Symbiosis in Fishes For Rachel for Everything Symbiosis in Fishes The Biology of Interspecific Partnerships

Ilan Karplus This edition first published 2014 © 2014 by Ilan Karplus

Registered Office John Wiley & Sons, Ltd, The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, UK

Editorial Offices 9600 Garsington Road, Oxford, OX4 2DQ, UK The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, UK 111 River Street, Hoboken, NJ 07030–57

For details of our global editorial offices, for customer services and for information about how to apply for permission to reuse the copyright material in this book please see our website at www.wiley.com/wiley-blackwell.

The right of the author to be identified as the author of this work has been asserted in accordance with the UK Copyright, Designs and Patents Act 1988.

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 the UK Copyright, Designs and Patents Act 1988, without the prior permission of the publisher.

Designations used by companies to distinguish their products are often claimed as trademarks. All brand names and product names used in this book are trade names, service marks, trademarks or registered trademarks of their respective owners. The publisher is not associated with any product or vendor mentioned in this book.

Limit of Liability/Disclaimer of Warranty: While the publisher and author(s) have used their best efforts in preparing this book, they make no representations or warranties with respect to the accuracy or completeness of the contents of this book and specifically disclaim any implied warranties of merchantability or fitness for a particular purpose. It is sold on the understanding that the publisher is not engaged in rendering professional services and neither the publisher nor the author shall be liable for damages arising herefrom. If professional advice or other expert assistance is required, the services of a competent professional should be sought.

Library of Congress Cataloging-in-Publication Data Karplus, Ilan. Symbiosis in Fishes : the biology of interspecific partnerships / Ilan Karplus. pages cm Includes bibliographical references and indexes. ISBN 978-1-4051-8589-9 (cloth) 1. Symbiosis. 2. Aquatic animals. 3. Fishes. 4. Invertebrates. I. Title. QH548.K37 2014 577.8′5–dc23 2013024351 A catalogue record for this book is available from the British Library.

Wiley also publishes its books in a variety of electronic formats. Some content that appears in print may not be available in electronic books.

Cover image: The goby Lotilia graciliosa associated with a burrowing alpheid shrimp. Photo by IKAN Underwater Archive. Cover design by Steve Thompson

Set in 10/11.5pt Minion Pro by SPi Publisher Services, Pondicherry, India.

1 2014 Contents

Preface x

Introduction 1

1 The Associations between Fishes and Luminescent Bacteria 6 Luminescent Bacteria 6 Symbiotic Luminescent Bacteria in Fish Light Organs 8 Flashlight Fishes 11 Taxonomy and Distribution 11 The Light Organs 13 The Eye and the Light Organ 17 Reproduction, Larval and Light Organ Development 18 The Photophobic Response 20 The Use of Light by Flashlight Fishes 21 School Formation 22 Territorial Defense 22 Sexual Signaling 22 Deep Sea Ceratioid Anglerfishes 24 Structure, Diversity and Distribution 24 Reproductive Strategies 25 Obligatory Sexual Parasitism 26 Temporary Associations 28 Facultative Sexual Parasitism 29 Light Organ Structure and Development: Light and the Mechanisms Controlling its Emission 29 The Use of Lures by Anglerfishes 34 Ponyfishes 37 Structure, Distribution and Taxonomy 37 The Light Organ System (LOS) and Diversity of the Generated Light Patterns 38 Disruptive Illumination 40 Discrete Projected Luminescence (DPL) 41 Ventral Body Flash 41 Opercular Flash 42 Buccal Luminescence 42 Sex-Specific Signaling 43 Inception of the Association between Luminescent Bacteria and Ponyfishes 43 Sexual Dimorphism of the LOS, Sex-Specific Signaling and the Role of Sexual Selection in the Evolution of Leiognathid Fishes 44 Specificity of the Partnerships between Luminescent Bacteria and Fishes 47 Optimization of the Benefits to Fishes from their Association with Bacteria 48 The Evolution of the Partnerships between Fishes and Luminescent Bacteria 49 References 52 vi Contents

2 The Associations between Fishes and Sponges 58 Sponges 58 Predator Deterrence by Sponges 59 Multiple Species Assemblages in Sponges 61 Obligatory Fish Symbionts and Adaptations for Living in Association with Sponges 62 Nutrition, Reproduction and Sponge Occupation by Obligatory Symbiotic Fishes 68 Partner Specificity and Sponge Sharing by Obligatory Symbiotic Fishes 69 Evolution of the Partnership Between Obligatory Fish Symbionts and Sponges 70 Sponges as Living Incubators of Fish Eggs 72 Facultative Partnerships Between Fishes and Sponges 74 References 75

3 The Associations between Fishes and Anthozoans 79 Sea Anemones 79 The Stinging Cells and their Release Mechanism 80 Obligatory Associations with Sea Anemones of Fishes of the Genera Amphiprion and Premnas 81 The Taxonomy, Distribution and Ecology of Host Sea Anemones and their Associated Fishes 81 The Protection of Anemone Fishes from Sea Anemones 86 Recognition, Attraction to and Selection of Sea Anemones by Anemone Fishes 93 Partner Specificity 104 Host Preference 106 Competitive Interactions 106 Stochastic Processes 109 Habitat Preference 109 Geographical Overlap 109 Protection from Sea Anemones 109 Species Coexistence 110 Adaptations of Anemone Fishes for Living with Sea Anemones 111 Protandric Sex Reversal 111 Monogamy and Mate Recognition 114 Step-fathering 118 Social Control of Growth and the Tolerance of Nonbreeders by the Breeders 118 Fish Territoriality, Aggression and the Sea Anemone 121 Limited Larval Dispersal and Natal Recruitment 124 Benefits and Costs to Anemone Fishes and Sea Anemones from being Associated and their Short-term Mutual Impacts 128 The Evolution of the Anemone Fish–Sea Anemone Partnership 134 The Facultative Associations Between Fishes and Sea Anemones 135 Protection from Sea Anemones 140 Partner Specificity 141 Settlement and Recruitment of D. trimaculatus to Sea Anemones 142 The Sharing of Sea Anemones with Anemone Fishes 143 Benefits and Costs to Facultative Fish Partners and Sea Anemones 144 The Associations Between Fishes and Scleractinian Corals 145 Scleractinian Corals 145 Microhabitat Selection by Coral Dwelling Fishes 146 Attraction of Pomacentrid Fishes to Corals 146 Attraction of Pomacentrids to Corals Inhabited by Conspecifics 150 Coral Occupation, Competiton and Coexistence of Coral dwelling Gobies 153 Adaptations to Habitat by Coral Dwelling Gobies 157 Contents vii

Small Size and Morphology 157 Noxious Skin 158 Hypoxia Tolerance and Air Breathing 158 Bidirectional Sex Reversal 159 Monogamy 161 Social Control of Growth 162 Multiple Species Assemblages Involving Coral Dwelling Gobies and Crustaceans 164 Benefits and Costs to Fishes and Corals for being Associated 169 Benefits to Fishes 169 Costs to Fishes 173 Benefits to Corals 177 Costs to Corals 180 Social Structure and Mating System Evolution in Coral Dwelling Damselfishes of the genus Dascyllus 181 References 186

4 The Associations between Fishes and Siphonophores 202 Siphonophores 202 Physalia physalis −the Portuguese Man-of-War 203 Fishes Associated with Siphonophores other than Physalia physalis 204 Fishes Associated with Physalia physalis 207 References 209

5 The Associations between Fishes and Scyphozoan Medusae 212 Scyphozoan medusae 212 Predation on Scyphozoan Medusae and their Structural and Behavioral Antipredator Defenses 214 Fishes Associated with Scyphozoan Medusae 215 The Protection of Fishes from Scyphozoan Medusae 217 Recognition and Attraction to Scyphozoan Medusae by Associated Fishes 217 Partner Specificity, Duration of the Medusa–Fish Bond and the Effects of the Medusae Size on the Associated Fishes 219 Benefits and Costs to Fishes and Medusae from being Associated 221 The Effects of Medusae on Fish Recruitment 225 The Association of Fishes with Floating Objects and the Fish–Medusa Partnership 226 References 227

6 The Associations between Fishes and Molluscs 230 The Association between Fishes and Cephalopods 230 Cephalopods 230 Octopus Dens, Foraging and Antipredatory Behavior 231 Scavenging Fishes Associated with Octopus Dens 233 Fishes Associated with Foraging Octopuses 234 Octopuses and Cleaning Symbiosis 238 Transport Associations between Octopuses and Fishes 239 Fishes Associated with Squid Schools 239 The Association between Fishes and Gastropods 241 Gastropods 241 Predation on Conchs, Antipredatory Strategies and Foraging in Conchs 241 The Association between Cardinal Fishes and Conchs 242 The Association between Nudibranchs and Gobiid Fishes 245 The Association between a Pearlfish and an Opisthobranch Gastropod 246 viii Contents

The Association between Fishes and Bivalves 246 Bivalves 246 The Glochidia Larvae of Freshwater Mussels and their Host Fishes 247 Bitterlings and their Freshwater Mussel Hosts 248 Attraction of the European Bitterling to Mussels and Choice of Oviposition Sites 250 Adaptations of Bitterling for Development Inside Freshwater Mussels 252 Male Reproductive Behavior and the Mussel 254 Female Reproductive Behavior and the Mussel 257 Host Utilization by Sympatric Bitterling Species 260 Costs and Benefits for the Mussel and Possible Coevolution of the Bitterling–Mussel Partnership 263 Pearl Fishes Associated with Bivalves 265 The Association of Snailfish and Red Hake with Sea Scallops 265 References 269

7 The Associations between Fishes and Crustaceans 276 The Associations between Fishes and Cleaner Shrimps 276 Cleaning Symbiosis and Shrimp 276 Taxonomy, Morphology, Coloration and Distribution of Cleaner Shrimp 276 Cleaner Shrimp Activity 287 Associations between Cleaner Shrimp and Sea Anemones 288 Communication between Fishes and Cleaner Shrimp 292 Removal of Parasites versus Mucus by Cleaner Shrimp 294 Costs and Benefits for Cleaner Shrimp and Fish Clients and the Proximate Mechanisms for Cleaning 296 The Evolution of the Cleaner Shrimp–Fish Partnership 298 Feeding associations between fishes and crustaceans 299 Mixed Species Schools of Fishes and Crustaceans 300 Liparid Fishes Associated with Lithodid Crabs 301 The Associations between Fishes and Burrowing Brachyuran Crabs 303 Gobiid Fishes Associated with Burrowing Thalassinid Shrimp 305 Thalassinid Shrimp and their Burrows 305 The Facultative Association of Clevelandia ios with Callianassa californiensis and Upogebia pugettensis 307 The Obligatory Association of the Blind Goby Typhlogobius californiensis with Callianassa affinis 311 The Obligatory Association of Austrolethops wardi with Neaxius acanthus 313 The Obligatory Association of Didogobius amicuscardis with Axiopsis serratifrons 314 Gobiid Fishes Associated with Burrowing Alpheid Shrimps 316 Systematics of Gobies and Shrimps 316 Biogeography 318 Diet and Feeding Behavior 319 Habitat Specificity 322 Population Structure and Dynamics 324 Burrow Structure, Construction and Dynamics 326 Activity Rhythms 330 Aggressive Behavior and Territoriality of Goby and Shrimp 334 Reproduction of Goby and Shrimp 336 Interspecific Communication 338 Communication under Natural Conditions in Indo-Pacific Partnerships 338 Warning Signal Generation by Indo-Pacific Gobies in Response to Predators and Models of Predators 340 Sequence and Information Analyses in Indo-Pacific Partnerships 342 Contents ix

Film Analysis of the Communication between the Goby Amblyeleotris steinitzi and the Shrimp Alpheus purpurilenticularis 344 Communication between Gobies and Shrimp in the Western Atlantic 347 Partner Specificity 349 Field Observations 349 Laboratory Experiments 350 The Mechanism Regulating Specificity 352 Goby–Shrimp Phylogeography 353 Costs and Benefits for Goby and Shrimp 357 Evolution 358 References 360

8 The Associations between Fishes and Echinoderms 371 The Association between Fishes and Sea Urchins 371 Sea Urchins 371 Sea Urchin Structural Defenses, Predation by Fishes and Antipredatory Strategies 371 Associated Fishes, their Size, Coloration and Sea Urchin Hosts 373 The Attraction of Associated Fishes to Sea Urchins 386 Benefits and Costs of the Fish–Sea Urchin Partnership 388 Partner Specificity in the Fish–Sea Urchin Association 390 The Evolution of the Fish–Sea Urchin Partnership 390 Mimicry of Sea Urchins by Fishes 391 The Association between Fishes and Crinoids 392 Crinoids 392 Predation on Crinoids by Fishes and Antipredatory Strategies of Feather Stars and Sea Lilies 393 Multiple Species Assemblages in Crinoids 394 Associated Fishes and Adaptations for Living with Crinoids 395 Attraction of Associated Fishes to Crinoids and Partner Specificity 398 Benefits and Costs of the Fish–Crinoid Partnership 398 Scarcity of Knowledge 398 The Association between Fishes and Sea Cucumbers 399 Sea Cucumbers 399 Predation on Sea Cucumbers by Fishes and their Structural and Behavioral Antipredatory Defenses 400 Fishes Associated with Sea Cucumbers and their Life Cycles 401 Host Location, Penetration and Occupation by Pearlfishes 405 Pearlfish Nutrition 406 Pearlfish Reproductive Biology 408 Ecology and Partner Specificity of Pearlfish–Holothurian Associations 410 Acoustic Communication in Pearlfishes 414 Morphological and Physiological Adaptations to Inquilism 415 Benefits and Costs of the Pearlfish–Sea Cucumber Partnership 417 The Evolution of the Partnership between Pearlfishes and their Hosts 417 The Association between Fishes and Sea Stars 418 Sea Stars 418 Sea Star Structural and Behavioral Antipredatory Defenses 419 Feeding Associations between Sea Stars and Fishes 420 Cardinal Fishes Sheltering among Sea Star Spines 421 Pearlfishes Associated with Sea Stars 421 References 423

Species Index 431 Subject Index 443 Preface

On a hot summer day more than forty years ago, I tropical Atlantic. During the following years that I was snorkeling at leisure in Marsa Murach, a shal- spent in the Red Sea, studying various aspects of the low sandy bay in the northern Red Sea during a communication systems and specificity of goby– break in a university field trip. I was suddenly shrimp partnerships, I encountered additional amazed by a strange couple consisting of a goby and ­symbiotic associations and the researchers studying a shrimp, which later I learned to recognize as the them. I was lucky to meet Hans and Simone Fricke large and beautiful Cryptocentrus caeruleopunctatus studying Amphiprion bicinctus and the cardinal and the blue and green mottled burrowing Alpheus fish Siphamia tubifer associated, respectively, with djiboutensis. The goby was perched in front of a bur- giant sea anemones and sea urchins. Likewise, I row on a small heap of sand while the shrimp con- met James Morin and his colleagues studying the tinuously excavated this burrow, moving in and out partnership­ between Photoblepharon steinitzi and of it similarly to a micro-bulldozer. During this luminescent bacteria, and Gadi Katzir studying entire operation the shrimp and goby remained very Dascyllus aruanus associated with branching corals. close to one another, with the shrimp touching the In later years, these associations became the topic of goby with its antenna. As I approached this strange many of the lectures that I delivered at the Hebrew couple, both disappeared into the same burrow with University in courses on fish behavior. Over these only a cloud of sand remaining visible to me. years I was looking, without success, for a book that Following this encounter I decided to investigate would cover most of the available information on this association and did so within the frame of my the partnerships between fishes, invertebrates and MSc and PhD dissertations carried out in the Red luminescent bacteria. Finally, I got such a book but Sea, as well as part of my postdoctoral studies in the only after writing it by myself. Introduction

Partnerships between fishes and invertebrates were published in 1997, The Diver­sity of Fishes, only a discovered quite early in the exploration of the aqua­ single page was devoted to these partnerships. There tic environment when coral heads, sea cucumbers is to date not a single book which focuses on the and sponges were removed from the water and associations between fishes and invertebrates. L.P. noticed to be occupied by a variety of fishes. With the Zann (1980) in his book Living Together in the Sea early use of SCUBA diving equipment many new discussed numerous marine associations between partnerships were discovered and in situ field studies fishes and inverteb­rates. However, this book was were carried out. These discoveries arose interest, published more than thirty years ago and was which persists till today, with regard to a variety of confined to the marine environment. questions, such as the function of these partnerships, This book provides the reader with a series of the way they are first established and how they updated reviews based on the available literature, evolved. Some associations (e.g., the anemone fish– including forgotten MSc and PhD dissertations, sea anemone partnership) have been the subject of which often contained extremely valuable infor­ research almost continu­ously since their discovery, mation. Topics are often discussed in a chronological whereas other partner­ships were only seldomly order, starting with the discovery of a partnership investigated­ (e.g., the associations between fishes and ending with the most recent findings in a and siphonophores). Reviews of the state of the particular area of research. art con­cer­ning some of the more intensively In this book the term “Symbiosis” is used in its investigated partnerships became gradually wide definition according to De Bary (1879), available, for example, the partnerships between namely the living together of different organisms, anemone fishes and sea anemones (Mariscal, 1966; with the relationship between them including Allen, 1972; Fautin, 1991), and between gobies and mutualistic, commensalistic and parasitic­ interac- burrowing alpheid shrimps (Karplus, 1987; Karplus tions. The parasitic relationship between bacteria, and Thompson, 2011). However, for most of the coelenterates, crustaceans and mollusks and their associations no reviews or only non-updated reviews fish hosts are not within the scope of this book are currently available. Moreover, the subject of and have been reviewed often in books dealing interspecific partnerships between fishes and inver­ with fish diseases and parasitiology. However, tebrates has not received the attention it deserves. included were those cases when the fish, the focus For example, in the excellent ichthyological text book of this book, was the parasite and its host an by G.S. Helfman, B.B. Collette and D.E. Facey, first invertebrate.

Symbiosis in Fishes: The Biology of Interspecific Partnerships, First Edition. Ilan Karplus. © 2014 Ilan Karplus. Published 2014 by John Wiley & Sons, Ltd. 2 Introduction

This book has been written for undergraduate partnership, and the evidence for coevolution and and graduate students as well as researchers and independent evolution. individuals who practice SCUBA or snorkeling and are interested in fishes and their symbiotic Chapter 2: Fishes associated with sponges. associations. Sources of information in the text Mechanical and chemical predator deterrence by have been cited in great detail to allow the readers sponges; multiple species assemblages in sponges; to go to a cited work and read the details of that adaptations of obligatory fish symbionts for living specific study. inside sponges, their reproduction and nutrition; The eight chapters of the book are each devoted the evolution of sponge dwelling in fishes in the to partnerships with different taxonomic groups, tropical Atlantic; sponges as living incubators of moving up the phylogenetic scale from lumines- fish eggs and the relationship between facultative cent bacteria to echinoderms. At the beginning fish symbionts and sponges. of each chapter is a short section that deals with the general characteristics of each group; it is Chapter 3: Fishes associated with anthozoans. The intended for those who are not familiar with stinging cells of coelenterates and their releasing invertebrate anatomy and physiology. Moreover, mechanisms; the protection of anemone fishes certain aspects of the biology of these groups are from sea anemones; visual and chemical­ attraction relevant to their relationship with associated of fishes to sea anemones and imprint­ing on sea fishes, such as the stinging cells and their release anemones; partner specificity and its regulation; mechanisms in giant sea anemones and the anemone sharing by anemone fishes; adaptations Cuvierian tubules in holothurians, are discussed of anemone fishes for living in sea anemones; as well. It is important to consider the defensive benefits and costs to anemone fishes and sea mechanisms of a host, such as the mechanical anemones of being associated; the evolu­tion of the and chemical antipredatory defenses of sponges, anemone fish–sea anemone partner­ship; faculta­ in order to better understand their association tive symbionts of sea anemones and their protection with fishes. Each chapter deals with the species from their host; sharing of anemones with anem­ involved in the partnerships, their geographic one fishes; the attraction to corals by associated distribution and habitats, and whether they are pomacentrids and gobiid fishes and the preference facultative or obligatory symbionts. Further are of corals occupied by conspecifics by pomacen­ addressed issues related to the sensory modalities trids; coexistence of coral dwelling gobies despite involved in the first formation and subsequent com­petitive interactions;­ adaptations of coral maintenance of the associations; partner speci- dwelling gobies to living inside corals; multiple ficity, its regulating mechanisms and the availa- species assemblages involving coral dwelling gobies ble information on what underlies that specificity; and crustaceans; benefits and costs to fishes and the costs and benefits to both fishes and inverte- corals from being associated; social structure brates from being associated and finally, the evo- and mating system evolution in coral dwelling lution of these partnerships and the possible damselfishes of the genus Dascyllus. involvement of independent and coevolutionary Chapter 4: Fishes associated with siphonophores. processes. The following main topics are included Antipredatory defenses of siphonophores and in the chapters of this book: Physalia physalis; fishes associated with Physalia and other siphonophores; protection of associa­ Chapter 1: Fishes associated with luminescent ted fishes from siphonophores and the costs and bacteria. Luminescent bacteria, the light producing benefits for fishes and siphonophores from being reaction, the genes involved and quorum sensing; associated; the difficulty in studying these bacteria contained in fish light organs and the partnerships. utilization of this light by fishes; the better known groups of fishes associated with luminescent Chapter 5: Fishes associated with scyphozoan bacteria – the flashlight fishes, deep sea anglerfishes medusae. Antipredatory defenses of medusae; and pony fishes; specificity of the symbiotic the protection of associated fishes from scypho­ associations and optimization of the benefits to zoan medusae; recognition and attraction to fishes from serving as hosts to luminescent medusae and floating objects by associated bacteria; the evolution of the fish–bacteria fishes; partner specificity and the duration of the Introduction 3

medusa–fish bond; benefits and costs to fishes fishes; host location, penetration and occupation; and medusae from being associated; the effect of pearl fish nutrition and reproductive biology; the medusae on fish recruitment. ecology and partner specificity of pearl fishes– holothurian associations; morphological and Chapter 6: Fishes associated with molluscs. physiological adaptations to inquilism; costs Octopus dens, foraging and antipredatory behav­ and benefits to pearl fishes and sea cucumbers ior of octopuses; fishes associated with octopus and the evolution of these partnerships; anti­ dens, foraging octopuses and squid schools; predatory defenses of sea stars; feeding associa­ conch antipredatory strategies and the association tions between fishes and sea stars; cardinal fishes of cardinal fishes with these gas­tropods; the sheltering among sea star spines; pearl fishes association between nudibranchs and gobiid associated with sea stars. fishes; bitterlings and their freshwater mussel hosts; male and female reproductive behavior I am grateful to all the individuals, some of whom and the mussel; host utilization by sympatric are no longer with us, who contributed over the bitterling species; costs and benefits for the mus­ years in the laboratory, on coral reefs, in the deep sel and the possible coevolution of the bitterling– sea, rivers and lakes to our knowledge of the symbi- mussel partnership; pearl fishes associated with otic relationships between fishes and invertebrates. bivalves; the association of inquiline snail fish Thanks to their efforts and accomplishments was it and red hake with sea scallops. possible for me to try to assemble some pieces of the Chapter 7: Fishes associated with crustaceans. puzzle of the symbiotic way of life. Cleaner shrimp taxonomy, coloration, distribu­ The information presented at the FishBase tion and cleaning activity; the association between web site (Froese and Pauly, 2012) was extremely cleaner shrimp and sea anemones; communica­ important for clarification of the taxonomic tion between cleaner shrimp and fishes; removal ­status of many of the fish species discussed in of parasites versus mucus by cleaner shrimp; this book and for information concerning their costs and benefits for cleaner shrimp and fish biology. I am very grateful for this important clients and the proximate mechanism for clean­ source of information. ing; the evolution of cleaner shrimp–fish partner­ Many people contributed in different ways to this ship; feeding associations between fishes and book and I am grateful to all of them. I am particu- crustaceans; mixed species schools of fishes and larity grateful to the late Professor Lev Fishelson, crustaceans; liparid fishes associated with litho­ who passed away several months ago. I have known did crabs; the associations of fishes with burrow­ Lev since I was thirteen years old and entered one ing brachyuran crabs and thalassinid burrowing day his recirculating aquaria rooms with my father shrimp; Gobiid fishes associated with burrow­ing and a small Dascyllus aruanus, which we had caught alpheid shrimp; biogeography; habitat specifi­ in the Red Sea. This fish was saved by Lev who city; population structure and dynamics; burrow introduced him into a colony of living corals in his structure, construction and dynamics; activity laboratory in Tel Aviv University. Over the years I rhythms, reproduction and aggressive behavior was inspired by Lev’s studies on reef fishes and we of gobies and shrimps; interspecific communi­ became very good friends. Lev encouraged me cation; phylogeography; costs and benefits to throughout the writing of this book and com- shrimps and gobies and the evolution of these mented on the entire manuscript, with our discus- partnerships. sions being extremely valuable for me. I would also like to thank Dr Assaf Barki of the Volcani Research Chapter 8: Fishes associated with echinoderms. Center, a former student and currently colleague Sea urchins structural defenses and antipredatory and good friend, for his comments on parts of the strategies; the attraction of fishes to sea urchins; book and the statistical analyses and graphics of the partner specificity and mimicry of sea urchins; laboratory studies carried out on flashlight fish by antipredatory defenses of crinoids; multiple spec­ Dr Gidi Sagi. I wish to thank Gidi for the time we ies assemblages in crinoids; adaptations of fishes spent together at his home discussing many aspects for living in crinoids and the scarcity of know­ of his field and laboratory studies on Photoblepharon ledge on these partnerships; sea cucumbers steinitzi and for allowing me to reanalyze parts of antipredatory defenses; the life cycle of pearl his data. My son Daniel Karplus did all the graphic 4 Introduction work in this book and I am very grateful for his Natural Sciences, Baton Rouge, U.S.A.; Coates, D., time and patience and the superb job that he did. Secretariat of the Convention on Biological My son Tal Karplus helped me in many ways such Diversity, U.S.A.; Colin, P., Coral Reef Research as obtaining permissions for use of figures and Foundation, Palau; Duffy, D., University of Hawaii, needed correspondence. It is a pleasure to thank U.S.A.; Dunlap, P., University of Michigan, U.S.A.; Miss Idith Sofer and Mrs Ruth Melchin from the Dworschak, P., Naturhistirisches Museum, Viena, Volcani Research Center Library who assisted me Austria; Elliott, J., University of Puget Sound, in literature searches and locating of articles often U.S.A.; Fishelson, L., Tel Aviv University, Israel; published in extremely hard to get journals. I would Fisher, R., University of Western Australia, also like to thank Mrs Miriam Schwimer of the Australia; Fricke, H.W., The Open University Faculty of Agriculture Rechovot Library of the Munich, Germany; Haag, W., U.S. Forest Service, Hebrew University. Miriam was always of good will U.S.A.; Hastings, J.W., Harvard University, U.S.A.; and spirit, locating through interlibrary loan Hattori, A., Shiga University, Japan.; Haygood, M., extremely hard to get references such as MSc and Oregon Health & Science University, U.S.A.; PhD dissertations published many years ago. Herring, P., University of Southhampton Water­ I am very grateful to the many colleagues who front Campus, UK; Hirata, T., Nagabori, Uwajima- have very generously allowed me to use their non- shi, Japan; Hobbs, J.P., University of Western published before photos and figures in this book Australia, Australia; Holbrook, S., University and their names are presented in alphabetical order: of California, U.S.A.; Jones, J.P., James Cook Arvedlund, M., Reef Consultants, Frederiksberg, University, Australia.; Lagardere, J.P., CNRS- Denmark; Colin, P., Coral Reef Research Founda­ Ifremer, France; Larson, H., Museum and Art tion, Palau; Darom, D., The Hebrew University, Gallery of the Northern Territory, Australia; Israel; Herler, J., University of Vienna, Austria; Martin, S.B., Fish & Wildlife Research institute, Janssen, J. and Gilmor, R., School of Freshwater Florida, U.S.A. Darom (Masry), D., The Hebrew Sciences, University of Wisconsin-Milwaukee, University, Israel; Masuda, R., Kyoto University, U.S.A.; Masuda, R., Maizuru Fisheries Research Japan; McCafferty, S., Wheaton College, U.S.A.; Station, Kyoto University, Japan.; Okuno, J., McCosker, J.E., California Academy of Sciences, Coastal Branch of Natural Museum and Institute, U.S.A.; Munday, P., James Cook University, Chiba, Japan.; Patzner, R. A., University of Australia; Murata, M., Osaka University, Japan; Salzburg, Austria.; Pietsch, T.W., University of Kanoh, Y., Osaka University of Economy and Law, Washington, U.S.A.; Randall, J.E., Bernice Bishop Japan; Katzir, G., Haifa University, Israel; Museum, Hawaii, U.S.A.; Reichard, M., Institute Kuwamura, T., Chukyo University, Japan; of Vertebrate Biology, Brno, Czech Republic; Sagi, McFall-Ngai, M., University of Wisconsin, U.S.A.; G., Moshav Herev Laete, Israel; Shpigel, M., The Ostlund-Nilsson, S., University of Oslo, Norway; National Marine Aquaculture Center, Elat, Israel; Parmentier, E., University of Liege, Belgium; Taylor, D. S., Brevard County Environmentalley Patzner, R.A., University of Salzburg, Austria; Peat, Endangered Lands Program, Florida, U.S.A.; Wirtz, P., S., Bakersfield College, U.S.A.; Pietsch, T.W., Universidade do Algarve, Portugal. University of Washington, U.S.A.; Planes, S., The majority of the figures and photos published University of Perpignan, France; Prachett, M., in this book have been originally published in James Cook University, Australia; Urbanczyk, H., ­scientific journals. I would like to thank all the University of Miyazaki, Japan; Roe, K.J., Iowa State authors listed in alphabetical order for kindly grant- University, U.S.A.; Ruber, L., Natural History ing me permission to use their figures, photos and Museum, Bern, Switzerland; Sargent, C., University tables: Able, K.W., Rutgers Univesity, U.S.A.; of Kentucky, U.S.A.; Schlichter, D., University of Aldridge, D., University of Cambridge, U.K.; Allen, Cologne, Germany; Schmitt, R., University of G., Western Australian Museum, Perth, Australia; California, U.S.A.; Senou, H., Kanagwa Prefectural Arvedlund, M., Reef Consultants, Fredriksberg, Museum of Natural History, Japan; Steene, R., Denmark; Barreiros, J.P., Universidade dos Acores, Cairns, Australia; Thompson, A., NOAA Fisheries, Portugal; Bilecenoglu, M., Adnan Menderes California, U.S.A.; Tyler, J., Smithsonian Institution, University, Turkey.; Brodeur, R., NOAA Fisheries, Washington, U.S.A.; Wong , M.Y.L., James Cook Newport, U.S.A.; Buston, P.M., University of University, Australia; Young, R.E., University of Boston, U.S.A.; Chakrabarty, P., Museum of Hawaii, U.S.A. Introduction 5

I wish to thank the many publishers of Scientific as mountain gorillas in Africa and the lemurs in journals and books, listed in alphabetical order, Madagascar. In this book, the reader is about to be who generously allowed me to reproduce figures, exposed to an enormous diversity and complexity photos and tables originally published in their of interspecific relationships, which are very much ­journals: Allen Press; American Association for still an enigma to us. I do sincerely hope that, like- the Advancement of Science; American Society wise, becoming aware of symbiotic relationships of Ichthyologists and Herpatologists; Brill; among aquatic organisms, often the product of Cambridge University Press; Carlsbergfondet; ­millions of years of evolution, will enhance a more Crawford House Publishing; Elsevier; Inter- cautious attitude and more concern in the Research Science Center, Masuda Kaiyo Produc­ ­preservation of our beautiful and much unknown tion Co. Ltd.; National Academy of Sciences of planet Earth. the U.S.A.; National History Museum of Los Ilan Karplus Angeles County; Naturalis Biodiversity Center; Nature Publishing; NOAA Pacific Islands Fisheries Science Center; Oxford Publishing; References Rosenstiel School of Marine & Atmospheric Science; University of Hawaii Press; University Allen, G.R. 1972. The Anemonefishes: their Classifi­ of Notre Dame; Science Publishers; Scientific cation and Biology. TFH Publications, Inc., NJ, U.S.A. American, Inc.; Sociedade Brasileira de Icti­olo­ De Bary, A. 1879. Die Erscheinung der Symbiose. Verlag gia; Societe Francaise d’Ichthyologie; Springer Von Karl J. Trubner, Strassburg. Netherlands; Springer Science + Business Media; Fautin, D.G. 1991. The anemonefish symbiosis: what Taylor & Francis; The Crustacean Society; The is known and what is not. Symbiosis 10: 23–46. Royal Society; West Pomeranian University of Froese, R. and Pauly, D. (eds) 2012. FishBase. www. fishbase.org (last accessed 30 October 2013). Technology in Szczecin; John Wiley & Sons, Ltd. Helfman, G.S., Collette, B.B. and Facey, D.E. 1997. The I am very grateful to Nigel Balmforth, Kelvin Diversity of Fishes. Blackwell Science, U.S.A. Matthews, and their colleagues at Wiley-Blackwell, Karplus, I. 1987. The association between gobiid as well as Kevin Dunn and Radjan LourdeSelva­ fishes and burrowing alpheid shrimps. Ocean­ radin for all their help and encouragement. ography and Marine Biology Annual Reviews 25: The inclusion of mistakes and the missing of 507–562. important information are inevitable in any written Karplus, I. and Thompson, A.R. 2011. The partnership text. I would greatly appreciate comments sent to me between gobiid fishes and burrowing alpheid ([email protected]) by individuals who have read shrimps. In: The Biology of Gobies (eds R.A. Patzner, parts of the text and have become aware of such J.L. Van-Tassel, H.K. Larson, and B.G. Kapoor). Science Publishers Inc., NH, U.S.A. shortcomings in order that these issues might be Mariscal, R.N. 1966. The symbiosis between tropical corrected in a possible second edition of this text. sea anemones and fishes: a review. In: The Galapagos The awareness and opposition of the public to (ed. R.I. Bowman). University of California Press, natural habitat destruction that endangers and Berkeley, CA, U.S.A., pp. 157–171. often leads to extinction of species is often enhanced Zann, L.P. 1980. Living Together in the Sea. T.F.H. by focusing on particularly attractive species, such Publications. ONE The Associations between Fishes and Luminescent Bacteria

Luminescent Bacteria ­photooxidation, hydrostatic pressure and the ability to grow in nutrient poor conditions (Yetinson and Luminescent bacteria are the most common and Shilo, 1979; Shilo and Yetinson, 1979; Herring, 1982; widely distributed of all luminous organisms, typi- Al Ali et al., 2010). Luminous Vibrionacea are found cally emitting a continuous blue-green light, peak- both free living, as saprophytes, and as gut symbionts ing around 490 nm. They occur most frequently in and symbionts contained in special light organs of the marine environment but are also found in the fishes and cephalopods. Five species of luminous freshwater and terrestrial environments (Nealson bacteria,­ Aliivibrio fisheri, Photobacterium leiognathi, and Hastings, 1990; Meighen, 1991). Currently, P. kishitanii, P. mandapamensis and Photodesmus twenty-five species of luminescent bacteria, belong- katoptron, and two groups of not yet identified bacte- ing to six genera and three families, have been iden- ria associated with deep sea anglerfishes and flashlight tified. Bacteria of the genus Schewanella (fam. fishes have been so far described from fish light organs Schewanellaceae) are commonly found free living (Table 1.1). The associations of these bacteria with in the freshwater environment. Members of the fishes is discussed throughout this chapter. A shift genus Photorhabdus (fam. Enterobacteriaceae) are from one niche to another (e.g., Photobacterium terrestrial, gut endosymbionts of nematodes of the leiognathi­ shifting from the nutrient-rich esophageal genus Heterorhabditis. The genera Photobacterium, light organ of ponyfish into the fish’s gut and, subse- Vibrio, Aliivibrio and Photodesmus are all marine quently, into the water column adapting a free living members of the Vibrionaceae, an ecologically mode in a starvation/survival habitat) is typical for diverse group of Gram-negative bacteria often asso- many of these symbiotic bacteria (Nealson and ciated with marine animals (Peat and Adams, 2008; Hastings, 1990; Urbanczyk et al., 2011). Hendry and Dunlap, 2011; Urbanczyk et al., 2011). The bacterial light producing reaction is Luminescent bacteria of the Vibrionaceae are ­catalyzed by the heterodimeric enzyme luciferase, found in all marine environments from the cold polar which ­consists of two similarly structured α and β seas to the warm tropics and from surface waters to subunits with molecular masses of 40 and 37 kDa, great depths. Both geographic, seasonal and depth- respectively. This enzyme oxidizes with atmos- related differences in luminescent bacterial abundance­ pheric oxygen (O2), a reduced riboflavin phos­ and species composition have been documented. phate (FMNH2) and a long chain fatty aldehyde The ­distribution patterns were suggested­ to be con- (RCHO) into an electronically excited flavin. With trolled by temperature, salt ­tolerance, resistance­ to the release of a blue-green light (490 nm), flavin

Symbiosis in Fishes: The Biology of Interspecific Partnerships, First Edition. Ilan Karplus. © 2014 Ilan Karplus. Published 2014 by John Wiley & Sons, Ltd. The Associations between Fishes and Luminescent Bacteria 7

Table 1.1 Fish hosts and symbiotic luminescent bacteria. Host Fish Depth and Temperature Luminescent Order and Family Bacteria Species Anguilliformes Shallow? Cool? Not identified Congridae Argentiformes Deep, cold Photobacterium kishitanii Opisthoproctidae Aulopiformes Deep, cold Photobacterium kishitanii Chlorophthalmidae Gadiformes Moderate to deep, cold Photobacterium kishitanii Macrouridae Aliivibrio fisheri Gadiformes Moderate to deep, cold Photobacterium mandapamensis Merculidae Gadiformes Moderate to deep, cold Photobacterium kishitanii Moridae Photobacterium mandapamensis Lophiformes Deep, pelagic, cold Not identified 9 families Berciformes Shallow to moderate, Not identified* Anomalopidae warm to cool Berciformes Shallow to moderate, Aliivibrio fisheri Monocentridae warm to cool Berciformes Deep, cold Photobacterium kishitanii Trachichthyidae Perciformes Shallow to moderate, Photobacterium kishitanii Acropomatidae cool to cold Photobacterium leiognathi Photobacterium mandapamensis Perciformes Shallow, warm Photobacterium mandapamensis Apogonidae Perciformes Shallow to moderate, Photobacterium leiognathi Leiognathidae warm to cool Photobacteriun mandapamensis Vibrio harveyi (?)

*Except for Photodesmus katoptron (Hendry and Dunlap 2011). Urbanczyk et al. 2011. Reproduced with permission of John Wiley & Sons.

­mononucleotide is produced together with water which probably resulted from gene duplication, since and a fatty acid (RCOOH): there is about 30% identity in the amino acid sequence between the α and β subunits of all bacte-

FMNH2 + RCHO + O2 → FMN + H2O + RCOOH rial luciferases. The order of the lux CDE genes cod- + hv (490 nm) ing for the fatty acid reductase complex is the same in all operons. The lux C and lux D genes, which code A fatty acid reductase complex containing three for fatty acid reductase and acyl-transferase polypep- enzymes, a reductase, a transferase and a synthetase, tides, respectively, flank the luciferase genes upstream catalyzes the acid with water back into the fatty alde- and lux E, which code for acyl-protein synthetase hyde substrate to facilitate the continuous produc- being downstream (Figure 1.1). Downstream of lux tion of light (Nealson and Hastings, 1990; Meighen, E is located lux G which specifies flavin reductase. 1991). Structural genes coding for enzymes involved Upstream of the lux operon of Aliivibrio fisheri are in the bioluminescent reaction of bacteria are located found the genes lux I and lux R, which are involved in the lux operon (Figure 1.1). The genes lux A and in quorum sensing,­ as discussed later. Similar ­content lux B code for the α and β subunits of luciferase, of the bacterial lux genes, their organization, and 8 Chapter 1

Photobacterium C D A B F E G Symbiotic Luminescent Bacteria phosphoreum in Fish Light Organs Photobacterium C D A B F G leiognathi The anatomical, physiological and behavioral Vibrio C D A B E G H harveyi ­expression of luminescence reaches its zenith in Vibrio fishes, being more diverse and complex than in any C D A B E G cholerae other group of organisms. Overall, the majority of Aliivibrio luminescent fishes produce their own light in R I C D A B E G fisheri intrinsic numerous photophores, whereas only a Shewanella minority forms symbiotic associations with lumi- R I C D A B E G hanedai nescent ­bacteria which they harbor in few light Photorhabdus C D A B E organs. Partnerships with luminous bacteria, the luminescens subject of this chapter are only formed by approxi- mately 500 species, constituting of less than 2% of all recognized fish species. However, these species Figure 1.1 Organization of the lux operons of are members of 21 families and 7 orders (Table 1.1). a variety of luminescent marine, freshwater and In contrast to the diversity of host fishes terrestrial bacteria including two light organ (Figure 1.2D, 1.2E, 1.2F), the symbiotic bacteria are symbionts. (Peat and Adams 2008. Reproduced few, relatively closely related, and found in three with permission of Springer Science+Business monophyletic groups (Figure 1.3). These bacteria Media B.V.) consist of one group of facultative symbionts of the genera Photobacterium (Figure 1.2A, 1.2B, 1.2C) homology, including those of the unculturable and Aliivibrio and two groups of not yet identified Photodesmus katoptron, support the view that bacteria apparently occurring obligatorily with ­bacterial luminescence arose apparently once flashlight fishes and deep sea anglerfishes. These (Meighen, 1991; Dunlap and Ast, 2005; Peat and symbionts were suggested to be unable to repro- Adams, 2008; Hendry and Dunlap, 2011). duce outside the fish light organs due to extreme Bacteria emit light in an energy conserving manner specialization and metabolic integration with their depending on population density, termed quorum hosts (Herring and Morin, 1978; Haygood, 1993; sensing. The bacteria release into the growth media a Urbanczyk et al., 2011). signal molecule, termed an inducer, which accumulates Light produced by bioluminescent organisms at and triggers luminescence only after a critical threshold the water surface is too dim to be functional in full concentration is crossed. Below a critical density the sunlight or moonlight. In shallow water ­lumi­nous generated light will probably not be seen by higher species of fishes are, therefore, crepuscular or organisms and is, therefore, nonbeneficial. The regula- ­nocturnally active (e.g., flashlight fishes) whereas tory mechanism of auto induction was first studied in deep water fishes (e.g., ceratioid anglerfishes) may Aliivibrio fischeri. This bacteria secrets a highly specific use light irrespective of the circadian light cycle auto inducer purified as β-ketocaproyle homoserine­ (Morin, 1981; Haygood, 1993). Luminescence in lactone. A. fischeri are usually nonluminous in sea water coastal water fishes is more often of a bacterial while free living when their concentration is usually than of an endogenous origin. This phenomenon below 102/ml. However, when a density of 107/ml is was suggested to be in some way related to the reached the inducer activates the lux R regulatory relative abundance of luminescent bacteria in gene, which in a simplistic manner acts as a specific coastal waters, particularly in the tropics and inducer of the lux operon, leading to the generation of ­subtropics, and their reduced abundance in oce- the enzymes involved in luminescence, as well as anic waters. However, some oceanic fishes which adding units of the inducer. In fish light organs, the occupy the bathypelagic (e.g., Ceratioidei), mes- concentrations of A. fischeri (e.g., 109–1010/ml) is by far opelagic (e.g., Ophistoproctidae) and benthopelagic above the threshold and light is continuously generated. (e.g., Macrouridae) zones, do harbor luminescent bac- The actual mechanism regulating quorum sensing is teria in their light organs (Herring, 1977, 1982). more complicated, involving more than one inducer There are no fishes with luminescent bacteria in and both ­negative and positive feedback loops (Ruby freshwater, except for few marine species such as and Nealson, 1976; Nealson and Hastings, 1990; Gazza spp. that enter estuaries and brackish water Meighen, 1991; Dunlap, 1999; Hoff, 2009). (Nicol, 1967). The Associations between Fishes and Luminescent Bacteria 9

A B C

D E

F

Figure 1.2 Luminescent bacteria photographed by the light they produce and their host fishes: A. Photobacterium mandapamensis; B. Photobacterium leiognathi; C. Photobacterium kishitanii D. Siphamia versicolor host of P. mandapamensis; E. Secutor megalolepis host of P. leiognathi; F. Chlorophthalmus albatrossis host of P. kishitanii (Urbanczyk et al. 2011. Reproduced with permission of John Wiley & Sons).

Luminescent bacteria are contained extra ­cellularly ­nutrition for growth and luminescence (Harvey, within the light organs, occupying the sheltered spaces 1952). Light organs may be either external, connected­ usually formed by parallel tubules or chambers. The directly via pores or ducts to the surrounding sea bacteria obtain from their host both oxygen­ and water, or internal, opening into the fish’s gut being only 10 Chapter 1

Facultative V. parahaemolyticus symbiots V. campbelli V. vulnificus Anomalopid symbionts V. harveyl V. alginolyticus V. orientalis P. phosphoreum P. steinetzli symboint P. leiognathi P. papebratus symbiont P. fischeri A. katoptron symbiont V. anguillarum K. alfredi symbiont

V. diazotrophicus

V. hollisae

V. cholerae M. johnsoni symbiont

C. couesi symbiont

Ceratioid symbionts E. coli

Figure 1.3 Phylogenetic relationships among luminous bacterial symbionts and other Vibrios based on parsimony analysis of small subunit rRNA sequences. Representative hosts are illustrated next to their respective symbionts. P. phosphoreum in the original figure was changed to P. kishitanii (Haygood 1993. Reproduced with permission of Taylor & Francis).

indirectly connected to the environment. There is an the appropriate species of bacteria, which has to be enormous diversity of both structure and location of contained exclusively in the light organ, whereas other light organs (Figure 1.4). For example, light organs of species should be excluded. Bioluminescent light must cardinal fishes of the genus Siphamia consist of two be maximized whereas bacterial growth must be different types. A single disc-shaped organ with com- strictly curtailed to save resources. Finally, the contin- plex masses of tubules located in the body cavity uously produced bacterial light has to be controlled in ­connected to the intestine, and paired gular sacs with order to effectively serve the fish. Means for light con- a simple chambered structure which ­protrude into the trol include structures such as shutters, rotatable light oral ­cavity (Fishelson, et al., 2005; Thacker and Roje, organs and chromatophores which may completely 2009). Fish light organs are ­fundamentally different block light passage. Accessory structures, such as dif- from intrinsic light ­producing­ photophores of fishes. ferentially reflective swim bladders, translucent tissues Light organs consist­ usually of one or two organs but and tubes lined with guanine crystals, may serve for never more than of four, whereas a single fish may light guidance, transmission and diffusion. carry thousands­ of photophores. Light organs are In most cases the presumed functions of light always directly or indirectly open to the exterior for emission have their basis in inference from morpho- ­discharge of surplus bacteria or inoculation, whereas logical and physiological characters and remain photophores are often closed. Internal light organs are conjectural. Experimental studies involving light mostly connected to the gut whereas photophores are emission and direct in situ observations are rare. usually not associated with the digestive tract (Herring, Even the monitoring of light emission in the field 1977). According to Herring (1977) fishes utilizing with submersibles and remotely operated vehicles is luminous bacteria in their light organs must cope with problematic, since many behaviors can only be several issues in order to maximize their benefit from observed unobtrusively (Widder, 2010). Despite the this association. Light organs must be infected with difficulty of studying bioluminescence in the marine The Associations between Fishes and Luminescent Bacteria 11

Anomalopid Leiognathid Esophagus Stomach Opisthoproctid Intestine Rectum Merlucciid Ceratioid Apogonid Morid

Monocentrid Anus Trachichthyid Macrourid Acropomatid

Figure 1.4 The locations, sizes and openings of the light organs of various groups of luminescent fishes presented in a single diagrammatic fish. (Hastings et al. 1987. Reproduced with permission of Springer Science+Business Media B.V.) environment and the little progress achieved, this (meaning abnormal eye). The name of the family topic has remained highly attractive over decades, was derived from the suboccular light organs pos- leading researchers to frequently review the sug- sessed by all flashlight fishes (Figures 1.5, 1.6, 1.7). gested functions for the emission of light (Nicol, The light organs are packed with a monomorphic 1967; Tett and Kelly, 1973; Herring and Morin, 1978; culture of bacteria that continuously emit a bright Morin, 1981, 1983; Young, 1983). One of the best blue-green light. The light from a single fish reviews, presented by Morin (1983), included three (e.g. Photoblepharon steinitzi) can be noticed major function categories of the emitted light, ­underwater from a distance of 10–20 meters and namely, predator evasion, prey capture and intraspe- from several hundred meters from the shore when cific communication. However, Morin (1983) in his watching a large school close to the water surface. review emphasized that multiple functions of a sin- Flashlight fishes control the passage of light from gle light organ is common among fishes. In view of their suboccular light organs by either a shutter the dearth of behavioral observations on light organ mechanism (e.g., Photoblepharon) or a rotational operation and function, the three better known mechanism (e.g., Anomalops) or a combination of groups of fishes using bacterial light, the flashlight the two (e.g., Kryptophanaron). In the shutter mech- fishes, the ponyfishes and the deep sea anglerfishes anism a black skin flap is lifted up, completely are discussed here. These fishes represent both shal- ­covering the light organ, thus blocking the passage low coastal and oceanic deep sea fishes, harboring of light (Figure 1.7E, 1.7F, 1.7G, 1.7H), while in the facultative and probably obligatory bacteria in rotational mechanism the light organ is turned ­external and internal light organs. downward facing the black-lined orbit (Figure 1.7A, 1.7B, 1.7C, 1.7D). Most flashlight fishes occur in the Indo-Pacific with only two species found in the New Flashlight Fishes World. Kryptophanaron alfredi was collected in ­different localities in the Western Central Atlantic Taxonomy and Distribution and Phthanophaneron harveyi has so far only been captured in the Eastern Pacific in the Gulf of Eight relative small species of flashlight fishes have California (Table 1.2). The relationship among the been described from the tropics and subtropics different flashlight fish genera was suggested on the (Table 1.2). These night active, dark colored fishes, basis of multiple morphological traits, such as body which are quite similar in their general appearance, scale rows, fin spines and number of vertebrae, and belong to the Beryciform family Anomalopidae traits related to the light occluding mechanism Table 1.2 Size, coloration, depth range and distribution of flashlight fishes. Species Maximal standard Light organ Coloration Depth range Distribution References or total length (mm) (% head length) (meters) Anomalops 3503 35.42 Black, the base of the 0–3651 Central Western 1 McCosker and katoptron (TL) dorsal anal and pelvic Pacific Ocean1 Rosenblatt, 1987 (Bleeker, 1856) fins white-grey1 2 Baldwin et al., 1997 3 FishBase Kryptophanaron 1253 36.3–44.72 Black, head and fins 27–2001 Western Central 1 Colin et al., 1979 alfredi (TL) darkest, white scales Atlantic (Jamaica, 2 Baldwin et al., 1997 (Silvester and Fowler, at the basis of the Puerto Rico, 3 FishBase 1926) second dorsal and Curacao, Grand anal fins1 Cayman, Bahamas) Parmops coruscans 66.52 35.63* Black, fins and lower 3501–4402 Tahiti1 and Fiji2 1 Rosenblatt and (Rosenblatt and (SL) part of head paler Johnson, 1991 Johnson, 1991) than the rest of the 2 Johnson et al., 2001 body1 3 Baldwin et al., 1997 Parmops 88.51 35.62* Black1 440–5501 Fiji1 1 Johnson et al., 2001 echinatus (SL) 2 Baldwin et al., 1997 (Johnson, Seeto and Rosenblatt, 2001) Phthanophaneron 2042 22.7–31.23 Black, lateral line 32–361 Eastern Pacific 1 Rosenblatt and harveyi (SL) scales lighter1 (Gulf of Montgomery, 1976 (Rosenblatt and California)1 2 McCosker and Montgomery, 1976) Rosenblatt, 1987 3 Baldwin et al., 1997 Photoblepharon 1203 48.62** Dark brown to black 0–501 Central and 1 McCosker and palpebratus (TL) with a conspicuous Western Pacific Rosenblatt,1987 (Boddaert, 1781) white spot on the Ocean 2 Baldwin et al., 1997 opercule1 3 FishBase Photoblepharon 1104 48.63** Dark brown to 0–2002 Red Sea1 1 McCosker and steinitzi grey black4 Comoro Island1 Rosenblatt, 19871 (Abe and Haneda, Maldive Islands4 2 Heemstra et al., 2006 1973) southern Oman and 3 Baldwin et al., 1997 Somalia4 4 FishBase Protoblepharon 229 14.5 Dark brown to black 274 Cook Islands Baldwin et al., 1997 rosenblatti (SL) (in alcohol) (Baldwin, Johnson and Paxton, 1997)

*Computed for the genus Parmops **Computed for the genus Photoblepharon The Associations between Fishes and Luminescent Bacteria 13

flashlight fishes remained unknown for a long time because of their occurrence at depths not accessible to scuba divers and in areas with hard bottoms that preclude collecting of fishes with nets and trawls. These species were only discovered after being unexpectedly found in the stomach of a grouper (i.e., Parmops coruscans) and in a prawn trap (Parmops echinatus). Some species of flashlight fishes, such as Kryptophanaron alfredi, were believed to be extremely rare, since only a single specimen was collected in 1907 at the water surface off Jamaica (Dahlgren, 1908) and the species was not encountered over the next seventy years. However, after research- ers realized that flashlight fish are active during Figure 1.5 Photoblepharon steinitzi (Repro­ moonless nights, additional specimens were col- duced with permission of D. Darom). lected from different localities in the Western Central Atlantic (Colin et al., 1979). In the future, more flashlight fish species and new records will probably be discovered with aid of manned subma- rines and scuba divers during dark night dives.

The Light Organs Early researchers who examined preserved speci- mens of flashlight fishes did not consider light organs as such. According to McCosker (1977), Boddaert in 1781 suggested that these structures protect the eyes of the fish from coral branches, while Lacepede in 1803 suggested that they serve for eye protection from solar radiation. Vorderman in1900 (cited in Harvey, 1922) was the first to report seeing light produced by the living fish. Harvey (1921, 1922) provided evidence that the light origi- Figure 1.6 Kryptophanaron alfredi (Reproduced nates from bacteria which occupy the light organs. with permission of P. Colin). He based his suggestions on microscopy of the bac- teria and several characteristics of the lumines- cence, such as continuous light emittance and the (Johnson and Rosenblatt, 1988; Rosenblatt and inhibiting effect of desiccation and potassium cya- Johnson, 1991; Baldwin et al., 1997; Johnson et al., nide on light production, which are typical for bac- 2001). Photoblepharon and Anomalops, genera with terial light. More recently, bacterial luciferase ­different light occluding mechanisms are maximally activity was detected in anomalopid light organ separated but also interconnected by several genera extracts (Leisman et al., 1980). The ultimate proof that possess both light occluding mechanisms that the bacteria are the source of light requires that (Figure 1.8). these bacteria are grown in a pure culture which Several species of flashlight fishes (e.g., Parmops luminesces. Our inability to rear these bacteria out- echinatus) were only collected from depths of sev- side the light organs was suggested to be due to the eral hundred meters (Table 1.2). Other species (e.g., bacteria’s obligate dependence on their host for the Photoblepharon steinitzi) ascend during moonless supply of factors essential for their growth nights from deep water in the Comoro Islands (Haygood, 1993). Inability to rear these bacteria (McCosker and Lagios, 1975) or dark caves in rela- hindered research related to their relationship with tive shallow water in the Red Sea (Fridman, 1972) to other bacteria, identification of the genes of their feed at the water surface on zooplankton. Some luminescent system and bacteria–host interactions. 14 Chapter 1

A E

B F

C G

D H

Figure 1.7 The rotational and shutter light occluding mechanisms of flashlight fishes. Left side: The rotational occluding mechanism of Anomalops katoptron (A); the exposed light organ (B); the downward rotation of the organ into a pouch (C); the occluded light organ (D). Right side: The shutter occluding mechanism of Photoblepharon palpebratus (E); the exposed light organ (F); the upward lifting of the shutter (G); the occluded light organ (H). (McCosker 1977. Reproduced with permission of Scientific American, Inc.) The Associations between Fishes and Luminescent Bacteria 15

on on on and function were described. The internal opaque lephar reflector, which is positioned above the pigment mops ophanar ob Anomalops Par PhthanophanerKrypt Phot cells, contains stacks of hexagonal guanine crys- tals positioned parallel to the reflectance surface. It reflects about 70% of the visible spectrum. The partly translucent external reflector is positioned at the lower ventral edge of the anterior face of the organ. It consists of a row of thin crystals lying obliquely to the reflecting surface. This reflector according to Watson et al. (1978) directs light out- ward and upwards and dims light emission ventrally. Light emission is augmented by about 35% due to light reflectance. Epithelial tubes of a length of about 0.5–1.0 mm and a diameter of about 30–40 µm lie Figure 1.8 Cladiogram of anomalopid fishes perpendicular to the sagital plane of the fish. The (Rosenblatt and Johnson 1991. Reproduced walls of the tubes, which are triangular with rounded with permission of Allen Press Publishing corners in cross-section, consist of a single layer of Services). epithelial cells. These tubes are packed with lumines- cent bacteria. Five to six tubes are clustered around a capillary forming a rosette-like structure. Whereas the top of the tubes are open, at their base are cells However, recently, culture independent methods, rich with mitochondria (Kessel, 1977), which are such as PCR amplification of DNA of bacterial cells, further discussed in the section dealing with bacte- which occur at high concentrations in the absence rial light intensity regulation by the fish. The bacteria of other bacteria in the light organs have led to new in the light organ of Photoblepharon steinitzi are rod insights (Haygood and Distel, 1993; Hendry and shaped, gram negative measuring 3–6 µm in length Dunlap, 2011). These exciting new discoveries are and 0.5 µm in width, often possessing 1–3 flagella of discussed in a later section dealing with the evolu- the polar sheated type. The anterior face of the light tion of the fish–bacteria symbiosis. organ is covered by a transparent perforated epider- Several early studies addressed the structure of the mal layer that transmits the light generated by the light organs of flashlight fishes (Dahlgren, 1908; bacteria. In this layer are several hundred pores Steche, 1909). A more recent detailed study of this structured like craters about 13–15 µm wide. Several structure in Anomalops katoptron was carried out parallel tubes open individually into a common by Bassot (1968) with aid of light microscopy. atrium, which itself is connected to the outside by a Subsequently, Kessel (1977) studied the ultrastructure pore. Surplus bacteria are released from the light of Photoblepharon steinitzi with aid of light, scanning organ through these pores into the surrounding and transmission electron microscopy. The structure water (Nealson et al., 1984). The light organ is sup- of the light organs of these two species were quite ported by a fibrocartilaginous cup. This cup articu- similar (Kessel, 1977). Anomalopid light organs are lates at the anterior end of the organ with a composed of four basic structural and functional fibrocartilaginous stalk that contains blood vessels elements: black pigment cells which block light and nerves. This stalk and several accompanying passage, guanine crystals that reflect light, epithelial ligaments form the only connection of the light tubules which contain luminescent bacteria in their organ with the rest of the fish’s body, being otherwise extra cellular lumen and a translucent epidermis that freely suspended in the lower part of the orbit. transmits the generated light. The entire light organ Newton Harvey, a pioneer of bioluminescence except for its anterior face is covered from the out- research (Harvey, 1952), raised the question of the side by layers of black pigment cells richly supplied different mechanisms controlling light passage by blood vessels and nerves. This layer prevents the (i.e., rotational vs shutter) in two related fishes, passage of light into the fish’s eyes and body. Anomalops katoptron and Photoblepharon palpe- Light reflectance was studied in detail in bratus. In his own words “why these two similar in Anomalops katoptron (Watson et al., 1978). In this most respects and especially in the general struc- species, two light reflectors differing in structure ture of the luminous organ, should have developed 16 Chapter 1 such totally different means of extinguishing the tional complex in which each genera exhibits light is a mystery” (Harvey, 1922). This enigma a slightly more intricate and integrated linkage was solved by Johnson and Rosenblatt (1988) in a ­system to effect the shutter erection” concomitantly detailed study of the functional anatomy of the with a gradual increase in the relative size of the light occluding mechanisms of all anomalopid light organ (Rosenblatt and Johnson, 1991; Baldwin fishes known at that time. Following their study, et al., 1997). The unnecessary complexity of the light occluding mechanisms were also described in shutter mechanisms is according to Johnson and newly discovered species of flashlight fishes Rosenblatt (1988) possibly a result of functional- (Rosenblatt and Johnson, 1991; Baldwin et al., 1997; morphological constraints imposed on the system Johnson et al., 2001). In their study Johnson and by the pre-existence of a rotational mechanism in Rosenblatt (1988) discovered that the light occlud- the ancestral flashlight fish. Finally, actual blinking ing mechanisms are only superficially different in flashlight fishes is more complicated than usually whereas in reality they are similar in many aspects. assumed. In Anomalops lateral and anterior move- The power for both mechanisms involves the ment of the stalk causes the entire organ to swing Adductor Mandibulae muscles transmitted through out and forward (Johnson and Rosenblatt, 1988). the same complex biomechanical linkage, which Moreover, the light from each organ can be inde- includes the Ethmomaxillary and Diogenes liga- pendently controlled and also “squinted” by par- ments, the latter present only in flashlight fishes. tially occluding the organ (Morin unpublished in The ligament of Diogenes inserts in Anomalops at Herring and Morin, 1978). the ventral lateral (outer) corner of the cartiloge- To better understand the relationship between nous cup, unlike Photoblepharon, where this liga- flashlight fishes and the bacteria which occupy their ment inserts at the ventral medial corner. Due to light organs, Meyer-Rochow (1976a) carried out a the differential attachment of this ligament a pull starvation experiment with seven Anomalops katop- on it results in rotation of the organ in Anomalops tron. These fish were deprived for four weeks of food but not in Photoblepharon. Only Anomalops ­possess and concomitantly the luminescence of the bacteria at the floor of its orbit a fibrocartilaginous rotation was monitored. Following one week of starvation pad. The cartilogenous stalk of Photoblepharon pos- light became dimer. After two weeks a black central sess a hook that is firmly attached to a moveable spot appeared in the light organ and finally after cartilaginous knob. Both these structures, which three weeks the light seemed to be extinguished to are essential for lifting the shutter, are missing in the human eye. Four weeks after the start of the Anomalops but are found in less developed forms in experiment, the fish showed no signs of malnutri- other species such as Kryptophaneron, which pos- tion. The light organs were dissected and were found sess both light occluding mechanisms. to contain fewer bacteria compared with fresh light According to Rosenblatt and Montgomery organs. Meyer-Rochow (1976a) concluded that the (1976) the light organs probably evolved in the twi- fish transfers via the blood capillaries either nutrients light zone habitat. Johnson and Rosenblatt (1988) essential for bacterial growth or substances directly stated that the ancestral flashlight fish probably related to light production such as long chain alde- controlled light passage through a forced rotational hydes. Flashlight fishes lose luminescence not only in mechanism similar to that of Kryptophaneron, pos- response to starvation and malnutrition but also in sessing also a skin flap at the base of the light organ response to an environmental temperature stress which may have been erectable or not. The light (Haygood, 1993), low ambient oxygen and prolonged occluding mechanisms diverged from the ancestral exposure to light (Herring and Morin, 1978). There form into two lines, namely the flipping rotational have been reports of flashlight fishes with extin- and the shutter mechanisms (Figure 1.7). Whereas guished light organs that regained luminescence Anomalops diverged from the early ancestral flash- 3–6 months following maintenance in isolation. light fish about sixty million years ago, extant flash- Moreover, fish with extinguished lights maintained light fishes with a shutter mechanism evolved only together with several fish with bright lights failed to within the last four to five million years (Wolf and do so (Haygood, 1993). Haygood (1993) suggested Haygood, 1991). The series of five genera, which on the basis of these facts that regaining lumines- starts with Protoblepharon followed by Parmops, cence is probably related to resumption of light pro- Phthanphanaron, Kryptophaneron and culminates duction by surviving bacteria and not due to new with Photoblepharon, provides “a rare illustration colonization of the organ. Flashlight fishes with of the gradual evolutionary elaboration of a func- extinguished lights were successfully maintained The Associations between Fishes and Luminescent Bacteria 17 for many months in captivity, provided they were Photoblepharon Anomalops fed under dim light conditions. This fact suggests A C that the major benefit for the fish from the bacteria, at least under captive conditions is the light they ­produce (Haygood, 1993). A B

The Eye and the Light Organ B An histological study of the eyes of two flashlight fishes Photoblepharon palpebratus and Anomalops katoptron by Meyer-Rochow et al. (1982) revealed in both species structural dim light adaptations, such as relative large eyes, pupils and lenses, and the presence of only rods in the retina. On the basis of C the ratio of nuclei in the various retinal layers of A these two species, Meyer-Rochow et al. (1982) sug- gested that Photoblepharon has a greater light sensi- A tivity, whereas Anomalops a better visual acuity. B These suggestions are in accord with his own field observations in the Banda Islands that the former C fish is more photophobic confining its activity to nights with very low levels of light. B The light emitted by the bacteria residing in the light organ of Photoblepharon steinitzi was found to correspond with the light to which this fish’s eye is most sensitive – λ max 496 nm (Girsch, 1976). This finding clearly supports the suggestion of Meyer- Rochow et al. (1982) that the eye and the light organ C A form a well-adjusted pair of organs in flashlight fishes. According to Morin (unpublished, cited in A Herring and Morin, 1978) the light organ of B Photoblepharon is flush with the eye emitting a dis- C crete beam laterally and somewhat anteriorly and B downward. In Anomalops the exposed light organ, which is flared out, partially overlapping with the eye, emits a more anteriorly-laterally directed beam. The angular distribution of light radiating from the organs of freshly killed Photoblepharon palpebratus and Anomalops katoptron was monitored in the lab- C oratory (Herring, 1982). Photoblepharon had a broad and uniform angular distribution whereas Anomalops had a more limited dorsal illumination (Figure 1.9). However, live Anomalops were sug- Figure 1.9 The angular distribution of light gested to possess some ability to alter the direction from the light organs of Photoblepharon palpebra- of the emitted light (Herring and Morin, 1978). tus (left) and Anomalops katoptron (right). The According to Morin (1981) the position of the light plane in which each pair of measurements was organ close to the eye reduces the parallax between made is indicated by the central diagram; the rela- the eye and the light source, allowing the fish to tive light output is indicated by the length of the detect the tapetal reflection of its crustacean prey. line in each direction; the longest line indicates Thanks to the bacterial light beam the fish is able to the direction of the maximum intensity and has feed on smaller zooplankton (e.g., crustaceans in the been normalized to the same length in each ­angular range 1–3 mm) than most nocturnal fishes (Morin ­diagram (Herring 1982. Reproduced with per- and Harrington, unpublished, cited in Morin, 1981). mission of Taylor & Francis). 18 Chapter 1

Howland et al. (1992) suggested that the spatial rela- Rico within less than 1 m3. The fish were “­blinking” tionship of the light organ to the pupil of the flash- their lights rapidly and swimming around one light fish is similar to that of some photoretinoscopes, another, suggesting a form of courtship or facilitating the detection of fishes due to their tapetal ­spawning aggregation (Colin et al., 1979). eye shine. Anomalops katoptron makes a retino- A single Photoblepharon palpebratus female can scope-like movement when rotating the light organ produce in a breeding season up to 1000 transpar- away from its pupil. According to these authors, at a ent spherical positively buoyant eggs 1.2 mm in four meter distance the eyeshine of a conspecific will diameter. After a short planktonic phase of 5–10 be four orders of magnitude brighter than that of hours, these eggs become negatively buoyant and reflections from adjacent surfaces. adhere to the substrate with the aid of a sticky adhe- sive substance that covers the vitaline membrane. Reproduction, Larval and Light Dissection of the ovary of a mature female revealed the presence of eggs of three size classes: large ripe Organ Development eggs about 1 mm in diameter, medium sized eggs Knowledge of the breeding season and reproductive (diameter of 0.2–0.6 mm) and tiny eggs (diameter behavior of flashlight fishes is mostly missing, below 0.2 mm) (Meyer-Rochow, 1976b). A similar ­fragmentary and mainly restricted to the genus structure of mature eggs has also been reported for Photoblepharon. In the Red Sea, the breeding season Anomalops katoptron (Harvey, 1922; Colin, 1988). of P. steinitzi occurs between July and September. According to Meyer-Rochow (1976b), the short Following daily exposure of a pair to 14 hours of light planktonic phase of the eggs may explain the limited­ during the winter, sexual maturation was attained in distribution of Photoblepharon palpebratus occur- January (Sagi, 1978). In the Banda Islands, mature ring in the Banda Islands but not in other localities eggs were striped from P. palpebratus and Anomalops in the South Moluccan Sea despite ­presence of katoptron in October (Harvey, 1922) and, in the case ­habitats suitable for these fish. of the former species, captured pairs also spawned Developmental series of flashlight fishes encom- in April (Meyer-Rochow, 1976b). Finally, ripe passing the ontogeny of the light organs are missing Kryptophanaron alfredi females were collected in (Baldwin and Johnson, 1995). Moreover, there are Puerto Rico in April and in January in the Cayman only few data on larval flashlight fishes. Colin (1988) Islands, with the former group shedding infertile described early pre-flexion Anomalops katoptron eggs three days after capture­ (Colin et al., 1979). larvae from hatching (2.6–3.3 mm NL -notocord Spawning of flashlight fishes has never been length) to an age of 132 hours post hatch (4.2 mm witnessed either in the field or captivity. Sexual NL) based on larvae reared in the laboratory. The dimorphism has only been reported for the genus larvae hatch with a fairly large yolk sac, unpig- Photoblepharon. Females of P. palpebratus possess mented eyes, and undeveloped mouth. Distinctive more rounded tail fin edges than males (Meyer- features of these pre-flexion larvae are slender body, Rochow, 1976b). In established pairs of P. steinitzi long straight gut, large pelvic fins and heavy pig- (Morin et al., 1975) and P. palpebratus (Meyer- mentation (Figure 1.10A, 1.10B, 12.10C). There Rochow, 1976b) females are larger than males. was no sign of luminescent bacteria in these lar- Pairs of P. palpebratus captured in shallow water in vae. A 5.8-mm (SL) post-flexion Anomalops katop- the Banda Islands successfully spawned at night in tron was collected in the Western North Pacific small aquaria. In the case of a single pair kept with (Konishi and Okiyama, 1997). The head of the larva a male or two pairs kept together, the male and was large, its length occupying nearly one-half of female of each pair were closely associated, swim- the body length. An unpigmented crescent-shaped ming one after the other in circles or figures of ­tissue lay beneath the eye, possibly representing eight with the female leading most of the time. the incipient luminous organ (Figure 1.10D). The proximity of additional individuals was Regretfully, the presence of bacteria inside this avoided by the pair but no aggressive interactions organ was not examined. A 6.2-mm (NL) larva of were witnessed. This is in contrast to P. steinitzi, Kryptophanaron alfredi in an advanced state of where pairs in shallow water defended territories Notochordal flexion was collected with a mid-water from conspecific intruders (Morin et al., 1975). trawl off the Bahamas from a water depth ranging Finally, a group of nine Kryptophanaron alfredi between 400 meter and the water surface. The larva was once observed close to the substrate and to ­possessed an anteriorly-directed rod-like projection one another at a depth of 36 meters off Puerto on each side of the snout (Figure 1.11A, 1.11B).