Revision of the genus Eunotia (Bacillariophyta) in the Antarctic Region. Plant Ecol. Evol., 147, 2, 256–284, 2014, 15 fig., 2 tables, https://doi.org/10.5091/plecevo.2014.930. [8.75] Vanormelingen, P., Chepurnov, V.A., Mann, D.G., Cousin, S., Vyverman, W., Congruence of morphological, reproductive and ITS rDNA sequence data in some Australasian Eunotia bilunaris (Bacillariophyta). Eur. J. Phycol., 42, 1, 61–79, 2007, https://doi. org/10.1080/09670260600942635. [8.76] Veselá, J. and Johansen, J.R., Three new Eunotia (Bacillariophyta) species from Acadia National Park, Maine, U.S.A. Phytotaxa, 175, 4, 181–200, 2014, https://doi.org/10.11646/ phytotaxa.175.4.1.

Motility in the Diatom Genus Eunotia  209 [8.77] Vinšová, P., Kopalová, K., Van de Vijver, B., Morphological observations on Pseudoeunotia linearis Carter (Bacillariophyta) and its transfer to the genus Eunotia. Bot. Lett., 163, 2, 117– 123, 2016, https://doi.org/10.1080/23818107.2016.1151826. [8.78] Vyverman, W., Sabbe, K., Mann, D.G., Kilroy, C., Vyverman, R., Vanhoutte, K., Hodgson, D., Eunophora gen. nov. (Bacillariophyta) from Tasmania and New Zealand: description and comparison with Eunotia and amphoroid diatoms. Eur. J. Phycol., 33, 95–111, 1998, https:// doi.org/10.1080/09670269810001736593. [8.79] Wetzel, C.E., Ector, L., Hoffmann, L., Bicudo, D.C., Colonial planktonic Eunotia (Bacillariophyceae) from Brazilian Amazon: Taxonomy and biogeographical considerations on the E. asterionelloides species complex. Nova Hedwigia, 91, 49–86, 2010, https://doi. org/10.1127/0029-5035/2010/0091-0049. [8.80] Zauer, C.M., Movement of Eunotia lunaris in connection with the problem of locomotion in diatom in general. Dokl. Akad. Nauk SSSR, 72, 1131–1133, 1950, (in Russian).

9 A Free Ride: Diatoms Attached on Motile Diatoms Vincent Roubeix1,2* and Martin Laviale3 1

Irstea, Aix Marseille Univ, RECOVER, Aix-en-Provence, France 2 Ifremer, BE, F‐44000 Nantes, France 3 Université de Lorraine, CNRS, LIEC, F-57000 Metz, France

Abstract

Periphytic diatoms are known to attach to a wide variety of submerged substrates. This review deals with a very particular substrate that consists of large benthic and motile diatoms. It is about passively motile epidiatomic diatoms. Their mode of attachment is similar to that of epiphytes on macrophytes, so they are also referred to as epiphytes, with the assumption of a negligible impact on the host’s fitness. Host-epiphyte relationships are quite specific and concern only a few freshwater species. It is not a random attachment to a neutral substrate, but rather an interaction between species that should at least benefit the epiphytes, in particular through a higher dispersal capacity. Current knowledge of this phenomenon is limited to observations. Further studies are needed to better characterize the nature of this original association between species. They should contribute to the poorly explored field of research concerning direct interactions between diatom species. Keywords:  Epidiatomic, epiphytes, phoresy, parasitism, symbiosis, mutualism, interactions, attachment

9.1 Introduction Many benthic diatoms are motile: they are able to glide on a substrate at velocities varying greatly with environmental conditions and among species [9.4] [9.23]. The dispersion of non-motile species is often limited to entrainment by the current. However, the adhesion capacity of diatoms can compensate for low mobility. Indeed, it is possible for diatoms to attach to more mobile organisms and take advantage of passive locomotion as epizoic diatoms living on macro-organisms such as turtles or crustaceans [9.17] [9.42] [9.52]. In a less known and more surprising way, some diatoms attach to large mobile benthic diatoms. It is this type of interspecific association, not common among diatoms, that will be presented in this review. The adhesion of diatoms is well known as epiphytes on aquatic macrophytes. Since diatoms are photosynthetic organisms, the small diatoms that attach to them could also be

*Corresponding author: [email protected] Vincent Roubeix: https://www.researchgate.net/profile/Vincent_Roubeix, https://orcid.org/0000-0002-3809-8678 Martin Laviale: [email protected], https://orcid.org/0000-0002-9719-7158 Stanley Cohn, Kalina Manoylov and Richard Gordon (eds.) Diatom Gliding Motility, (211–222) © 2021 Scrivener Publishing LLC

211

212  Diatom Gliding Motility called epiphytes, by analogy with epiphytes on macrophytes. However, the host is not a plant but a protist. The epiphytes could then be described as epidiatomic. An example is provided by small diatoms attached to the cingulum of a large marine planktonic species of the genus Arachnoidiscus [9.47]. If the host is motile, like many benthic diatoms, epiphytes could also be considered phoronts. While epiphytism is associated with the plant kingdom, phoresy refers to an interaction between animal species in which a phoretic species attaches to a host species to promote its dispersion for part of its lifetime [9.60]. The terms epiphytism and phoresy are used when the association of the two organisms has a low impact on the fitness of the host, unlike parasitism or mutualism. We choose to use the term epiphyte, as other authors have done previously [9.12] [9.48], or more precisely, the expression “passively motile epidiatomic diatoms,” to designate diatoms attached to motile diatoms. However, since the cost or benefit for the host of attached diatoms is not yet known, the terms ectoparasites or mutualist ectosymbionts cannot be excluded. The adhesion of epiphytes to the frustule of their host is a condition for transport and its different modalities will be presented. Not all small diatoms can attach to all large motile diatoms. Among the many combinations between two species that would be possible, it seems that only a limited number occurs in biofilms. This selection of hosts and epiphytes will be referred to as host-epiphyte interaction specificity. Finally, it is important to consider the evolutionary advantage that the epidiatomic habitat can provide to epiphytes, as well as the possible cost or benefit of the presence of epiphytes for the host. Epidiatomic diatoms have been observed on three genera of motile benthic diatoms with fibulate raphe: Nitzschia Hassal, Cymatopleura Ehrenberg and Surirella Turpin [9.12] [9.48] [9.50]. The examples presented in this review will focus on the freshwater host species Nitzschia sigmoidea (Nitzsch) W. Smith [9.26]. It is an exceptionally large pennate species (up to 500 µm long) with a sigmoid shape, moving quite fast (at about 20 µm.s-1) and well known to often exhibit epiphytes. Among these, several species have been identified, including Pseudostaurosira parasitica (W. Smith) E. Morales [9.12] [9.36] [9.51] (Figure 9.1), Fallacia helensis (Schulz) D.G. Mann [9.44] (Figure 9.2), Amphora copulata (Kützing) (a)

(b)

50 µm

(c)

Figure 9.1  Gliding cells of Nitzschia sigmoidea with stalked epiphytes of Pseudostaurosira parasitica: a single epiphyte in connective view (a), in valve view (b), and two epiphytes attached to the same frustule (c), which can be seen in movement [9.39].

Diatoms Attached on Motile Diatoms  213

50 µm

(a)

(b)

Figure 9.2  Cells of Nitzschia sigmoidea with adnate epiphytes of Fallacia helensis (OM): single epiphyte in connective view and host in valve view (a), four epiphytes in valve view and host in connective view (b), which can be seen in movement [9.43]. (a)

(b)

50 µm

(c)

100 µm

Figure 9.3  Cells of Nitzschia sigmoidea with many epiphytes (OM): epiphytes of Amphora copulata on a still gliding host (a), epiphytes of Amphora copulata on a dividing host (b), epiphytes of Amphora copulata and Pseudostaurosira parasitica on the same host (c), which can be seen in movement [9.41].

Schoeman & R.E.M. Archibald [9.12] [9.46] [9.48] (Figure 9.3) and Cocconeis pediculus Ehrenberg [9.15] [9.25].

9.2 Adhesion and Distribution of Epidiatomic Diatoms on Their Host The adhesion of epidiatomic diatoms do not differ from that of benthic diatoms in general. Thus, there are two main patterns of attachment to the substrate: the entire surface of a valve is fixedly applied to the substrate (i.e., adnate diatoms), or the diatom is attached to the substrate by a mucilaginous pad or stalk (i.e., stalked diatoms). For adnate diatoms, fixation is probably by the raphe, which also enables individuals to glide, since adhesion

214  Diatom Gliding Motility (a)

(e)

(c)

(b)

(d)

(f)

Figure 9.4  Focus on epiphytes of Nitzschia sigmoidea (SEM). Adnate, Fallacia helensis, and stalked, Pseudostaurosira parasitica, epiphytes on the same host (a), two cells of Pseudostaurosira parasitica still associated after division (b), two superimposed cells of Fallacia helensis after division and a third single one attached on the edge of the frustule (c), apex of a cell of Pseudostaurosira parasitica with the mucilaginous pad secreted for adhesion (d), individual of Amphora copulata (e) and internal view of one valve of Amphora sp., probably Amphora copulata var. epiphytica Round & Kyung Lee, considering the almost circular areolae on the ventral side (f). Scale bars indicate 10 µm, except in 9.4d (=5 µm).

and mobility are linked by the secretion of mucilage adhering to the substrate [9.31] [9.59]. Both patterns of adhesion are found in the epiphytes of N. sigmoidea. The species C. pediculus and F. helensis are adnate while P. parasitica is attached to its host by a pad (Figure 9.4a). This last araphid species has apical pore fields on its valves from which the pads are secreted (Figure 9.4d) [9.12]. The pad or the stalk is a flexible fixation that causes the epiphytes to move as the host glides [9.39]. The fixation of A. copulata is particular. The ventral connective face is in contact with the substrate as well as the two raphes in ventral position on each valve. The dorsal margin of the valves is very deep, giving individuals a curved appearance on their host (Figure 9.4e,f). Small mucilaginous pads visible at the valve apices strengthen adhesion [9.48], so that the epiphytes do not move. In natural populations of a host species such as N. sigmoidea, the number of epiphytes varies greatly between individuals. Some have none and others are literally covered with them (Figure 9.3). Several factors can influence the number of epiphytes. First, it is necessary to consider the rate of colonization of hosts, which must depend on the number of epiphytes available in the host’s microenvironment and therefore on the probability of meeting the host. Then, if it is assumed that epiphytes can reproduce on the host, the age of the host must also be important. Finally, it can be assumed that, since the valves of a mother cell are shared between daughter cells after asexual reproduction, epiphytes are also transmitted to the offspring. Therefore, there would be some heredity in the rate of epiphytism. Consideration should also be given to the possibility of an epiphyte leaving its host (1) actively, to attach to another type of substrate or to hitch a ride on another host, as a phoront, or (2) passively, by scraping against an object or selective grazing by protozoa.

Diatoms Attached on Motile Diatoms  215 Research on the possibility of epiphytes to leave their host is needed. The non-adherent valve of a biraphid species such as F. helensis and the free end of a frustule of P. parasitica potentially producing a mucilaginous pad, are adherent surfaces that can allow an individual to transfer from one host to another host or substrate by simple contact. In the asexual reproduction of diatoms, a mother cell gives rise to two daughter cells. So, the question arises whether both daughter cells of an epidiatomic diatom can remain attached to the host. In the case of adnate species, reproduction results in two superimposed cells (Figure 9.4c). The individual in the lower position, thus attached to the host, can keep its place while the individual above must glide to another location on the host or attach to another substrate. For stalked species such as P. parasitica (Figure 9.4b), daughter cells have identical or close attachment points on the host, which can generate rosette-like colonies after several reproductions [9.12]. Daughter cells may also remain associated only by their apex most distant from the host, resulting in an association of small chains of cells attached to the host (Figure 9.3c, [9.41]). Microdistribution studies on macrophytes have shown the preferential location of epiphytic microalgae on parts of their host [9.13] [9.14]. Although this question has not been addressed precisely concerning epidiatomic diatoms, observation of individuals on N. sigmoidea does not show preferred attachment sites or less occupied areas. The fact that all motile host diatoms have fibulae [9.48] suggests the hypothesis that fibulae could facilitate attachment. However, this hypothesis is not supported by the distribution of epiphytes on N. sigmoidea. The valve and connective faces of the hosts appear to be equally colonized (Figure 9.3) even though the connective faces (with a sigmoid shape) are wider and flatter than the valve faces. The adnate epiphytic diatoms are oriented by the raphe, which determines the direction of their movement, as well as the orientation of the frustule, which minimizes the frontal surface, i.e., the surface that the frustule presents to water when the frustule or water is in movement (lentic or lotic context) and which determines the drag force. It is striking to note that on the frustules of N. sigmoidea, the vast majority of epiphytes are oriented along the longitudinal axis and the direction of movement of their host [9.48] (Figures 9.2 and 9.3). We cannot at this time attribute this alignment to flow induced alignment [9.20], so how it is achieved requires further investigation.

9.3 The Specificity of Host-Epiphyte Interactions It has been long debated whether macrophytes constitute a neutral substrate for epiphytes [9.5] [9.11] [9.16]. Some studies concluded that different macrophyte species were equivalent substrates for epiphytes [9.10] while others showed variations in epiphyte communities between host species, suggesting the existence of host-epiphyte interactions [9.13]. The same question can be raised concerning analogous diatom-diatom interactions. Some observations are in favor of a neutral substrate. Indeed, a study showed that two populations of N. sigmoidea from a Mediterranean river and a tributary located three kilometers upstream, had different epiphytes [9.42]. In the upstream site, the epiphytes belonged to the species P. parasitica and F. helensis, whereas in the downstream site, only A. copulata was found in the epidiatomic habitat (Figure 9.5). This suggests that N. sigmoidea can host any epidiatomic species present in the environment. As water quality was very different between the two rivers, it can be assumed that the epiphytic flora on N. sigmoidea

45 40 35 30 25 20 15 10 5 0

Upstream

Amphora copulata

Pseudostaurosira parasitica

Downstream

Fallacia helensis

% occurrence on N. sigmoidea

216  Diatom Gliding Motility

Figure 9.5  Variations in the specific composition of epiphytes on Nitzschia sigmoidea between two sampling sites located on two connected rivers (up- and downstream sites), expressed as the occurrence of three epidiatomic species on frustules of N. sigmoidea. In fact, through the observation of fresh material, N. sigmoidea could not be strictly distinguished from the close species N. vermicularis. Both species were present in each site (see [9.44] for details).

simply reflects environmental conditions and that epidiatomic species have more stringent requirements than their host. Another observation also suggests that the substrate is neutral. In the downstream river where only the species A. copulata was reported on N. sigmoidea, some individuals of A. copulata were found attached to filaments of Melosira varians that were abundant in the sample [9.46]. Apparently, epiphytes could opportunely attach to another diatom species even if it is not motile. However, while the interactions may not be species-specific, some associations between species seem to be privileged. Indeed, epiphytes of A. copulata on N. sigmoidea have been reported in different contexts, including French and English rivers and a German lake [9.12] [9.46] [9.47]. In addition, a statistical study on a database of 200 samples from western France revealed that co-occurrences of the two species were much more frequent than chance would allow given the prevalence of each species in the database [9.46]. This suggests that the host-epiphyte association of the two species may be significant at the regional scale, whereas it has only been observed at local scale. The study of specific associations from a database allows an interesting change of scale, but it faces problems related to the subtle taxonomy of diatoms based on the shape and ornamentation of their frustule. The precise identification of a species often requires the use of scanning electron microscopy, which is still rarely used in routine surveys. Thus, the species A. copulata and A. pediculus can easily be confused [9.48]. Individuals of A. copulata epiphytes on N. sigmoidea have a morphological characteristic: almost circular areolae on the ventral part of the valve (Figure 9.4f). Round and Lee [9.48] have made a variety from this particularity: A. copulata var. epiphytica. The specificity of the association could be better appreciated if this variety could be distinguished from others in the databases. More generally, epidiatomic diatoms, currently known on various substrates, such as Cocconeis pediculus, may in the future reveal the existence of particular forms that attach more specifically to a diatom host. Finally, the consideration of a single species seems to invalidate the hypothesis of neutrality. Gyrosigma attenuatum (Kützing) Rabenhorst is a diatom species that has obvious

Diatoms Attached on Motile Diatoms  217 (a) (b)

NSIO

GYAT

100 µm

GYAT

100 µm

NSIO

Figure 9.6  Sigmoid frustules of Nitzschia sigmoidea (NSIO) and Gyrosigma attenuatum (GYAT) (OM, H2O2 treated material). Two species co-occurring in rivers samples with similar abundance, valve length and motility. However, Gyrosigma attenuatum was never seen with epiphytes.

similarities with N. sigmoidea. It is very large, highly motile and sigmoid in shape (Figure 9.6). It was present in both upstream and downstream sites of the previously reported study [9.41], with about the same abundance as N. sigmoidea. The two species probably have very similar ecological niches. However, none of the three epidiatomic diatoms identified at these sites have been observed on any individual of G. attenuatum. This may be due to the absence of fibules and the central position of the raphe on the valves of Gyrosigma. This finding also suggests the existence of a mechanism of epiphyte attraction specifically in N. sigmoidea. These mechanisms of search could involve chemical communication between the host and its epiphytes. The two partners could communicate through the production of bioactive metabolites. Indeed, diatoms are known to produce a wide range of such compounds, which are potentially involved in intercellular communication [9.30]. Alternatively, or even complementarily, it is possible that the two partners may interact through specific receptors localized on their respective frustules [9.35] [9.61]. How such receptors could function has yet to be ascertained. For this purpose, the use of atomic force microscopy (AFM) should help in deciphering the molecular mechanisms governing the adhesion between the two partners [9.27].

9.4 Cost-Benefit Analysis of Host-Epiphyte Interactions From the point of view of epidiatomic diatoms, various benefits of attachment to a motile diatom host can be identified. They are those of epiphytes and phoronts in the plant and animal kingdom respectively. As described for plant communities, diatom growth form is an important trait supporting competition for light between individuals. Being on top of

218  Diatom Gliding Motility the host should therefore give the epiphyte a competitive advantage, as it escapes the shade cast by other organisms closer to the substrate and has priority access to the light source. When fixed only by one apex, an individual of N. sigmoidea ensures, by the length of its frustules, a significant elevation of its epiphytes above the base substrate. Moreover, N. sigmoidea is easily carried away by the current and is often found in plankton samples [9.38]. This greatly increases the ability of epiphytes to disperse by drifting in rivers or through slower water motion in lakes. Also, at the substrate microscale, the relatively high speed of motion of the host facilitates the dispersion of epiphytes. In addition, the motility of the host must provide other benefits to the epiphytes. The motion of diatoms is directed towards the resources necessary for their reproduction. Epiphytic diatoms can take advantage of it to avoid being buried by sediments and to move to areas where light is more favorable [9.32]. Chemotaxis towards nutrients, such as silica [9.6], can also be beneficial for epiphytes. Finally, the size and speed of a large benthic diatom such as N. sigmoidea are a protection against a number of grazers, such as amoebas or ciliates, which cannot easily catch or ingest such a prey. Epiphytes, as long as they do not interfere with the motion of their host, may also benefit from this protection. The assembly of individuals of the same species form colonies preserved from sedimentation and grazing, thanks to an increase in size that benefits each cell. Size is indeed a key parameter in the ecology of microorganisms. Similarly, epiphytism may have evolved as a way for epidiatomic diatoms to artificially increase their size or even combine the physiological advantages of a small size (higher production rate and surface-to-volume ratio) with those of a large size [9.7]. But this is likely to be at the expense of the host. The possible costs for the host are first an unavoidable competition for resources with the epiphytes they carry. The latter take up the same nutrients and probably use the same light [9.37]. Diatoms capture light and nutrient resources through the entire surface of their cell. Epiphytes, especially adnate forms, may be a stronger form of competition for their diatom host than any other non-epiphyte species because of their covering position on the frustule of their host, which gives them priority access to resources. It is likely that the presence of epiphytes affects the physiology of the host. For instance, both light quantity and quality should be modified for the host, similar to what has been described for biofilm understory (self-shading in relation to biofilm development [9.28] or epiphytes blanketing on macrophytes [9.49] [9.53]). This should induce a physiological response (i.e., photo-­ acclimation) of the host in order to adjust to these lower light conditions. This response usually involves changes in pigmentation and/or chloroplast structure, resulting in modifications of photosynthesis efficiency. No experimental evidence has been put forward to support this hypothesis so far. However, the existence of single-cell imaging tools, such as the ones based on the monitoring of in-vivo chlorophyll fluorescence (e.g., [9.2] [9.19] [9.56]) or hyperspectral reflectance [9.21] [9.34] [9.54] should make it possible soon. If the host’s fitness suffers too much from its epiphytes, they can be considered as parasites. Physiological studies are needed to determine whether epiphytes do photosynthesis, whether they feed on exudates released by their host [9.9] or whether they are able to take nutrients from their host, as do chytrids, for example [9.3]. It has been shown that diatoms can dissolve glass to produce silicic acid necessary for their growth. It is possible that epiphytes can collect silicon from the frustule of their host for their own consumption. It is also questionable whether there are mechanisms by which host diatoms could get rid of their epiphytes, or whether some protozoan grazers are able to remove epiphytes from a diatom

Diatoms Attached on Motile Diatoms  219 host. If host cells are able to produce allelopathic molecules that inhibit epiphyte growth [9.1] [9.22] [9.29] [9.30] [9.55], a relatively higher host growth could gradually release the population from its epiphytes. Another disadvantage for the host is the increased drag that can reduce motility and adhesion in a stream [9.57], and the possible interference epiphytes may cause with the functioning of the raphe for locomotion. However, the comparative observation of Nitzschia individuals with and without epiphytes did not show a clear difference in motion speed [9.44]. Individuals swung from one face to the other without slowing down regardless of the presence of epiphytes on the face in contact with the glass substrate. The possible benefits for the host are less obvious. Epiphytes could act as a protective filter against UV radiation or high light level as it was hypothesized for epiphytes covering aquatic plants [9.8]. Indeed, these conditions are known to induce oxidative stress in phototrophs, which can lower photosynthetic efficiency (e.g., [9.28]). Nutrient flows are generally considered from the host to the epiphytes [9.9] [9.33], but production by epiphytes of substances useful to the host (photosynthate) could also be envisaged. This type of chemical transfer appears more realistic with adnate epiphytes that have a cell exchange surface in common with the host, than with stalked epiphytes. In such cases, it would be interesting to know if the pores of an epidiatomic diatom overlap the pores of its host. The mutual benefits of the two species would constitute a form of mutualist symbiosis, as found among microalgae, between diatoms and cyanobacteria [9.18] or dinoflagellates [9.24]. Considering the nitrogen-fixing cyanobacteria that some diatoms may harbor in the conopeum of their frustule [9.58], e.g., F. helensis, and the possibility that some epidiatomic diatoms may themselves host diatoms [9.12], this mobile association of species observed in biofilm could even be a tripartite symbiosis.

9.5 Conclusion The phenomenon of diatom transport by benthic diatoms raises many questions. It has not yet been much studied, probably because it is difficult to identify diatoms when they are combined and because the usual preparation of samples necessary for diatom identification (hot oxidation) separates the epiphytes from their hosts. The recent research on this topic has been limited to observations but it contributed to a better description of the interactions (species combinations, attachment and mobility). Experimental approaches based on single-cell imaging should help to better understand the precise nature of these associations and test the many hypotheses advanced in this review. Be it epiphytic, phoretic, parasitic or mutualistic, attachment allows small diatoms to reach a relatively high speed in the biofilm. This is also an example of direct, not only competitive interactions between diatoms, and suggests the existence of other less obvious interactions that may be underestimated and remain to be discovered.

References [9.1] Allen, J.L., Ten-Hage, L., Leflaive, J., Allelopathic interactions involving benthic phototrophic microorganisms. Environ. Microbiol. Rep., 8, 5, 752–762, 2016.

220  Diatom Gliding Motility [9.2] Baker, N.R., Chlorophyll Fluorescence: A Probe of Photosynthesis In Vivo. Annu. Rev. Plant Biol., 59, 1, 89–113, 2008. [9.3] Bertrand, C., Coute, A., Cazaubon, A., Fungal parasitism of the diatom Asterionella formosa Hassall (Bacillariophyceae) by Chytridiomycota. Ann. Limnol.-Int. J. Lim., 40, 1, 63–69, 2004. [9.4] Bertrand, J., Mouvements des diatomées VI. Les efforts pendant le déplacement apical Mesures, analyses, relations: longueur, vitesse, force [Diatom movement VI. Strains during apical displacement. Measurements, analysis, relationships: length, speed, force] [French]. Cryptogam. Algol., 20, 1, 43–57, 1999. [9.5] Blindow, I., The composition and density of epiphyton on several species of submerged macrophytes: The neutral substrate hypothesis tested. Aquat. Bot., 29, 2, 157–168, 1987. [9.6] Bondoc, K.G., Heuschele, J., Gillard, J., Vyverman, W., Pohnert, G., Selective silicate-­directed motility in diatoms. Nat. Commun., 7, 10540, 2016. [9.7] Brown, J.H. and Sibly, R.M., Life-history evolution under a production constraint. Proc. Natl. Acad. Sci. U. S. A., 103, 47, 17595–17599, 2006. [9.8] Burkholder, J.M., Interactions of benthic algae with their substrata, in: Algal Ecology: Freshwater Benthic Ecosystems, R.J. Stevenson, M.L. Bothwell, R.L. Lowe (Eds.), pp. 253–297, Academic Press, San Diego, California, USA, 1996. [9.9] Burkholder, J.M. and Wetzel, R.G., Epiphytic alkaline phosphatase on natural and artificial plants in an oligotrophic lake: Re-evaluation of the role of macrophytes as a phosphorus source for epiphytes. Limnol. Oceanogr., 35, 3, 736–747, 1990. [9.10] Cattaneo, A. and Kalff, J., Seasonal changes in the epiphyte community of natural and artificial macrophytes in Lake Memphremagog (Que. & Vt.). Hydrobiologia, 60, 2, 135–144, 1978. [9.11] Cejudo-Figueiras, C., Alvarez-Blanco, I., B′cares, E., Blanco, S., Epiphytic diatoms and water quality in shallow lakes: The neutral substrate hypothesis revisited. Mar. Freshwater Res., 61, 12, 1457–1467, 2010. [9.12] Chang, T.-P. and Steinberg, C., Epiphytische diatomeen auf Cymatopleura und Nitzschia [German]. Diatom Res., 3, 2, 203–216, 1988. [9.13] Comte, K. and Cazaubon, A., Structural variations of epiphytic diatom communities on three macrophytes in a regulated river (Durance), in South-East of France. Ann. Limnol.-Int. J. Lim., 38, 4, 297–305, 2002. [9.14] Comte, K., Fayolle, S., Roux, M., Quantitative and qualitative variability of epiphytic algae on one Apiaceae (Apium nodiflorum L.) in a karstic river (Southeast of France). Hydrobiologia, 543, 37–53, 2005. [9.15] Ehrenberg, C.G., Die Infusionsthierchen als vollkommene Organismen - Ein Blick in das tiefere organische Leben der Natur, nebst einem Atlas von vierundsechszig colorirten Kupfertafeln, gezeichnet vom Verfasser [The Infusoria animals as complete organisms - a view into the deeper organic life of nature, together with an atlas of sixty-four colored copper plates, drawn by the author] [German], Verlag von Leopold Voss, Leipzig, Germany, 1838. [9.16] Eminson, D. and Moss, B., The composition and ecology of periphyton communities in freshwaters. Br. Phycol. J., 15, 4, 429–446, 1980. [9.17] Fayolle, S., Moriconi, C., Oursel, B., Koenig, C., Suet, M., Ficheux, S., Logez, M., Olivier, A., Epizoic algae distribution on the carapace and plastron of the European pond turtle (Emys orbicularis, Linnaeus, 1758): A study from the Camargue, France. Cryptogam. Algol., 37, 4, 221–232, 2016. [9.18] Foster, R.A., Kuypers, M.M.M., Vagner, T., Paerl, R.W., Musat, N., Zehr, J.P., Nitrogen fixation and transfer in open ocean diatom-cyanobacterial symbioses. ISME J., 5, 9, 1484–1493, 2011. [9.19] Goessling, J.W., Su, Y., Cartaxana, P., Maibohm, C., Rickelt, L.F., Trampe, E.C.L., Walby, S.L., Wangpraseurt, D., Wu, X., Ellegaard, M., Kühl, M., Structure-based optics of centric

Diatoms Attached on Motile Diatoms  221 diatom frustules: modulation of the in vivo light field for efficient diatom photosynthesis. New Phytol., 219, 1, 122–134, 2018. [9.20] Gordon, R., Björklund, N.K., Robinson, G.G.C., Kling, H.J., Sheared drops and pennate diatoms. Nova Hedwigia, 112, Festschrift for Prof. T.V. Desikachary, 287–297, 1996. [9.21] Gowen, A.A., Feng, Y., Gaston, E., Valdramidis, V., Recent applications of hyperspectral imaging in microbiology. Talanta, 137, 43–54, 2015. [9.22] Gross, E.M., Feldbaum, C., Graf, A., Epiphyte biomass and elemental composition on submersed macrophytes in shallow eutrophic lakes. Hydrobiologia, 506, 1, 559–565, 2003. [9.23] Harper, M.A., Movements, in: The Biology of Diatoms, D. Werner (Ed.), pp. 224–249, Blackwell Scientific Publications, Oxford, UK, 1977. [9.24] Hehenberger, E., Burki, F., Kolisko, M., Keeling, P.J., Functional relationship between a dinoflagellate host and its diatom endosymbiont. Mol. Biol. Evol., 33, 9, 2376–2390, 2016. [9.25] Jahn, R., Kusber, W.H., Romero, O.E., Cocconeis pediculus Ehrenberg and C. placentula Ehrenberg var. placentula (Bacillariophyta): Typification and taxonomy. Fottea, 9, 2, 275– 288, 2009. [9.26] Knattrup, A., Yde, M., Lundholm, N., Ellegaard, M., A detailed description of a Danish strain of Nitzschia sigmoidea, the type species of Nitzschia, providing a reference for future morphological and phylogenetic studies of the genus. Diatom Res., 22, 1, 105–116, 2007. [9.27] Laviale, M., Beaussart, A., Allen, J., Quilès, F., El-Kirat-Chatel, S., Probing the Adhesion of the Common Freshwater Diatom Nitzschia palea at the Nanoscale. ACS Appl. Mater. Interfaces, 11, 51, 48574–48582, 2019. [9.28] Laviale, M., Prygiel, J., Lemoine, Y., Courseaux, A., Creach, A., Stream periphyton photoacclimation response in field conditions: Effect of community development and seasonal changes. J. Phycol., 45, 5, 1072–1082, 2009. [9.29] Leflaive, J. and Ten-Hage, L., Algal and cyanobacterial secondary metabolites in freshwaters: a comparison of allelopathic compounds and toxins. Freshwater Biol., 52, 2, 199–214, 2007. [9.30] Leflaive, J. and Ten-Hage, L., Chemical interactions in diatoms: role of polyunsaturated aldehydes and precursors. New Phytol., 184, 4, 794–805, 2009. [9.31] Lind, J.L., Heimann, K., Miller, E.A., vanVliet, C., Hoogenraad, N.J., Wetherbee, R., Substratum adhesion and gliding in a diatom are mediated by extracellular proteoglycans. Planta, 203, 2, 213–221, 1997. [9.32] McLachlan, D.H., Brownlee, C., Taylor, A.R., Geider, R.J., Underwood, G.J.C., Light-induced motile responses of the estuarine benthic diatoms Navicula perminuta and Cylindrotheca closterium (Bacillariophyceae). J. Phycol., 45, 3, 592–599, 2009. [9.33] McRoy, C.P. and Goering, J.J., Nutrient transfer between the seagrass Zostera marina and its epiphytes. Nature, 248, 173, 1974. [9.34] Méléder, V., Laviale, M., Jesus, B., Mouget, J.L., Lavaud, J., Kazemipour, F., Launeau, P., Barillé, L., In vivo estimation of pigment composition and optical absorption cross-­section by spectroradiometry in four aquatic photosynthetic micro-organisms. J. Photochem. Photobiol. B: Biol., 129, 115–124, 2013. [9.35] Molino, P.J., Chiovitti, A., Higgins, M.J., Dugdale, T.M., Wetherbee, R., Diatom Adhesives: Molecular and Mechanical Properties, in: Biological Adhesives, A. Smith (Ed.), Springer, Cham, 2016. [9.36] Morales, E., Pseudostaurosira parasitica, 2010, https://diatoms.org/species/pseudostaurosira_ parasitica. [9.37] Phillips, G.L., Eminson, D., Moss, B., A mechanism to account for macrophyte decline in progressively eutrophicated freshwaters. Aquat. Bot., 4, 103–126, 1978. [9.38] Robson, S., 2017. https://diatoms.org/species/nitzschia_sigmoidea.

222  Diatom Gliding Motility [9.39] Roubeix, V., Epidiatomic diatoms (1). figshare. Figure [Movie]. https://doi.org/10.6084/ m9.figshare.10288718.v1, 2019. [9.40] Roubeix, V., Epidiatomic diatoms (2). figshare. Figure [Movie]. https://doi.org/10.6084/ m9.figshare.10289045.v1, 2019. [9.41] Roubeix, V., Epidiatomic diatoms: An insight into interaction specificity. Phycol. Res., 2020, https://doi.org/10.1111/pre.12420 [9.42] Roubeix, V., Attia, L., Chavaux, R., Very, F., Olivier, A., Ector, L., Vassal, V., Specificity of diatom communities attached on the carapace of the European pond turtle (Emys orbicularis). Adv. Oceanogr. Limnol., 12, 1, 2021. [9.43] Roubeix, V. and Chalie, F., Cells of Fallacia helensis attached on a moving individual of Nitzschia sigmoidea. figshare. Figure [Movie]. https://doi.org/10.6084/m9.figshare.7861883. v1, 2019. [9.44] Roubeix, V. and Chalié, F., Identification of an epiphytic diatom on Nitzschia sigmoidea (Bacillariophyceae). Bot. Lett., 166, 2, 207–211, 2019. [9.45] Roubeix, V. and Coste, M., AME 80:55-59, Supplement 1 [Movie]. https://www.int-res.com/ articles/suppl/a080p055_supp/, 2017. [9.46] Roubeix, V. and Coste, M., A case of close interspecific interactions between diatoms: Selective attachment on a benthic motile species. Aquat. Microb. Ecol., 80, 1, 55–59, 2018. [9.47] Round, F.E., Crawford, R.M., Mann, D.G., The Diatoms, Biology & Morphology of the Genera, Cambridge University Press, Cambridge, UK, 1990. [9.48] Round, F.E. and Lee, K., Studies on freshwater Amphora species IV. The Amphora epiphytic on other diatoms. Diatom Res., 4, 2, 345–350, 1989. [9.49] Sand-Jensen, K., Revsbech, N.P., Barker Jørgensen, B., Microprofiles of oxygen in epiphyte communities on submerged macrophytes. Mar. Biol., 89, 1, 55–62, 1985. [9.50] Skuja, H., Taxonomie des Phytoplanktons einiger Seen in Uppland, Schweden [German], in: Symboliae Botanicae Upsalienses, vol. 9, pp. 1–399, 1948. [9.51] Smith, W., Synopsis of British Diatomaceae, John Van Voorst, London, England, 1856. [9.52] Thiery, A. and Cazaubon, A., Epizootic algae and protozoa on fresh-water branchiopods (Anostraca, Notostraca and Spinicaudata) in Moroccan temporary ponds. Hydrobiologia, 239, 2, 85–91, 1992. [9.53] Tóth, V.R., The effect of periphyton on the light environment and production of Potamogeton perfoliatus L. in the mesotrophic basin of Lake Balaton. Aquat. Sci., 75, 4, 523–534, 2013. [9.54] Vallotton, V., Angel, B., Mccall, M., Osmond, M., Kirby, J., Imaging nanoparticle-algae interactions in three dimensions using Cytoviva microscopy. J. Microsc., 257, 2, 166–169, 2015. [9.55] Vanelslander, B., Paul, C., Grueneberg, J., Prince, E.K., Gillard, J., Sabbe, K., Pohnert, G., Vyverman, W., Daily bursts of biogenic cyanogen bromide (BrCN) control biofilm formation around a marine benthic diatom. Proc. Natl. Acad. Sci. U. S. A., 109, 7, 2412–2417, 2012. [9.56] Villareal, T.A., Single-cell pulse amplitude modulation fluorescence measurements of the giant diatom Ethmodiscus (Bacillariophyceae). J. Phycol., 40, 6, 1052–1061, 2004. [9.57] Wahl, M., Ecological lever and interface ecology: epibiosis modulates the interactions between host and environment. Biofouling, 24, 6, 427–438, 2008. [9.58] Wehr, J.D., Sheath, R.G., Kocoiolek, J.P., Freshwater Algae of North America: Ecology and Classification, 2nd, Academic Press, Amsterdam, 2015. [9.59] Wetherbee, R., Lind, J.L., Burke, J., Quatrano, R.S., The first kiss: Establishment and control of initial adhesion by raphid diatoms. J. Phycol., 34, 1, 9–15, 1998. [9.60] White, P.S., Morran, L., de Roode, J., Phoresy. Curr. Biol., 27, 12, R578–R580, 2017. [9.61] Wigglesworth-Cooksey, B. and Cooksey, K.E., Can diatoms sense surfaces? State of our knowledge. Biofouling, 5, 3, 227–238, 1992.

10 Towards a Digital Diatom: Image Processing and Deep Learning Analysis of Bacillaria paradoxa Dynamic Morphology Bradly Alicea1,2*, Richard Gordon3,4, Thomas Harbich5, Ujjwal Singh6, Asmit Singh6 and Vinay Varma7 OpenWorm Foundation, Boston, Massachusetts, USA Orthogonal Research and Education Laboratory, Champaign, Illinois, USA 3 Gulf Marine Specimen Laboratory, Panacea, Florida, USA 4 Department of Obstetrics and Gynecology, Wayne State University, Detroit, Michigan, USA 5 Am Brudenrain, Weissach im Tal, Germany 6 Indraprastha Institute of Information Technology (IIIT-Delhi), Delhi, India 7 Amrita Vishwa Vidyapeetham University, Coimbatore, Tamil Nadu, India 1

2

Abstract

Recent years have witnessed a convergence of data and methods that allow us to approximate the shape, size, and functional attributes of biological organisms. This is not only limited to traditional model species: given the ability to culture and visualize a specific organism, we can capture both its structural and functional attributes. We present a quantitative model for the colonial diatom Bacillaria paradoxa, an organism that presents a number of unique attributes in terms of form and function. To acquire a digital model of B. paradoxa, we extract a series of quantitative parameters from microscopy videos from both primary and secondary sources. These data are then analyzed using a variety of techniques, including two rival deep learning approaches. We provide an overview of neural networks for non-specialists as well as present a series of analysis on Bacillaria phenotype data. The application of deep learning networks allows for two analytical purposes. Application of the DeepLabv3 pre-trained model extracts phenotypic parameters describing the shape of cells constituting Bacillaria colonies. Application of a semantic model trained on nematode embryogenesis data (OpenDevoCell) provides a means to analyze masked images of potential intracellular features. We also advance the analysis of Bacillaria colony movement dynamics by using templating techniques and biomechanical analysis to better understand the movement of individual cells relative to an entire colony. The broader implications of these results are presented, with an eye towards future applications to both hypothesis-driven studies and theoretical advancements in understanding the dynamic morphology of Bacillaria. *Corresponding author: [email protected] Richard Gordon: [email protected] Thomas Harbich: [email protected] Ujjwal Singh: [email protected] Asmit Singh: [email protected] Vinay Varma: [email protected] Stanley Cohn, Kalina Manoylov and Richard Gordon (eds.) Diatom Gliding Motility, (223–248) © 2021 Scrivener Publishing LLC

223

224  Diatom Gliding Motility Keywords:  Deep learning, biomechanics, computational biology, big data biology, motion analysis

10.1 Introduction “I still remember, as many years ago, when I found the Bacillaria paradoxa near Greifswald many years ago, that I stood as if clinging to the microscope and could not turn my back on the strange spectacle that presented itself to me…. they are glued together as if they were an organism, and yet each moves for itself next to the other!” – translated from Max Schultze [10.48]

Creating digital instantiations of a model organism is of great potential to well-established communities centered around model organisms such as Caenorhabditis elegans [10.46]. The opportunity for creating a digital model of a non-model organism is potentially greater. In this chapter, we will introduce a data-intensive approach to modeling the behaviors and dynamic phenotypes associated with the colonial diatom Bacillaria paradoxa (Figure 10.1). Using image processing techniques, we can construct a computational model of Bacillaria colonies. This simple phenotypic model provides a means to capture the dynamics of movement across different moments in time. Inferring the movement of these colonies reveals the diversity and complexity of behavior exhibited in a simple organismal colony. The value of this approach is enhanced by the nature of the Bacillaria literature. While the literature is quite old (Bacillaria was first observed by Otto Friedrich Müller in 1783 [10.47] [10.54]), work to date has focused mostly on taxonomy and cellular structural biology. The partial synchrony of Bacillaria colonies [10.18] indicates that the behavior of a colony may be greater than the sum of its individual cells (i.e., that each colony is a multicellular organism). Cellular biology work that has been done on the movement of Bacillaria [10.27] [10.40] [10.60] is neither computational in nature nor at the whole-organism level. There has been some computational work conducted on diatom morphogenesis [10.4] [10.5] [10.12] [10.21–10.23] [10.25] [10.31] [10.38] [10.39] [10.59] which is expanding due to their usefulness in nanomaterials [10.7] [10.8] [10.20] [10.32] [10.35], medical applications [10.55], and many other fields [10.49]. Yet despite these studies, there is little computational work integrating structural morphology with diatom motility.

10.1.1 Organism Description Bacillaria paradoxa, synonymous with Bacillaria paxillifera [10.29], is a diatom in the Bacillariaceae family which has been subdivided into three species: B. paxillifera, B. kuseliaeand and B. urvemillerae, with B. paxillifera further divided into four varieties: var. czarneckii, var. pacifica, var. tropica and var. tumidula [10.29]. As the distinctions are mostly made at the SEM level of resolution, we will adopt the blanket designation B. paradoxa in this study (Figure 10.1). Diatoms are a group of eukaryotic microalgae whose ornate cell walls are composed primarily of amorphous silica. They exhibit a unique life history [10.44]. Cells of Bacillaria (sometimes called filaments) are elongated and motile, sliding along each other, in stacked colonies that curve slightly out of the plane. Cells are rectangular in girdle view (as seen in colonies, as their valves face each other), and lanceolate in valve view (Figure 10.2). The raphe system is slightly keeled and runs from pole to pole with no central nodule [10.43]. Two large plate-like chloroplasts are present, one near each end of the cell. The nucleus is located centrally. Cells are yellow-brown in color. Fibulae are strong, and the valve surface is covered in transverse parallel structures called striae [10.29].

Towards a Digital Diatom  225

3 1

5

2

7

8

4

6

Figure 10.1  Drawing adapted from O.F. Müller (1783, translated in [10.54]), who was the first to characterize Bacillaria colonies. Examples 1 through 8 show the various states of expansion and contraction (dynamic phenotypes) of colonies.

Bacillaria cells are arranged in parallel stacks, and these parallel stacks form a colony. These stacks form early in the life history of a colony by cell divisions perpendicular to the valves. The stacked colony moves by consecutive pairs of individual cells sliding against each other [10.54]. When these sliding movements occur in temporal order across the colony [10.18], the synchronized movement is like the sliding of a deck of cards [10.34] and results in a large extension of the whole colony. The continual folding and unfolding of a colony of cells stacked in parallel results in cyclic gliding movements. Each cell is an intrinsic oscillator [10.14], so the partial synchronization may be due to entrainment of these oscillators by an unknown mechanism, perhaps light piping within each colony [10.19] [10.16]. The mechanism of gliding movement is still being worked out [10.19]. Observations of high accelerations of single diatoms [10.15] suggest a motor that moves with explosive force at a molecular scale [10.45]. This may be the basis of the often-­ observed jerky motion of diatoms, hints of which have been seen in Bacillaria [10.30].

27

28

rf

29

30

226  Diatom Gliding Motility

10 µm

(a)

1 µm

(b)

1 µm

(c)

1 µm

2 µm

(d)

(e)

Figure 10.2  Bacillaria close-up images of single cells using scanning electron microscopy (SEM). (a) a whole valve seen from the inside. (b) close up of the same, middle section. The horizontal slit is the raphe. It lacks a central node. (c) Tip of the inside of a valve. (d) Middle section of a valve, exterior view. Note the raphe is a slit through the whole valve. (e): External view of the tip of a valve [10.33]. (Reprinted with permission of Amgueddfa Cymru, National Museum Wales.)

Towards a Digital Diatom  227

10.1.2 Research Motivation Diatoms (and Bacillaria in general) are an excellent model system for understanding phenotypic growth and dynamics [10.3]. Unfortunately, there is no model for its behavioral and dynamic phenotypic properties. It is our contention that digital technologies such as image processing techniques, machine learning techniques, and open data repositories allow us to construct a digital model that can be used to uncover previously unexplored relationships. Our data rely on the ability to generate microscopy movies from cultured specimens. Some of these movies are publicly available in places such as YouTube or in the Supplemental Materials of academic papers. Fortunately, Bacillaria is relatively easy to isolate and culture, and we have collected primary microscopy data as well. Enabled through the application of deep learning [10.11] [10.36] [10.51] and biomechanical techniques, we postulate that a digital Bacillaria is not only possible but highly useful to the scientific community. Our purpose here is twofold: to review key concepts related to our computational approach, and to present the technical details of building the digital Bacillaria. First, we will provide an overview of neural networks and machine intelligence. Then the methods used for extracting data from individual video images will be presented, along with more detailed descriptions for methods employed in the data analysis. The chapter will conclude with an analysis using several techniques for image processing and discovering the computational features that define a Bacillaria colony. These include the shape parameters of the cell, techniques to extract intracellular features, and techniques to approximate colony movement. Our techniques range from formal deep learning techniques for extracting morphological features to the creation of masks and templates to quantify images extracted from time series. Description of Neural Networks. Neural networks are computer programs assembled from many computational units that behave as an adaptive system. This adaptive system is inspired by the nervous system: providing the systematic power and computational capabilities of a computer with the densely reticulating connectivity of a model of a biological brain. The rationale for using neural networks (sometimes referred to a connectionist approach) in a computational setting is to simulate the pattern recognition and decision-making properties of biological learning. Computers and brains have much in common, but there exist important differences. The most fundamental difference is that computers and minds think and reason (such as this exists in machines) in entirely different ways. Computational networks are also wired in very simple ways, usually in serial chains. Each link in this chain is connected to maybe four or five others in arrangements known as logic gates. By contrast, the brain has complex entities called neuronal cells which are densely interconnected in complex, parallel ways. Each neuron can be connected with up to 10,000 of its neighbors. These connection patterns lead to functional distinctions which are currently the subject of great debate [10.62]. Input units are designed to receive various forms of information from the outside world that the network will attempt to learn about, recognize, or otherwise process information. Other units reside on the opposite side of the network and signal how it responds to the information it’s learned; those are known as output units. Situated between input units and output units are one or more layers of hidden units, which, together, form the majority of the network. Most neural networks are fully connected, which means each hidden

228  Diatom Gliding Motility unit and each output unit is connected to every unit in the adjacent layers on either side. Connections between one unit and another are represented by a number called a “weight,” which can be either positive (if one unit excites another) or negative (if one unit suppresses or inhibits another). The higher the weight, the more influence one unit has on another. This corresponds to the way actual brain cells trigger one another across tiny gaps called synapses, with excitatory or inhibitory connections. Neural networks learn things through training and testing. During both training (when the ANN learns associations) and testing (when the ANN makes associations), patterns of information are fed into the network via the input units. In turn, layers of hidden units are activated, which ultimately trigger the output units. This common design is called a feedforward network. Each unit receives inputs from the units to its left, and the inputs are multiplied by the weights of the connections they travel along (Figure 10.4). Every unit adds up all the inputs it receives. When the sum is more than a certain threshold value, the unit “fires” and triggers all connected units. For a neural network to learn, there has to be an element of feedback involved—just as children learn by being told what they’re doing is right or wrong. In fact, we all use feedback, all the time. Neural networks learn things through a process of feedback called backpropagation (or backprop). This involves comparing the output a network produces with the output it was meant to produce, and then modifying the weights of the connections between the units in the network accordingly. In time, backpropagation causes the network to adapt to the desired output (or learn), reducing the difference between actual and intended output to the point where the two coincide, so the network figures things out exactly as it should. Once the network has been trained with enough learning examples, it reaches a point where you can present it with an entirely new set of inputs that it has never seen before and see how it responds. For example, suppose you’ve been teaching a network by showing it many pictures of chairs and tables, represented in some appropriate way it can understand, and telling it whether each one is a chair or a table. As an example, suppose we train the model with 25 images of chairs and 25 images of tables. Depending on how completely the model is trained, new examples will be categorized as either a chair or a table. This is generalized from its past experience rather than being generated de novo. The inputs to a network are essentially binary numbers: each input unit is either switched on or switched off. Using five input units, we can feed in information about different features of an object using binary strings. For example, a chair might conform to five feature categories: possessing a back, possessing a top, having soft upholstery, allows for comfortable sitting, and storage capacity. This results in a series of categorical responses: Yes, No, Yes, Yes, No (10110). For a table, these responses might be: No, Yes, No, No, Yes (01001). During the learning phase, the network is simply looking at binary strings (e.g., 10110 and 01001), to determine what represents a chair and what represents a table.

10.2 Methods 10.2.1 Video Extraction Microscopy videos of Bacillaria were obtained from a number of sources (private archives and YouTube). The video frames were then processed to obtain segmented features and

Towards a Digital Diatom  229 numeric variables. Specifically, we decomposed video files of Bacillaria microscopy images into its component frames. The number of frames depends on the sampling rate for the given video. Deep Learning models DeepLabv3 (a TensorFlow library reviewed in [10.10] and OpenDevoCell (built with DeepLearning4J; see [10.13]) were then used to analyze these images. Due to the heterogeneous sources of acquisition, information was extracted in a manner that allows for unsupervised classification of the Bacillaria colony. The dataset for the DeepLabv3 analysis consisted of around 20,000 frames of secondary microscopy data, which contains much variety related to Bacillaria movement. The data for the OpenDevoCell and time-lapse analyses comes from microscopy data collected specifically for these analyses. Image Skeleton Creation. The pre-masked images (skeletons) derived for our primary data were created in GIMP 2.10 [10.17]. First, we converted each raw image to an indexed (1-bit) imaged with no color dithering. Next, we converted the resulting binary map to an RGB (red, green, blue) indexed image. Select the cell area (RGB value 0,0,0) by color, and change to RGB value to (0,217,0). Maintaining the selection by color, edit the stroke selection function to a width of 1.0 pixels for a thin skeleton, and 5.0 pixels for a thick skeleton. This ensures separation between edges that are close together while also remaining selectable by the segmentation algorithm. Once the boundaries (which were colored as 0,217,0) have been selected, select the area inside the boundary and change to RGB value 0,0,0. These transformations should result in a black cell surface area with a light green boundary. The final step is to change the background color (select the background by color) to RGB value 0,0,0. To create a thick skeleton from a thin skeleton, select the thin skeleton by color and then select the border function. The border width should be set to 4, hard border, and filled with RGB value 0,217,0. The pseudo-code for GIMP script-fu is located on Github (https://github.com/devoworm/ Digital-Bacillaria/tree/master/Image-Skeletons). Image Tracking for Movement. We also employ image tracking for the primary microscopy data. The tracking of a partial image (template) of a diatom can be used under certain conditions to obtain its trajectory. In particular, a movement of the diatoms in a plane perpendicular to the optical axis is essential. Sufficiently differentiated structures (chloroplasts) are required. The boundaries between neighboring diatoms are not considered. Image tracking is done using Tracker 5.1.3 (https://www.softpedia.com/get/ScienceCAD/Douglas-Tracker.shtml), an image analyzer that provides information regarding the motion of features between movie frames. For the tracking of paths in videos, various algorithms have been developed and implemented [10.61]. Each method has specific applications and strengths. It is often assumed that the objects to be tracked differ significantly from a sufficiently homogeneous background, yet this is not the case for Bacillaria paxillifer colonies. Image tracking allows for cells in the colony to be tracked without complete separation from its background. Our ad-hoc method of feature selection defines cells as an ellipse registered with horizontal and vertical axes (Figure 10.3a). For sake of consistency, all video frames are rotated by 31 degrees and cropped. This was done primarily to make alignment of the ellipse and the diatom identical. This allows for templates of different size in addition to a direct comparison between cells using the x-axis. The tracking procedure proceeds by

230  Diatom Gliding Motility

(a)

Diatom #3

(b)

Diatom #2 57.5 µm

Diatom #4

59.14 µm

Diatom #1 25

50

Figure 10.3  Demonstration of the image tracking procedure. (a) definition of tracked feature (white ellipse within a cell). (b) labeled (numbered) cells with relative measurements provided in red. The determined coordinates refer to a target in the middle of the template. The target can be moved and placed on the apex of the diatom being tracked. Then the coordinates of the apex are captured. This position is indicated in Figure 10.3a by a mark and a vertical line. Image scale: 38.36 μm per cm, or 0.325 μm per pixel. Scale bars 50 µm.

numbering the cells within a colony (see Figure 10.3b). The cell labeled “1” indicates the reference cell that is fixed relative to the substrate. Diatoms labeled “2” and “3” are then tracked as shown in the bottom panel of Figure 10.3. As these numeric values refer to the position of a particular cell relative to a reference cell (position of cell #2 relative to cell #1), an offset value must be added in order to optimize initial positions relative to movement.

10.2.2 Deep Learning Deep learning [10.2] [10.56] is implemented using a pre-trained model called DeepLabv3 and open-source software with a web interface called OpenDevoCell (based on Deeplearning4J). DeepLabv3 (Google, Mountain View, California, USA) is a package for TensorFlow, and Deeplearning4J (Eclipse Foundation, Ottawa, Canada), a Java-based library that works with TensorFlow. OpenDevoCell is open-source software located on GitHub (https://github.com/devoworm/GSOC-2019/tree/master/OpenDevoCell) and as a web-based application (https://open-devo-cell.herokuapp.com). The DeepLab model [10.10] is implemented in TensorFlow, a platform for implementing machine and deep learning algorithms. The DeepLab pre-trained model is based on deep learning, which utilizes a neural network of many layers. DeepLab is also specialized for extracting features that conform to a bounding box, as it has been trained on both “hard” object classes (objects embedded in an obscuring background) and datasets such as ImageNet and JFT-300M [10.53]. Bounding boxes of different sizes and orientations not only enable a heuristic means of segmenting individual cells in a Bacillaria colony, based on visual inspection it fits the cell shape and nature of variability quite well. The DeepLab v3 algorithmic analysis is based on the idea of dilated convolution, where the input signal is sampled in alternative ways [10.53]. Ultimately, the algorithm attempts to find a trade-off between localization (using a small sampling aperture of image pixels) and context assimilation (using a large sampling aperture of image pixels). This optimal tradeoff in spatial resolution determines which pixels are labeled as individual cells and which pixels are labeled as background or boundary. This aligns with the nature of a dataset that contains many instances of dynamic Bacillaria colonies, including changes in movement

Towards a Digital Diatom  231 phase, spatial orientation, and image opacity (e.g., the presence of algae that might mask the cells). For the DeepLab analysis, all images are rectified to a common orientation and hand-labeled with information about the desired properties of bounding box edges. We then obtain XML records for each image and randomly partition these data into two subsets: a training set (train.csv) and a testing set (test.csv). These files are used to generate a record file (TFRecord) which is used to move data in and out of the deep network. More generally, deep learning is an instance of the neural network approach and relies upon a network with a large number of hidden layers relative to a standard neural network. We discuss neural networks in more detail in a previous section of this chapter. In general, a deeper network with more hidden layers translates into a larger feature space. This provides a user with models that have better resolution, but also models that have a greater potential for error [10.26]. To promote reproducibility, a tutorial for DeepLabv3 implementation is available on GitHub [10.50], and the software implementation is available on GitHub (https://github.com/devoworm/Digital-Bacillaria). Bounding Box Method (DeepLab v3). Now we turn to the segmentation and measuring of individual Bacillaria cells. The first step is to employ methods for identifying the boundaries of cells, in particular, to distinguish the boundaries from the area of other cells and the image background. This is done by defining a bounding box and then applying a cell segmentation method to a single image (Figure 10.4). We define five points on each cell: a) the two ends of the cell (1 and 2 in Figure 10.4), b) the midpoint of the line formed between points 1 and 2 (3 in Figure 10.4), and c) the edges of a cell (4 and 5 in Figure 10.4) defined by drawing a line perpendicular to the axis defined by points 1, 2, and 3. Centroids are then calculated from these data in a post-processing step by finding the midpoints between and maximum values for the x (centroid x) and y (centroid y) axes.

1

5 4

3

2

Figure 10.4  A diagram showing the five points on a sample cell (two ends, midpoint of the transverse line, and edges of the cell).

232  Diatom Gliding Motility Soft Bounding Box Method (OpenDevoCell). The OpenDevoCell platform was originally trained using pixel labeling and segmentation masks. Pixel level labeling requires a lot of effort and time, and does not always provide great results. A lot of data that we have is partially labeled, where instead of each pixel being given a zero or one label, each pixel has a membership in a macro category. The data provides spatial information about these categories in the form of (x,y,z) coordinates or (r,θ) (polar coordinates). One way of approaching the problem is by broadly defining potential features in the form of boxes and refining the boxes using a deep convolutional neural network (CNN) [10.28] to get a semantically segmented image where the segmented features are defined by distinguishing between specific labels. The OpenDevoCell technique uses region proposal methods to generate segmentation masks. The candidate segments are used to update the deep CNN. The semantic features learned by the network are used to generate better candidates and proceed as an iterated procedure. This method was originally applied to Caenorhabditis elegans embryogenesis data by utilizing the spatial locations and making the bounding boxes. With ground-truth bounding boxes, we can find the candidate masks that overlap the most with the bounding boxes. An error/cost function is used to maximize the overlap. For application to Bacillaria colonies, we convert our dataset to skeletons using a procedure implemented in GIMP 2.10. These skeletons are pre-masks that mimic the initial condition of the Caenorhabditis elegans embryos, which was a series of high-resolution images in which membrane expression of a GFP marker was used to define cell boundaries. Noise Reduction (DeepLabv3). Since edge detection is susceptible to noise in the image [10.63], the DeepLabv3 pre-trained model uses a noise reduction algorithm. The first step is to remove the noise in the image with a 5×5 Gaussian kernel which results in spatial smoothing [10.52]. To find the intensity gradient of a given image, smoothed (locally averaged) images are then filtered with a Sobel kernel [10.58] in both horizontal and vertical directions to get the first derivatives in the horizontal direction (Gx) and vertical direction (Gy). From these two images, we can find an edge gradient and direction for each pixel as follows:



G( x ) = Gx2 + G y2

θ A = tan −1

Gy Gx

(10.1) (10.2)

The gradient direction is always perpendicular to edges. It is rounded to one of four angles representing vertical, horizontal and two diagonal directions. Non-maximum Suppression (DeepLabv3). The DeepLabv3 pre-trained model also utilizes non-maximum suppression. After getting gradient magnitude and direction, a full scan of an image is done to remove any unwanted pixels which may not constitute an edge. For this, at every pixel, a pixel is checked if it is a local maximum in its neighborhood in the direction of the algorithmic gradient (see Figure 10.5).

Towards a Digital Diatom  233

C

A

B

C

A

GRADIENT DESCENT

EDGE

B GRADIENT DESCENT

EDGE

Figure 10.5  Point A is on the edge (vertical direction). The gradient direction is normal to the edge. Points B and C are in gradient directions. So, point A is checked with point B and C to see if it forms a local maximum. If so, it is considered for the next stage, otherwise, it is suppressed (set to zero). The result is a binary image with “thin edges.”

Hysteresis Thresholding (DeepLabv3). DeepLabv3 also uses hysteresis thresholding [10.57] for image segmentation (Figure 10.6). Hysteresis refers to the retention of low threshold edges that are associated with high threshold edges. This stage of processing decides which edges in the image are most likely to be actual edges. Hysteresis thresholding relies on two threshold values, Vmin and Vmax. Any edges with intensity gradient more than Vmax are sure to be edges, while those below Vmin are sure to be non-edges and thus discarded. Those that lie between these two thresholds are classified edges or non-edges based on their connectivity. Hysteresis occurs when the number of edges identified by this method are fewer than those defined by Vmin, but greater than those defined by Vmax. The edge A is above the Vmax, so considered as “sure-edge.” Although edge C is below Vmax, it is connected to edge A, so that is also considered as a valid edge and we get that full curve. But edge B, although it is above Vmin and is in the same region as that of edge C, it is not connected to any “sure-edge,” so it is discarded (Figure 10.6). It is very important that we have to select Vmin and Vmax accordingly to get the correct result. This stage also removes small pixels noises on the assumption that edges are long lines and ultimately produces strong edges in the image. Although this is quite an advanced technique, we are still not able to detect the cells when a colony stretches out during its course of the movement (see examples in Figure 10.1).

A C

Vmax

B Vmin

Figure 10.6  Diagram showing an example of hysteresis thresholding and the labeled edge relative to the “sure edge” threshold (Vmax).

234  Diatom Gliding Motility

10.2.3 DeepLabv3 Analysis A number of methods were considered in the course of segmenting images for analysis by the DeepLabv3 model. The results for two of these (watershed and Canny edge detection) are presented and contrasted here. The watershed and Canny edge detection analyses are done in OpenCV (Intel Corporation, Santa Clara, CA). Two methods are presented as a means of comparing performance on static images, and then these methods are contrasted with the deep learning results. DeepLabv3 Model Dataset. Our dataset used as input to the pre-trained model has been extracted from YouTube videos of Bacillaria colonies. The image backgrounds are normalized and rotated to be in horizontal alignment. The data are normalized using a z-score transform (or y-score transform for images with an n < 6). This creates a coordinate space that is based on individual colonies relative to the mean and standard deviation of the full dataset. The full dataset of images and segmented cells (N = 65, n = 810) is also paired down to a dataset of selected samples (N = 46, n = 599). The selected dataset is also analyzed using a principle component analysis (PCA) using Scilab 6.0 [10.37]. Feature Detection Methods. The watershed method [10.6] is based on the concept of geographic watersheds, or drainage basins. The relative contrast of pixels across the image is used to define the watersheds (high-intensity regions) and the boundaries between watersheds (low-intensity regions). This is done by treating image intensities for each pixel as part of an elevation map [10.42], which results in a binary classification of the image. The watershed algorithm is particularly good at finding contours between distinct regions of an image. Thresholding of overall image intensity was done using a marker-based approach (optimization via trial-and-error). Canny edge detection [10.9] operates on a noise-filtered intensity gradient derived from the original image data. As with the watershed method, Canny edge relies on a series of thresholding and filtering techniques to determine the strength of potential edges in the image. One of these is to rely on averaging and signal suppression to classify all potential edges into a set of four angular orientations (0, 45, 90, 135) across a 180-degree arc. Canny edge detection also involves a hysteresis step, which is employed to deal with ambiguous components of the signal relative to an intensity threshold [10.24]. Primary Dataset. We collected microscopy movies of movement for a single Bacillaria colony using light microscopy. These data were converted into still images, which are transformed and segmented using machine learning techniques. Microscopy is conducted using a Zeiss standard upright microscope. All images are brightfield, 40x plain objective. Images are extracted from videos representing 8x time-lapse. Data are collected from a colony transferred to a slide from culture.

10.2.4 Primary Dataset Analysis We will also present an analysis of a primary dataset. Part of this involves masking and segmentation of microscopy images using the OpenDevoCell platform. The other part of this analysis involves using a time-lapse approach to track the motion of cells over time. We are

Towards a Digital Diatom  235 able to approximate patterns of movement and acceleration by identifying labeled features across images over time. OpenDevoCell. We analyzed the primary data using the OpenDevoCell application (https://open-devo-cell.herokuapp.com/). For this analysis, there is a masking step and a segmentation step. OpenDevoCell is a Java-based deep learning platform that uses the TensorFlow library Deeplearning4j. Time-Lapse Analysis. We also analyzed the primary data using time-lapse techniques. The time-lapse is created using VirtualDub, version 1.10.4.35491 (http://www.virtualdub.org/). The Bacillaria strains contained in our primary data (videos) have been harvested from the Neckar river in Germany (49°04’41.8”N 9°09’17.9”E). Samples were collected on September 14, 2019. The average size of each cell (filament) is approximately 81 µm.

10.2.5 Data Availability Select unprocessed (raw) data are available at our GitHub repository (https://github.com/ devoworm/Digital-Bacillaria) processed numeric and image data (numeric tables and skeletonized images), and select video files are available on the Open Science Framework https://github.com/devoworm/Digital-Bacillaria/find/master.

10.3 Results 10.3.1 Watershed Segmentation and Canny Edge Detection Neither the Watershed Segmentation nor the Canny Edge Detection approaches provide very strong performance. The desired feature (closed boundary around each cell) was not detected, as interference from noise in the form of other algae or unclear cell boundaries dominates the analysis. Examples of these results are shown in Figure 10.7.

25

50

Figure 10.7  An example of feature identification performance for the Watershed Segmentation algorithm (left, red boundary) and Canny Edge Detection algorithm (right, white boundary). Image scale: 38.36 μm per cm, or 0.325 μm per pixel. Scale bars 50 µm.

236  Diatom Gliding Motility

10.3.2 Deep Learning The results for the pre-trained model (DeepLabv3) were much more accurate. Based on feature training using the bounding box method demonstrated in Figure 10.8, we are able to reconstruct several parameters of the individual cells which suggest an accurate reconstruction of the source image. We can also see the improved accuracy of the correct performance in Figure 10.9. If the source image is not too blurry (out of focus, dominated by artifacts), the model can be trained and features detected without too much difficulty. An analysis of the pre-trained model outputs is shown in Figures 10.9, 10.10, and 10.11. Figure 10.9 shows the distribution of cell sizes across the full dataset, ranked from smallest to largest. There is a long tail of very large cells (750 to 800 pixels2) that represents filaments much larger than their neighbors. This could be due to errors in the segmentation process of bounding boxes, but could also represent two cells lumped as one, or cell division in the process. Figure 10.10 also uses a rank ordering from largest to smallest instance but breaks Figure 10.9 down into the lengths and widths of each bounding box. In this graph, we see that the length is generally larger than the width as expected, but that there is a more normal distribution of bounding box lengths. This could mean that there is natural variation in the length, but this could also be partially due to resolution loss across the image or the truncation of cells at the edge of the image. Figure 10.11 contains two graphs representing different aspects of the selected dataset. The first relationship (top) is a bivariate plot of all centroid locations among the selected data, similar to the rank-order analysis of bounding box areas shown in Figure 10.9. In the second relationship (bottom), a principal component analysis (PCA) is conducted and represented by plotting the first two principal components. Both plots use a normalized coordinate system (see Methods). Figure 10.12 shows how the pre-trained model can be optimized. In this case, we tried two different optimizations to extract the bounding boxes from random images in the training set. Each optimization (as well as the final segmentation) provides a trade-off between the number of boxes and the accuracy of the segmentation set. The final segmentation included elements of each optimization, which is a trade-off between a greater number of boxes and

25

50

Figure 10.8  An example of feature identification training (purple rectangles) for the deep learning approach on a single set of cells. Notice the resolution of the colony. An example of correct performance is shown in Figure 10.5. Image scale: 38.36 μm per cm, or 0.325 μm per pixel. Scale bar 50 µm.

Towards a Digital Diatom  237 10000 9000

Bounding Box Size (area)

8000 7000 6000 5000 4000 3000 2000 1000 0

0

100

200

300

400

500

600

700

800

900

Rank Order (size)

Figure 10.9  Rank-order analysis of bounding box (cell) sizes (area) across the dataset. The area is measured in pixels squared. Image scale: 38.36 μm per cm, or 0.325 μm per pixel.

350

300

Bounding Box Size (area)

250

200

150

100

50

0

1

41

81

121 161 201 241 281 321 361 401 441 481 521 561 601 641 681 721 761 801 Rank Order (size)

Figure 10.10  Rank-order analysis of height (blue) and width (red) of bounding boxes (cells) across the dataset. The area is measured in pixels squared. Image scale: 38.36 μm per cm, or 0.325 μm per pixel.

an increase in false positives. Even the most optimized result will produce false positives and false negatives due to differences in the resolution of cell boundaries across the image. Although this optimal trade-off between feature number and accuracy was used in our analysis, Figure 10.13 demonstrates how difficult it is to achieve a perfectly accurate segmentation. In Figure 10.13, four examples of segmented images are shown with centroids of the segmented cells plotted in a bivariate coordinate space on the left and the original image on the right. For images a, c, e, and g a normalized coordinate space (see Methods)

238  Diatom Gliding Motility 10000 9000

Bounding Box Size (area)

8000 7000 6000 5000 4000 3000 2000 1000 0

0

100

200

300

400

500

600

700

800

900

Rank Order (size) 600

PC2 (spatial position)

500

400

300

200

100

0

0

100

200

300 PC1 (spatial position)

400

500

600

Figure 10.11  (Top) location of centroids in normalized coordinate space in selected dataset for static analysis. (Bottom) First two principal components from PCA analysis (PC1 represents horizontal position, while PC2 represents vertical position) of coordinates representing the x,y position for all four edges of each bounding box using the selected datasets for static analysis. Image scale: 38.36 μm per cm, or 0.325 μm per pixel.

is used for both the x and y axes. While it is hard to see the degree of concurrency, there is a correspondence between the centroid locations and the cells as stacked in their respective colonies. In general, images such as a and c are harder to resolve than images such as e and g. This is likely due to the configuration of the colony as a “V” (as is the case with e and g) shape versus a more irregular shape. Now we will use the OpenDevoCell model to demonstrate what happens when a model trained for one type of biological system (nematode embryos) is utilized to identify cells in another context (Bacillaria colonies). This analysis does not demonstrate the efficacy of the OpenDevoCell model, rather, it is to show how we might go beyond the bounding box method to more general implementations. Generally, models that learn from data are

Towards a Digital Diatom  239 8

7

6

Frame

5

4

3

2

1

0 0

11

22

33

44

55

66

77

88

95

Boxes

Figure 10.12  An example of feature identification optimization procedures implemented in DeepLabv3. GRAY: no optimization applied, RED: Optimization #1, BLUE: Optimization #2. Given an initial number of training frames (y-axis), the non-optimized procedure (originally detected) will yield a certain number of boxes (x-axis). Applying various optimization procedures generally leads to a decreased number of boxes per frame for both low and high numbers of boxes.

generalized to a specific set of features. Our purpose here is to see if there are any similarities between nematode embryogenesis and diatom functional morphology. The first step in this demonstration is to create skeletons of the original images in order to mimic the resolution of a suitable image for OpenDevoCell, which has been trained to recognize GFPlabeled cell membrane boundaries (between cells). We show the original images along with their corresponding image skeletons in Figure 10.13 using three exemplar images from the primary data described in the Methods section. The pre-masking exercise shown in Figure 10.14 demonstrates how difficult it is to define to distinguish an outer edge versus the contours and internal features of Bacillaria cells. Yet along with the templating analysis featured in Figure 10.15, we demonstrate that intracellular features can be identified and potentially used as quantitative features. Using the DeepLabv3 pre-trained model, we can derive bounding boxes with significant variability in how they map to the original image. In this case, we create image skeletons that are interpretable by the OpenDevoCell model. As the OpenDevoCell model has been optimized for fluorescent images, these image skeletons are bright green and can be separated from both the background and extraneous noise. When these skeletons are presented to the OpenDevoCell model, it can provide spatially referenced segmentation of Bacillaria colonies. These objects can also be co-registered with bounding boxes yielded from the DeepLab analysis. In this case, a model that works quite well for embryos performs more like the unsupervised models shown in Figure 10.7. One reason why the DeepLab model may be more successful at defining bounding boxes in this context could be the ability to recognize generic rectangles, features which are helpful for identification but not descriptive of subtle changes

240  Diatom Gliding Motility (a)

(b)

1 0.8 0.6 0.4

Y

0.2 0 -0.2 -0.4 -0.6 -0.8 -1 -3

-2.5

-2

-1.5

-1

-0.5

0

0.5

1

1.5

2

X

(c)

25

50

25

50

(d)

1 0.5

Y

0 -0.5 -1 -1.5 -2 -1.5

(e)

-1

-0.5

0

0.5

X

1

1.5

1

(f)

0.5 0

Y

-0.5 -1

25

-1.5 -2 0.2

(g)

50

0.3

0.4

0.5

0.6

0.7

X

0.8

0.9

1

1.1

1.2

2.5

(h)

2

Y

1.5 1 0.5 25

0 -0.5 -1.2

-1

-0.8

-0.6

-0.4

X

-0.2

0

0.2

0.4

50

Figure 10.13  Four examples of how the identified features map to two different images (a, c, e, g) of a Bacillaria colony. Points (b, d, f, h) represent the centroids for all bounding boxes identified in images a, c, e, and g, respectively. Image scale: 38.36 μm per cm, or 0.325 μm per pixel. Image scale: 38.36 μm per cm, or 0.325 μm per pixel. Scale bars 50 µm.

Towards a Digital Diatom  241

25

25

25

50

25 50

25

50

25

50

(a)

50

50

(b)

25

50

25

25

50

50

(c)

Figure 10.14  Three examples of how images of a Bacillaria colony are converted into a skeleton image. (Top Row) light microscopy images, (Middle Row) thin skeletonization based on a procedure implemented in GIMP, (Bottom Row) thick skeleton based on a procedure implemented in GIMP. Image scale: 38.36 μm per cm, or 0.325 μm per pixel. Image scale: 38.36 μm per cm, or 0.325 μm per pixel. Scale bars 50 µm.

242  Diatom Gliding Motility in structure or shape. This is particularly relevant in terms of cells that fall partially out of frame or detail within what is captured as a bounding box. While difficult to obtain, the objects segmented by this implementation of OpenDevoCell are more detailed and less square than those extracted by the DeepLab analysis. We propose that a combination of OpenDevoCell and pre-masking may be useful in revealing potential intracellular features. This is particularly true for capturing variations within cells, which allows for these features to be mapped to the bounding box segmentation data. This would provide a nice balance between accuracy and detail, but also yields inconsistent results for features that change position between frames over time. 70

Position (µm)

50 30 10 -10 0

100

200

300

400

500

600

-30 -50

(a)

Time (s)

80

Position (µm)

60 40 20 0 -20

0

200

400

600

800

1000

-40 -60 -80

10 8 6 4 2 0 -2 -4 -6 -8 -10

Velocity (µm/s)

-70

(b)

Time (s)

Velocity (µm/s)

15

10

5

-70

-50

-30

-10

0

Position (µm) 10

30

50

70

-5

-10

-15

(c)

80 60

Position (µm)

40 20 0 -20

0

200

400

600

800

1000

-40 -60 -80

Time (s)

(d)

Figure 10.15  Examples of relative movement of cells in a sample colony. (a) Comparisons between changes of position for cell #2 relative to cell #1 (red) and cell #3 relative to cell #2 (blue). (b) Comparisons between changes of position (red) and changes of velocity (blue) for cell #2 relative to cell #1. (c) a phase diagram of the data shown in b, velocity versus position. (d) comparison between changes of position for cell #2 relative to cell #1 (red) and sine wave (black dashed). The oscillation period in frame c is 62.56 seconds.

Towards a Digital Diatom  243 Given our limitations in acquiring feature/background separation for sequential samples, our final analysis is to apply motion tracking to the Bacillaria colonies across frames. This can be done using template analysis (see Methods). The templating method shown in Figure 10.3 shares commonalities with the DeepLab model in how cells are segmented and represented. In both cases, the centroid approximation of an identified feature is used as a reference point. Yet for purposes of approximating motion over time, the most important information for the analysis of the movement is the positions of individual cells in a Bacillaria colony relative to neighboring diatoms as a function of time. Highlights of our analysis for single-cell motion over time, particularly with respect to neighboring cells, is shown in Figure 10.15. At the level of individual cells, the movement of the colony appears to be oscillatory. Comparisons of cell #2 and cell #3 reveal oscillations that are slightly out-of-phase (Figure 10.15a). As we might expect, changes in velocity for the oscillation of a single cell are noisier than changes in position for that same cell (Figure 10.15b). Yet the velocity function shows a relatively smooth transition between two extremes. When the entire chain comes to rest, we observe a dampening of both the position and acceleration time series. In Figure 10.15c, the absolute value of velocity increases linearly with time until the diatoms in consideration lie next to each other with their apices, then decreases linearly until they slow down. Linear increases and decreases of speed over time means that the curve of the positions is composed of parabolic segments. For this comparison of position and velocity, we are able to replicate the findings of [10.60]. An exception to this are the ranges in the proximity of the reversal points, in which this linearity is not given. In our analysis, the diatoms behave as if velocity is proportional to the length of their common contact surface.

10.4 Conclusion This chapter introduces a new approach to understanding biological processes, more generally and specifically, Bacillaria colony morphology and movement. We employ image processing and machine learning techniques to segment images and extract quantitative parameters. These data can then be used to both infer the phenotypic structure of a colony and movement patterns of these colonies. Of particular interest is the combination of multiple image processing and deep learning techniques with a biomechanical analysis. Hopefully, this will provide guidance towards the future development of digital models. This is the first attempt at characterizing individual colonies and cells using advanced computational techniques. Future refinements of analytical techniques and input data will allow us to build better models, bridging the gap between prior work and emerging methods. In addition, we provide our own innovations to the study of Bacillaria. While our approach provides unique information about the process of colony form and function, we are also in a position to develop and clarify potential theoretical arguments regarding organismal phenotypes, movement, and behavior. Even when the various aspects of feature selection are optimized (as shown in Figure 10.12), false positives are occasionally included in the output. This is true for all the techniques presented here. For example, the segmentation process can introduce errors when the input data is highly variable. Despite our attempt to normalize both input and output

244  Diatom Gliding Motility data, the normalization procedures make certain assumptions about the data that do not fully capture natural variation. More generally, the shape of colonies observed in our input data does not represent every configuration found in nature. While we expect our analyses to be robust to variations on observed shapes and movement dynamics, we cannot assure that this is the case. Manipulations such as tracking and segmentation of diatoms colonies work best when the colony lies in the focal plane perpendicular to the optical axis for a sufficiently long time. This is likely to be the case for both the tracking experiments and the secondary data. Since Bacillaria colonies often move in three dimensions, diatoms are often seen from different perspectives and overlaps may occur. Although it is possible to force the colonies between a slide and a cover glass into this position, the movement is often influenced by adhesion to the confining surfaces. We must also consider the role of hydrodynamic flow in distorting the movement function within and between measurements. Yet, since Bacillaria lives at low Reynolds numbers, it is likely to have minimal impact on the results. In general, analysis is most accurate in cases where small colonies are in continuous contact with the visible surface. These issues might be overcome with a larger and more diverse training set [10.36] [10.51]. The fact that the number of cells in a colony is constant from one frame to the next (unless a cell division has occurred) has not yet been incorporated. The lack of interpolation for cells that are truncated by the edge of the microscopy image frame is another issue. While this method is able to generate useful quantification of the image set, we still arrive at a number of issues with mapping the functional phenotype of a Bacillaria colony to a digital representation. This is a meta-issue when compared to false-positive identification, and so may require a new concept to characterize the relative imprecision of the mapping between image and digital representation. Better methodology with respect to the feature space (more subtle components of colony morphology) would also be a helpful future advance. This study also makes several contributions to both computational biology and machine learning literature. The first is to create a digital model of several parameters that are implemented by an existing general-purpose pre-trained model. This has largely been successful and has allowed us to extract several key parameters that describe the phenotype. We are also able to compare these results with models trained for other specific biological tasks. This was less successful but does guide us toward future work. This digital model will be made available as an open-access model and can be updated with improved data and methods. Another contribution is to create a dynamic model of the Bacillaria that will allow us to predict movement and other deformations of the phenotype. This will help us characterize not only modes of movement, but also any potential collective behaviors that require coordinated decision-making between cells. A secondary theme of our study are cases in which deep learning techniques succeed or fail at capturing the desired features. In the Results section, unsupervised techniques are ruled out as inadequate, while the more technically sophisticated deep learning techniques are shown to yield results of varying utility. Of particular interest is the level of generalizability for such models. The optimal model should not be limited by biological variability, but should also be able to identify uniquely biological features, even when they mimic mechanical features. One solution to this are toy models, which can be used to capture complex processes in a simplified and mathematically tractable model [10.1]. Results of the comparison between the DeepLabv3 and OpenDevoCell models suggest the need for a specialized pre-trained model [10.41] optimized for the shape, movement patterns,

Towards a Digital Diatom  245 and intracellular contours of a Bacillaria colony (see Figure 10.2). Specialized pre-trained models have been created for a host of specific types of systems, such as linguistic and object recognition and transfer, so creating a model specialized for the analysis of dynamic biological systems is both desirable and attainable.

Acknowledgments The authors would like to thank members of the DevoWorm group, in particular, the two working groups within DevoWorm (Digital Bacillaria and DevoWormML) that made this study happen. Thanks also go to the diatom community for access to movies and other information on an otherwise obscure organism, and the open-source software community for developing the tools used in the presented analyses. Special thanks go to the Google Summer of Code program for funding and enabling a portion of this work. Additional thanks go to Matt Ashworth for providing additional Bacillaria paradoxa specimens from Florida, USA.

References [10.1] Alicea, B. and Gordon, R., Toy Models for Macroevolutionary Patterns and Trends. Biosystems, 122, 25–37, 2014. [10.2] Angermueller, C., Pärnamaa, T., Parts, L., Stegle, O., Deep learning for computational biology. Mol. Syst. Biol., 12, 7, 878, 2016. [10.3] Annenkov, V., Seckbach, J., Gordon, R. (Eds.), Diatom Morphogenesis [DIMO, Volume in the series: Diatoms: Biology & Applications, series editors: Richard Gordon & Joseph Seckbach, In preparation], Wiley-Scrivener, Beverly, MA, USA, 2019. [10.4] Bentley, K., Clack, C., Cox, E.J., Diatom colony formation: A computational study predicts a single mechanism can produce both linkage and separation valves due to an environmental switch. J. Phycol., 48, 3, 716–728, 2012. [10.5] Bentley, K., Cox, E.J., Bentley, P.J., Nature’s batik: A computer evolution model of diatom valve morphogenesis. J. Nanosci. Nanotechnol., 5, 1, 25–34, 2005. [10.6] Bertrand, G., On topological watersheds. J. Math. Imaging Vision, 22, 2-3, 217–230, 2005. [10.7] Bradbury, J., Nature’s nanotechnologists: Unveiling the secrets of diatoms. PloS Biol., 2, 10, 1512–1515, 2004. [10.8] Bradbury, J., Nature’s nanotechnologists: Unveiling the secrets of diatoms, in: Handbook of Nanotechnology, E. Preston (Ed.), pp. 16–23, Palm Bay, FL., Apple Academic Press, 2010. [10.9] Canny, J., A computational approach to edge-detection. IEEE Trans. Pattern Anal. Mach. Intell., 8, 6, 679–698, 1986. [10.10] Chen, L.-C., Papandreou, G., Kokkinos, I., Murphy, K., Yuille, A.L., DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs. IEEE Trans. Pattern Anal. Mach. Intell., 40, 4, 834–848, 2018. [10.11] Ching, T., Opportunities and obstacles for deep learning in biology and medicine. J. R. Soc. Interface, 15, 141, 2018. [10.12] Cox, E.J., Willis, L., Bentley, K., Integrated simulation with experimentation is a powerful tool for understanding diatom valve morphogenesis. BioSystems, 109, 3, Special Issue on Biological Morphogenesis, 450–459, 2012. [10.13] DeepLearning4J Documentation, URL: https://deeplearning4j.org/docs/latest/, Accessed 11/21/2019.

246  Diatom Gliding Motility [10.14] Drum, R.W., Gordon, R., Bender, R., Goel, N.S., On weakly coupled diatomic oscillators: Bacillaria’s paradox resolved. J. Phycol., 7, Suppl., 13–14, 1971. [10.15] Edgar, L.A., Diatom locomotion: computer assisted analysis of cine film. Br. Phycol. J., 14, 83–101, 1979. [10.16] Ghobara, M.M., Mazumder, N., Vinayak, V., Reissig, L., Gebeshuber, I.C., Tiffany, M.A., Gordon, R., On light and diatoms: A photonics and photobiology review [Chapter 7], in: Diatoms: Fundamentals & Applications [DIFA, Volume 1 in the series: Diatoms: Biology & Applications, series editors: Richard Gordon & Joseph Seckbach], J. Seckbach and R. Gordon (Eds.), pp. 129–190, Wiley-Scrivener, Beverly, MA, USA, 2019. [10.17] GIMP Documentation Team, GNU Image Manipulation Program User Manual, 2019, URL: https://docs.gimp.org/2.10/en/, Accessed 11/21/2019. [10.18] Gordon, R., Partial synchronization of the colonial diatom Bacillaria “paradoxa”. Res. Ideas Outcomes (RIO), 2, e7869, 2016. [10.19] Gordon, R., The whimsical history of proposed motors for diatom motility, in: Diatom Gliding Motility [DIGM, Volume in the series: Diatoms: Biology & Applications, series editors: Richard Gordon & Joseph Seckbach], S.A. Cohn, K.M. Manoylov, R. Gordon (Eds.), WileyScrivener, Beverly, MA, USA, 2019, In preparation. [10.20] Gordon, R. and Aguda, B.D., Diatom morphogenesis: Natural fractal fabrication of a complex microstructure, in: Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, Part 1/4: Cardiology and Imaging, 4-7 Nov. 1988, New Orleans, LA, USA, Institute of Electrical and Electronics Engineers, New York, pp. 273–274, 1988. [10.21] Gordon, R. and Drum, R.W., The chemical basis of diatom morphogenesis. Int. Rev. Cytol., 150, 243–372, 421-422, 1994. [10.22] Gordon, R., Losic, D., Tiffany, M.A., Nagy, S.S., Sterrenburg, F.A.S., The Glass Menagerie: Diatoms for novel applications in nanotechnology. Trends Biotechnol., 27, 2, 116–127, 2009. [10.23] Grachev, M.A., Annenkov, V.V., Likhoshway, Y.V., Silicon nanotechnologies of pigmented heterokonts. BioEssays, 30, 4, 328–337, 2008. [10.24] Green, B., Canny Edge Detection Tutorial, http://dasl.mem.drexel.edu/alumni/bGreen/ www.pages.drexel.edu/_weg22/can_tut.html, 2002. [10.25] Gutiérrez, A., Gordon, R., Dávila, L.P., Deformation modes and structural response of diatom shells. J. Mater. Sci. Eng. Adv. Technol., 15, 2, 105–134, 2017. [10.26] He, K.M., Zhang, X.Y., Ren, S.Q., Sun, J., Ieee, Deep residual learning for image recognition, in: 2016 IEEE Conference on Computer Vision and Pattern Recognition, IEEE, New York, pp. 770–778, 2016. [10.27] Heintzelman, M.B., Cellular and molecular mechanics of gliding locomotion in eukaryotes. Int. Rev. Cytol., 251, 79–129, 2006. [10.28] Hoo-Chang, S., Roth, H.R., Gao, M., Lu, L., Xu, Z., Nogues, I., Yao, J., Mollura, D., Summers, R.M., Deep Convolutional Neural Networks for Computer-Aided Detection: CNN Architectures, Dataset Characteristics and Transfer Learning. IEEE Trans. Med. Imaging, 35, 5, 1285–1298, 2016. [10.29] Jahn, R. and Schmid, A.-M.M., Revision of the brackish-freshwater diatom genus Bacillaria Gmelin (Bacillariophyta) with the description of a new variety and two new species. Eur. J. Phycol., 42, 3, 295–312, 2007. [10.30] Kapinga, M.R.M. and Gordon, R., Cell motility rhythms in Bacillaria paxillifer. Diatom Res., 7, 2, 221–225, 1992. [10.31] Lobel, K.D., West, J.K., Hench, L.L., Computational model for protein mediated biomineralization of the diatom frustule. Mar. Biol., 126, 3, 353–360, 1996. [10.32] Losic, D. (Ed.), Diatom Nanotechnology: Progress and Emerging Applications, Royal Society of Chemistry, London, 2018.

Towards a Digital Diatom  247 [10.33] Mann, D.G., Kelly, M., Jüttner, I., Bacillaria paxillifera (O.F.Müller) T.Marson; 1901; 254, in: Freshwater Diatom Flora of Britain and Ireland, I. Jüttner, H. Bennion, C. Carter, E.J. Cox, L. Ector, R. Flower, V. Jones, M.G. Kelly, D.G. Mann, C. Sayer, Turner J.A. (Eds.), Cardiff, Wales, UK, Amgueddfa Cymru - National Museum Wales, 2019, https://naturalhistory. museumwales.ac.uk/diatoms/browsespecies.php?-recid=4557. [10.34] McKay, C., How to do Card Tricks: How to Spread & Turnover a Deck of Playing Cards [movie], https://www.youtube.com/watch?v=UuuHeKckPfI, 2007. [10.35] Mishra, M., Arukha, A.P., Bashir, T., Yadav, D., Prasad, G., All new faces of diatoms: Potential source of nanomaterials and beyond. Front. Microbiol., 8, 1239, 2017. [10.36] Moen, E., Bannon, D., Kudo, T., Graf, W., Covert, M., Van Valen, D., Deep learning for cellular image analysis. Nat. Methods, 16, 1233–1246, 2019. [10.37] Nagar, S., Introduction to Scilab: For Engineers and Scientists, APress, New York, NY, USA, 2017. [10.38] Pappas, J.L., Tiffany, M.A., Gordon, R., The uncanny symmetry of some diatoms and not of others: A multi-scale morphological characteristic and a puzzle for morphogenesis [DUNC], in: Diatom Morphogenesis [DIMO, Volume in the series: Diatoms: Biology & Applications, series editors: Richard Gordon & Joseph Seckbach], V. Annenkov, J. Seckbach, R. Gordon (Eds.), Wiley-Scrivener, Beverly, MA, USA, 2019, in press. [10.39] Parkinson, J., Brechet, Y., Gordon, R., Centric diatom morphogenesis: A model based on a DLA algorithm investigating the potential role of microtubules. Biochim. Biophys. Acta – Mol. Cell Res., 1452, 1, 89–102, 1999. [10.40] Poulsen, N.C., Spector, I., Spurck, T.P., Schultz, T.F., Wetherbee, R., Diatom gliding is the result of an actin-myosin motility system. Cell Motil. Cytoskeleton, 44, 1, 23–33, 1999. [10.41] Rajaraman, S., Antani, S.K., Poostchi, M., Silamut, K., Hossain, M.A., Maude, R.J., Jaeger, S., Thoma, G.R., Pre-trained convolutional neural networks as feature extractors toward improved malaria parasite detection in thin blood smear images. PeerJ, 6, e4568, 2018. [10.42] Rosebrock, A., Watershed OpenCV, https://www.pyimagesearch.com/2015/11/02/ watershed-opencv/, 2015. [10.43] Round, F.E., Crawford, R.M., Mann, D.G., The Diatoms: Biology and Morphology of the Genera, Cambridge University Press, Cambridge, UK, 1990. [10.44] Sabater, S., Diatoms, in: Encyclopedia of Inland Waters, K. Tockner and G.E. Likens (Eds.), pp. 149–156, Amsterdam, Netherlands, Elsevier Science, 2009. [10.45] Sabuncu, A.C., Gordon, R., Richer, E., Manoylov, K.M., Beskok, A., The kinematics of explosively jerky diatom motility: A natural example of active nanofluidics [JRKY], in: Diatom Gliding Motility [Volume 3 in the series: Diatoms: Biology & Applications, series editors: Richard Gordon & Joseph Seckbach], S.A. Cohn, K.M. Manoylov, R. Gordon (Eds.), WileyScrivener, Beverly, MA, USA, 2019, In press. [10.46] Sarma, G., Lee, C.W., Portegys, T., Ghayoomi, V., Alicea, B., Cantarelli, M., Palyanov, A., Khayrulin, S., Idili, G., Gleeson, P., Gordon, R., Watts, M., Hasani, R., Lung, D., Currie, M., Jacobs, T., Gingell, S., Gerkin, R.C., Larson, S.D., Openworm: Overview and recent advances in integrative biological simulation of Caenorhabditis elegans. Philos. Trans. R. Soc. B, 373, 20170382, 2018. [10.47] Schmid, A.M.M., The “paradox” diatom Bacillaria paxillifer (Bacillariophyta) revisited. J. Phycol., 43, 1, 139–155, 2007. [10.48] Schultze, M.J.S., Die Bewegung der Diatomeen/The movement of diatoms. Schultze‘s Arch. Mikrosk. Anat., 1, 376–402, 1865. [10.49] Seckbach, J. and Gordon, R. (Eds.), Diatoms: Fundamentals & Applications [DIFA, Volume 1 in the series: Diatoms: Biology & Applications, series editors: Richard Gordon & Joseph Seckbach], Wiley-Scrivener, Beverly, MA, USA, 2019.

248  Diatom Gliding Motility [10.50] Singh, U., Singh, A., Alicea, B., TensorFlow Implementation using Pre-trained Models (MobileNet/DeepNet v3), https://github.com/devoworm/Digital-Bacillaria/blob/master/Tutorials/ TensorFlow-using-DeepLab.md, 2019. [10.51] Tang, B., Pan, Z., Yin, K., Khateeb, A., Recent Advances of Deep Learning in Bioinformatics and Computational Biology. Front. Genet., 10, 214, 2019. [10.52] Trabelsi, A. and Savaria, Y., A 2D Gaussian smoothing kernel mapped to heterogeneous platforms. IEEE 11th International New Circuits and Systems Conference, 2013. [10.53] Tsang, S.-H., Review: DeepLabv3 — Atrous Convolution (Semantic Segmentation). Toward Data Science Medium blog, January 19, https://towardsdatascience.com/reviewdeeplabv3-atrous-convolution-semantic-segmentation-6d818bfd1d74, 2019. [10.54] Ussing, A.P., Gordon, R., Ector, L., Buczkó, K., Desnitskiy, A.G., VanLandingham, S.L., The Colonial Diatom “Bacillaria paradoxa”: Chaotic Gliding Motility, Lindenmeyer Model of Colonial Morphogenesis, and Bibliography, with Translation of O.F. Müller (1783), “About a peculiar being in the beach-water” [BPAR], p. 5, A.R.G. Ganter Verlag Kommanditgesellschaft, Ruggell, 2005. [10.55] Uthappa, U.T., Brahmkhatri, V., Sriram, G., Jung, H.Y., Yu, J.X., Kurkuri, N., Aminabhavi, T.M., Altalhi, T., Neelgund, G.M., Kurkuri, M.D., Nature engineered diatom biosilica as drug delivery systems. J. Controlled Release, 281, 70–83, 2018. [10.56] Webb, S., Deep learning for biology. Nature, 554, 7693, 555–557, 2018. [10.57] Wikipedia, Edge detection, https://en.wikipedia.org/wiki/Edge_detection, 2019. [10.58] Wikipedia, Sobel operator, https://en.wikipedia.org/wiki/Sobel_operator, 2019. [10.59] Willis, L., Page, K.M., Broomhead, D.S., Cox, E.J., Computational models of the formation of diatom pore occlusions. Plant Ecol. Evol., 143, 3, 297–306, 2010. [10.60] Yamaoka, N., Suetomo, Y., Yoshihisa, T., Sonobe, S., Motion analysis and ultrastructural study of a colonial diatom, Bacillaria paxillifer. Microscopy (Oxford, England), 65, 3, 211– 221, 2016. [10.61] Yang, H., Shao, L., Zheng, F., Wang, L., Song, Z., Recent advances and trends in visual tracking: a review. Neurocomputing, 74, 18, 3823–3831, 2011. [10.62] Zador, A.M., A critique of pure learning and what artificial neural networks can learn from animal brains. Nat. Commun., 10, 3770, 2019. [10.63] Ziou, D. and Tabbone, S., Edge detection techniques-an overview. Pattern Recognition and Image Analysis C/C of Raspoznavaniye Obrazov I Analiz Izobrazhenii, vol. 8, pp. 537–559, 1998.

11 Diatom Triboacoustics Ille C. Gebeshuber1*, Florian Zischka1, Helmut Kratochvil2, Anton Noll3, Richard Gordon4,5 and Thomas Harbich6 Institute of Applied Physics, Vienna University of Technology, Vienna, Austria, Europe 2 Department of Integrative Zoology, University of Vienna, Austria, Europe 3 Austrian Academy of Sciences, Acoustics Research Institute, Vienna, Austria 4 Gulf Specimen Marine Laboratory and Aquarium, Panacea, Florida, USA 5 C.S. Mott Center for Human Growth and Development, Department of Obstetrics and Gynecology, Wayne State University, Detroit, Michigan, USA 6 Independent Researcher, Am Brüdenrain, Weissach im Tal, Germany 1

Abstract

The aim of this work was to develop a method to record low-level sounds underwater in order to listen to possible sounds related to the gliding movement of raphid, motile diatoms, inspired by their jerky, high acceleration movements. Different techniques concerning the gathering and handling of diatoms and the possibilities of recording sounds related to their movement are presented. A model was created to get a rough estimation of the expansion speed of mucopolysaccharide filaments. In a series of initial experiments, a hydrophone was used to get an idea of the acoustic situation. Furthermore some attempts to increase the density of raphid diatoms in a given volume were made. Though with these rough measurements no sounds could be detected, alternatives and advice on how to improve the experiment for future research are provided. Keywords:  Diatom, pennate, benthic, locomotion, tribology, acoustics, hydrophone, snail

Glossary accretion Accumulation of material. bulk modulus K  Describes the change of pressure that is necessary to cause a change of volume of a body. *Corresponding author: [email protected] Ille C. Gebeshuber: [email protected], https://orcid.org/0000-0001-8879-2302, http://www.ille.com Florian Zischka: [email protected], https://www.researchgate.net/profile/Florian_Zischka2 Helmut Kratochvil: [email protected], https://www.nature.com/articles/srep44526, https:// zoology.univie.ac.at/people/staff/helmut-kratochvil/, https://hekratochvil.hpage.com/wissenschaftliches.html Anton Noll: [email protected], https://www.researchgate.net/profile/Anton_Noll Richard Gordon: [email protected] Thomas Harbich: [email protected], https://www.researchgate.net/profile/Thomas_Harbich, https:// diatoms.de/en/ Stanley Cohn, Kalina Manoylov and Richard Gordon (eds.) Diatom Gliding Motility, (249–282) © 2021 Scrivener Publishing LLC

249

250  Diatom Gliding Motility chemotaxis  Movement or orientation of organisms caused by chemical stimulus. diatom  Single-celled alga that forms an outer shell out of hydrated silicon dioxide. ephemeral Volatile, vanishing quickly. epipelic  Residing at the interface of water and sediments (mud, clays and silt). helictoglossa Internal, distal termination of the raphe. hydrophone  The underwater equivalent of a microphone. Used for recording sounds underwater. inertial force Resistance of an object to change in its velocity. millidyne A unit of force. 1 mdyne = 10−8 N mucopolysaccharide  Acid polysaccharide that protrudes from the raphe in the form of mucus filaments. They attach to the substratum and flow along the raphe. Through that the motive force for diatom locomotion could be generated. pennate  Regarding diatoms, species that form usually bilaterally symmetric shells typically elongated parallel to the raphes. Many pennate diatoms can use the raphe to move along a solid substrate though there are also pennate diatoms without a raphe (araphid). photophobia  When organisms react strongly to changes in light intensity, ­avoiding light. phototaxis Movement or orientation of organisms caused by light stimulus. plasmalemma  Cell membrane that separates the interior of a cell from the outside environment. protoplast  Plant cell composed of nucleus, cytoplasm and plastids without a cell wall. raphe  A slit in the shell of diatoms which is connected to diatom motility. raphe-sternum  Thickened silica typically located along the apical axis of diatoms. Contains the raphe. Reynolds number The Reynolds number is used for flow patterns in different fluid flow situations. At a low Reynolds number the flow is dominated by laminar flow, at a high Reynolds number turbulence occurs. The Reynolds number is the ratio of inertial force to viscous force:



Re =

inertial force v ρl = viscous force η

sound pressure The variations of pressure of a medium that occur due to propagation of sound waves through that medium, because sound waves are pressure waves. triboacoustics  The phenomenon of noise generated by friction, lubrication and wear.

Diatom Triboacoustics  251 vibrometer  A measuring instrument for quantifying mechanical oscillation. Interference of laser light is used to measure the frequency and amplitude of an oscillation. viscous force Resistive force on an object inside of a fluid due to the friction between the layers of the fluid.

11.1 State-of-the-Art 11.1.1 Diatoms and Their Movement Diatoms are single-celled algae with an outer shell made from hydrated silicon dioxide. There are up to 200,000 different species in all kinds of forms and shapes. Raphid, motile diatoms - as the name suggests - have slits called raphes in their shells [11.28]: “The raphe, as an organelle for the motion of the cells is commonly found in many pennate diatoms, but it occurs in very different structures. [...] Basically the raphe is a gap-shaped breach of the cell wall of more or less complexity.” ([11.28], translated from German)

The raphe allows the diatoms to stick to surfaces and move along them. Previous work of Harper and Harper [11.26], where the adhesive forces of diatoms sticking on substrates as well as the tractive forces in the direction of motion were measured, showed that there is clearly a close relation between locomotion and adhesion. “In a total of over 500 observations, whenever a diatom moved it was adhering to the substrate.” The amount of force is heavily dependent on the species considered and can reach from few to several hundreds of millidynes. “However, strong adhesion does not prevent movement: Amphora ovalis (Kützing, 1844) cells were able to move normally while exerting adhesions of over 400 millidynes” [11.26]. “There are also diatoms that do not separate after asexual reproduction, but adhere together and form chain-like colonies” [11.25]: Bacillaria paxillifer (O.F. Müller) Hendy (1951) even forms motile chains that can move around through water [11.1]. Sizes of diatoms range from 4–5 µm to up to 500 µm. “The highest speeds of locomotion occur in tidewater-diatoms. The maximum speed of Navicula radiosa (Kützing, 1844) amounts to 20 µm/s. Pinnularia nobilis (Ehrenberg, 1843) was able to cover a distance of 14 mm in 20 min, which equals almost 12 µm/s.” ([11.28], translated from German). The gliding movement of raphid, motile diatoms started to be the subject of research over 200 years ago [11.42] and over time various theories that tried to explain the mechanism were developed [11.22]. “According to some, the raphe is occupied by streaming cytoplasm, others proposed small flagella that protrude through the raphe slits” [11.36]. “Recent ideas have in common that the movement of the cell relative to the sub-stratum is considered to be mediated by the secretion of material from the raphe” [11.36]. In a model by Gordon, it is proposed that the hydration of mucopolysaccharide, the material the raphe is filled with and which is left behind as a mucilage trail, would provide sufficient motive force to explain the gliding motility of raphid diatoms [11.24] [11.21]. “Diatom movement appears to be smooth over short periods of time between reversals or stopping, but is in fact jerky, sudden accelerations and decelerations alternating with periods when the diatom is stationary or moving with constant velocity” [11.36].

252  Diatom Gliding Motility Edgar [11.15] analyzed the speed and acceleration of several species of epipelic river diatoms using data from motion picture films of moving cells, examined frame-by-frame, and showed that large changes of speed occur within one tenth of a second, which is to be expected at that scale, because of the small inertial forces. Therefore, high speed cinematography is necessary to study the movements [11.15]. However, even at 890 frames per second, large accelerations between frames still occurred, pushing the limits of ordinary light microscopy [11.37]. Therefore, new approaches are needed to reach an understanding of what causes these huge, short-term accelerations, which is why we turned to triboacoustics. Considerations of the jerky movement of a diatom in a highly viscous situation were made by Edgar [11.16]. Therefore, the effect of external forces on a moving diatom were discussed. The Reynolds number (Re) is the expression of the ratio of inertial force to viscous force.



Re =

inertial force v ρl = viscous force η

(11.1)

where: v = velocity, ρ = density of the fluid, l = size of body, η = viscosity of the fluid [11.16].

11.1.2 The Navier-Stokes Equation The Navier-Stokes equation is the general equation of motion for the volume element dV of a viscous, flowing fluid. It results from Newton’s second law, the basic equation of motion in classical mechanics: F = m · a The equation of motion for one mass element ∆m = ∙ ∆V of a flowing medium is:



F = Fp + Fg + FR = ∆mr = ρ ⋅ ∆V ⋅

du dt

(11.2)

dr where u = is the velocity of flow of the volume element dV [11.7]. dt With the terms: dFR = η∆udV dFp = −grad(p) · dV dFg = ρgdV

(force of friction) (force of pressure) (gravity)

of the single forces and the acceleration:



du ∂ u = + (u ⋅∇)u dt ∂ t

(11.3)

the equation of motion becomes the Navier-Stokes equation:



∂  ρ  + u ⋅∇  u = − grad p + p ⋅ g + ηDu  ∂t 



(11.4)

For ideal liquids (η = 0) it becomes the Euler equation. The friction term η∆u turns the Euler equation (differential equation of first order) into an equation of second

Diatom Triboacoustics  253 order and thereby complicates solving it. On the right side of the Navier-Stokes equation there are the forces and on the left side the movement caused by those forces [11.7]. All dimensions of length can be scaled to one standard length l, all times to one standard time T and then all velocities u can be expressed as functions of l/T:





t = t ′ ⋅T ∇′ ∇= L

u = u′ ⋅

l T 2

 l p = p′ ⋅ ⋅ρ T

 ∂ ∂ ∂ where t′, u′, ∇′ = l ⋅  and p′ are nondimensional quantities. , ,  ∂ x ∂ y ∂ z  Through that, the Navier-Stokes equation (without the gravity term) becomes:



∂u′ 1 + (u′ ⋅ ∇′)u′ = −∇′p′ + ∆′u′ Re ∂t ′

(11.5)

with the nondimensional Reynolds number:



Re =

ρ ⋅l2 ρ ⋅v ⋅l = η ⋅T η

(11.6)

v = l/T has the dimension of a velocity. It defines the velocity of flow averaged over the length l. In ideal liquids η = 0 and therefore Re = ∞. In fluid dynamics that means that for viscous liquids with η = 0, currents are only similar when they take place in vessels with similar ratio of dimension and when they have the same Reynolds number Re. [11.7].

11.1.3 Low Reynolds Number The following words from the wonderful publication Life at Low Reynolds Number by E.M. Purcell [11.33] lead to a better understanding of the meaning of situations at very low Reynolds number: “The Reynolds number for a man swimming in water might be 104. For a goldfish it might get down to 102. At very low Reynolds number of about 10-4 or 10-5 inertia is totally irrelevant. [...] As an example an animal of about a micron (= 1 µm) in size may move through water, where the kinematic viscosity is 10-2 cm/s, at a typical speed of 30 m/s. If the driving force for the movement of that animal suddenly ceases, it will only coast for about 0.1 Å and it takes about 0.6 s to slow down. This makes clear what low Reynolds number means. Inertia plays no role whatsoever. If you are at very low Reynolds number, what you are doing at the moment is entirely determined by the forces that are exerted on you at that moment, and by nothing in the past” [11.33].

254  Diatom Gliding Motility Also, according to Purcell [11.33], at low Reynolds number a living being can’t shake off its environment. “If it moves, it takes it along; it only gradually falls behind. [...] In that context diffusion is very important, because at low Reynolds number stirring isn’t very good.” Purcell also showed that the transport of wastes away from the animal and food to the animal is entirely controlled locally by diffusion: “It can thrash around a lot, but the fellow who just sits there quietly waiting for stuff to diffuse will collect just as much.” Also, an increased velocity of the moving animal is not beneficial for gaining more nutrients: “To increase its food supply by 10% it would have to move at a speed of 700 µm/sec, which is 20 times as fast as it can swim. The increased intake varies like the square root of the bug’s velocity so the swimming does no good at all in that respect. But what it can do is find places where the food is better or more abundant.” Therefore, it has to move far enough to outrun diffusion. At typical diffusion constant D and speed v that minimum distance to outswim diffusion D/v is about 30 µm. It has been shown that this is just about what swimming bacteria were doing [11.33].

11.1.4 Reynolds Number for Diatoms According to Edgar, the Reynolds number of a diatom 10 · 10 · 100 μm3 in volume, moving at 10 µm/s is in the region of 10−4, which is very low. “A low Reynolds number ( 1 (see [11.8]). The water is therefore not as densely packed in the bound state as in the bulk of the water. Eq. (11.11) is used (see Deng et al. [11.8]) for the case of complete hydration, so that V means the maximum volume (V = VM). Obviously, the equation is also correct before the beginning of hydration (V = Vd). For volumes between the extremes, Eq. (11.11) would also apply, assuming that the inflowing water locally leads to complete hydration, so that an outer fully hydrated region and an inner water-free region exist. Since this cannot be assumed in principle, Eq. (11.11) is only considered to be a linear interpolation for all volumes. For the volume of water it holds, Vw = NVM, where VM is the volume of one mole of liquid water. From Eq. (11.11) it follows:

N=



V − Vd α VM

(11.12)

Using Eq. (11.10) we get:



V − Vd  d  V = α VM Ak  cs −  dt α VMV 

(11.13)

An approximation was used that Vd is not time-dependent and that c = N/V. Until now, the geometric shape of the fibril was not explicitly taken into account. It is assumed that

260  Diatom Gliding Motility the cylinder has a radius of r and the height of h. Water absorption should only take place through the curved surface area and expansion should only take place radially. If you insert V = r2πh, Vd = r2πh and A = 2rπh in Eq. (11.12), you get an equation with r as the only time-dependent variable (dV/dt = Adr/dt):



v=

 d r 2 − rd2  r = k  α VM cs −   r2  dt

(11.14)

If the radius of the cylinder without water content rd and at saturation with water rd are used as model parameters, cs is not an independent parameter. From Eq. (11.12) we get results for saturation with water (V = Vm) and with cs = N/V (N total amount of water absorbed):

rm2 − rd2 cs = α VM rm2



(11.15)

If rd ≪ rm, then cs ≈ 1/(αVm) is valid as expected. So far it has not been determined in which state the expansion will start. It should be assumed that the fibril initially contains no water, so that the start condition at time t = 0 is r0 = r(0) = rd and V0 = V(0) = Vd. With this definition and with Eq. (11.15), Eq. (11.14) it is taking the form:



v=

 r2 r2  d r = k  02 − 02  r rm  dt

(11.16)

The initial velocity v(0) results with r(0) = r0 from Eq. (11.16) giving



v(0) = k

rm2 − r02 rm2

(11.17)

The calculation of the quantity v(0) does not yet make use of the assumption of homogeneous water density in the fibril over the period of expansion, because this homogeneity is given at the beginning. A reaction-diffusion equation should also provide this initial value. In [11.8] k = 2.3 ·10–5 m/s is used (adopted from [11.41]). Using the model parameters for rm and r0 (Table 11.1) it follows from Eq. (11.17):

v(0) = k · 0.96 ≈ 2.21 · 10−5 m/s

(11.18)

If this speed would last until reaching the maximum radius (12.5 nm), this radius would be reached in 0.57 ms. This is very long compared to the estimated duration of the ejection, so that almost the entire expansion of the fibril takes place after the ejection. However, the form of Eq. (11.16) shows that the velocity decreases very rapidly with the radius. In the above approximation, at twice the radius of the starting value r0 it is only 1/4 of the speed at the start.

Diatom Triboacoustics  261 Eq. (11.16) can be analytically integrated after separation of variables. The inverse function t(r) reads:

t = k−1



rm2  rm rm + r  ln − C0   r02  2 rm − r

(11.19)

with the integration constant C0:

C0 = −r0 +



rm  rm + r0  ln 2  rm − r0 

(11.20)

Numerical Results For the volume, the length of the fibril h is also needed (Table 11.1). From the range 0.3−3 μm the value 3 µm was chosen in order to describe the optimum case of observability. In Figure 11.1, the graphs according to Eq. (11.19) show the expected qualitative behavior: In the selected time period, after a steep ascent, a slow increase of the radius is seen. The initial radius is doubled in approx. 0.284 ms. The doubling of the surface ( 2r0 ) is already achieved after approx. 0.071 ms. The fast decrease of the speed with increasing radius leads to the fact that the 10-fold initial radius is never reached. As the radius asymptotically approaches the maximum radius rm, this radius is never matched. As a measure for the duration of the expansion, you can define, for example, after which time 90% of the radial expansion is completed. If the characteristic radius rk is defined by (rk – r0)/(rm – r0) = 0.9 the characteristic radius is reached after 9.05 · 10−3 seconds. For the question of sound generation, it is important that rapid expansion only occurs as long as the radius of the fibril is a few nm.

11.2.2.2 Sound Generation Model Assumption One of the main assumptions was that the fibrils are ejected at an enormous speed and therefore their expansion along their entire length starts as soon as they are in the water 1.6E-21

1.2E-08

1.4E-21 1.2E-21 Volume (m3)

Radius (m)

1.0E-08 8.0E-09 6.0E-09 4.0E-09

8.0E-22 6.0E-22 4.0E-22

2.0E-09 0.0E+00 0

1.0E-21

2.0E-22 0.005

0.01

0.015 0.02 Time (m)

0.025

0.03

0.0E+00 0

(a) Expansion of the radius

Figure 11.1  Expansion of a cylindrical mucopolysaccharide fibril.

0.005

0.01

0.015 0.02 Time (m)

0.025

(b) Expansion of the volume

0.03

262  Diatom Gliding Motility body. Apart from a possible sound development due to the ejection itself, the sound is generated by the expansion of the cylindrical fibril. This expansion starts with an initial velocity of v(0) = 0.96 · k. With the selected model parameters, the expansion initially occurs at 2.21 · 10−5 ms−1. After that, the speed drops very rapidly. Therefore, a wide frequency spectrum can be expected. Theoretically it is infinitely broad according to the model; because the movement starts immediately, practically an infinitely steep edge is not to be expected in nature. In the following estimation we therefore work with the assumption that the wavelengths of the sound are large compared to the expansion of the sound-emitting body. A wave of the wavelength of λ corresponds to a frequency of cL/λ, where cL is the speed of sound in water (about 1500 m/s). If we would consider frequencies where the wavelengths are in the range of the length of the fibril (0.3 to 3 µm [24]), we would also have to include frequencies in the range of 2.5 · 108 to 2.5 · 109 hertz. Velocity Potential and Boundary Condition When using a description of the sound by a velocity potential Φ, so that the velocity of the particles in the wave vs is given by the gradient of this scalar potential

vs = ∇Φ



(11.21)

the following equation must be solved:

DΦ −



1 ∂2Φ =0 cL2 ∂t 2

(11.22)

Considering an expanding body with a solid surface, the normal component of the liquid velocity at the surface must be equal to the normal component of the velocity of the body at the surface [11.31]:

∂Φ =v ∂n



(11.23)

In the case of the expanding fibril, however, water from the water body flows into the fibril, which reduces the speed of the water pressed outwards, as shown in Figure 11.2. Instead of Eq. (11.23) we write:

∂Φ = v − vw ∂n



(11.24)

Here vw is the velocity with which the water flows into the fibril. From Eq. (11.11) it follows that



α

dVw dV dr = = A = Av dt dt dt

(11.25)

Diatom Triboacoustics  263 V Vw V Vw V

Figure 11.2  Schematic drawing of the expanding fibril.

As the volume change of the water is given by the flow of water through the surface (dVw/dt = Avw), the following applies:

dVw = α −1v dt

(11.26)

∂Φ α −1 = v − vw = v ∂n α

(11.27)

vw = A−1

Eq. (11.24) therefore becomes



Calculation of the Velocity Potential In the vicinity of the body (distance small compared to the wavelength) the second term in Eq. (11.22) can be neglected and the Laplace’s equation ΔΦ = 0 applies. In [11.31] it is shown that for large distances compared to the size of the body but small compared to the wavelength a general solution of the Laplace’s equation exists which has the form



Φ=−

a 1 + K∇ R R

(11.28)

R is the distance to the sound emitting object, where the coordinate origin is somewhere inside the body. The first term a/R only occurs when the body is pulsating, whereby 4πa represents the flow of liquid through a closed sphere around the body, so that



4 πa = (v − vw )A =

α −1 α −1  vA = V α α

(11.29)

By looking at the outgoing spherical wave (R ≫ l), the solution for Eq. (11.22) is given:



Φ=−

α − 1 V (t − R /cL ) 4πR α

(11.30)

264  Diatom Gliding Motility and for the velocity of the particles in the wave



v s = ∇Φ =

α − 1 1   R V  t −  nˆ α 4πcL R  cL 

(11.31)

where nˆ is the unit vector in radial direction. Eqs. (11.30) and (11.31) are adopted from [11.31], whereby only the pre-factor has been modified. The factor (α – 1)/α) in Eqs. (11.30) and (11.31) shows that sound generation only occurs if α > 1. If α = 1, then the radius of the fibril would grow to its maximum size, but water would enter to the same extent as its diameter increases. An impulse would not be transferred to the surrounding water. As model parameter α = 4.7 was used. Thus (α − 1)/α ≈ 0.79. Altering this factor would not change the result by orders of magnitude. Calculation of the Sound Pressure For the question of which signal is produced by a microphone, the sound pressure of the pulse is determined.

p = ρ0cLvs



(11.32)

where ρ0 is the density of the water. Using Eq. (11.31) one gets



p = ρ0c L

α − 1 1   R V t −  α 4π cL R  cL 

(11.33)

By expressing the radius r in the solution t(r) according to Eq. (11.19) by the volume V and deriving it with respect to V, one obtains:



2

r V −V V = 2k π h 02 m rm V

(11.34)

A second differentiation yields



r2 V + Vm / V V = 2k π h 02 V rm V

(11.35)

where V is given by Eq. (11.34). For a radius r with r0 ≤ r ≤ rm (rm is never reached) the time at which this given radius is achieved can be calculated with the help of Eq. (11.19). Furthermore, r also gives the volume V. Using Eqs. (11.34) and (11.35), the time derivatives of the volume can be calculated and finally, with Eq. (11.33), the desired pressure p. Assuming a distance between sound source and microphone of R = 1 cm and with the model parameters given above, this leads to this time dependence of p (Figure 11.3): The maximum pressure is about 6.24 · 10−11 Pa. Since the retardation R/cL in Eq. (11.33) does not play a role for a single pulse, the shift in the time axis is not shown.

Diatom Triboacoustics  265 7.0E-11 6.0E-11

Sound pressure (Pa)

5.0E-11 4.0E-11 3.0E-11 2.0E-11 1.0E-11 0.0E+00 0.0E+00

1.0E-04

2.0E-04

3.0E-04

4.0E-04

5.0E-04

Time (s)

Figure 11.3  Time dependence of the sound pressure p.

1.2E-09

Sound pressure (Pa)

1.0E-09 8.0E-10 k = 1.0E-4

6.0E-10

k = 5.0E-5 k = 2.30E-5

4.0E-10

k = 1.0E-5

2.0E-10 0.0E+00 0.0E+00

5.0E-05

1.0E-04

1.5E-04

2.0E-04

2.5E-04

3.0E-04

Time (s)

Figure 11.4  Time dependence of the sound pressure p for different mass transfer coefficients k.

It should be noted that the mass transfer coefficient k, which is not known exactly, has a large effect on p(t). In Figure 11.4, in addition to the model parameter of 2.3 · 10−5 m/s, other values covering one decade were used. Finally, it should be mentioned that the sound absorption by the water has no relevant impact considering the small distances between diatoms and hydrophone. If a diatom moves on the substrate, however, the sound wave is created between the valve and the substrate. The sound can at least partially reach a hydrophone by diffraction around the valve (wavelengths in the detectable range are large compared to the diatom’s extension). If the

266  Diatom Gliding Motility raphe system opposite the substrate is active, a sound wave generated there can, however, propagate freely into the water body. Note on Pulse Superposition and Sound Spectrum For a single and a periodic pulse, its spectrum can be obtained by Fourier transformation. However, this is not of primary importance for observation with a hydrophone. A sequence of discrete single pulses can be recognized if the single pulse is perceptible by a “clicking” sound.1 However, if there are a lot of pulses per time, the probability of overlapping pulses increases. By superposition, higher pressures are then achieved than with a single pulse. With the pulse shape of p(t) shown above, two pulses must follow in a time interval of no more than 1.81 · 10−4 s, in order for the sum level to be about 10% above the maximum value of a single pulse. For a periodic pulse sequence, this is the case at 5500 pulses per second. As a rule, it can also be said for randomly distributed events that high pulse peaks rarely occur with significantly fewer than 5500 events on average per second. With a significantly higher number of pulses per time unit, one can profit from the superposition and achieve higher output voltages at the hydrophone. In Section 11.4 (Conclusions and Outlook) a pulse sequence of 400 pulses per second for a single diatom is estimated. These pulses should not overlap significantly in view of the short duration of the pulses. A positive effect on the observability beyond that of a single pulse can be expected from 5500/400 ≈ 14 active diatoms in the vicinity of the microphone. If the pulse sequence of a diatom is regular (no strong temporal fluctuations between single pulses), a clear peak in the spectrum at the frequency of the pulse sequence should be visible even in the case of many diatoms. Accordingly, this would show up in the autocorrelation function of pressure versus time. †

11.2.3 Gathering Diatoms Keeping diatoms is not that simple. Many species change their behavior when cultivated or kept in captivity because of the change in environment. Sometimes the diatoms even change their form. This can even lead to the point where the species cannot be identified anymore. If there is no sexual reproduction the diatoms get smaller and smaller until normally after a few months they are nonviable and the population dies out [11.39]. Nevertheless, there are a few raphid species that can be kept for years because they do not need sexual reproduction for a very long time, e.g., Nitzschia palea. That species is very active, durable, grows very fast and is therefore suitable and highly recommended for experiments. Other active species are diatoms from the Navicula genus. Pinnularia are also a suitable option, although they are rather sedate in their movement [11.39]. Edgar stated that: “Observations show that the large, bulky cells (e.g., Pinnularia, Cymatopleura) move more slowly than the flatter species (e.g., many Navicula spp. and Nitzschia spp.)” [11.16]. In order to keep diatoms, it is important that there is not too much water over the layer of diatoms in the container (as seen in Figure 11.5) to enable gas exchange also with the container closed, e.g., during transport. Of course, not closing the container airtight can be This can be compared to the observation of raindrops by the sound of their impact on a roof, which are statistically independent and produce a so-called shot noise. However, the pulses of a single diatom are not statistically independent.

1 †

Diatom Triboacoustics  267

air water diatoms

Figure 11.5  Schematic drawing of a jar with diatoms.

the preferred option, but it should be covered lightly to prevent entry of foreign substances or organisms from contaminating the water. Many diatoms can live well at a temperature of about 20°C. The container with the diatoms should ideally be placed at a window facing north, because direct sunlight should be avoided [11.39]. In general, there are two different ways of obtaining diatoms: Purchasing diatoms and catching diatoms in the wild. Both have advantages and disadvantages

11.2.3.1 Purchasing Diatom Cultures Diatoms can be purchased online. As mentioned before, it should be considered that many diatom cultures that are kept in captivity for longer periods of time lose their typical morphology. Nevertheless, the advantages are that only one specific species can be obtained and the amount of foreign substances and organisms would be minimal. At the University of Göttingen (Germany) there is an institution for research and cultivation of diatoms, where some species are offered. See also [11.38] [11.43].

11.2.3.2 Diatoms from the Wild One advantage of diatoms that are harvested from the wild is that they are generally more active and vital. On the other hand, one disadvantage is that there are many different species and so it is hard to determine the one at hand. Single diatoms of the desired species can be extracted with capillary pipettes, but the species would first need to be identified. Another option to isolate raphid, motile diatoms could be to set a light spot to one area (perhaps on a microscope slide), so that motile diatoms would move there [11.39]. Redfern has already described another method to isolate Navicula and other test objects, using fine hairs [11.34]. Further research and experiments were made with diatoms from the wild exclusively (as shown in Figure 11.6.), because of their activeness and the possibility of obtaining them easily with little time investment, and performed some quick, rough measurements. For gathering diatoms from the wild, a few different techniques were tried out: Mud from the Bottom of a Body of Water Because they can be found in almost every water body, diatoms can be obtained just by collecting mud, sand or other kinds of substrates from the bottom. For collecting raphid

268  Diatom Gliding Motility

50µm

50µm

Figure 11.6  Raphid diatom (to our best knowledge), caught by FZ; Note the visibility of the chloroplasts inside the diatoms; the scale bars are estimated.

diatoms, it is advised to do so on the side of a river or creek, where the current is not too strong, but strong enough for non-raphid diatoms to get flushed away. In the case presented here, the mud from a pondside was collected in Natschbach, Austria on June 14, 2019 and observed under the optical microscope. Placing a Substrate in a Water Body Another option specifically for gathering raphid diatoms is to place some kind of substrate for diatoms to move up to in some body of water. Therefore, different kinds of substrates were placed in a small creek in Natschbach, Austria, with not too strong current and left there for a few weeks. This was done first with microscope slides and then with plastic foil, which could later be crumpled up to increase the surface of the substrate and thereby also the density of raphid diatoms. Stones from Underwater For this technique stones from underwater are collected. Stones with golden-brown film on them usually work well for obtaining diatoms (as displayed in Figure 11.7.). The film is brushed off into a container with an old toothbrush and then washed away with a little bit of water from the same origin as the diatoms. The water should then be of a light-brown color. To receive diatoms, only the film from the upside of the stone needs to be scraped off. This is also a technique specifically for gathering raphid, motile diatoms [11.9]. Comparison of the Different Methods Samples of diatoms collected with the different methods were observed under an optical microscope from Budapest Telescope Center (BTC), model BIM313T. In Figure 11.8a there are diatoms that were brushed off stones from underwater. This has proven to be the best method for obtaining raphid diatoms while at the same time mostly avoiding other organisms, plants or material. It can clearly be seen that there is the least amount of foreign substances. Figure 11.8c shows mud from the edge of a pond. In this case, although many raphid diatoms were collected, many other organisms could also be observed. On the right side of Figure 11.8, substrates that were placed in a small creek and left there for a few weeks are shown. Here the enormous growth of algae is clearly visible.

Diatom Triboacoustics  269

Figure 11.7  Stones from underwater with golden-brown film on them, collected by FZ in Natschbach, Austria, on May 5, 2019.

50µm

50µm (a) Diatoms brushed off stones

100µm

(b) Microscope slide

50µm (c) Mud from a pondside

(d) Plastic foil

Figure 11.8  Comparison of diatom samples obtained with different methods; the scalebars are estimated.

In Figure 11.8b, the substrate was a microscope slide made out of glass, which usually is a good substrate for raphid diatoms. Therefore, this sample contained many raphid diatoms, but also a lot of foreign substances. In Figure 11.8d, the substrate was a piece from the plastic foil that was placed in the creek. Here the ratio of raphid diatoms to foreign substances is the worst of all methods that were used. Pennate diatoms usually cannot stick well to plastic, so plastic is not a suitable substrate.

11.2.4 Using a Hydrophone to Detect Possible Acoustic Signals from Diatoms 11.2.4.1 First Setup For a first measurement, to detect possible acoustic signals related to diatom movement, diatoms were first scraped off stones from underwater and then mud and sand from a

270  Diatom Gliding Motility riverside were collected. For every measurement with diatoms, a reference measurement was performed. It is important that the reference container is of the same material and shape as the container with diatoms. Furthermore, it was filled with water from the same origin as the diatoms, directly from the top of the small creek, where the water is pumped from the bottom of the pond. Therefore, there probably were not that many live diatoms as in the containers where diatoms were gathered with the different methods presented earlier. The two containers had roughly the same level of liquid. The containers used were glass jars with a diameter of about 10 cm. In Figure 11.9 the setup for these measurements is shown. On the left there is the power supply for the hydrophone, in the middle the recorder and on the right the jar with diatoms and the hydrophone and the reference jar. Here the diatoms scraped off stones are recorded. Also, different materials to muffle ambient noise can be seen in Figure 11.9. Usually textiles or artificial fur work well. In this case, vibration-insulating mats (as seen in yellow) were also used. The material these mats are made from is called “Sylomer” from Getzner Werkstoffe GmbH in Vorarlberg, Austria. There are different kinds available for different weight forces (force per area). The hydrophone used was a Brüel & Kjær hydrophone, Type 8106, with the hydrophone power supply also from Brüel & Kjær, Type 2804. The recorder used was a Tascam DR-100 MKIII linear PCM recorder. This hydrophone Type 8106 is a low-noise hydrophone, designed for the measurement of weak, underwater signals. It has a frequency range from 3 Hz to 80 kHz and a receiving sensitivity of −173 dB re 1 V/μPa. The measurements with the first setup were performed in an office with a surrounding background sound level of 39 dBA, re 20 µPa, RMS fast, where “dBA” means “decibel according to evaluation curve A,” which takes into account the human hearing, “re” stands for “relative to.” So “relative to 20 µPa” is the sound pressure reference value in the level measurement, which corresponds to the human hearing threshold at 1000 Hz. This has to be stated in this context, because it specifies the reference level. Otherwise, this declaration would be ambiguous. “RMS fast” means that the measuring steps were performed at intervals of 135 ms.

Figure 11.9  Setup of first measurements.

Diatom Triboacoustics  271

Figure 11.10  Measurement of jar with diatoms (right), reference jar (left).

Figure 11.10 shows the measurement of the diatoms that were collected from a river, together with mud, sand and possibly many foreign organisms and substances. Nevertheless, a rough measurement was performed.

11.2.4.2 Second Setup In a second attempt, we tried to increase the density of raphid diatoms. To do so, transparent glass balls with a diameter of about 4 mm [11.3] were filled in jars so that the bottom was covered with them. The glass balls are made from soda-lime glass, the prevalent type of glass. Diatoms were collected by scraping them off stones from the same creek as before and put inside the jars with the glass balls, as seen in Figure 11.11. For the measurements, the content of all the jars could be filled in one container, and with the much higher surface there would hopefully also be a higher density of raphid diatoms. A reference jar with roughly the same amount of glass balls and the same level of water from the same creek was also prepared. The hydrophone and its power supply were the same as in the first setup, but the recorder was a Marantz model No. PMD660. Additionally, we tried to record the noises made by two adult great pond snails—Lymnaea stagnalis (Linnaeus, 1758). These animals have a radula, which they use to scrape algae off a surface and eat them. This usually produces a munching sound. The snails were put in an extra jar with water from the same origin. The rest of the setup remained the same as for the diatoms. This time measurements were performed in a soundproof room in Althanstraße 14, 1090 Vienna, Austria. Its internal dimensions are 3 m · 3 m · 3 m. Beyond there is 1 m of silencers

272  Diatom Gliding Motility

Figure 11.11  Jars with diatoms and little glass balls.

out of foam material on the ceiling, on the floor and on all four walls. Behind that there is a double wall filled with sound-insulating material. The whole cube is mounted on buffers out of rubber and also acts as a Faraday cage. At the time of the measurements 3 observors were present in the room, which could have been an additional source of noise, but was considered neglectible, as the goal was only to perform rough measurements.

11.3 Results and Discussion 11.3.1 Spectrograms Figures 11.12 and 11.16 show the results of the setup shown in Figure 11.10 and the results of the reference measurement with the same water but without diatoms. The diagrams shown here are spectrograms, the illustrations of the frequency spectrum of a signal, as they are used for analysis of acoustic signals. They serve as an overview illustration of sound signals and cannot be used to read exact amplitude values. On the abscissa the time in [s] and on the ordinate the frequency in kilohertz [kHz] is displayed. The sound level in decibel [dB] is displayed through color, where brighter colors indicate stronger sound levels. In the following spectrogram, 40 dB are displayed (−82 dB…−42 dB). This makes 0.634 dB per color grade. The color range is ordered as follows: black (≤ −82 dB), dark blue, light blue, green, yellow, white, red (≥ −42 dB). Exact values and differences in the sound levels can be read off the averaged spectra. All averaged spectra were created in the same way. For each of the two recordings a longterm averaging and a short-term averaging were performed. Comparing the short- and long-term averaged spectra shows that the spectrum is quite stable over the whole period of time (short converges to long). Also, besides short disturbances or pulses, no recognizable signal is visible. Differences between the averaged spectra of the two recordings can be

Diatom Triboacoustics  273 “/190509_Versuch5- Ausschnitt.wav/test” (Ch. 1): Amp-Spg: range=-82..-42dB (0.634921 dB kHz

60

40

20

“/190509_Versuch5- Ausschnitt.wav/test” (Ch. 1): max=1, 0dB=1, range=+/-0.0143886, dur=

1.0

1.5

2.0

2.5

3.0

s

Figure 11.12  Spectrogram of measurement with diatoms.

explained as follows: Firstly, it is a matter of two recordings that were performed separately. Secondly, the recordings have different lengths and the pulses are not evenly distributed. Lastly, the horizontal lines in the spectrogram/peaks in the spectrum differ, showing the same frequency but different sound levels. Analyzing the recorded sounds and creating the spectrograms was done with the program STx. It is freeware and can be downloaded from the website of the Austrian Academy of Sciences [11.2]. Some characteristics of the spectrogram of the measurement with diatoms (Figure 11.12) can be summarized as follows: Throughout the whole signal: • Tone at 16.4 kHz, 10 dB above the background • Tone at ca. 24 kHz, 7 dB above the background (occasionally interrupted or AM modulated) • Noise band between 44 kHz and 64 kHz At some locations: • Tone at ca. 7200 Hz, slightly modulated; The level is difficult to measure, ca. 3–8 dB above the background. For example, between 2.75 s and 3 s: a tone at about 7100 Hz (+8 dB) and a modulated/varying tone between 4800 Hz and 5200 Hz • There are similar signals between 2.4 s – 2.5 s and 1.85 s – 1.95 s and possibly 0.6 s – 0.95 s.

274  Diatom Gliding Motility In Figure 11.13, both stationary tones (16.4 kHz and 24 kHz) and the noise band are clearly visible. The small peaks at around 5 kHz and 7 kHz are caused by the sporadic signals. In Figure 11.14 the stationary tone at 16.4 kHz and the localized tone at ca. 7100– 7200 Hz are clearly visible. In this range (Figure 11.15) there is no occurrence of the sporadic tones, and the spectrum is therefore similar to the averaged spectrum over the whole signal (Figure 11.13), although the variance is higher, since the averaged signal is shorter. The same kind of considerations were made for the reference measurement. Throughout the whole signal: • Tone at ca. 16.4 Hz, 13 dB above the background • Tone at ca. 24 kHz, 7 dB above the background (occasionally interrupted or AM modulated) fft: length=9600 samples, df=20Hz phase: -Pi..+Pi db -100 -105 -110 -115 20000

40000

60000

Hz

Figure 11.13  Averaged spectrum over the whole spectrogram (0.5 s – 3.5 s). fft: length=9600 samples, df=20Hz phase: -Pi..+Pi db -100 -105 -110 -115 20000

40000

60000

Hz

60000

Hz

Figure 11.14  Averaged spectrum over the range 2.75 s – 3 s. fft: length=9600 samples, df=20Hz phase: -Pi..+Pi db -100 -105 -110 -115 20000

Figure 11.15  Averaged spectrum between 1 s – 1.5 s.

40000

Diatom Triboacoustics  275 “/190509_Versuch6_Referenz - Ausschnitt.wav/test” (Ch. 1): Amp-Spg: range=-82..-42dB (0. kHz

60

40

20

“/190509_Versuch6_Referenz - Ausschnitt.wav/test” (Ch. 1): max=1, 0dB=1, range=+/-0.014

1.0

1.5

2.0

2.5

3.0

s

Figure 11.16  Spectrogram of reference measurement.

• Tone at ca. 32.8 Hz, 10 dB above the background • Noise band between 44 kHz and 64 kHz At some locations: • At ca. 1.2 s there is a short broadband disturbance. • At ca. 1.65 s – 1.75 s (just before the relative strong 1.3 kHz tone) there is a short tone at ca. 4000 Hz (possibly with parts at lower frequencies). Comparison with the first signal of the measurement with diatoms: • All three tones are also present in the first sound file (Figures 11.12–11.15), though with lower amplitudes (particularly the 32.8 kHz tone). • The tones at 7 kHz and 5 kHz found in the first sound file are not found here. In Figure 11.17 the three static tones (16.4 kHz, 24 kHz, 32.8 kHz) and the noise band are clearly visible. In Figure 11.18 a tone with slowly rising frequency can be seen at ca. 12.5 kHz to 13.5 kHz. Figure 11.19 shows a relatively strong tone at ca. 1300 Hz (not completely stable frequency). Figure 11.20 shows only the “background signal.” It is similar to the average of the complete signal, but with higher variance.

276  Diatom Gliding Motility fft: length=9600 samples, df=20Hz phase: -Pi..+Pi db -100 -105 -110 -115 20000

40000

60000

Hz

60000

Hz

60000

Hz

60000

Hz

Figure 11.17  Averaged spectrum over whole spectrogram (0.5 s – 3.5 s). fft: length=9600 samples, df=20Hz phase: -Pi..+Pi db -100 -105 -110 -115 20000

40000

Figure 11.18  Averaged spectrum between 0.5 s – 0.65 s. fft: length=9600 samples, df=20Hz phase: -Pi..+Pi db -100 -105 -110 -115 20000

40000

Figure 11.19  Averaged spectrum between 1.85 s – 2 s. fft: length=9600 samples, df=20Hz phase: -Pi..+Pi db -100 -105 -110 -115 20000

40000

Figure 11.20  Averaged spectrum between 2.5 s – 3.0 s.

Apart from background noise in both measurements no sounds could be recorded. The spectrograms of the measurements with diatoms and the reference measurements show no great differences. They also do not look exactly the same, which is due to the fact that they were not recorded at the same time, which would require two

Diatom Triboacoustics  277 hydrophones. The recordings of diatoms scraped off stones, as well as the measurement of the containers with diatoms on glass balls, all led to the same results. The recordings of Lymnaea stagnalis were a challenge, because they did not seem to be very hungry at the time of the experiment. Only one quiet munching sound could be recorded. For this experiment with the snails, it might be best to put the hydrophone under water for some time, so that algae can grow on it. Then the snails could scrape them right off the hydrophone, which would produce a strong signal.

11.3.2 Discussion At the Institute of Sound Research of the Austrian Academy of Sciences, the comparative sound recordings (with diatoms and reference recording without diatoms) were subjected to a detailed analysis. Averaged spectra were taken from equally long sections of both signals. These showed background signals that could not be distinguished between the two signals. Stable partials that occur in the background have the character of technical faults. Likewise, faults are to be detected, but these occur only for a short period of time and therefore cannot originate from the expected organic source. In summary, it can be stated that both recordings have a very similar background signal, which is present over the entire time. This can be shown by the overall average spectra (Figures 11.12 and 11.16) and by the short-time averages (Figures 11.15 and 11.20), which have the same mean but a higher variance (because of the shorter average time). The first recording (measurement with diatoms) contains some short localized signals with low frequency components. Because the signals have similar frequency components, they may be caused by the same (probably mechanical) source. The second recording (reference measurement) also contains some short low-frequency signals, but they are different from that in the first signal. The stable/continuous tones at 16.4 kHz, 24 kHz and 32.8 kHz are probably caused by electrical devices (monitor?). The results from these experiments could have different causes. It is possible that there simply were not enough diatoms. For recording sounds with hydrophones, it would be best if the diatoms stuck directly to the surface of the hydrophone and move alongside it. This is not very likely because most hydrophones usually are coated with elastomers, which is probably not a suitable substrate for raphid diatoms. Furthermore, the sounds of the diatoms could also be too quiet to be recorded with that kind of hydrophone. The problem here is that hydrophones need to be of a certain size to be very sensitive. At the same time, it would be best to get as close to the diatoms as possible, which would require a small hydrophone, but with smaller size its sensitivity decreases. Finally, it is also possible that there simply are no sounds related to raphid diatom movement.

11.4 Conclusions and Outlook We have put forward the hypothesis that diatoms are driven by the explosive hydration of the mucopolysaccharide microfilaments released from raphes, whose diameter may be estimated from micrographs in [11.13] at 50 nm. If a diatom is moving at 20 μm/s, it would then be releasing 20,000/50 = 400 filaments per second. If these are indeed explosions, they would then be occurring in the frequency range of 0.4 kHz. However, an individual explosion could

278  Diatom Gliding Motility last for an even shorter period of time. If a filament exits the raphe at the speed of a discobolocyst (a projectile launching organelle), estimated at 260 m/s [11.21], and we assume a filament length equal to the cross-sectional length of a raphe (0.3−3 μm in [11.24]), then a single explosion could last as little as 1 ns, providing acoustic frequencies in the range of 106 kHz. The latter is beyond the frequency range of the hydrophone we used and might explain our negative result. There is a possibility that the diatom trail itself damps the sound of hydration explosions, as analogously suggested for instrument vibrations [11.37]. Of course, all the experiments performed were rough, basic approaches and could be refined tremendously. With two hydrophones of the same kind available, the reference measurement could be done at the same time as the measurement of the jar with diatoms. Then the signal of the reference measurement could be subtracted and only sounds that are present in the jar with diatoms would be shown, assuming that the noise is from the environment or the observer and not from the hydrophone itself. A soundproof room is possibly the best environment to perform these kinds of measurements, but with two hydrophones it might not be necessary, because most of the background noise would be cancelled out. Another issue is the density of diatoms, which should definitely be increased, because the more diatoms per volume unit there are in the container, the more likely it is to detect possible sounds. To accomplish that, while at the same time maintaining the diatoms’ activeness, it could be worth trying to put active, vital diatoms from the wild into one container together with great numbers of diatoms from cultures. Increasing the surface could also be done by constructing some kind of inset out of microscope slides, or, even more ideally, coverslips on which the diatoms could attach. Because coverslips are made out of very thin glass, they would amplify potential vibrations. The glass balls that were used to increase the surface (as presented in Subsection 11.2.4.2), are out of solid glass. A proposal to improve the experiment would be to use hollow glass balls, like Christmas tree balls, but much smaller. Of course, they would float on the water because of buoyancy, and therefore would have to be held underwater, e.g., with a net or sintered or glued to a surface. Another promising approach could be to experiment with colonies of Bacillaria, the earliest known genus of diatoms. Bacillaria cells live in colonies where the cells are connected to long strands and can move parallel to each other [11.28]. This would provide a high density of diatoms, which means higher sound levels can be expected. Sounds could also occur in Diatoma colonies. This genus forms chains or even bundles and sometimes two parallel diatoms open up to form a “V” so that the chain reaches a new state of equilibrium. This process happens very fast and could possibly produce an acoustic signal. A suggestion for a setup for future measurements would be to grow a very dense colony of diatoms, preferably of a species that is very active. Ideally, they are bred directly on the hydrophone, because the acoustic signal would decrease tremendously with increasing distance. Then possible changes in the signal could be detected when the light is turned off and on again, because diatom movement is in most cases correlated with light intensity. An alternative proposal to the presented method could be to use a vibrometer. A drop of water with diatoms would have to be applied onto a very thin bar, e.g., an AFM cantilever, which could maybe be stimulated by the vibrations of the diatoms. This could be measured via interference. Of course, a reference measurement with the same amount of water but without diatoms would also have to be performed.

Diatom Triboacoustics  279 As another approach an optical hydrophone could be used, e.g., from XARION Laser Acoustics GmbH in Vienna. Their hydrophone output signal is analogue (max ±7.5 V at 50 Ohm), therefore there is no software, but it can be connected with any measurement device. According to Wolfgang Rohringer, the lowest sound pressures detectable at the moment are at ~ 50 μPa/Hz, in a frequency range from 10 Hz to ~ 3 MHz. In case acoustic analysis of diatoms succeeds in the future, an interesting application could be to track the sounds of different activities in water, especially of moving diatoms. Through that perhaps conclusions about diatoms or other active organisms inside the water body could be made. Acoustic methods for analysis and control of friction processes are widely used in tribology. Approaches on triboacoustic monitoring of friction were made by Dykha et al. [11.12] [11.11]. There have already been proposals to measure vibrations with MEMS (microelectromechanical systems) sensors, e.g., by Looney. The sounds produced by the repetitive mechanical motion of mechanical parts can be used to observe machine health [11.32]. If measuring low intensity sound signals of that kind succeeds, it could be applicable to ensure the proper functionality of machines. The measured acoustic signals could serve as a check if a device is working correctly [11.40] [11.20]. We hope that these rough initial measurements of sounds of underwater creatures will stimulate further research along these lines, and that future scientific approaches can build upon our undertakings! Shrimp and other marine organisms produce and use sound that propagates distances many times their sizes. Perhaps there is yet a definitive chapter on diatom ecoacoustics [11.18] to be written.

Acknowledgements FZ: I am very grateful for all the wonderful people around me, who provided me with the motivation to carry on through all the different work stages and supported me especially when things seemed to be stuck. During the past year I worked on this subject with Alexander Müller. We did a lot of research, developed many ideas and also performed the measurements together. Thank you for that! A big thank you goes out to my parents who luckily decided to build a pond in the garden many years ago which provided me with a sheer inexhaustible source of vital diatoms to play with. I would really like to express my gratitude to Wolfi Murnberger who provided his optical microscope and spent a lot of time with me, watching the mesmerizing gliding movement of diatoms. I also want to thank Michael Schagerl, Manfred Kaltenbacher and Michael Kloster for the helpful communications per email and/or in person.

References [11.1] Alicea, B., Gordon, R., Harbich, T., Singh, A., Varma, V., Mehan, P., Singh, U., Machine learning segmentation of diatom movies: Towards a digital Bacillaria paradoxa [BCLA], in:

280  Diatom Gliding Motility Diatom Gliding Motility [DIGM, Volume in the series: Diatoms: Biology & Applications, series editors: Richard Gordon & Joseph Seckbach] [In preparation], S.A. Cohn, K.M. Manoylov, R. Gordon (Eds.), p. #s, Wiley-Scrivener, Beverly, MA, USA, 2020. [11.2] Austrian Academy of Sciences Acoustics Research Institute, Sound tools extended (stx) intelligent sound processing, URL https://www.kfs.oeaw. ac.at/stx/ (last accessed on 2021/01/14), 2020. [11.3] Bastelwelt Creativ, Glaskugeln mini, klar, leicht bläulich schimmernd, 4 mm, 100 gr.„ URL https://www.bastelweltcreativ.de/glaskugeln-miniklar-leicht-blaeulich-schimmernd/p-8760.html (last accessed on 2021/01/14), 2019. [11.4] Bondoc, K.G., Heuschele, J., Gillard, J., Vyverman, W., Pohnert, G., Selective silicatedirected motility in diatoms. Nat. Commun., 7, 10540, 2016. [11.5] Bondoc, K.G., Kiel, C., Vyverman, W., Pohnert, G., Searching for a mate: Pheromonedirected movement of the benthic diatom seminavis robusta. Microb. Ecol., 72, 2016. [11.6] Cussler, E., Diffusion: Mass Transfer in Fluid Systems, Cambridge Series in Chemical Engineering, pp. 237–273, Cambridge University Press, Cambridge, England, 2009. [11.7] Demtröder, W., Experimentalphysik 1, Springer, Berlin, Germany, 2005. [11.8] Deng, W., Jeng, D., Toorop, P., Squire, G., Iannetta, P., A mathe- matical model of mucilage expansion in myxospermous seeds of capsella bursa- pastoris (shepherd’s purse). Ann. Bot., 109, 419–427, 2012, www.aob.oxfordjournals.org. [11.9] diatomp, Diatomeen Teil 1: Einführung and das Sammeln von Diatomeen, URL https:// www.youtube.com/watch?v=KOFd_EDAIeQ (last accessed on 2021/01/13), 2019. [11.10] Drum, R.W. and Hopkins, J.T., Diatom locomotion: An explanation. Protoplasma, 62, 1–33, 1966. [11.11] Dykha, A., Zaspa, Y., Slashchuk, V., Triboacoustic control of fretting. J. Frict. Wear, 39, 169– 172, 2018. [11.12] Dykha, A., Zaspa, Y., Vychavka, Y., Tribo-acoustic analysis of the processes of dynamic friction, MOTROL. Comission of Motorization and Energetics in Agriculture, vol. 19(2), pp. 11–14, 2017. [11.13] Edgar, L., Mucilage secretions of moving diatoms. Protoplasma, 118, 44–48, 1983. [11.14] Edgar, L. and Zavortink, M., The mechanism of diatom locomotion. II: Identification of actin. Proc. R. Soc London Ser. B, Biol. Sci. (1934-1990), 218, 345–348, 1983. [11.15] Edgar, L.A., Diatom locomotion: Computer assisted analysis of cine film. Br. Phycol. J., 14, 1, 83–101, 1979, URL https://doi.org/10.1080/00071617900650111. [11.16] Edgar, L.A., Diatom locomotion: A consideration of movement in a highly viscous situation. Br. Phycol. J., 17, 243–251, 1982. [11.17] Edgar, L.A. and Pickett-Heaps, J.D., The mechanism of diatom loco- motion. I. @ an ultrastructural study of the motility apparatus. Proc. R. Soc London Ser. B, Biol. Sci., 218, 1212, 331–343, 1983, URL http://www.jstor.org/stable/35706. [11.18] Farina, A., Ecoacoustics: A quantitative approach to investigate the ecological role of environmental sounds. Mathematics, 7, 1, 21, 2019. [11.19] Freeman, S., Freeman, L., Giorli, G., Haas, A., Photosynthesis by marine algae produces sound, contributing to the daytime soundscape on coral reefs. PloS One, 13, 10, e0201766, 2018. [11.20] Geisel, T., Horton hört ein Hu!, Rogner und Bernhard bei Zweitausendeins, Hamburg und Affoltern a. A., 2003. [11.21] Gordon, R., A retaliatory role for algal projectiles, with implications for the mechanochemistry of diatom gliding motility. J. Theor. Biol., 126, 4, 419–436, 1987, URL http://www.sciencedirect.com/science/article/ pii/S0022519387801492.

Diatom Triboacoustics  281 [11.22] Gordon, R., The whimsical history of proposed motors for diatom motility, in: Diatom Gliding Motility [DIGM, Volume in the series: Diatoms: Biology & Applications, series editors: Richard Gordon & Joseph Seckbach], S.A. Cohn, M. Manoylov, R. Gordon (Eds.), p. #s, Wiley-Scrivener, Beverly, MA, USA, 2021. [11.23] Gordon, R., Björklund, N., Robinson, G., Kling, H., Sheared drops and pennate diatoms. Nova Hedwigia, 112, Festschrift for Prof. T.V. Desikachary, 287–297, 1996. [11.24] Gordon, R. and Drum, R., A capillarity mechanism for diatom gliding locomotion. Proc. Natl. Acad. Sci. U. S. A., 67, 1, 338–334, 1970. [11.25] Harbich, T., Introduction to motility, URL https://diatoms.de/en/diatoms/introductionto-motility (last accessed on 2021/01/14), 2019. [11.26] Harper, M. and Harper, J., Measurements of diatom adhesion and their relationship with movement. Br. Phycol. Bull., 3, 2, 195–207, 1967. [11.27] Hopkins, J. and Drum, R., Diatom motility: An explanation and a problem. Br. Phycol. Bull., 3, 1, 63–67, 1966, URL https://doi.org/10.1080/00071616600650081. [11.28] Kalbe, L., Kieselalgen in Binnengewässern: Diatomeen, VerlagsKG Wolf, Magdeburg, Germany, 2005. [11.29] Kratochvil, H., email communication (2019/11/28), 2019. [11.30] Kratochvil, H. and Pollirer, M., Acoustic effects during photosynthesis of aquatic plants enable new research opportunities. Sci. Rep., 7, 44526, 2017. [11.31] Landau, L. and Lifshitz, E., Fluid Mechanics, Second Edition, Volume 6 (Course of Theoretical Physics S), Elsevier, Amsterdam (NL), 1987. [11.32] Looney, M., An introduction to mems vibration monitoring. Analog Dialogue, 48, 06, 1–3, 2014. [11.33] Purcell, E.M., Life at low Reynolds number. Am. J. Phys., 45, 1, 3–11, 1977, URL https://doi. org/10.1119/ 1.10903. [11.34] Redfern, P., Mode of isolating naviculae and other test objects. Q. J. Microsc. Sci., 1, 3, 235– 236, 1853. [11.35] Rogers, P. and Trivett, D., Hydrophone, AccessScience, (McGraw-Hill Education), New York City, New York, USA, 2014, https://www.accessscience.com/content/hydrophone/330500 (last accessed 2021-01-13. [11.36] Round, F., Crawford, R., Mann, D., Diatoms: Biology and Morphology of the Genera, Cambridge University Press, Cambridge, England, 1990, URL https://books.google. de/ books?id=xhLJvNa3hw0C. [11.37] Sabuncu, A., Gordon, R., Richer, E., Manoylov, K., Beskok, A., The kinematics of explosively jerky diatom motility: A natural example of active nanofluidics, in: Diatom Gliding Motility [Volume in the series: Diatoms: Biology & Applications, series editors: Richard Gordon & Joseph Seckbach], S.A. Cohn, K.M. Manoylov, R. Gordon (Eds.), pp. 33–63, Wiley-Scrivener, Beverly, MA, USA, 2021. [11.38] SAG, The culture collection of algae at Goettingen university, URL https://www.uni-goettingen. de/en/www.uni-goettingen.de/ de/184982.html (last accessed on 2021/01/14), 2019. [11.39] Schagerl, M., oral communication (2018/12/11), 2018. [11.40] Seuss, Horton Hears a Who!, Random House, Gütersloh, Germany, 1954. [11.41] Spiazzi, E. and Mascheroni, R., Mass transfer model for osmotic dehydration of fruits and vegetables—I. Development of the simulation model. J. Food Eng., 34, 4, 387–410, 1997, URL http://www.sciencedirect.com/science/article/ pii/S0260877497001027. [11.42] Ussing, A., Gordon, R., Ector, L., Buczkó, K., Desnitskiy, A., VanLandingham, S., The Colonial Diatom “Bacillaria paradoxa”: Chaotic Gliding Motility, Lindenmeyer Model of Colonial Morphogenesis, and Bibliography, with Translation of O.F. Müller (1783), “About a

282  Diatom Gliding Motility peculiar being in the beach- water”, A.R.G. Ganter Verlag Kommanditgesellschaft, Ruggell, 2005. [11.43] Utex, Culture collection of algae, https://utex.org (last accessed 2021-01-13), 2019. [11.44] Wang, J., Cao, S., Du, C., Chen, D., Underwater locomotion strategy by a benthic pennate diatom Navicula sp. Protoplasma, 250, 1203–1212, 2013. [11.45] Harper, M.A., Movements, in: The Biology of Diatoms, D. Werner (Ed.), pp. 224–249, Blackwell, Oxford, 1977.

12 Movements of Diatoms VIII: Synthesis and Hypothesis1

*

Jean Bertrand Saint Jean de Braye, France

Abstract

The author proposes a hypothesis that incorporates two main phenomena: propulsion by mic­ rofibrils on the one hand and by mucilage (mucus) on the other hand. After traversing the raphe, the tiny fibrils would be pushed out of the diatom by the mucus and execute one after another a vertical retractile movement. These microfibrils are thought to be connected to the actin micro­ filaments located under the raphe. The interactions between myosin filaments and actin filaments would create sliding of the actin filaments and consequentially the contraction of the microfibrils. These contractions seem to be generated by an impulsive force in the form of internal “wave trains” activated by light and/or calcium ions acting on actomyosin bundles. This reaction would act in a synchronous way or independently in the two directions or in reverse on the actomyosin bundles longitudinally situated under the raphes. The extreme ends of the microfibrils are inclined as a result of the adhesion to the base of the adjacent microfibrils. These would retract either by the shortening of the actin and myosin microfilaments. After the passage of the wave, the mucus would be expelled into the raphe and force the microfibrils to return to their original position until the next wave’s passage. On the semi-raphe length, one or several waves can be generated, producing external “wave trains.” The rapid but momentary slant of the apical part of the microfibrils and their adherence to the substrate allow for the gliding movement. The adhesion point on the substratum seems to be set by the expulsion of the mucus. Keywords:  Bacillariophyceae, diatoms, movements, hypothesis

12.1 Introduction The author does not give a complete history of diatom motility research, as there have been previous reviews, especially the comprehensive one by [12.14].

Email: [email protected] Bertrand, J. (2008) Mouvements des diatomées VIII: synthèse et hypothèse [Diatom movements VIII: Synthesis and hypothesis, French]. Diatom Research 23(1), 19-29. Translation prepared by Richard Gordon, Martin Laviale, and Karen K. Serieyssol in consultation with Jean Bertrand. **

1

Stanley Cohn, Kalina Manoylov and Richard Gordon (eds.) Diatom Gliding Motility, (283–294) © 2021 Scrivener Publishing LLC

283

284  Diatom Gliding Motility This work began with my compilation of all observable modes of movement of live diatoms [12.4]. Only forward/reverse/linear movement along the longitudinal/apical axis of pennate diatoms (the apical translational movement) had been studied to 1992. But all of the other modes of movement must be taken into account for a coherent explana­ tion of diatom/substrate relationships. Some typical and fundamental movements have been identified and studied by observing living diatoms and via theoretical analysis of biomechanics: The equilibrium dynamics of Rhoicosphenia abbreviata Agardh [12.3] The vertical polar pivoting of Eunotia pectinalis (Kützing) Rabenhorst [12.5] The transapical movement of Cocconeis pediculus Ehrenberg [12.6] The polar pivoting of Gomphonema acuminatum Ehrenberg [12.7]. This list is not exhaustive, as other diatom genera, such as Nitzschia and Surirella, show some/certain transapical movements that are difficult to interpret. Among the hypotheses that have been put forward to explain these movements, the two main ones, however, only explain the apical translational movement:2 †

a)

The expulsion of mucus, which is also used for adhesion [12.19] [12.37] [12.41] b) Propulsion by microfibrils [12.11] [12.14] [12.21] Our first goal was to try to distinguish the actions of microfibrils and mucus. The action of mucus in adhesion and its continuous production during movement has been demon­ strated by the presence of traces3 identified by staining [12.37]. The propulsive action of the mucus during apical translational movement was the most discussed aspect of diatom movement and led to many controversies [12.11] [12.14] [12.19] [12.21] [12.41] that have not been resolved. Microfibrils were first described in a transmission electron microscopy (TEM) study of longitudinal and transverse sections  [12.11] [12.12]. But their role has never been clearly demonstrated, as critical observations have only been done for a limited number of types of diatom movement. In this context, our goal was to try to distinguish the actions and, if possible, modes of action of the mucus and microfibrils, independently or simultaneously, not just in the case of apical displacement, but also of the more rarely described “exceptional” movements. ‡

12.2 Review of the Conditions Necessary for Movements Although microfibrils were observed unambiguously, their action has not been observed and confirmed during movements. It was demonstrated  [12.3] [12.5] [12.6] that the Martin Laviale consulted with Jean Bertrand: A third hypothesis would be from  Harper and Harper 1967 [12.19] who suggested an expulsion of a continuous ribbon of mucus. This hypothesis has to be discarded, according to [12.4] But they are many more explanations. This explains the third hypothesis alluded to in the text below. 3 The diatom trail [12.20] [12.40].

2

†

‡

Movements of Diatoms VIII  285 attraction of organic particles was real and operates at a significant distance, 4  µm in Rhoicosphenia abbreviata, 10 to 12 µm in Eunotia pectinalis and 12 to 14 µm in Cocconeis pediculus in the complete absence of diatom movement. Mucus is an inert material, therefore it seems unrealistic to assign this to organic particle attraction, as well as the aspiration of mucus [12.19] that was never confirmed. On the other hand, the hypothe­ sis of the mircofibril action, whether actual or apparent, as well as the microfibril length outside of the frustule, seems to be confirmed by the following three examples. The length outside the frustule and its ability to contract, however, is implied in the move­ ments of transapical raphe diatoms with the connective face positioned on the substrate. In this position the raphe is about 10 microns off the substrate for certain genera, like Pinnularia. However, the contact with the substrate is essential for ensuring transapical movement, hence the microfibril length and contractibility are necessary. The contract­ ibility is equally important in the case of vertical polar pivoting, at least when the diatom is suspended under the coverslip [12.5]. Therefore, the cell must have a strong force in order to return to the horizontal position. Regarding vertical pole pivoting on the micro­ scopic slide, the problem is clearly not understood, as was demonstrated by [12.41]. There could be mucus accumulation under an extremity, thereby causing the pivoting. However, this only works in this case. This movement is apparently impossible when the diatom slides under the coverslip to return to the horizontal position. Moreover, the horizontal pivoting cannot be attributed to the secretion of mucus, as in the case of apical pivoting of the desmid genus, Closterium (personal unpublished observations), especially due to the high speed at which these movements occur in diatoms. However, the mucus, by its persistence at any point on the frustule and being simultaneously in contact with the substrate, participates and allows the horizontal pivoting. However, it is difficult to prove and it was the polar pivoting of Gomphonema acuminatum  [12.7] that led to the development of the following hypothesis along with apical displacement elements. In this study, with a diatom situated on organic matter in the water and performing simultaneous rotations and displacements, the need for longitudinal action of the micro­ fibrils is seen as necessary. This action was observed by [12.28] and discussed by [12.12] in the case of organic matter transported by the raphe along the frustule. However, these findings and the hypothesis of  [12.14] were in contradiction with biomechanical princi­ ples [12.3] and the rapid longitudinal displacement of microfibrils in the raphe. It should be noted that all these described movements [12.4] are fully consistent with the forces gen­ erated by diatoms, whether for moving or pivoting [12.8]. It was shown that the necessary forces were well within the physical capabilities of diatoms.

12.3 Hypothesis The microfibrils exert forces and cause the displacement of organic material (or the diatom) by an undulating longitudinal wave motion by the microfibrils associated with the expul­ sion of mucus. One could imagine this undulating motion as a series of waves traveling along the microfibrils aligned with the entire length of the raphe (Figure 12.1) and moving down each raphe half from the center towards the apex, or vice versa. This wave would be

286  Diatom Gliding Motility Sens de progression l’onde n1

Sens de progression l’onde n2

Chloroplastes

Sens de progression l’onde n3 Noyau

Diatomée en vue connective

Figure 12.1  Waves formed by the microfibrils. The waves for each half-raphe vary independently in frequency and intensity. Translation of labels: Sens de progression l’onde = Direction of wave progression; Chloroplastes = Chloroplasts, Noyau = Node; Diatomée en vue connective = Girdle view of diatom.

generated by the successive and progressive inclination of the microfibril bases. The apical extremities of the microfibrils straightened, adhere to the substrate, and propel the diatom or the organic matter in contact with it. When the microfibrils tilt, they create the appear­ ance of a contracting. The tilting would be achieved by a system analogous to that of flagella of Eugleniods or cilia of Ciliates but adapted to the diatom cell structure. Microfibrils would be adherent one to another at a point close to their base [12.11] [12.12, Volume 2]. Each microfibril pair would form a rigid triangle, with the contact point being near the base. They thus form an articulated undeformable triangle, with the width of the base equal to the diameter of microfibril. The geometry tells us that if one shortens one side of a triangle without touching the other two, the angle of tilting of the microfibril pair is a function of the shortening and the interval of the two points situated at the base. Simple geometrical calculations lead to adopting the following formula:



f = (e2 – rc2 + 2L. rc)/2rc

(12.1)

where f = angular displacement of the top, e = distance between the bases of two microfi­ brils, rc = shortening of a microfibril, L = length of the microfibrils (Figure 12.2). It demon­ strates that the length between the base and the contact point between the two microfibrils has no influence on the rotation angle of the pair. This angle is a function only of the amount of shortening and the distance between the two base points. If the distance between these two base points is equal to shortening, then the rotation angle reaches 90°. This short­ ening would be generated by the differential sliding of the actin-myosin complex  [12.1] [12.14] [12.15] [12.38] [12.39, pp. 139-140] situated underneath the raphe. The pivoting of myosin heads could be initiated by a depolarization current, itself initiating a wave train (Figure 12.3), that runs longitudinally under the raphe. Perhaps this is started by introduc­ tion of calcium [12.10] as in the reversal of Paramecia cilia [12.1] [12.9] or by the action of light [12.9]. This depolarization current could act on each raphe half independently of one another on the same valve face or on opposite sides, in the same direction or in the opposite direction (Figure 12.1).

Movements of Diatoms VIII  287 Sens du mouvement des microfibrilles

f

rc

Point de liaison Base des microfibrilles

L l e 1

2

3

Figure 12.2  Movements of the microfibrils after their shortening: Position 1: Microfibrils at rest; Position 2: rc, shortening of one of the two microfibrils; Position 3: The two microfibrils, bonded at the point of connection, slanted at an angle according to the value of the shortening. Translation of labels: Sens du mouvement des microfibrilles = Direction of microfibril movement; Point de liaison = Contact point; Base des microfibrilles = Microfibril base. Top: Variation of f as a function of rc. Bottom: General formula f = (e2 – rc2 + 2L. rc) / 2rc. Setting L = 1, f = (e2 – rc2 + 2rc)/2rc. Setting L = 1, e = rc, then f = 1 = L = 90°.

Phase de repos

Phase de contraction des microfibrilles (raccoucissement relatif) Actine Myosine ?

Repos

Contraction

Génération d’impulsions par dépolarisation des charges électriques au niveau du plasmalème

Figure 12.3  The slope of myosin heads generates the sliding of the actin filaments which causes their shortening and the retraction of the microfibrils. Translation of labels: Phase de contraction des microfibrilles (raccourcissement relatif) = Microfibril contraction phase (relative shortening); Phase de repos = Resting phase; Génération d’impulsions par dépolarisation des charges électriques au niveau du plasmalème = Depolarization pulse generation by electric charges at the cell membrane level.

After a rest period for recharging energy [12.4] [12.13], the diatom expels mucus through­ out the whole raphe length accompanied by the creation and growth of microfibrils (Figure 12.4). Microfibrils then leave the raphe through all four raphe halves. Next, a depolariza­ tion current initiated by light [12.9] then passes through the raphe halves or synchronously

288  Diatom Gliding Motility Expulsion du mucus Inclinaison et coup de fouet des microfibrilles

Contraction

Redressement des microfibrilles

Repos

Diatomée Train d’ondes de contraction

Figure 12.4  Birth, growth and slope of the microfibrils in relation to the wave trains of contraction. At rest, there is expulsion of mucus. Translation of labels: Inclinaison et “coup de fouet” des microfibrilles = Tilt and “whiplash” of microfibrils; Expulsion du mucus = Expulsion of mucus; Redressement des microfibrilles = Recovery of microfibrils; Repos = Rest; Diatomée = Diatom; Train d’ondes de contraction = Contraction of wave train.

in the same direction or simultaneously or independently in opposite directions, thereby generating wave trains. The depolarization wave forces the myosin heads to rotate, causing the retreat of one microfibril with respect to the next one (Figure 12.4). The adhesion of the microfibrils, one to the other at a point close to their base, causes the abrupt toppling of two adjacent microfibrils and in this way one after another all the way to the raphe extremi. The brutal toppling generates a “whiplash” on microfibrils that tilt and propel the diatom in the opposite direction of this action (Figure 12.4). However, after the passage of the depolar­ ization wave, the myosin heads are still in the contracted state and in order to start a new contraction it is necessary that they return to their original position, as in muscle fibers. However, in this case, there are opposing contrary fibers that allow the “reset” of the group. In the case of diatoms, there do not appear to be myosin filaments oriented in the opposite direction, at least to our knowledge. It is the expulsion of an “outburst” mucus after each wave passage that would “reset” the system by straightening the microfibrils (Figure 12.4), ready for the next cycle.

12.4 Analysis – Comparison with Observations 12.4.1 Translational Apical Movement This is the one that was the most observed of movements and has guided all hypotheses since those of  Müller [12.29–12.35] and Lauterborn [12.27]. Martens [12.28] made sys­ tematic observations on the movement of organic matter on the raphe by noting the com­ plexity of motion, synchronized or not, or in opposition or not on tandem raphe halves

Movements of Diatoms VIII  289 or one frustule to another. Edgar  [12.12] noticed the extremely small contacts between the diatom and the substrate. These observations were confirmed [12.4] by using video microscopy. Through contraction waves at the base on the microfibrils, synchronized or not between the raphe halves, the ability of the diatom to move either along a raphe half with a vertical tilting of 8 to 10°, as with Navicula radiosa, or organic matter moving in opposite directions along the raphe half, on same valve face, or two movements simultaneously, were demon­ strated. Also, organic matter could be transported in a different direction on the other valve opposite the substrate (Figure 12.5). The wave train caused by the microfibrils allows dia­ tom movement when they are in contact with a point of the organic matter. In the hypoth­ esis described above, the absence of longitudinal movement of the microfibrils causes no mechanical friction in the raphe, contrary to the hypothesis of [12.14]. Similarly, the pres­ ence of fissures oriented perpendicular to the raphe, at the apical extremities and central nodules found in most diatom genera [12.16] [12.17], is no longer an obstacle, despite the speed of diatom movement, since there is no longitudinal displacement of the microfibrils. The displacement speed of the diatom can be as great as the energy transfer allows, since the speed depends only on the frequency of the wave trains depolarization and wavelength. This wavelength must be only a few µm long to explain, for example, the movement of dia­ toms with a very small length such as Navicula seminulum (3 µm minimum). In this case, it would be only one raphe half that acts. It is obvious that every diatom species must have its own system of wave generation as diatoms have different displacement speeds that are independent of their length [12.2] [12.5] [12.8] [12.13]. In addition, the exudation of mucus that allows adhesion [12.41] of diatoms on the glass slide or under the coverslip is in part designed to reset the microfibrils. The presence of mucus as discontinuous traces deposited on the glass slide [12.37] and residual isolated points on the frustule [12.12] confirms this opinion for us.

Émission de mucus

Matière organique entraînée

Sens de progression l’onde

Résidu de mucus

Sens du mouvement de la diatomée Sens de progression l’onde

Figure 12.5  Displacement of the diatom on a half-raphe with organic matter drive on another half-raphe. Translation of labels: Émission de mucus = Mucus emission; Matière organique entraînée = Attached organic material; Sens de progression l’onde = Direction of wave travel; Résidue de mucus = Residue mucus2; Sens du mouvement de la diatom = Direction of diatom movement.

290  Diatom Gliding Motility

12.4.2 The Transapical Toppling Movement This movement has been reported only two or three times [12.18] [12.19] [12.26] [12.37] in order to show that the diatom needed a support to ensure its movement. All diatoms per­ form these lateral toppling movements, which are imperative due to their morphology for Rhoicosphenia abbreviata [12.3], while others, such as Navicula radiosa, use it for finding a spatial position, allowing better contact of the raphe with the substrate [12.4]. This movement was not explained and could not be under the assumptions of  [12.14]. On one hand, the dia­ toms must have or generate long microfibrils corresponding to the half-width of the diatom; thus, for certain Pinnularia, the length must be 10 to 15 µm in order to make contact with the substrate. On the other hand, it is also necessary that the microfibrils contract or tilt, exert­ ing a tensile force while adhering to obtain lateral toppling. If the movement of microfibrils should be synchronous in the same direction or symmetric in the raphes, the result would be a transapical toppling, as was demonstrated with Rhoicosphenia abbreviata [12.3]. In the tran­ sapical pivoting of Cocconeis pediculus, the author was unable to determine from the three proposed options, 1: Il s’agit des trois théories testées : a) propulsion par des microfibrilles : Drum et Hopkins (1966) [12.11]; b) Glissement sur un ruban de mucus; Harper (1967) [12.19]; c) Propulsion par de microfibrilles glissantes dans le raphé: Edgar & Piketts-Heaps (1983) [12.14]. As part of the above proposed hypothesis, it is clear that it is wave generation by the shortening contraction between the anchor points of the microfibrils and raphe, which causes the transapical toppling. Every contraction wave shortens a network of microfibrils and each time decreases the distance between the raphe and a substrate contact point.

12.4.3 Diverse Pivoting  Two main categories of pivoting [12.4] were distinguished: apical and median. Only the median pivoting is discussed in detail because apical rotation is normally deduced from the former. However, the vertical pivoting of Eunotia pectinalis [12.5] will be more fully discussed as it is more complex. The median pivoting mandatorily includes solv­ ing the problem of apical displacement. In fact, this movement encompasses curved tra­ jectories such as those shown by Cymbella tumida, Rhoicosphenia abbreviata, and Eunotia bilunaris [12.4], including the sudden pivoting of Gomphonema acuminatum in the aquatic space [12.4] [12.7]. As part of the new hypothesis, the following mode of action was pro­ posed to solve this pivoting problem. There are two components: a)

a fixed, resistant point creating friction or immobilization of the frustule at a specific point. b) a continuous movement created by the raphe microfibrils. If the resistance point is on the edge of the diatom and away from the raphe, the movement would be slow. If the fixed point is near the raphe, the movement would be extremely fast. On the other hand, if in place of the fixed point there is a weak resistance formed by a drop of mucus, allowing slipping, then the diatom follows a curve as shown in [12.37] and [12.4] [12.36] [12.37]. The curvature orientation, parallel or opposite the raphe curvature, is directly related to the position of the resistance point (mucus) on the frustule rel­ ative to the raphe. The “S” movement on the substrate would be due to

Movements of Diatoms VIII  291 the lateral rocking of the diatom, linked to the rounded frustule, which would offer alternative resistance on either side of the raphe. The apical pivoting described by [12.36] [12.37] for the first time is thus reduced to the presence of a fixed mucus point near the end of the diatom raphe and the continuous action of the microfibrils. It can then perform horizontal, vertical or in any other orientation in the space provided if there is free space around the diatom. The rotation on an extremity of the diatom when pivoting vertically [12.4] is due, in this case, to the laterally off-axis stress exerted by the microfibrils located opposite the apical contact point. Some vertical pivoting seems more difficult to justify, such as Eunotia pectinalis [12.5], where two contradictory hypotheses are presented. However, it was impossible to be able to determine which best meet the observations. In this movement, it is argued that the microfibril action exercises either a continuous translation along the length of the short raphe or it contracts. The particular morphology of this species’ raphe led us to adopt only the contraction by pulsation of the microfibrils under the assumption that was proposed. However, a key point that needs to be checked is the physical ability of the microfibrils to develop the force needed to make these efforts and especially those necessary for vertical pivoting. Ishijima et al. [12.23] and Kuo and Sheetz [12.25] experimentally determined by two different methods that the efforts made by the contraction of the actin-myosin mole­ cule amounts to 1.9 ± 0.4 pN on average. They confirm the previous force measurements: 1-2 pN for muscle fibers  [12.22], 1 pN for a sperm flagellum  [12.24]. It is starting with this value of 1.9 pN that a comparison of the forces needed to pivot Navicula radiosa and Gomphonema acuminatum was calculated in the framework of measured forces developed during displacement studies  [12.8]. Assuming, as previously proposed, that the wave­ length contraction should not exceed 1 μm, the effective active length of a half-wave will be 0.5 μm. Moreover, the actin microfibril diameter would be about 2 nm [12.1]. It would take 500 microfibrils to make 1 μm, therefore only 250 microfibrils would be needed for this action. Assuming that there is only one actin-myosin contractile molecule per microfibril (an unlikely extreme minimum case), the microfibril group would thus develop a force of 4.75 × 10-10 N. The force calculated for pivoting of Navicula radiosa is 1.38 × 10-10 N, thus the microfibril force ratio necessary for pivoting is 3.4. For the polar pivoting of Gomphonema acuminatum the ratio is still 1.45. This shows consistency with the ratio of 3.14 calculated from the force of a unitary constant [12.8] for the same diatom taken in the same condi­ tions. This allows us to conclude that our hypothesis is likely correct.

12.5 Conclusion This hypothesis provides only a portion of the answers to questions posed by diatom move­ ment. It is based on the most complete observations possible and it seems likely that a significant portion of our deductions is consistent with the reality of the phenomenon. However, we are conscious that in everything related to the internal functioning of diatoms, our assumptions are hazardous and are based only on limited experimental confirmations.

292  Diatom Gliding Motility Nevertheless, all the described phenomena are explained by the existing functions. They are present in many zooplankton or phytoplankton and with specific adaptations may suitably apply to diatoms. Nevertheless, it is essential to continue this research in order to increase our knowledge of the internal functioning of the diatoms, i.e., to confirm the existence of depolarization currents; the actual presence of contractile molecules either individually or in pairs; the presence of contraction waves; the link between actin fibers and microfibrils; and the combined action of the mucus and microfibrils; etc.

Acknowledgments The author gratefully thanks the reviewers who indicated important modifications and made useful suggestions. My thanks also go to Dr. Pierre Marsot for his translations, to my wife for the ungrateful job of reviewing this document and all those who, during the 15 years of my research, have encouraged and supported me.

References [12.1] Berkaloff, A., Bourguet, J., Favard, P., Lacroix, J.C., Biologie et physiologie cellulaires. Tome 1. Micofilaments cytoplasmiques, Hermann éditeur des Sciences et des Arts, Paris, 1981. [12.2] Bertrand, J., La vitesse de déplacement des diatomées [Speed of diatom motion]. Diatom Res., 5, 2, 223–39, 1990. [12.3] Bertrand, J., Mouvements des diatomées. I. L’équilibre dynamique chez Rhoicosphenia abbreviata [Diatoms movements .I. The dynamic stability of Rhoicosphenia abbreviata]. Cryptogam. Algol., 12, 1, 11–19, 1991. [12.4] Bertrand, J., Mouvements des diatomées. II - Synthèse des mouvements [Diatoms move­ ments II - Synthesis of movements]. Cryptogam. Algol., 13, 49–71, 1992. [12.5] Bertrand, J., Mouvements des diatomées. III. Le pivotement polaire vertical de Eunotia pectinalis (Kütz) Rab. Essai de quantification des forces [Diatoms movements. III. Vertical polar pivoting of Eunotia pectinalis (Kütz) Rab - Essay on forces quantification]. Cryptogam. Algol., 14, 4, 157–72, 1993. [12.6] Bertrand, J., Mouvements des diatomées. IV. Le mouvement transapical de Cocconeis pediculus Ehrenberg [Diatom movements. IV. Transapical movement of Cocconeis pediculus Ehrenberg]. Cryptogam. Algol., 16, 1, 1–20, 1995. [12.7] Bertrand, J., Mouvements des diatomées. V. Le pivotement polaire de Gomphonema acuminatum Ehrenberg/Movements of diatoms V. The median polar pivoting of Gomphonema acuminatum Ehrenherg. Ann. Limnol.-Int. J. Lim., 33, 4, 211–22, 1997. [12.8] Bertrand, J., Mouvements des diatomées VI. Les efforts pendant le déplacement apical Mesures, analyses, relations: longueur, vitesse, force [Diatom movement VI. Strains during apical displacement. Measurements, analysis, relationships: length, speed, force]. Cryptogam. Algol., 20, 1, 43–57, 1999. [12.9] Cohn, S.A., Spurck, T.P., Pickett-Heaps, J.D., High energy irradiation at the leading tip of moving diatoms causes a rapid change of cell direction. Diatom Res., 14, 2, 193–206, 1999. [12.10] Cooksey, B. and Cooksey, K.E., Calcium is necessary for motility in the diatom Amphora coffaeformis. Plant Physiol., 65, 129–31, 1980. [12.11] Drum, R.W. and Hopkins, J.T., Diatom locomotion: an explanation. Protoplasma, 62, 1, 1–33, 1966.

Movements of Diatoms VIII  293 [12.12] Edgar, L.A., Diatom Locomotion [Ph.D. Thesis], Bristol University, Bristol, 1979a. [12.13] Edgar, L.A., Diatom locomotion: computer assisted analysis of cine film. Br. Phycol. J., 14, 1, 83–101, 1979b. [12.14] Edgar, L.A. and Pickett-Heaps, J.D., The mechanism of diatom locomotion. I. An ultrastruc­ tural study of the motility apparatus. Proc. R. Soc B: Biol. Sci., 218, 331–43, 1983. [12.15] Edgar, L.A. and Zavortink, M., The mechanism of diatom locomotion. II. Identification of actin. Proc. R. Soc B: Biol. Sci., 218, 345–48, 1983. [12.16] Germain, H., The central nodule of Nitzschia obtusae Grunow, in: Proceedings of the Eighth International Diatom Symposium, Paris, August, 1984, M. Ricard (Ed.), Köeltz, Koenisgtein, pp. 227–35, 1984. [12.17] Germain, H., Dissemblance entre les côtés externes et internes de la valve observée en MEB chez quelques diatomées pennées. Cryptogam. Algol., 10, 173–79, 1988. [12.18] Harper, M.A., Movements, in: The Biology of Diatoms, D. Werner (Ed.), pp. 224–49, Blackwell Scientific Publications, Oxford, 1977. [12.19] Harper, M.A. and Harper, J.T., Measurements of diatom adhesion and their relationship with movement. Br. Phycol. Bull., 3, 2, 195–207, 1967. [12.20] Hopkins, J.T., The diatom trail. Microscopy, 30, 209–17, 1967. [12.21] Hopkins, J.T. and Drum, R.W., Diatom motility: an explanation and a problem. Br. Phycol. Bull., 3, 1, 63–67, 1966. [12.22] Huxley, H.E., The mechanism of muscular contraction. Sci. Am., 213, 6, 18–27, 1965. [12.23] Ishijima, A., Doi, T., Sakurada, K., Yanagida, T., Sub-piconewton force fluctuations of acto­ myosin in vitro. Nature, 352, 6333, 301–06, 1991. [12.24] Kamimura, S. and Takahashi, K., Direct measurement of the force of microtubule sliding in flagella. Nature, 293, 5833, 566–68, 1981. [12.25] Kuo, S.C. and Sheetz, M.P., Force of single kinesin molecules measured with optical twee­ zers. Science, 260, 5105, 232–34, 1993. [12.26] Küster, E., Die Gallertbildungen der Amphipleura rutilans [The gelatinous formations of Amphipleura rutilans]. Arch. Protistenkunde, 88, 211–35, 1937. [12.27] Lauterborn, R., Untersuchungen über Bau, Kernteilung und Bewegung der Diatomeen [Investigations on Morphogenesis, Nuclear Division and Movement of the Diatoms], W. Englemann Pub, Leipzig, 1896. [12.28] Martens, P., La locomotion des diatomées [The locomotion of diatoms]. Cellule, 48, 279– 306, 1940. [12.29] Müller, O., Die Ortsbewegung der Bacillariaceen betreffend [Concerning the locomotion of the Bacillariaceae]. Ber. Dtsch. Bot. Ges., 11, 571–76, 1893. [12.30] Müller, O., Die Ortsbewegung der Bacillariaceen betreffend. II [Concerning the locomotion of the Bacillariaceae. II]. Ber. Dtsch. Bot. Ges., 12, 136–43, 1894. [12.31] Müller, O., Die Ortsbewegung der Bacillariaceen betreffend. III [Concerning the locomo­ tion of the Bacillariaceae. III]. Ber. Dtsch. Bot. Ges., 14, 54–64 + Tafel III-IV, 1896a. [12.32] Müller, O., Die Ortsbewegung der Bacillariaceen. IV [Concerning the locomotion of the Bacillariaceae. IV]. Ber. Dtsch. Bot. Ges., 14, 111–28 + Tafel VIII, 1896b. [12.33] Müller, O., Die Ortsbewegung der Bacillariaceen betreffend. V [Concerning the locomotion of the Bacillariaceae. V]. Ber. Dtsch. Bot. Ges., 15, 70–86, 1897. [12.34] Müller, O., Die Ortsbewegung der Bacillariaceen betreffend. VI [Concerning the locomo­ tion of the Bacillariaceae. VI]. Ber. Dtsch. Bot. Ges., 26, 676–85, 1908. [12.35] Müller, O., Die Ortsbewegung der Bacillariaceen betreffend. VII [Concerning the locomo­ tion of the Bacillariaceae. VII]. Ber. Dtsch. Bot. Ges., 27, 27–43 + Tafel II, 1909. [12.36] Nultsch, W., Studien über die Phototaxis der Diatomeen. Arch. Protistenkunde, 101, 1–68, 1956.

294  Diatom Gliding Motility [12.37] Nultsch, W., Die Bewegung der Diatomeen [The movement of diatoms]. Mikrokosmos, 46, 220–27, 1957. [12.38] Poulsen, N.C., Spector, I., Spurck, T.P., Schultz, T.F., Wetherbee, R., Diatom gliding is the result of an actin-myosin motility system. Cell Motil. Cytoskeleton, 44, 1, 23–33, 1999. [12.39] van den Hoek, C., Mann, D.G., Jahns, H.M., Algae: An Introduction to Phycology, Cambridge University Press, Cambridge, 1995. [12.40] Wetherbee, R., Diatom Trail Formation, 2004, J. Phycology, USA, http://www.botany. unimelb.edu.au/RW/media/trails.html. [12.41] Wetherbee, R., Lind, J.L., Burke, J., Quatrano, R.S., The first kiss: establishment and control of initial adhesion by raphid diatoms. J. Phycol., 34, 1, 9–15, 1998.

13 Locomotion of Benthic Pennate Diatoms: Models and Thoughts Jiadao Wang*, Ding Weng, Lei Chen and Shan Cao State Key Laboratory of Tribology, Department of Mechanical Engineering, Tsinghua University, Beijing, China

Abstract

Diatoms are single-celled autotrophic organisms, whose origin can be traced back to the Jurassic Era. They are one of the most important species in the marine ecosystem, playing an important role in the global silicon and carbon cycles. Benthic diatoms move and reposition on surfaces within the sediment to select suitable sites (based on external conditions such as light, nutrients) for proliferation. Many previous studies have been done on diatom locomotion, developing concepts such as the Edgar model to explain the motility, but our observation of periodic pits left in the mucilage trail cannot be explained yet by such models. In this chapter, the physiological structure of diatoms was studied through experimental observations to discover a possible organelle responsible for the pits left over by locomoted diatoms in their mucilage trail. The morphology and mechanical properties of the mucus were then further analyzed. On the basis of the above research, a new hypothesis based on extended structures for diatom locomotion has been proposed and supported in kinematics and dynamics analyses. The locomotion parameters (angle, distance and steps) of diatoms were obtained quantitatively. Moreover, the locomotion and aggregation of diatoms were simulated by the NetLogo system, in which the parameters of perception threshold and angle between diatoms were investigated. However, this new hypothesis of diatom locomotion still leaves some issues that have not yet been clarified, which could be a source for interesting research and discoveries in the future. Keywords:  Aggregation, cell motility, diatom, motility, mucilage, pits, van der Waals force

13.1 Diatom Structure 13.1.1 Ultrastructure of Frustules In this chapter, we mainly focus on Navicula sp., a raphid benthic genus of diatoms, which has the ability of underwater locomotion on substrates [13.11] [13.23]. Unless otherwise specified, all diatoms in this chapter refer to unidentified Navicula sp. diatoms that were collected from Jiaozhou Bay of Qingdao (36°N; 120°E), China.

*Corresponding author: [email protected] Stanley Cohn, Kalina Manoylov and Richard Gordon (eds.) Diatom Gliding Motility, (295–334) © 2021 Scrivener Publishing LLC

295

296  Diatom Gliding Motility

Figure 13.1  Appearance of Navicula sp. diatoms under an optical microscope. Cells are about 14 µm long [13.24].1 (a)

6/19/2009 HV Spot Mag WD Pressure Det Temp 4:02:16 PM 15.0 kV 4.0 14000x 9.9 mm 350.0 Pa GSED …

(b)

−2.0µm− A

10/23/2009 HV Spot Mag WD Pressure Det Temp 1:44:46 PM 15.0 kV 4.0 15000x 10.2 mm 80.0 Pa LFD …

−2.0µm− 091023

Figure 13.2  SEM Image of Navicula sp. in girdle band face (a) and the frustule view (b).2

Photographs of the Navicula sp. diatoms are shown in Figure 13.1 (optical microscope) and Figure 13.2 (SEM). Because the frustules are transparent [13.22], its transmittance makes the inside ultrastructure and many of the internal organelles visible (e.g., chloroplast, lysosomes, etc.). As shown in Figure 13.3, the diatoms we have used have an average aperture of the central macropore of 100-200 nm, with a spacing of the central macropore of 200-300 nm (Figure 13.3a). The diameter of the micropore on the sieve membrane is about 10 nm, which is composed of an approximate regular hexagon (Figure 13.3c,d). This composite secondary pore structure enlarges the specific surface area of the silicon wall, thus facilitating sufficient material exchange between diatoms and the outside world [13.10], and alters some physical properties of the silicon wall, such as flexibility, which will be discussed later. Editors: The first number in Figures, Tables, and Equations refers to the section number of this chapter. Editors: enhanced using ImageJ v2,0,0: Process: Enhanced contrast: Saturated pixels 0.3%: √Equalize histogram. Other images are just footnoted “Editors: enhanced.”

1 2

Locomotion of Benthic Pennate Diatoms  297 (a)

(b)

(c)

200 nm

0.2 µm

(e)

(f)

105.62 nm

190.29 nm 369.72 nm

121.73 nm

174.30 nm

500 nm

(d)

5 0

.0 1.5

20 nm

2.0 nm

0.5

1.0

1.5

2.0

525

350 350

525

175 700 nm

Figure 13.3  Pore array and pore structure on the frustule. (a-c) Meso-porous structure in the frustule, (d) nanoporous strucuture inside mesopores. AFM mapping of pores around ridge (e) and mesopores (f).3 (a)

(b)

(c)

37°

Figure 13.4  The bending ability of the diatom wall during the re-positioning process, scale bar is 5 μm. (a-c) The diatom indicated by the arrow attaches to the wall and the diatom on the right side is changing its orientation by rotating in situ, which the maximum bending angle is about 37 degrees in (b).

13.1.2 Bending Ability of Diatoms Observations of locomoting diatoms show that Navicula sp. diatom in our study exhibited a distinct curvature of its frustules when it rotates in situ, as in Figure 13.4. The diatom indicated by the arrow attaches to the wall with a strong force, and the diatom on the right side is changing its orientation by rotating in situ, from Figure 13.4a to Figure 13.4c. Due to the extrusion between the two, the right diatom shows the maximum bending angle at Figure 13.4b, which is about 37 degrees. Considering that the main component of the frustule is silicon dioxide, and its elastic modulus is about 70 GPa, we believe that the special micro-nano composite porous structure of the frustule significantly changes its bending properties, allowing bending in what would be an otherwise fairly rigid structure. The natural autofluorescence of chlorophyll a [13.21] allowed us to simultaneously observe Editors: enhanced.

3

298  Diatom Gliding Motility (a)

DC Power

(b)

Carbon Electrode

Tungsten Wire

(c)

Figure 13.5  Equipment for analyzing bending ability of diatoms. See text for description. (a) Experimental set up of electrode corrosion for preparing tungsten needles, (b) characterization of an as-prepared tungsten needle, (c) experiment set up of bending ability characterization inside the SEM equipped with a micromanipulator and a force sensor.4

the chloroplast structure along with the frustule bending by using a 543 nm incident light source. In order to quantify the bending capacity of diatom frustules in our study, the relationship between bending deformation and stress of diatom was further measured by scanning electron microscopy and force detector, as shown in Figure 13.5, and the relationship between bending deformation and stress of the frustules could be measured. Tungsten needles with tip diameter less than 5 μm were prepared by electrode corrosion, as shown in Figure 13.5a. Briefly, tungsten wire was cut into 3-4 cm pieces, ground with sandpaper, then subjected to electrolysis corrosion until the diameter was < 5 µm. (Figure 13.5b). Then, using a SEM configured with a micromanipulator (MM3A-EM, Kleindiek Nanotechnik GmbH), diatoms were individually adhered onto a tungsten needle tip by gently approaching the diatom with the needle, then pressing down to make the tungsten Editors: enhanced.

4

Locomotion of Benthic Pennate Diatoms  299 needle fully contact with the diatom. Diatoms became attached to the needle via weak electrostatic interactions. The diatom was then transferred in the SEM to a second tungsten needle containing an SEM glue. The SEM beam was focused on the bonding site for 1 hour to complete curing of the glue, which had a bonding strength of 2 mN upon curing. A force sensor was then placed against the bound diatom in order to ascertain force measurements (Figure 13.5c). As shown in Figure 13.6a, the force detector initially contacted the diatom silicified wall but without exerted pressure the silicon wall kept its original shape. When the vertical force was gradually given, the silicon wall exhibited a distinct bending deformation, as shown in Figure 13.6b. At this time, the perpendicularly measured pressure was 1.8 μN, the bending angle was about 26 and the silicon wall was bent without obvious fracture. The measurement results agreed with observation of a locomoting diatom (Figure 13.4). Comparing the SEM results with the bending effect observed in the actual locomoting diatom (Figure 13.4), the lateral force of the bent diatom should also be of the magnitude of 2 μN. It is clear that the adhesion of diatom to the substrate wall is sufficient to support the lateral friction at that magnitude, which may be an important contributor to diatom gliding. In order to estimate the bending characteristics of the frustule, we can model the frustule as a simplified cantilever beam. As shown in Figure 13.7, assuming that the cross section of the circular ring representing the diatom is De = 4 μm, Dn = 3.8 μm (in which the thickness of the frustule is 100 nm), and the length of the cantilever is L = 15 μm, the inertia moment of the cross section of the circular ring can be calculated as [13.9]:

I=



4 π De4   Dn   −24 4 1 −     = 2.33 × 10 m 64  De 

(13.1)

According to the deflection equation of the free end of the cantilever beam:

FL3 w( L ) = 3EI



(13.2)

Taking the concentration force F as 1.8 µN, the deflection of the free end is approximately

w = L × tan θ = 7.32 μm (a)

2 1.8

(b)

(c)

1.6 1.4

26°

1.2

F/µN

1.8µN

1 0.8 0.6 0.4 0.2

11:29:30

11:29:00

11:28:30

11:28:00

11:27:30

11:27:00

11:26:30

11:26:00

11:25:30

11:25:00

11:24:30

11:24:00

11:23:30

0 –0.2

Ub=2.5V Gain=100 Scaling: F(U)=1µN/V x U

Figure 13.6  The relationship between bending deformation and stress of frustules. The scale bar is 5 μm. (a) original position, (b) bending deformation with an angle of 26 degree, (c) forces measured with deformation.

300  Diatom Gliding Motility L

F De Dn θ

Figure 13.7  Simplifying the frustule bending into a simple cantilever beam system.

where θ = 26°. It can be obtained from Eq. (13.2):



E=

FL3 = 119 MPa 3wI

(13.3)

This calculated elastic modulus E (119 MPa) of the silicon wall is much smaller than that of bulk silicon dioxide (around 70 GPa). Therefore, the hypothesis is that the structure of the composite porous structure on the silicon wall of these diatoms has physical properties significantly altered from bulk silicon dioxide that enhances its bending ability. In conclusion, Navicula sp. diatoms in our study have a far smaller bending modulus than bulk SiO2, which could be due to the composite porous structure on the silicon wall or other properties of the frustule matrix. Of course, the rigidity varies between different species and the degree of silicification could be affected by various conditions. The testing protocol described could be used to help determine the frustule flexibility of various species.

13.2 Models for Diatom Locomotion 13.2.1 Edgar Model for Diatom Locomotion It is generally believed that diatom locomotion is controlled by actin-myosin motor system mediated by extracellular proteoglycan. The cytoskeletal protein actin is often comprised of fibers and is involved in many physiological processes, such as the formation of cellular processes (pseudopodia), cell division and intracellular transport. Since cables of actin filaments have been observed running parallel to and underlying the raphe, and mucus has been found secreting from the raphe [13.4], an early model was proposed by Gordon in which the mucilage secretion generated a net propulsion force to move the diatoms ([13.13], updated model presented in this volume [13.12] using actin cables to localize the secretions). However, this model was not considered to describe the active regulated movement of diatoms well, and a subsequent model was proposed by Lesley A. Edgar, referred to in this article as the Edgar model [13.5–13.8] [13.20]. As diagramed in Figure 13.8, the model requires the presence of one or more transmembrane proteins that connect the actin cables underneath the cell membrane with the mucus outside the cell membrane. Molecular

Locomotion of Benthic Pennate Diatoms  301 Gliding

Actin Filament

Molecular Chaperone

Transmembrane Protein

Mucilage trail Mucus molecule

Figure 13.8  Classical Edgar model. See text for description of components.

Figure 13.9  Pits found in the mucilage trails. Scale bar equals 2 μm.5

motors such as myosin, acting along the actin cables, facilitate the transport of the transmembrane proteins and the associated mucilage down the raphe [13.4]. When the diatom glides, the connecting mucilage strands are dragged down the raphe canal, and because the bottom of secretory mucus is firmly in contact with the substratum, the diatom crawls in the opposite direction of the motor protein. At the same time, the surface of the silica wall is covered with a hydrophobic grease film, and the diatom can slide along the mucus. When the mucus reaches the tail end of the diatom, it is cut off by the constricted opening of the raphe and left on the substratum to form a mucilage trail [13.19]. Drug studies have shown that both latrunculin and butanedione monoxime, which inhibit myosin, can effectively inhibit the locomotion of diatoms [13.17] [13.20]. Although the Edgar model has been generally accepted [13.7], we have observed a characteristic of mucilage trails that cannot be explained by the model. As shown in Figure 13.9, microscopic analysis has shown that some regularly spaced pits were present in the Editors: enhanced.

5

302  Diatom Gliding Motility

Figure 13.10  Diatoms locomoting while raised at an angle of inclination.7

mucilage trail left behind after the diatoms’ locomotion.6 The Edgar model does not include any components that would explain such periodic pits in mucilage trails. †

13.2.2 Van der Waals Force Model (VW Model) for Diatom Locomotion 13.2.2.1 Locomotion Behavior of Diatoms Some diatoms that we observed did not have the valve face directly in contact with the substratum during their locomotion, but showed motility while the cell displayed a certain inclination angle, as shown in Figure 13.10. The outline of the silica wall and the shape of the chloroplast indicated that arrowed diatom was tilted. Figure 13.11a showed the state of a diatom crawling normally (valve face and raphe in direct contact with the substratum), and Figure 13.11b showed the state of a diatom when it is crawling obliquely. When the diatom locomoted obliquely, it was difficult to ensure that the diatom secretion spanned the distance between the cell and the substratum because the raphe was far from the substratum. In another group of observations, 5-10 μm diameter silica particles were randomly dispersed in a sample of motile diatoms to observe the interaction between the particles and diatoms during movement. As shown in Figure 13.12, the same diatom can crawl along the raphe (Figure 13.12a-b) or the girdle band (Figure 13.12c-f), although it is possible that the cell is making some mucilage contact through the raphe and making contact with relatively immotile beads. Although some of the electron micrographs of mucilage/cell connections show the length of some of the mucilage strands to be far in excess of the small raphesubstratum distance for cells in direct contact, allowing for the possibility that the cells can be pulled even while on their girdle side [13.14], we believe that may not be happening 6

Editors: This might be due to periodic reshaping of a cylinder of fluid, as reviewed in: Gordon, R., Goel, N.S., Steinberg, M.S. and Wiseman, L.L. (1972) A rheological mechanism sufficient to explain the kinetics of cell sorting. J. Theor. Biol. 37(1), 43-73. 7 Editors: enhanced. †

Locomotion of Benthic Pennate Diatoms  303 (a)

(b)

Actin

Raphe

Mucus Obstacle

Figure 13.11  Cross-section view of diatom locomotion, (a) normal locomotion, (b) inclined locomotion.

(a)

200 px

(b)

200 px

(d)

200 px

(c)

200 px

(e)

200 px

(f)

200 px

Figure 13.12  The same diatom crawls along the raphe (a-b) and its girdle band (c-f) surfaces.8

here. If the cell were moving solely by short-range raphe-based mucilage, then the mucilage secreted by the diatoms would first adhere to the particle, and their locomotion would be around the particles. However, detailed image comparisons show that diatoms in girdle view can not only move, but can transport beads short distances in the same direction. For example, in Figure 13.12e-f the diatom generally had a displacement towards the upper left direction, while the silica particle also had a displacement to the upper left. If the diatom locomoted around the surface of the particle, we believe the particle should show the opposite movement trend to the direction of the diatom movement. Figure 13.13 further demonstrates that diatoms can glide on a surface while situated with their girdle bands in closest contact.

Editors: enhanced.

8

304  Diatom Gliding Motility

Figure 13.13  Diatoms crawl with the girdle band facing the substrate. (a)

(b)

(c)

(d)

Figure 13.14  The circular structures in the body of some locomoting diatom cells. (a-d) are serial captures of the same diatom while locomoting, which two circular structures were observed to move within the cell body at high-frequency vibration at the microscale.9

13.2.2.2 Moving Organelles and Pseudopods We also investigated the internal structure of cells by observing a large number of moving diatoms in vivo that had been placed on acid-cleaned glass slides. As shown in Figure 13.14, two circular structures, about 1 μm in diameter, were observed inside the diatoms, generally located at either end of the long apical axis. The circular structures were observed to move within the cell body at high-frequency vibration at the microscale. In most cases, however, the circular structure in crawling diatoms is not obvious, as shown in Figure 13.15. From our laser scanning confocal microscope (LSCM) observations, the proportion of gliding diatoms with a discernable circular structure was about 6.5% (16 out of 248 motile diatoms). In contrast, the proportion of static diatoms having such structures was less than 1%. While we suggest that there may be a correlation between diatom locomotion and circular structure, this is not conclusive. However, we feel this structure was often not visible because the refractive index of the circular structure may be extremely similar to the general protoplasm in the diatom, making observation under the light microscope difficult. In order to better understand the chemical composition of the circular structure, whole cell staining of diatoms was carried out using the actin-binding probe FITC-Phalloidin (Enzo, Switzerland) to determine if actin was present in these structures. Filamentous actin exists widely in eukaryotic cells and contributes to many cell processes such as cytoplasmic division and intracellular transport of substances. After staining, diatoms were washed by

Editors: enhanced.

9

Locomotion of Benthic Pennate Diatoms  305 (a)

(b)

Figure 13.15  Obscure circular structures in locomoting diatom cells. Circular structures could be observed in some diatoms (a) but not always could be found (b) even under LSCM.10

Figure 13.16  F-actin (green) stained by FITC-Phalloidin. Red color is due to autofluorescence of the diatom chloroplasts. Scale bar equals 5 μm.11

PBS and observed with a 488 nm laser LSCM (laser scanning confocal microscope). As shown in Figure 13.16, a high concentration of phalloidin aggregated into circular structures at both ends of the long axis of diatom, along with lower-level staining along the raphe. According to Mann [13.18], these circular structures are almost certainly volutin granules which are associated with a microfilamentous ring, which may help form the contractile ring during cell division. As phalloidin strongly stains F-actin, we suggest that

Editors: enhanced. Editors: enhanced.

10 11

306  Diatom Gliding Motility F-actin is rich around/in these volutin granules at the ends of the diatoms. A hypothesis of our model is that the F-actin associated with these granules could act in concert with molecular motors to produce motile force at the ends of a diatom. When a 543 nm laser was used as the incident light source, chlorophyll A emitted red light due to autofluorescence, so the position of chloroplasts could be clearly located within the cell. Figure 13.17a–c show a time series of a moving diatom. The diatom encountered obstacles during its crawl, causing it to slow down and eventually stop. Although the size of the obstacle was much larger than that of diatom, it still showed some displacement, indicating that the diatom was generating significant motile force. Figure 13.17d–f shows that the circular structure within the diatom moved backward along the long axis of diatom and gradually squeezed against the chloroplast. Such analysis of the motion of moving cells led us to three conclusions: 1) the circular structures often were observable within the cell body of actively moving diatoms; 2) the circular structure could produce deformation of structures within the cell; 3) there might be “space competition” between chloroplast and the circular structures when diatoms crawl if the movement of the circular structures is related to the diatom motility. Diatoms undergoing movement were further analyzed using LSCM Z-stack observations at different focal planes. Figure 13.18 showed the result of tomography of the diatom from bottom plane (Z axis coordinate is 1.82 μm) to top plane (Z axis coordinate is 7.60 μm). Based on the focal planes, the clearest position of the circular structure was very close to the focal plane of the obstacle, indicating that the two were at nearly the same level within the cell. In other words, the most prominent area of the circular structure was on the side of the frustule near the substratum.

(a)

(b)

(c)

(d)

(e)

(f)

Figure 3.17  Change of circular structures and space competition with chloroplasts of locomoting diatoms after encountering obstacles. (a-c) a diatom was approaching an obstacle, in which circular structures were clearly seen and chloroplasts was stained with red color. (d-f) Since the diatom can not move the obstacle, it chose to return and the circular structure was squeezing against the chloroplasts to get itself backward.

Locomotion of Benthic Pennate Diatoms  307 (a)

(b)

(c)

(d)

Figure 13.18  Z-stack scanning of a diatom from bottom plane (Z axis coordinate is 1.82 μm) to top plane (Z axis coordinate is 7.60 μm), corresponding from (a) to (d). (a)

mucilage trails

(b)

mucilage trails

(c)

232.69 nm

glided diatom

527.43 nm

217.87 nm

Figure 13.19  Mucilage trails under SEM, in which pits could be clearly seen in (b) and (c). Scale bars equal to 5 μm (a) and 1 μm (b, c).12

13.2.2.3 Chemical Properties of Mucilage Trails Extracellular polymeric substance (EPS) secreted by diatoms is mainly composed of biological macromolecules with high water content, so it is difficult to be observed by SEM after conventional freeze-drying treatment. Chemical fixation was used to protect the structure as much as possible. As described previously [13.14], a drop of diatom culture solution was placed on a silicon wafer for 1 hour to allow the diatoms to glide and produce mucilage trails. A few drops of Alcian blue were added to the surface of the silicon wafer, after which 2% glutaraldehyde was added to react for 1 minute to fix the biological material. A mixture of osmium tetroxide (1% volume fraction) dissolved in 0.1M arsenite was then placed on the surface of silicon wafer for 45 minutes to fix any membranous material. Finally, the silicon wafer was washed with deionized water, dehydrated in ethanol and air-dried at room temperature. After this chemical fixation, the mucilage trails left behind by diatoms could be observed by SEM, as shown in Figure 13.19. The average width of the trails was about 450-550 nm, and there were small circular pits with diameters of about 200-250 nm scattered on it. The approximate width of mucilage trails was similar in size to the width of the raphe, which was about 160 nm. A preliminary energy spectrum analysis (EDAX – Energy Dispersive X-ray Analysis) of mucilage trails was made using the SEM. As shown in Figure 13.20, the energy spectrum of mucus and background silicon wafers were qualitatively compared. It was found that the mucus has a higher C and O concentration. The very high concentration of Si signal of the mucilage is likely a background signal due to the silicon wafer used as the substrate. Editors: enhanced.

12

308  Diatom Gliding Motility Elements

9/1/2011 HV Spot Mag WD Pressure Det Temp 8:13:23 PM 15.0 kV 4.0 29000x 10.2 mm … ETD …

2.0µm S+5

Wt%

At%

Mucus

Wafer

Mucus

Wafer

C

5.92

2.13

12.72

4.84

O

1.33

0.27

2.14

0.47

Mg

0.31

0.21

0.33

0.23

Si

91.47

96.77

84.08

93.97

S

0.31

0.24

0.25

0.21

Cl

0.66

0.38

0.48

0.29

Figure 13.20  EDAX of mucilage trails and silicon substrate.13

Considering the large accumulation of F-actin in diatoms (i.e., in the circular structures) and the presumed requirement for actin and diatom locomotion, it is preliminarily speculated that actin aggregates with a diameter of about 100 nm extend out of the frustule from the raphe, thus leaving a series of circular pits on the mucilage trails. The circular structure is the “storage pool” for these actins. Analysis of the composition of EPS and mucilage trails by Raman spectroscopy was used to further understand the characteristics of diatoms’ secretion. Stainless steel surfaces of 3 cm by 4 cm were used as substrates and placed in diatom culture medium for about 4 hours to allow diatoms to adhere. Subsequently, the culture dishes and stainless steel samples were scanned by a Raman spectrometer (LabRAM HR 800, Horiba Jobin Yvon), with an argon ion laser (514 nm, 10 mW) for 30 s using a spectral resolution of 1 cm-1. Both the mucilage trail of diatom (the area after the diatom) and the EPS of diatom, (using the area near the aggregation of diatoms) were scanned. As shown in Figure 13.21, for the Raman spectra of mucilage trails and EPS, the main spectral bands were concentrated in 400-3000 cm-1. The bands of the two kinds of secretion were generally the same, with peaks such as: 599 cm-1 and 594 cm-1 representing phenylalanine, 968 cm-1 and 966 cm-1 representing carotenoids, 1090 cm-1 and 1088  cm‑1 representing polysaccharides, 1612 cm-1 and 1619 cm-1 representing tyrosine, among other peaks. Composition and classification of the various spectral bands are summarized in Table 13.1. However, some bands are unique to EPS, including: 1437 cm-1, likely representing deformed -CH2- structures, 1655 cm-1 representing C=O structures and 2882 cm-1 and 2936 cm-1 representing the extensional structure of -CH2- and -CH3-, respectively. Table 13.2 lists the properties of the aromatic amino acids, phenylalanine and tyrosine, both of which appear to be contained in diatom mucilage. Since phenylalanine tends to be hydrophobic (repelled by water environments) and tyrosine is hydrophilic (associates well with water environments, the mucilage is likely amphipathic, which partly explains why EPS can adhere to both hydrophilic and hydrophobic surfaces. These attractions are likely relatively weak, as both amino acids in diatom EPS have a large number of nitrogen and oxygen atoms (as well as the OH group on the tyrosine) that act as sites of hydrogen bond formation. Editors: enhanced.

13

Locomotion of Benthic Pennate Diatoms  309 1437

1655

2936

a.u.

SM 595

1339

966

1612

1090

TM-1 TM-2 TM-3 500

1000

1500 2000 Wavenumber (1/cm)

2500

3000

Figure 13.21  Raman spectra of mucilage trail. SM denotes the spectra from secreted mucilage, while TM-1, TM-2, and TM-3 represent the spectra from three representative mucilage trials.

Table 13.1  Composition and category of material corresponding to Raman spectrum band. Wave number (cm-1)

Location

Corresponding materials

Category

435

Both

SiO2

SiO2

594-599

Both

Phenylalanine

Proteins

966-968

Both

Carotenoids

Carotenoids

1088-1090

Both

Stretching of C-H of carotenoids

Carotenoids

1339-1344

Both

Stretching of C-N of Chlorophyll A

Chlorophyll A

1437

EPS

Deformation of -CH2-

Proteins

1612-1619

Both

Tyrosine

Proteins

1655

EPS

C=O

Polysaccharides

2882 & 2936

EPS

Stretching of -CH2- & -CH3

Polysaccharides

Overall, both secretions contain proteins and polysaccharides, with phenylalanine and tyrosine as the main components. However, there are some differences between the two components. The EPS has a small amount of unique polysaccharide structure in addition to the protein components, which may be the fundamental reason for its greater adhesion.

310  Diatom Gliding Motility Table 13.2  Related properties of amino acids in diatom EPS. Phenylalanine (Phe)

Tyrosine (Tyr)

Formula

C6H5CH2CH(NH2)COOH

C9H11NO3

Molecular Weight

165

181

Structure

O

O

OH

OH NH2

HO

Hydropathy index

2.8

−1.3

Number of hydrogen-bonded donors

2

3

Number of hydrogen bond receptors (nitrogen and oxygen)

3

4

NH2

13.2.2.4 Mechanical Properties of Mucilage Trails In order to better understand the structure and mechanical properties of mucilage trails, the fresh mucus secreted by locomoting diatom was analyzed by atomic force microscopy (AFM). Several drops of diatom-containing culture solution were dripped onto a glass slide, and allowed to incubate for 30 minutes. A crawling diatom on the surface was then located and tracked. Because the mucus was invisible without dyeing or fixation, the AFM tip could only be tentatively probed into the area where the diatom had just glided. When the position of the mucilage trail was found by analyzing the force signal from the tip, the force curve of the lattice was measured perpendicular to the gliding direction to obtain data measuring the force characteristics of the mucilage secreted by the diatom. The radius of the AFM probe used in this experiment was about r=5 nm. After obtaining the lattice measurement of the force curve near a typical mucilage trail, the maximum adhesion at each measuring point was counted and the mean values were taken to draw the adhesion distribution map as shown in Figure 13.22. The width of each square grid represents 100 nm, so the width of the mucilage trail area measured was about 500-600 nm, which is consistent with the observation of SEM (Figure 13.19). From the graph, the adhesion at different locations was significantly different, and the general trend was that the greatest adhesion was along the central line of the mucilage trail, whose value was greater than 7 nN. From the central line to the edge, the adhesion gradually decreased, and it was about 2 nN on the edge. The topology from AFM (Figure 13.23) showed that the mucilage trail was thick in the middle and thin at the sides, the maximum height was 200 nm. It inferred that thicker parts gave out greater adhesion, and vice versa. Figure 13.24 displays a typical force curve measured on a mucilage trail by AFM. Figure 13.24a was the force curve between the probe and the glass substrate in the culture medium, and there was only a slight attachment (about 0.3 nN) during the separation. However, when the same probe was used to measure the mucilage trail, its maximum adhesion increased

Locomotion of Benthic Pennate Diatoms  311 12 11 10 9 8 7 6 5 4 3 2 1

7 6 5 4 3 2 1 1

2

3

4

5

6

7

8

9 10

(nN)

Figure 13.22  Lattice map of adhesion of the mucilage trail.

(a)

(b) 300 nm

300 nm

250

250 5 µm 4

150

300 nm 150

3 2

0 5 µm 4

1

3

5 µm

200

100

4

300 nm

3

150

50 0

0

2

5 µm 4

1

3

200 150 100 50 0

2

2 1

1

0 0

0 0

Figure 13.23  Topology of a piece of mucilage trail under AFM Scanning: (a) & (b) are two randomly selected sections.

significantly to about 5 nN. As shown in Figure 13.24b, two attachments are noted during the return journey, marked at point 1 and 2 respectively. At point 1, it was considered that the probe separated from the wall. At this time, the adhesion force was the greatest and recorded as Fsmax (the maximum force on the substratum). At point 2, it was considered that the probe separated from the trail mucilage. At this time, the adhesion force was recorded as Fmmax. The statistical result of 60 force curves shows that Fsmax mainly distributes in the range of 2-9 nN with an average value of 4.51±1.72 nN, and Fmmax mainly distributes in the range of 0-2.5 nN with an average value of 1.31±0.24 nN. Since the thickness of the mucus was about 200 nm, it could be inferred from Figure 13.24b that the tip of the probe had been completely immersed in the secretion when measuring the mucus force curve, so the adhesion force at point 1 was the interaction force between the tip and the substratum with the mucus as the medium. Because the physical parameters of the two media (such as Hamaker constant,14 dielectric constant, etc.) were different, the van der Waals force and electric double layer force between the probe and the wall were changed, thus the final measured adhesion was different. In other words, Fsmax was closely related to the properties of the medium near the wall. ‡

Editors: https://en.wikipedia.org/wiki/Hamaker_constant.

14 ‡

312  Diatom Gliding Motility (a) 35

(b)

30

30

25 20

20

Force (nN)

Force (nN)

25

15 10

15 10 5

5

0

0

-5

-5 -50

0

50

100 150 Distance (nm)

200

1

-10 -50

250

0

2 50

100 150 Distance (nm)

200

250

Figure 13.24  Typical AFM measurements in culture medium (a) and mucus (b).

The adhesion forces of the mucilage trails and the mucilage within the EPS were compared using AFM, as shown in Figure 13.25. Differences between these two types may mainly come from two factors: 1) as shown in Table 13.1, there are differences in the composition of the two kinds of secretions, especially in the viscous mechanical properties; 2) the differences in the spatial configuration of the molecular structure within the two kinds of secretions could lead to different steric forces, and thus different force characteristic curves [13.3]. For the mucilage in the trail, as shown in Figure 13.26a, when the probe contacts the base, the distance shown in the abscissa is zero, and the corresponding adhesion is zero. The tip of the probe started to return, and the detected force had a linear relationship with the distance, which indicates that the elastic deformation of the substrate surface occurs. Upon further return, as the cantilever beam at the tip moved up within the mucilage, the adsorption force between the tip and the substrate was generated and reached the highest value, as shown at point A. Subsequently, the probe kept moving away from the substrate, and the force reading decreased rapidly to near zero. Such a rapid change at very short distances could be regarded as a process of desorption between the probe and the substrate. However, the probe was still immersed in the mucus. As the probe continued to move away from the substrate, the adhesion remained almost zero within a certain distance, as shown in the corresponding interval of point B in the figure. When the probe tip reached point C, it detached from the mucilage and produced a second adhesion force

250

20

200 Force (nN)

(b) 300

25

Force (nN)

(a) 30

15 10 5

100 50

0

0

-5 -10 -50

150

-50 0

50

100 150 Distance (nm)

200

250

-2

-1

0

1 2 3 Distance (nm)

Figure 13.25  (a) Force curve against mucilage trails, (b) force curve against EPS.

4

5

Locomotion of Benthic Pennate Diatoms  313 (b) Cantilever Deflection

Cantilever Deflection

(a)

(A)

(B) (C) (D)

Separation Distance

(A)

(B)

(C)

(D)

(E)

Separation Distance

Figure 13.26  Schematic diagram of return process after contact between probe tip and wall. (a) A typical force curve against the trail mucilage, (b) a typical force curve against EPS.

signal with a smaller value. For the EPS, as shown in Figure 13.26b, unlike the force curve with the trail mucilage, the adhesion force almost linearly correlated with the distance from the maximum value to zero in a certain range during the whole return journey of the probe away from the substrate, until the moment the EPS was pulled off. It is worth noticing that the adhesion force between the probe and the mucus was almost zero in a considerable distance away from the substrate except for the two instants of the substrate and mucus detachment, while in the EPS, the adhesion force on the probe in the whole return process was increasing. In fact, the maximum adhesion of the EPS was about 7.1 times of that of the mucus, and the action distance of the EPS was about 25.8 times of that of the mucus. Moreover, besides proteins, the EPS also contained a large number of carbohydrates, so it was presumed that there were proteoglycan polymers in the adherent mucus that contributed in the adhesion force. The relative molecular weight of these polymer monomers could reach the order of 2×106, and it could be composed of hundreds of monomers in a huge polymer matrix. In conclusion, the total steric force in the EPS is far greater than that in the mucus. It is not difficult to understand due to the much smaller amount of mucilage and the thinner thickness (no more than 200 nm) within the diatom trail, in which only a limited amount of polymer can be spatially accommodated. Thus, the morphology and structure of the mucilage in the trail are likely relatively loose and the tangling formed between the polymers is simple, as illustrated in Figures 13.26 and 13.27. When it contacted with the AFM probe, the polymer gradually stretched with the probe return process, but the steric force between the polymer molecules was relatively small, so the adhesion force was almost zero at a considerable distance near point B in Figure 3.26a, and the low entanglement of the mucus made its action distance quite short. In contrast, for the EPS, as shown in Figure 3.26b, because of its large secretion amount, a large number of adhesive molecules accumulate in a limited space, so it is more complexly tangled and highly compressed. When the probe was lifted in the EPS, closely packed polymers attracted each other, so it maintained a large adhesion force and long action distance. Therefore, the polymers and material EPS play a far greater role in overall diatom attachment than the mucilage within the trails produced by movement.

314  Diatom Gliding Motility (a)

(b)

Substrate

Substrate

Figure 13.27  Schematic diagram of polymer morphology of the mucus (a) and the EPS (b). Gliding Diatom Actin

Girdle band Mucus

h d Mucus Substrate

Figure 13.28  Function of the mucus.

13.2.2.5 VW Model for Diatom Locomotion In the marine environment, van der Waals electrostatic forces and hydrogen bonding between materials are often significantly decreased compared to freshwater because of the much larger concentrations of ions, resulting in the far greater ionic strength in the seawater. Therefore, it is reasonable to infer that one result of diatoms mucilage secretion is the enhancement of adhesion between the diatoms and substrates, providing a more adherent matrix through which to glide (see Figure 13.28). In addition, under our actin protrusion model (see below) the actin contraction is equivalent to the detachment process of the AFM probe and the substratum that produced force. Under this scheme, our model for locomotion has been named the Actin-Pin model, in which the force produced by extension and retraction of actin protrusions provide the force for movement [13.24]. Based on our work of observations and analysis of Navicula sp. collected from Jiaozhou Bay of Qingdao, China, we propose this model of diatom movement as follows. Actins, concentrated in the cell, could extend out of the frustule from the raphe and contact the secreted mucilage adhered to the substratum. As the actin protrusions extend, the diatom then generates static friction in the horizontal direction as the driving force of locomotion through the positive pressure from the actin extensions on the mucilaginous substrate. It is important to state, however, that there is no known motile generation in other organisms that uses extracellular actin filaments to drive motility,15 or that use extension and retraction of filaments through the cell membrane. Nor have any actin filaments, as suggested in our model, been found protruding through the raphe fissure. It is possible that the actin §

Editors: But see: Goldberg, M.B. (2001). Actin-based motility of intracellular microbial pathogens. Microb. Mol. Biol. Rev. 65(4), 595-626.

15 §

Locomotion of Benthic Pennate Diatoms  315 could still be associated with the circular structure simply because it is a site of actin storage, or due to some type of artifact in the phalloidin staining, and other than the evidence of the pits, no special residue was found throughout our observations. Nonetheless, we feel it is important to explain our model so that researchers will be better able to test its validity with further observations and experiments. In the Actin-Pin model, the role of the secreted mucilage and two or more actins extending through the frustule at the raphe is essential [13.24]. The protractile actin serves as the functional structure providing the force of diatom locomotion. Two or more groups of actin extending at different positions within the cell coordinate with each other and produce adhesion and positive pressure through force balance between the two extended sets of actins, generating static friction as the driving force. Because it is difficult for diatoms to generate a large positive pressure in a liquid environment where gravity and buoyancy are almost balanced, the secreted mucilage significantly increases the adhesion between actins and the underlying substratum by changing the medium in between them. The process could be further regulated as the circular structures containing the presumptive protractile actins can move horizontally along the raphe, possibly driven by the actin cables underlying the raphe using molecular motors, producing the gliding via alternating motions of the actins. The dynamics of the Actin-Pin model is further elaborated in Figure 13.29, which shows the step diagram of a diatom locomoting to the right. The red parts are the circular structures, from which the orange actin is extended. The blue layer between the diatom and the substrate is the secreted mucilage, and the blue arrow is the resulting force between the diatom and the substrate. The systematic protrusions from the actins through the raphe would thereby produce the pits within the mucilage trails that we have observed. In summary, when the diatom comes into contact with the substrate, it secretes mucilage through the raphe, at which time two (or more) actin fiber protrusions (containing more than one actual actin filament) extend through the raphe, extending and inserting into the mucus. In the case of a diatom locomoting to the right, as shown in Figure 13.29a, the actin protrusion on the right side is squeezed downward and gets pushed upward instantaneously as a result of force balance. The actin protrusion on the other side is subjected to an upward force greater than its maximum adhesion to the mucus, thus gets detached from the substrate, leaving a pit on the mucilage trail. At the same time, actins have good deformability and can be easily detached by changing their shape to reduce the adhesion force between them and the substrates. Then, as shown in Figure 13.29b, the actins on the right maintain a firm adherence to the substrate, and the left circular structure moves to the right in the diatom for a distance. Then the left circular structure extends more actins and squeezes down again, as shown in Figure 13.29c. There are three effects here. First of all, the left actin squeezes down to produce positive pressure, which balances the adhesion force between the right actins and the substrate. Secondly, the left actins maintain a firm adherence to the substrate and keeps relatively static, and at the same time create static friction under the positive pressure to overcome the external resistance and move the diatom to the right (Figure 13.29d). Then, similar to Figure 13.29a, the right actins detach from the substrate and, as shown in Figure 13.29e, the right circular structure moves to the right, extends actin and squeezes EPS downward again, so that the diatom returns to the state of Figure 13.29a and completes a motion cycle. Under cryo-TEM we also observed that the membrane of some diatoms was found with distinct protuberant structures, as shown in Figure 3.30, extending into the area of the raphe. It may be that the raphe part of the protrusion extends out of the frustule in living

316  Diatom Gliding Motility

(a)

Circular Structure Diatom

Mucus

Actin (b)

(c)

Drag Force

(d)

(e)

Figure 13.29  Scheme of a cycle of repeatable steps (a-e) when a diatom is locomoting to the right in VW model. Briefly, two circular structures alternately use actins as anchors and changing their positions inside the diatom thus generate forces to achieve the locomotion, please refer to the detail description in the text.

diatoms, and appears to be inside of the frustule in our images due to water loss during sample fixation and preparation. We can use estimates of intermolecular forces to determine the forces that might be present on a protrusion extending into the mucilage onto the substratum. Generally speaking, the causes of adhesion force include electrostatic attraction, intermolecular forces (van der Waals forces) and hydrogen bonding. Because diatoms are making attachments under the water, many of the weaker electrostatic forces are likely mitigated by the polar water molecules. In terms of van der Waals force, a protrusion with radius r = 5 nm can obtain the maximum adhesion with average F = 4.5 nN, which can be simplified to a sphere near an infinite plane. When the distance between the two is very small (D = 1 nm), significant van der Waals attraction can be obtained:

Locomotion of Benthic Pennate Diatoms  317

Figure 13.30  The processes on the cell membrane of diatom contacting the frustule at the raphe.16

F=



Ar 6D 2

(13.4)

where A is the Hamaker constant. By substituting r, F and D, we can get the interfacial relationship between the probe and the silica substrate in the mucus, A = 5.4×10‑18 J, which is significantly larger than the Hamaker constant (3.7×10-20 J) measured in water [13.188]. Based on this, the van der Waals force between the actin and the substrate can be estimated. Because of the large surface area of the actin protrusion (according to Figure 13.18, radius r = 100 nm), the van der Waals force between the protrusion and the substrate is assumed to be a two-plane model.

F A 2 = 6π D 3 πr



(13.5)

Therefore, the van der Waals force can be calculated to be 9.0 µN. Actually, more than one actin may be extended, so the maximum adhesion force obtained by diatoms is very likely to be larger than that calculated by this method. Editors: enhanced.

16

318  Diatom Gliding Motility Considering that the underwater friction coefficient is usually 0.01-0.2 [13.15], when the vertical adhesion (i.e., positive pressure) reaches 9.0 µN, the corresponding friction force is about 0.1-1.8 µN. The friction force is the maximum lateral force that an extended actin protrusion can withstand. The bending force corresponding to the bending degree of a diatom during the locomotion (Figure 13.7) is also around 1.8 µN (Figure 13.6), which is in the same order of magnitude as the calculated friction force. Thus, the above locomotion model seems able to generate a force on the order of magnitude observed from the frustule deformation. As mentioned above, the function of the mucilage secreted during locomotion also needs to be further clarified. In our new hypothetical locomotion model, diatom gliding is powered by actin protrusions which are introduced into the mucilage. High-speed crawling of diatoms can thus be realized by a rapid frequency of protrusion into the mucilage, possibly correlated with the high-frequency vibration of the circular structures seen inside the diatom. Although each locomotion cycle can only advance the cell by hundreds of nanometers, as shown in Figure 13.28, the locomotion speed seen by many diatoms (several µm/s) can be supported by increasing the frequency of the protrusion extension and retraction cycle. According to our Actin-Pin model, the turning of a diatom can also be achieved using two actin protrusions. One of the protrusions would act as the center of rotation, attaching firmly to the substrate, while the other extends out and adheres to the substrate at a direction which would create the turning during locomoting. As for the detachment of the protrusions, this could be achieved in two ways, one is to detach one actin protrusion by pressing down on another of the extended actin protrusions. Alternatively, the contact area can be decreased by deforming the actin protrusion itself, so as to reduce the adhesion force and achieve detachment. The new model also supports the ability to create the various speeds, paths, and characteristics of diatom movement, by regulating the frequency, direction, and extension of actin protrusions. As shown in Figure 13.11, when the protruding part is the actin rather than the underlying mucilage, the former can create multi-dimensional and multi-directional forces such as tension, pressure and torque, thus generating the locomotion of diatoms in various positions. It should also be pointed out that this new model based on actin protrusions is similar to the movement of some other species in nature, such as amoeba, in which the movement requires control of the actin-based extension and retraction of lamellipodia and filopodia. In the case of amoeba, the protuberances (pseudopods) on the cell membrane constantly change shape, attaching and detaching, generating the movement of an amoeba (although amoeboid movement also requires the coordination with myosin-based contractions within the cell cortex during movement, a phenomenon not yet observed in diatoms). Overall, we believe the new Actin-Pin locomotion model, based on the circular structures within the diatoms, and the periodic pits observed within the mucilage trails of diatoms, can better explain the motility of gliding diatoms. The forces involved in such a model appear rational with regards to motion system dynamics. Currently, not all organelles or chemical mechanisms involved in such a model have been determined. It is hoped that the presentation of this model will make a positive contribution to the locomotion mechanism of diatoms by stimulating further research and testing. We further believe that such a model could provide the basis for developing a good bionic prospect for under­ water vehicles.

Locomotion of Benthic Pennate Diatoms  319

13.3 Locomotion and Aggregation of Diatoms 13.3.1 Locomotion Trajectory and Parameters of Diatoms To better understand the characteristics of diatom motility, we investigated the path trajectories of the cells. The locomotions of diatoms were determined by observing motile diatoms that had adhered to glass slides immersed in diatom culture medium for 2 hours, and rinsed to remove diatoms that had not attached to them. Diatoms were observed under the light microscope and images taken every 15 seconds. The locomotion trajectory of a diatom was determined according to the position of the diatom in each image. As shown in Figure 13.31, each section of the line represents the displacement of diatoms during one 15 s interval. Because the distance between individual diatoms was far greater than the size of the cells, we assumed the movements to be independent (not involving the interaction between diatoms) and endogenous (determined by the mechanism of diatom locomotion in each individual cell). The relative parameters of the path trajectories were calculated to quantify the main characteristics of movements for individual cells. These measured characteristics of each 15 s step were: the migration angle of each step and the migration step size of each step. The total sample size of measured steps exceeded 1000 steps. The migration angle of each step was determined as shown in Figure 13.32, as the angle α defined by the defined path of the last two steps. 9-out

8-in 6-end 7-begin

9-in 8-end

5-begin

6-begin

7-end

5-lost

12-begin 11-begin

10-end

12-out

200µm 11-out

10-in

Figure 13.31  Locomotion trajectory of several diatoms. Each number represents the trail of a separate diatom.

320  Diatom Gliding Motility 2 α

3

1

Figure 13.32  Determination of locomotion angle for each step. Positions 1, 2, and 3 represent 3 consecutive positions observed for a motile cell. When diatoms reach position 2, 15 s after position 1, the path direction is defined from the center point of the cell at each time point. After the cell moved to position 3 after another 15 s, the angle of turning from position 2 to position 3 was defined as α, in which α > 0 was defined as turning in a clockwise direction as observed. 80 70

Frequency

60 50 40 30 20 10 0 -200

-150

-100

-50 0 50 Migration angles

100

150

200

Figure 13.33  Distribution of locomotion angle.

The histogram obtained from our measurements is shown in Figure 13.33, insignificantly different from a normal distribution using the Jarque-Bera test.17 Therefore, the dfittool module of MATLAB was used to fit the data to a normal distribution; the mean value was -4.4° and the standard deviation was 73.9°. In conclusion, the angle value of diatom locomotion fit well to a normal distribution defined by Eq. (13.6). ¶

1 α + 4.4  2  73.9 

−  1 f (α ) = e 2 73.9 × 2π





(13.6)

As such, the movement of these diatoms basically showed a short-term path curvature that was on average fairly linear, and no bias toward turning clockwise or counterclockwise. We can also determine the scalar distance of diatom movement in the 15 s intervals by determining the net distance of diatom locomotion (λ) at each step, which is defined as:

Editors: https://en.wikipedia.org/wiki/Jarque–Bera_test.

17 ¶

Locomotion of Benthic Pennate Diatoms  321

  λ = | xi (t ) − xi (t − δ t )|



(13.7)



where xi(t) represents the position of the “i”th diatom at time t, and δt represents the time interval (δt = 15 s). The histogram of such scalar step sizes is shown in Figure 13.34. This distribution, rejected as a normal distribution, can obviously be regarded as an exponential distribution. The probability density function fitted to an exponential distribution by MATLAB is as follows:

f(λ) = 31.4 × e−31.4×λ

(13.8)

The data thus suggests that the average locomotion length of a diatom is 31.4 µm per step (15 s), generating an average 15 s step velocity of about 2 µm/s. The actual acceleration rates and extents of diatom movement could be much higher than the above values, because these are 15 s interval values, and the length measured within 15 seconds is the overall displacement during that time rather than the actual distance traveled during a pulse of movement which might have occurred in much less than the 15 seconds. The energy required for motility is defined by the number of steps that a diatom takes for its movement. Based on the experimental observation, the number of continual locomotion steps in any particular trip before a diatom stops is random, so there is no such obvious distribution rule as the locomotion angle and step length. Therefore, our approximation assumption is that the locomotion energy of diatoms obeys a uniform distribution. The maximum locomotion energy is set to be energy-max » 80 step equivalents because the average number of locomotion steps measured is 35.4±27.8. We assume that regardless of the path, the average energy used by a diatom is the same per step, so that the probability density function of the number of locomotion steps is:

100

Frequency

80

60

40

20

0

0

20

40 60 Migration step lengths

Figure 13.34  Distribution of locomotion length.

80

100

322  Diatom Gliding Motility

 1/80 f (energy ) =   0



0 ≤ energy ≤ 80 elsewhere

(13.9)



In order to determine the role of the substratum material on the locomotion behavior of diatoms, we studied their motility on silicon, stainless steel and PE (polyethylene) hydrophobic membranes for comparison, using the same method as that for the glass surface mentioned above. The motile rate of diatoms on glass slides was the highest. For example, 38.9% of the diatoms on the glass slide crawl faster than 4 µm/s, while we found no diatoms moving above this rate on the silicon slide, 3.1% of the cells moving that fast on the hydrophobic surface and 14.9% of the cells moving that fast on the stainless steel surface (Figure 13.35). Based on their average speeds, the relative abilities of diatoms to move on different surfaces was: glass slide > stainless steel slide > silicon slide > PE hydrophobic surface. We also found a correlation of motile ability with the observed contact angle, with the measured contact angle being about 15° for glass, 40° for stainless steel, 50° for silicon and 90° for PE hydrophobic film. Therefore, the locomoting rate of diatoms is significantly Velocity Distribution on Different Surfaces

% of cells observed to be in the velocity range

110.0% 100.0% 90.0% 80.0% 70.0% 60.0% 50.0% 40.0% 30.0% 20.0% 10.0% 0.0% Glass Slide

Silicon Wafer

>4 µm/s

Surfaces

3~4 µm/s

Hydrophobic PE 2~3 µm/s

1~2 µm/s

Velocity Distribution Ratio >4 μm/s

3~4 μm/s

2~3 μm/s

1~2 μm/s

Stainless Steel VMR > 10, e.g., Figure 13.40b) and moderate aggregation (10 > VMR > 1, e.g., Figure 13.40c). Because the VMR value of diatom

328  Diatom Gliding Motility (a)

(b)

(c)

Figure 13.40  Simulated results of diatom adhesion from NetLogo. (a) very strong aggregation, which VMR>20, (b) strong aggregation, which 20>VMR>10, and (c) moderate aggregation, which 10>VMR>1.

adherence measured in our culture experiment is 12.2 ± 3.5, we consider normal diatom aggregation in a well-established culture to be in the category of “strong aggregation.” By changing the perception parameters (between perception thresholds of 0, 1, 3, 5 and 10, and perception angles of 0, 45, 90 and 135 degrees, the simulated VMR was compared with the measured VMR; the resulting curves are shown in Figure 13.41. Figure 13.41a was obtained by introducing a proliferation of diatoms. It shows that when the perception setting is very strong (perception threshold ≤ 3), as long as the perception angle is not zero, the simulation generates excessive aggregation (VMR > 16), while when the perception is very weak (perception threshold ≥ 10), the simulation generates great uniformity (VMR < 9) that is significantly different from the actual measured VMR. Therefore, our simulation model suggests that there is intercellular perception among diatoms, with a moderate degree of perception. Based on the parameters used when our simulation results are within the confidence interval of actual observed distributions, (e.g., the relative EPS secretion of 3 units and the mucilage secretion of 0.1 units), we believe that: 1) the contribution of mucilage to diatom aggregation is very low; 2) allowing for the diffusion and dissolution of EPS, an average of 16-17 adherent diatoms are necessary to “attract” a nearby locomoting diatom. With our parameters, the calculation here is that the perception threshold needs

(a)

(b)

60

0° 45° 90° 135°

40 30 20

0° 45° 90° 135°

18 16 Average VMR value

Average VMR value

50

20

14 12 10 8 6 4

10

2 0

0

2

4 6 Perception-threshold

8

10

0

0

2

4 6 Perception-threshold

8

10

Figure 13.41  Comparisons between simulated and experimental values of VMR under different perception thresholds either with (a) or without (b) the introduction of cell division, using different perception angles. The two horizontal dashed lines correspond to the 95% confidence intervals from the solid line representing the VMR value determined directly from diatom cultures.

54.6***±10.6

135°

21.4***±3.4

20.5***±3.3

* P