Coping with Life on Land: Physiological, Biochemical, and Structural Mechanisms to Enhance Function in Amphibious Fishes

Andy Joseph Turko

A Thesis presented to The University of Guelph

In partial fulfilment of requirements for the degree of Doctor of Philosophy in Integrative Biology

Guelph, Ontario, Canada

ABSTRACT

COPING WITH LIFE ON LAND: PHYSIOLOGICAL, BIOCHEMICAL, AND STRUCTURAL

MECHANISMS TO ENHANCE FUNCTION IN AMPHIBIOUS FISHES

Andy Joseph Turko Advisor:

University of Guelph, 2018 Dr. Patricia A. Wright

The invasion of land by fishes was one of the most dramatic transitions in the evolutionary history of vertebrates. In this thesis, I investigated how amphibious fishes cope with increased effective gravity and the inability to feed while out of water. In response to increased body weight on land (7 d), the gill skeleton of Kryptolebias marmoratus became stiffer, and I found increased abundance of many proteins typically associated with bone and cartilage growth in mammals. Conversely, there was no change in gill stiffness in the primitive ray-finned fish

Polypterus senegalus after one week out of water, but after eight months the arches were significantly shorter and smaller. A similar pattern of gill reduction occurred during the tetrapod invasion of land, and my results suggest that genetic assimilation of gill plasticity could be an underlying mechanism. I also found proliferation of a gill inter-lamellar cell mass in P. senegalus out of water (7 d) that resembled gill remodelling in several other fishes, suggesting this may be an ancestral actinopterygian trait. Next, I tested the function of a calcified sheath that I discovered surrounding the gill filaments of >100 species of killifishes and some other percomorphs. I found no evidence that this calcification evolved to provide support in amphibious fish out of water. Instead, my experimental data suggests that the calcified sheath

maintains the position of gill filaments during aquatic ventilation. The role of gill mechanics has largely been neglected in previous studies of respiratory function, but my data suggests that filament stiffness may be critically important. Finally, I tested the hypothesis that prolonged survival out of water is enabled by slow metabolism and conservation of energy stores in K. marmoratus, which cannot feed in air. I found that low metabolism prolonged survival out of water by almost two weeks and this phenotype had increased fecundity in microcosms that were intermittently dry for half of a year. There was no obvious trade-off in fully aquatic environments. Overall, this thesis integrates physiological, ecological, and evolutionary perspectives to provide new insight into how amphibious fishes survive out of water.

ACKNOWLEDGEMENTS

First and foremost, I want to thank my supervisor Dr. Pat Wright for all the wisdom she has shared over the past 10(!) years of working together. Pat – I can’t thank you enough for giving me my first opportunity in science as a naïve undergrad, and for the countless opportunities that have followed. You let me follow my interests wherever they led, and that has made my graduate experience more fulfilling than I could have ever hoped. I have learned so much from you about how to be a thoughtful, rigorous scientist, and about how to succeed in the modern scientific world. It has been an absolute privilege to work with you.

Thank you so much to my committee members, Drs. Douglas Fudge, Todd Gillis, and

Graham Scott, who were so helpful all along the way. My thesis was shaped in large part by your ideas, expertise, equipment, and experimental know-how. I appreciate all of our great discussions and look forward to many more.

I have been lucky to work with fantastic collaborators each of my thesis chapters. To Drs.

Frank Smith, Roger Croll, and Matt Stoyek – my time in Halifax was an all-time research highlight. Thank you so much for sharing your microgravity expertise for being so much fun to be around. Dr. Dietmar Kültz – thank you for conducting the gill proteomics analysis, your data made a huge contribution to my understanding of gill remodelling. To Dr. Em Standen – thank you for sharing your Polypterus expertise, for housing me in Ottawa, and for all the good times in Belize. Priyam Maini, I really appreciate all of our thoughtful discussions, and of course your attention to detail analyzing Polypterus gills. Drs. Trina Du and Javier Santos-Santos, thank you both also for your help interpreting micro-CT scans. Bianca Cisternino – you have impressed me so much over the years we have worked together. You were instrumental keeping the microcosms running on schedule, and I am indebted to you for all your hard work analyzing gill

iv filament calcification. Dr. Ryan Earley - thank you for helping plan the microcosm experiments, the countless hours you spent analyzing microsatellites, and for all the stimulating conversations we have had over the years. You have taught me so much, and I look forward to learning more.

Justine Doherty – working with you to understand emersion tolerance in rivulus has been both rewarding and fun. I really appreciate all of your insight, and of course your long hours in the lab measuring metabolic rate and energy stores. Irene Yin-Liao – thank you for your many, many contributions, it has been such fun to work alongside you. Kelly Levesque – your help designing and running the microcosms was invaluable, I really enjoyed working with you. Perryn Kruth – you are a master of gill dissection. Thank you so much for all your hard work.

My co-conspirators in the Wright lab have been a major source of inspiration, knowledge, and good times. Dr. Chris Cooper, Hadi Dhiyebi, Kelly Regan, Mike Wells, Kristin

Bianchini, Cayleih Robertson, Kelly Levesque, Quentin Heffel, Tessa Blanchard, Giulia Rossi, and Louise Tunnah – I could not ask for better friends and colleagues. Matt Cornish, Mike

Davies, Abiran Sritharan, Sean Lee, Sydney Baxter, Andrea Dobrescu, and so many undergraduate volunteers – thanks for keeping the fish alive.

Dr. Scott Taylor – your passion for rivulus and all things mangrove have been such a source of wisdom and inspiration. I have enjoyed every minute we spent together in the field, and look forward to many more. Drs. Bill Milsom and Jonathan Wilson, thank you both for your insight regarding gill structure and function, I have learned so much. I would also like to thank the extraordinary members of the Southern Ontario Killifish Society who have shared their discoveries and their fish. I received generous tissue donations from many people – thank you

Hans Behr, Brittney Borowiec, Felix Breden, Peter DeSouza, Tyrone Genade, Brian Glazier,

Kate Gould, Christopher Martin, Jackie Matsumoto, Karen Murray, Jason Podrabsky, Kristina

Pohl, David Reznick, Nick Sakich, and Shayla Raycroft-Tuttle.

I received financial support to conduct this research from the Natural Sciences and

Engineering Research Council of Canada, Ontario Graduate Scholarship program, Canadian

Society of Zoologists, and the Journal of Experimental Biology Travelling Fellowship program.

A special thank you to my mom and dad, and to Clare and Joe, for a lifetime of support and encouragement. Zoey – you are always there for me. I love you.

STATEMENT OF AUTHORSHIP AND CONTRIBUTIONS TO SCIENCE

All material contained in this thesis were authored by Andy Turko or as described in the author list of each chapter. Author contributions are described below:

Chapter 2: AJT and PAW conceived the direction of the paper, AJT performed the literature review and wrote the draft manuscript, PAW edited the manuscript. Both authors approved the final version.

Chapter 3: Conceptualization: AJT, DF, PAW; Methodology: AJT, DK, DF, RPC, FMS, MRS;

Validation: AJT, DK; Formal analysis: AJT, DK; Investigation: AJT, DK, MRS; Resources: DK,

RPC, FMS, PAW; Data curation: DK; Writing - original draft: AJT; Writing - review & editing:

AJT, DK, DF, RPC, FMS, MRS, PAW; Visualization: AJT; Supervision: DF, RPC, FMS, PAW;

Project administration: PAW; Funding acquisition: AJT, DK, RPC, PAW.

Chapter 4: AJT, PAW and EMS conceived the study and designed the experiments. EMS and

PM collected and analyzed micro-CT data. AJT performed the mechanical testing experiments.

AJT and PM analyzed the histological data. AJT wrote the draft the manuscript; all authors revised the manuscript and provided final approval.

Chapter 5: AJT conducted the experiments, analyzed the data, and wrote the draft manuscript.

BC analyzed the gill photographs. PAW contributed to the overall conception of the study and design of all experiments and edited the manuscript.

Chapter 6: AJT, JED, RLE, and PAW conceived the study. AJT, JED, IY-L, KL, PK, JH, and

RLE carried out the experiments, AJT and JED analyzed the data. AJT wrote the draft manuscript.

vii

TABLE OF CONTENTS

Abstract ...... ii Acknowledgements ...... iv Statement of Authorship and Contributions to Science ...... vii Table of Contents ...... viii List of Tables ...... xii List of Figures ...... xiii Chapter 1: General Introduction ...... 1 1.1 Introduction ...... 2 1.2 Differences between air and water ...... 3 1.3 Gill structure and function...... 5 1.4 Thesis overview...... 7 1.5 References ...... 8 Figures ...... 14 Chapter 2: Evolution, ecology and physiology of amphibious killifishes (Cyprinodontiformes) ...... 17 Abstract ...... 18 2.1 Introduction ...... 19 2.2 Diversity and evolution ...... 20 2.3 Behaviour and ecology ...... 24 2.3.1 Emersion stimuli ...... 25 2.3.2 Terrestrial locomotion ...... 31 2.4 Physiology ...... 34 2.5 References ...... 41 Figures ...... 55 Supplementary Material ...... 59 Chapter 3: Skeletal stiffening in an amphibious fish out of water is a response to increased body weight ...... 60 Abstract ...... 61 3.1 Introduction ...... 62 3.2 Materials and Methods ...... 64

viii

3.2.1 Animals ...... 64 3.2.2 Terrestrial acclimation ...... 64 3.2.3 Mechanical testing ...... 65 3.2.3 Microgravity simulation ...... 66 3.2.4 Proteomics ...... 67 3.2.5 Histology ...... 69 3.2.6 Calcium content of gill baskets ...... 70 3.2.7 Statistical analyses ...... 70 3.3 Results ...... 71 3.3.1 Mechanical properties of gill arches ...... 71 3.3.2 Proteomic responses to terrestrial acclimation ...... 72 3.3.3 Morphological changes after terrestrial acclimation ...... 73 3.4 Discussion ...... 75 3.4.1 Weight responsiveness in an amphibious fish ...... 75 3.4.2 How are gill arches stiffened? ...... 77 3.4.3 Evolution of weight sensing in vertebrates ...... 81 3.5 References ...... 83 Figures ...... 95 Supplementary Material ...... 103 Chapter 4: Gill remodelling during terrestrial acclimation in the primitive amphibious fish Polypterus senegalus ...... 109 Abstract ...... 110 4.1 Introduction ...... 111 4.2 Materials and Methods ...... 113 4.2.1 Experimental Animals ...... 113 4.2.2 Experimental Protocol ...... 114 4.2.3 Series I – Short-term terrestrial acclimation ...... 114 4.2.4 Series II – Long-term terrestrial acclimation ...... 117 4.2.5 Statistical analysis...... 118 4.3 Results ...... 118 4.3.1 Short-term terrestrial acclimation ...... 118 4.3.2 Long-term terrestrial acclimation ...... 119 4.4 Discussion ...... 120

4.4.1 The inter-lamellar cell mass ...... 121 4.4.2 Plasticity of gill filaments and arches ...... 122 4.4.3 Phenotypic plasticity and evolution ...... 125 4.5 References ...... 126 Figures ...... 134 Supplementary Material ...... 138 Chapter 5: Fish respiratory function depends on gill filament calcification ...... 139 Abstract ...... 140 5.1 Introduction ...... 141 5.2 Methods ...... 143 5.2.1 Experimental animals ...... 143 5.2.2 Staining and clearing ...... 144 5.2.3 Phylogenetic analysis ...... 145 5.2.4 Plasticity of filament calcification ...... 146 5.2.5 Gill basket resistance ...... 147 5.3 Results ...... 148 5.4 Discussion ...... 150 5.4.1 Phylogenetic patterns of filament calcification ...... 151 5.4.2 Calcified filaments improve respiratory function ...... 152 5.4.3 Filament calcification in amphibious killifishes ...... 154 5.4.4 Perspectives ...... 155 5.5 References ...... 157 Figures ...... 164 Supplementary Material ...... 170 Chapter 6: Prolonged survival out of water is linked to a generally slow pace of life in a selfing amphibious fish ...... 174 Abstract ...... 175 6.1 Introduction ...... 177 6.2 Materials and Methods ...... 179 6.2.1 Animals ...... 179 6.2.2 Emersion tolerance ...... 180 6.2.3 Metabolic rate ...... 180 6.2.4 Energy reserves and consumption ...... 181

6.2.5 Life history traits ...... 182 6.2.6 Microcosms ...... 183 6.3 Results ...... 185 6.4 Discussion ...... 187 6.4.1 Metabolism and emersion tolerance ...... 188 6.4.2 Emersion tolerance trade-offs ...... 190 6.4.3 Conclusions and perspective ...... 194 6.5 References ...... 196 Figures ...... 207 Supplementary Material ...... 214 Chapter 7: General Discussion ...... 221 7.1 Major findings ...... 222 7.2 Thesis limitations ...... 226 7.3 Future directions ...... 231 7.4 Conclusions ...... 235 7.5 References ...... 236

LIST OF TABLES

Table 3.S1: Abundances of all collagen isoforms identified in the gills of Kryptolebias marmoratus acclimated to water or 14d in air…………………………………………….. 103

Table 5.S1: Species, sample sizes (n), calcification (proportion of filament length), standard error (sem), and acquisition information for all fishes used for phylogenetic analysis…… 170

Table 6.S1. Body composition data used to calculate energy use in three distinct genetic lineages of Kryptolebias marmoratus after 21 d of terrestrial acclimation………………………… 214

xii

LIST OF FIGURES

Figure 1.1. Vertebrate phylogeny indicating clades that contain amphibious fishes...... 14

Figure 1.2. Representative gill arch from Goodea gracilis with anatomical features labelled. ... 16

Figure 2.1. Phylogenetic placement of cyprinodontiform families containing amphibious species (after Costa, 2012)...... 55

Figure 2.2. Phylogenetic distribution of amphibious fishes within the cyprinodontiform families (a) Rivulidae (after Costa, 2011a, 2013) and (b) Fundulidae (after Ghedotti & Davis, 2013). .... 56

Figure 2.3. Photographs of amphibious killifishes out of water...... 57

Figure 2.4. Capillaries (green) extend between the scales of Kryptolebias marmoratus into the epidermis...... 58

Figure 3.1. Mechanical testing of mangrove rivulus gill arches...... 95

Figure 3.2. Increased gill stiffness after terrestrial acclimation is reversible in Kryptolebias marmoratus...... 96

Figure 3.3. Increased stiffness of K. marmoratus versus Danio rerio gill arches...... 97

Figure 3.4. Relative abundances of structural proteins that significantly changed expression after 2 weeks of terrestrial acclimation...... 98

Figure 3.5. Increased density of collagen molecules in K. marmoratus gill arches after terrestrial acclimation...... 99

Figure 3.6. Increased collagen density in mangrove rivulus gill filaments after terrestrial acclimation...... 101

Figure 3.7. Calcium content of K. marmoratus gill arches...... 102

Figure 3.S1. Increased gill stiffness (calculated as a spring constant from force/length curves) after terrestrial acclimation is reversible in K. marmoratus...... 104

Figure 3.S2. Activity of mangrove rivulus in the simulated microgravity experiment...... 105

Figure 3.S3. Increased gill stiffness (calculated as a spring constant from force/length curves) after terrestrial acclimation in K. marmoratus does not occur in simulated microgravity...... 106

Figure 3.S4. Increased gill stiffness (calculated as a spring constant from force/length curves) of K. marmoratus versus D. rerio gill arches...... 107

Figure 4.1. Representative light micrographs of Polypterus senegalus gills stained with hematoxylin and eosin...... 134

xiii

Figure 4.2. Mechanical properties of Polypterus senegalus gill filaments after short-term (7 d) terrestrial acclimation...... 135

Figure 4.3. Mechanical properties of Polypterus senegalus gill arches after short-term (7 d) terrestrial acclimation...... 136

Figure 4.4. Morphology of Polypterus senegalus gill arches after long-term (8 month) terrestrial acclimation...... 137

Figure 4.S1. Representative image of a picrosirius red stained cross-section through the first gill arch, under polarized light, of a control Polypterus senegalus reared in water...... 138

Figure 5.1. Representative gill filaments stained with Alizarin red and Alcian blue, each demonstrating a calcified sheath surrounding the base of the supportive cartilage rod...... 164

Figure 5.2. The degree of filament calcification in Kryptolebias marmoratus depends on the position along the gill arch...... 166

Figure 5.3. Ancestral state reconstruction of gill filament calcification in the Cyprinodontiformes and several outgroup taxa...... 167

Figure 5.4. Representative calcein-labelled gill of Kryptolebias marmoratus...... 168

Figure 5.5. Resistance of the gill basket in Kryptolebias marmoratus before (control) and after decalcification at five ventilatory frequencies (Vf) maintained with a peristaltic pump...... 169

Figure 5.S1. Gill filament calcification in Kryptolebias marmoratus after 28d of acclimation to water versus air, or fresh versus 35‰ seawater...... 173

Figure 6.1. Survival (proportion) in different strains (Belize, Florida, Honduras) of Kryptolebias marmoratus out of water...... 207

Figure 6.2. Fulton’s condition factor in different strains (Belize, Florida, Honduras) of Kryptolebias marmoratus in water under normal conditions (control) and after 21 d out of water (terrestrial)...... 208

Figure 6.3. Mean ± SE rates of O2 consumption in different strains (Belize, Florida, Honduras) of Kryptolebias marmoratus (A) over 7d of terrestrial acclimation, and (B) after 21 d out of water...... 209

Figure 6.4. Mean ± SE rates of fuel use after 21 d of terrestrial acclimation in 3 strains (Belize, Florida, Honduras) of Kryptolebias marmoratus...... 210

Figure 6.5. Reproductive measures of 3 strains (Belize, Florida, Honduras) of Kryptolebias marmoratus over the first 18 months of life...... 211

xiv

Figure 6.7. Number of embryos recovered from microcosms after 12 months...... 213

Figure 6.S1. Body mass (A) and standard length (B) as a function of age in two strains of Kryptolebias marmoratus...... 215

Figure 6.S2. Energy and water content in 3 strains (Belize, Florida, Honduras) of Kryptolebias marmoratus under normal housing conditions (control) and after 21 d terrestrial acclimation. 216

Figure 6.S3. Change in body mass of 3 strains (Belize, Florida, Honduras) of Kryptolebias marmoratus after 21 d terrestrial acclimation...... 217

Figure 6.S4. Mean ± SE time spent moving (%) over a 1 h interval in different strains (Belize, Florida, Honduras) of Kryptolebias marmoratus over 21 d of terrestrial acclimation...... 218

Figure 6.S5. Body mass (A), gonad mass (B), and liver mass (C) as a function of body length in Belize and Honduras strains of Kryptolebias marmoratus after 12 months in fully aquatic (control) or periodically drained (fluctuating) microcosms...... 219

Figure 6.S6. Calculated total length of gill filaments in Belize and Honduras strains of Kryptolebias marmoratus after 12 months in fully aquatic (control) or periodically drained (fluctuating) microcosms...... 220

CHAPTER 1: GENERAL INTRODUCTION

1.1 Introduction

Major habitat transitions are rare evolutionary events (Vermeij and Dudley, 2000). These transitions are probably infrequent because phenotypes that are well suited to any given environment are often poorly suited to environments with different abiotic or biotic characteristics. However, organisms that successfully invade novel habitats can take advantage of unoccupied ecological niches, and these pioneers often drive adaptive radiations (Schluter,

2000). Thus, understanding how biochemical, physiological, and morphological traits can either prevent or enable habitat transitions is of fundamental interest to many biological disciplines.

One of the most dramatic habitat transitions is the switch between aquatic and terrestrial environments, due to the dramatically different physical properties of air and water (Denny,

1993; Sayer, 2005). Given these challenges, the invasion of land by early lobe-finned fishes has been called “the greatest step in vertebrate history” (Long and Gordon, 2004), and ultimately resulted in the adaptive radiation of tetrapods (Clack, 2002). These are not the only fishes to colonize land, however. Amphibious lifestyles have independently evolved at least 20 times in extant ray-finned fishes, and there are currently over 200 species known to leave water (emerse) as a regular part of their natural history (Ord and Cooke, 2016; Wright and Turko, 2016; Figure

1.1). Extant amphibious fishes face many of the same challenges as our tetrapod ancestors and may provide our best tool for understanding the physiological adaptations required to transition from water to land (Graham and Lee 2004). The reasons why fishes emerse are diverse (Sayer and Davenport, 1991; Turko and Wright, 2015 [Chapter 2]). Some amphibious fishes leave water to avoid unfavourable abiotic conditions such as hypoxia or high temperatures (e.g. Regan et al.

2011; Gibson et al., 2015; Livingston et al. 2018), others to escape aquatic predation (Baylis,

1982; Ord et al., 2017), and some fishes are stranded without water during the tropical dry

2 season (Bruton, 1979; Taylor et al., 2008). Maximum survival time out of water also varies widely among species, but the physiological mechanisms that enable long-term emersion tolerance are not well understood.

1.2 Differences between air and water

The fundamental physical differences between water and air pose several challenges for amphibious fishes (Bliss, 1979; Dejours, 1988). Air is approximately 800 times less dense than water, and thus fishes in air receive much less buoyant support from the surrounding environment than when in water (Dejours, 1988). This means that fishes that are effectively weightless in water must contend with increased effective body weight on land due to the increased apparent force of gravity (Denny, 1993). Coping with increased body weight on land is a major challenge for extant amphibious fishes, and this increased weight is also thought to have driven the evolution of stronger skeletal and muscular systems in terrestrial tetrapods to provide support and facilitate locomotion (Clack, 2002). Proximate changes in the mechanical loading of tissues (e.g. exercise or spaceflight) can also affect the musculoskeletal system of some terrestrial animals (Turner, 1998). However, fishes evolved under fully aquatic, effectively weightless conditions and therefore some researchers have hypothesized that fishes may not be able to respond to changes in body weight (Horn, 2005). Prior to this thesis, no studies had investigated the consequences of increased apparent gravity in amphibious fishes, which experience increased body weight while on land [Chapters 3, 4, 5].

The physical differences between air and water also make it difficult for many amphibious fishes to feed while on land. In water, prey is typically suspended in the water column and a fish can position itself directly in line with the food item for capture. On land, however, food is

3 mostly present on the substrate, and without a neck fishes have limited ability to move the mouth dorso-ventrally thus making prey acquisition difficult (Heiss et al., 2018). Orienting the mouth towards prey is also difficult because depression of the hyoid, necessary for suction feeding, pushes the mouth and head up and away from the substrate. Furthermore, generating enough suction to capture prey is nearly impossible for fishes in air, given its low density and viscosity

(Alexander, 1970). Even if prey can be captured without suction, swallowing also requires fluid transport within the mouth and bucco-pharyngeal cavity. Some amphibious fishes have overcome these challenges and can feed on land (e.g. hydrodynamic tongues in mudskippers,

[Michel et al.,2015]; flexible “necks” in eel-catfish [Van Wassenbergh et al., 2006] and reedfish

[Van Wassenbergh et al., 2017]), but most cannot feed while out of water (Heiss et al., 2018). A few amphibious fishes, such as Protopterus aethiopicus, Synbranchus marmoratus, and

Lepidogalaxias salamandroides survive extended periods (months-years) out of water using metabolic rate suppression or aestivation to conserve energy stores (Guppy and Withers, 1999).

Other amphibious fishes maintain or slightly increase metabolic rate out water (Graham, 1997;

Martin, 2014; Wright and Turko, 2016). Among these non-aestivating amphibious fishes, the length of time different species can survive out of water varies dramatically from hours (Graham et al., 1985) to several months (Taylor, 1990). The physiological basis of this variation in survival time is not well understood. Considering that most of these fishes cannot feed without water, one hypothesis is that prolonged survival on land is enabled by large energy stores or intrinsically slower metabolic rates [Chapter 6].

1.3 Gill structure and function

The gills of amphibious fishes are thought to be especially sensitive to the effects of increased effective gravity (Graham, 1997). In water-breathing fishes, gills provide the main site of respiration, osmo- and iono-regulation, and nitrogen waste excretion (Hughes, 1984; Wilson and Laurent, 2002). Without the buoyant support of water, however, gills typically become sub- or non-functional as the tissues collapse and coalesce, resulting in reduced functional surface area. To maintain physiological regulation of homeostasis, amphibious fishes must therefore either use adaptive specializations to maintain gill function, or these physiological processes can be shifted away from the gills to alternative organs.

Gill structure and function is relatively conserved among fishes. In Actinopterygian fishes, the gill basket is comprised of four pairs of branchial arches, and each arch receives structural support from a series of three bones (the epibranchial, ceratobranchial, and hypobranchial) which are joined dorsally by the pharyngobranchials and ventrally by the basibranchials (Hughes, 1984,

Olson, 2000). Extending from the medial side of the gill arch are the bony or cartilaginous gill rakers, which are used for feeding (Kahilainen et al., 2011). Two rows of gill filaments extend perpendicularly from the lateral side of each gill arch, collectively forming a holobranch

(Hughes, 1984; Figure 1.2). Each gill filament is supported by a cartilage rod running its length.

The core of these rods is filled by a series of chondrocytes, which produce the collagen and other glycoproteins that form the surrounding cartilage. Sometimes, the proximal portion of the cartilage rod is calcified (Conway and Mayden, 2009; [Chapter 4]). This calcified cartilage has only been reported in a few teleost species and the functional significance has never been investigated [Chapter 4]. The gill filaments are connected to the gill arch by muscle, and this musculature can also be used to adjust the position of the gill filaments (Hughes, 1984). The

5 dorsal and ventral surfaces of each filament bear the gill lamellae, which protrude perpendicularly from the filament and are typically free along the distal edge (Figure 1.2). The lamellae roughly resemble thin plate or leaf-like structures and consist of two layers of thin epithelium with circulating blood in between (Wilson and Laurent, 2002). The lamellae are the main site of exchange between blood and the external environment, as overall lamellar surface area is large and diffusion distances are short (typically 0.5-3 µm; Hughes, 1984). Some structural support of the gill lamellae is provided by pillar cells that span the epithelial layers, but proper orientation of the lamellae also usually requires buoyant support from the environment

(Laurent and Dunel, 1980; Olson, 2000).

Some amphibious fishes have specialized gill morphology that is thought to enhance function out of water. For example, the mudskipper Periophthalmodon schlosseri has branched gill filaments that do not completely coalesce when out of water, which maintains more surface area than typical straight filaments (Low et al., 1988). Furthermore, the lamellae of these mudskippers and a few other amphibious fishes are fused distally, which probably also serves to maintain surface area on land (Daxboeck et al., 1981, Low et al., 1988). This functional hypothesis has never been directly tested, however, and knowledge of morphological specializations in amphibious fishes is limited to studies of single species or small groups of closely related taxa.

Another possible strategy that could be used by amphibious fishes to maintain gill function out of water is the use of phenotypic plasticity. Phenotypic plasticity refers to the environmental sensitivity of a phenotypic trait, or the ability of an organism to modify its phenotype to suit environmental conditions (DeWitt et al., 1998; West-Eberhard, 2003). This definition includes both developmental plasticity, irreversible changes in phenotype caused by environmental conditions experienced early in life, as well as phenotypic flexibility, or reversible changes in

6 phenotype that occur in mature organisms. There are only a few examples of phenotypic plasticity in amphibious fishes in response to terrestrial acclimation. The mangrove rivulus

Kryptolebias marmoratus reversibly fills the spaces between gill lamellae with epithelial cells, which may reduce evaporative water loss or provide structural support while on land (Ong et al.,

2007; LeBlanc et al., 2010). African lungfish Protopterus annectens also fill the inter-lamellar spaces, using mucous, when out of water (Sturla et al., 2002). Phenotypically plastic responses in other parts of the gills, such as the arches or filaments, have not been investigated in amphibious fishes out of water. However, other environmental conditions (e.g. hypoxia) are known to have large effects on gill size and shape (Chapman et al., 2000; Crispo and Chapman, 2010). Thus, one central goal of this thesis was to determine the scope and direction of plasticity in amphibious fish gills out of water.

1.4 Thesis overview

The overall objective of this thesis was to understand how amphibious fishes have evolved to cope with increased effective gravity while out of water. I emphasized the functional morphology of the gills, given their importance to many physiological processes, and the trade- offs in structure and function caused by the physical differences between air and water. This thesis is comprised of a review paper (Chapter 2) and four original studies (Chapters 3, 4, 5, and

6). In Chapter 2, I synthesized the literature on amphibious killifishes, an order in which amphibious lifestyles have independently evolved several times. This order also contains the mangrove rivulus K. marmoratus, the model organism I studied for most of the original work included in this thesis. In Chapter 3, I used a combination of simulated micro-gravity, mechanical testing, histology, and proteomics to investigate if, and then how, the gills of K.

7 marmoratus respond to increased effective weight during terrestrial acclimation. Next, in

Chapter 4, I extended this work on weight responsiveness to the most basal clade of ray-finned fishes by studying the response of Polypterus senegalus acclimated to 1 week and 8 months out of water using histology, mechanical testing, and micro-computed tomography. In Chapter 5, I investigated the presence of a calcified “sheath” that I discovered surrounding the base of gill filaments in the amphibious K. marmoratus. I used a comparative phylogenetic analysis

(surveyed > 100 species), acclimation experiments, and experimental removal of the calcification to test two functional hypotheses: (1) that the calcification serves to support the gills while out of water, or (2) calcification maintains the position of the gill filaments during aquatic ventilation.

Finally, in Chapter 6 I investigated why some amphibious fishes can survive out of water for longer than others by comparing isogenic, effectively “clonal” strains of K. marmoratus.

Specifically, I tested the hypothesis that long-term survival out of water is ultimately limited by energy reserves in fishes that cannot feed during emersion, and thus tolerance to would be conferred by low metabolic rates enabled by reduced investment in energetically expensive tissues (e.g. gills, gonads).

1.5 References

Alexander, R. (1970). Mechanics of the feeding action of various teleost fishes. Journal of

Zoology, 162, 145–156.

Baylis, J. R. (1982). Unusual escape response by two cyprinodontiform fishes, and a bluegill

predator’s counter-strategy. Copeia, 1982, 455–457.

Bliss, D. E. (1979). From sea to tree: saga of a land crab. American Zoologist, 19, 385-410.

Bruton, M. N. (1979). The survival of habitat desiccation by air breathing clariid catfishes.

Environmental Biology of Fishes, 4, 273-280.

Chapman, L. J., Galis, F. and Shinn, J. (2000). Phenotypic plasticity and the possible role of

genetic assimilation: Hypoxia‐induced trade‐offs in the morphological traits of an African

cichlid. Ecology Letters, 3, 387-393.

Clack, J. A. (2002). Gaining ground: the origin and evolution of tetrapods. Bloomington,

Indiana: Indiana University Press.

Conway, K. W. and Mayden, R. L. (2009). Gill-filament ossifications: a possible morphological

synapomorphy uniting the families Balitoridae and Cobitidae (Ostariophysi: Cypriniformes).

Journal of Fish Biology, 75, 2839-2844.

Crispo, E. and Chapman, L. J. (2010). Geographic variation in phenotypic plasticity in response

to dissolved oxygen in an African cichlid fish. Journal of Evolutionary Biology, 23, 2091-

2103.

Daxboeck, C., Barnard, D. K. and Randall, D. J. (1981). Functional morphology of the gills of

the bowfin, Amia calva L., with special reference to their significance during air exposure.

Respiration Physiology, 43, 349-364.

Denny, M. W. (1993). Air and Water: The Biology and Physics of Life’s Media. Princeton, NJ:

Princeton University Press.

Dejours, P. (1988). Respiration in Water and Air: Adaptations-Regulations-Evolution.

Amsterdam: Elsevier Science Ltd.

DeWitt, T. J., Sih, A. and Wilson, D. S. (1998). Costs and limits of phenotypic plasticity. Trends

in Ecology and Evolution, 13, 77-81.

Gibson, D. J., Sylvester, E. V., Turko, A. J., Tattersall, G. J. and Wright, P. A. (2015). Out of the

frying pan into the air - emersion behaviour and evaporative heat loss in an amphibious

mangrove fish (Kryptolebias marmoratus). Biol. Lett. 11, 20150689.

Graham, J. B. (1997). Air-breathing Fishes: Evolution, Diversity, and Adaptation. San Diego,

CA: Academic Press.

Graham, J. B. and Lee, H. J. (2004). Breathing air in air: in what ways might extant amphibious

fish biology relate to prevailing concepts about early tetrapods, the evolution of vertebrate

air breathing, and the vertebrate land transition? Physiological an Biochemical Zoology, 77,

720-731.

Graham, J. B., Jones, C. B. and Rubinoff, I. (1985). Behavioural, physiological and ecological

aspects of the amphibious life of the pearl blenny Entomacrodus nigracans gill. Journal of

Experimental Marine Biology and Ecology, 89, 255-268.

Guppy, M. and Withers, P. (1999). Metabolic depression in animals: physiological perspectives

and biochemical generalizations. Biological Reviews, 74, 1-40.

Heiss, E., Aerts, P., and Van Wassenbergh, S. (2018). Aquatic–terrestrial transitions of feeding

systems in vertebrates: a mechanical perspective. Journal of Experimental Biology, 221,

jeb154427.

Horn, E. R. (2005). Gravity effects on life processes in aquatic animals. Advances in Space

Biology and Medicine, 10, 247-301.

Hughes, G. M. (1984). General anatomy of the gills. In: Fish physiology, Vol 10A (Eds. W. S.

Hoar, D. J, Randall). New York: Academic Press. pp 1–72.

Kahilainen, K. K., Siwertsson, A., Gjelland, K. Ø., Knudsen, R., Bøhn, T. and Amundsen, P.- A.

(2011). The role of gill raker number variability in adaptive radiation of coregonid fish.

Evolutionary Ecology, 25, 573–588.

Laurent, P. and Dunel, S. (1980). Morphology of gill epithelia in fish. American Journal of

Physiology Regulatory, Integrative and Comparative Physiology, 238, R147-R159.

LeBlanc, D. M., Wood, C. M., Fudge, D. S. and Wright, P. A. (2010). A fish out of water: gill

and skin remodeling promotes osmo-and ionoregulation in the mangrove killifish

Kryptolebias marmoratus. Physiological and Biochemical Zoology, 83, 932-949.

Livingston, M. D., Bhargav, V. V., Turko, A. J., Wilson, J. M. and Wright, P. A. (2018).

Widespread use of emersion and cutaneous ammonia excretion in Aplocheiloid killifishes.

Proceedings of the Royal Society B, 20181496.

Long, J. A. and Gordon, M. S. (2004). The greatest step in vertebrate history: A paleobiological

review of the fish-tetrapod transition. Physiological and Biochemical Zoology, 77, 700–719.

Low, W. P., Lane, D. J. W. and Ip, Y. K. (1988). A comparative study of terrestrial adaptations

of the gills in three mudskippers: Periophthalmus chrysospilos, Boleophthalmus boddaerti,

and Periophthalmodon schlosseri. Biological Bulletin, 175, 434-438.

Martin, K. L. (2014). Theme and variations: amphibious air-breathing intertidal fishes. Journal of

Fish Biology, 84, 577-602.

Michel, K. B., Heiss, E., Aerts, P. and Van Wassenbergh, S. (2015). A fish that uses its

hydrodynamic tongue to feed on land. Proceedings of the Royal Society B, 282, 20150057.

Olson, K. R. (2000). Respiratory system. In: The Laboratory Fish (Ed. G. K. Ostrander).

California: Academic Press. pp. 357-368.

Ong, K. J., Stevens, E. D. and Wright, P. A. (2007). Gill morphology of the mangrove killifish

(Kryptolebias marmoratus) is plastic and changes in response to terrestrial air exposure.

Journal of Experimental Biology, 210, 1109-1115.

Ord, T. J. and Cooke, G. M. (2016). Repeated evolution of amphibious behavior in fish and its

implications for the colonization of novel environments. Evolution 70, 1747-1759.

Ord, T. J., Summers, T. C., Noble, M. M. and Fulton, C. J. (2017). Ecological release from

aquatic predation is associated with the emergence of marine blenny fishes onto land.

American Naturalist, 189, 570–579.

Regan, K. S., Jonz, M. G. and Wright, P. A. (2011). Neuroepithelial cells and the hypoxia

emersion response in the amphibious fish Kryptolebias marmoratus. Journal of

Experimental Biology, 214, 2560–2568.

Sayer, M. D. J. (2005). Adaptations of amphibious fish for surviving life out of water. Fish and

Fisheries, 6, 186-211.

Sayer, M. D. J. and Davenport, J. (1991). Amphibious fish: why do they leave water? Reviews in

Fish Biology and Fisheries, 1, 159-181.

Schluter, D. (2000). The Ecology of Adaptive Radiation. Oxford: Oxford University Press.

Sturla, M., Paola, P., Carlo, G., Angela, M. M. and Maria, U. B. (2002). Effects of induced

aestivation in Protopterus annectens: a histomorphological study. Journal of Experimental

Zoology, 292, 26-31.

Taylor, D. S. (1990). Adaptive specializations of the cyprinodont fish Rivulus marmoratus.

Florida Scientist, 53, 239–248.

Taylor, D. S., Turner, B. J., Davis, W. P. and Chapman, B. B. (2008). A novel terrestrial fish

habitat inside emergent logs. American Naturalist, 171, 263-266.

Turko, A. J. and Wright, P. A. (2015). Evolution, ecology and physiology of amphibious

killifishes (Cyprinodontiformes). Journal of Fish Biology, 87, 815-835.

Turner, C. H. (1998). Three rules for bone adaptation to mechanical stimuli. Bone, 23, 399-407.

Van Wassenbergh, S., Herrel, A., Adriaens, D., Huysentruyt, F., Devaere, S., and Aerts, P.

(2006). A catfish that can strike its prey on land. Nature 440, 881.

Van Wassenbergh, S., Bonte, C. and Michel, K. B. (2017). Terrestrial capture of prey by the

reedfish, a model species for stem tetrapods. Ecology and Evolution, 7, 3856-3860.

Vermeij, G. J. and Dudley, R. (2000). Why are there so few evolutionary transitions between

aquatic and terrestrial ecosystems? Biological Journal of the Linnean Society, 70, 541-554.

West-Eberhard, M. J. (2003). Developmental Plasticity and Evolution. Oxford, UK: Oxford

University Press.

Wilson, J. M. and Laurent, P. (2002). Fish gill morphology: inside out. Journal of Experimental

Zoology, 293, 192-213.

Wright, P. A. and Turko, A. J. (2016). Amphibious fishes: evolution and phenotypic plasticity.

Journal of Experimental Biology, 219, 2245-2259.

Figures

Figure 1.1. Vertebrate phylogeny indicating clades that contain amphibious fishes. Lineages that contain at least one amphibious species are highlighted in bold and accompanied by a sketch of a representative amphibious species from that group; fully aquatic groups are labelled in blue. The scale bar indicates time since divergence (Mya, millions of years ago). Figure reproduced from

Wright and Turko (2009), Journal of Experimental Biology 219, 2245-2259.

Figure 1.2. Representative gill arch from Goodea gracilis with anatomical features labelled.

Tissue was stained with Alcian blue (cartilage) and Alizarin red (bone). Inset box shows filaments and lamellae enlarged 3-fold for clarity.

CHAPTER 2: EVOLUTION, ECOLOGY AND PHYSIOLOGY OF AMPHIBIOUS KILLIFISHES

(CYPRINODONTIFORMES)

Published as: Turko, A. J. and Wright, P. A. (2015). Evolution, ecology and physiology of

amphibious killifishes (Cyprinodontiformes). Journal of Fish Biology, 87, 815-835.

Abstract

The order Cyprinodontiformes contains an exceptional diversity of amphibious taxa, including at least 34 species from six families. These cyprinodontiforms often inhabit intertidal or ephemeral habitats characterized by low dissolved oxygen or otherwise poor water quality, conditions that have been hypothesized to drive the evolution of terrestriality. Most of the amphibious species are found in the Rivulidae, Nothobranchiidae and Fundulidae. It is currently unclear whether the pattern of amphibiousness observed in the Cyprinodontiformes is the result of repeated, independent evolutions, or stems from an amphibious common ancestor.

Amphibious cyprinodontiforms leave water for a variety of reasons: some species emerse only briefly, to escape predation or capture prey, while others occupy ephemeral habitats by living for months at a time out of water. Fishes able to tolerate months of emersion must maintain respiratory gas exchange, nitrogen excretion and water and salt balance, but to date knowledge of the mechanisms that facilitate homeostasis on land is largely restricted to model species. This review synthesizes the available literature describing amphibious lifestyles in cyprinodontiforms, compares the behavioural and physiological strategies used to exploit the terrestrial environment and suggests directions and ideas for future research.

2.1 Introduction

The earliest terrestrial vertebrates were not only able to exploit new ecological opportunities on land, but also faced many challenges (Clack, 2002). Fossils of early tetrapods have provided insight into the morphological adaptations required for the invasion of land, but provide less information regarding the behavioural and physiological adaptations that are necessary. Extant amphibious fishes, despite their relatively distant phylogenetic relationship to modern tetrapods, have probably faced many of the same selection pressures and thus provide another excellent system for understanding the evolution of terrestriality (Graham & Lee, 2004).

For the purpose of this review, amphibious fishes are defined as species that are found out of water as part of their normal life history (Gordon et al., 1969; Sayer, 2005). This broad definition includes fishes that leave water (emerse) for brief intervals (seconds to minutes) as well as those that remain in terrestrial environments for prolonged periods of days, weeks or even months.

Amphibious fishes are able to exploit a diversity of resources and habitats not available to fully aquatic relatives (Sayer & Davenport, 1991; Graham, 1997; Sayer, 2005). Some species of amphibious cyprinodontiforms are able to escape predation and interspecific competition by moving overland to ephemeral pools of water not available to most fishes, while other species leave the water to snatch terrestrial prey or avoid deteriorating water conditions. These fishes must be able to navigate terrestrial habitats while avoiding water loss and managing gas exchange, osmotic balance and nitrogenous waste excretion. The goals of this review are thus to summarize what is known about amphibious lifestyles in cyprinodontiforms, compare the strategies used by the various amphibious species for exploiting the terrestrial environment and propose ideas and directions for future research. Although the basic biology of many of these fishes is poorly known (Costa et al., 2010), two amphibious cyprinodontiforms, the mummichog

Fundulus heteroclitus (L. 1766) and mangrove rivulus Kryptolebias marmoratus (Poey 1880), have emerged as popular model organisms for research (Burnett et al., 2007; Earley et al., 2012;

Wright, 2012). These two species will be highlighted in this review.

2.2 Diversity and evolution

The order Cyprinodontiformes is large (c. 1000 species, eight families) and includes species from Europe, Africa, Asia and the Americas (Nelson, 2006). Amphibious habits have been described in at least 34 species of cyprinodontiforms, distributed among six families and in both the suborders Aplocheiloidei and Cyprinodontoidei (Fig. 2.1). Two members of the

Cyprinodontidae emerse, Cyprinodon dearborni Meek 1909 and Jordanella floridae Good &

Bean 1879. Anablepidae also contains two amphibious species, Anableps anableps (L. 1758) and

Anableps microlepis Müller & Troschel 1844, as does Poeciliidae, Gambusia affinis (Baird &

Girard 1853) and an unknown Poecilia species. The family Nothobranchiidae contains at least five amphibious species in both the genus Fundulopanchax and within the large and much- disputed Aphyosemion. These species include Fundulopanchax gardneri (Boulenger 1911),

Aphyosemion ahli Myers 1933, Aphyosemion cyanostictum Lambert & Géry 1968, Aphyosemion franzwerneri Scheel 1971, Aphyosemion fulgens Radda 1975, Aphyosemion georgiae Lambert &

Géry 1968 and an undescribed Aphyosemion species Lobaye popular in the aquarium hobby and a member of the Aphyosemion elegans species group (Collier, 2006).

The family Rivulidae contains at least 13 amphibious species, all formerly considered members of the genus Rivulus (Parenti, 1981; Huber, 1992). This is probably an underestimate of the number of amphibious rivulids, however, as numerous species have anecdotally been noted in the aquarium hobbyist literature to be strong jumpers that may emerse (Huber, 1992). In his

20 comprehensive book about the genus Rivulus, Huber (1992) characterized the jumping tendencies of 48 species, an amazing 79% of these (38 of 48 species) were considered to be excellent jumpers. In recent years, a great deal of effort has been spent revising the formerly paraphyletic Rivulus, and under the most recent classification scheme (Costa, 2011a) amphibious species are now found in four genera split between two subfamilies: the Kryptolebiatinae and the

Rivulinae [Fig. 2.2(a)]. Kryptolebias contains two amphibious species, K. marmoratus and

Kryptolebias sepia Vermeulen & Hrbek 2005. Two more amphibious species are found within

Laimosemion, Laimosemion geayi (Vaillant 1899) and Laimosemion kirovskyi (Costa 2004). The remaining amphibious rivulids are found in the sister genera Cynodonichthys and Anablepsoides.

These species include Cynodonichthys brunneus (Meek & Hildebrand 1913), Anablepsoides beniensis (Myers 1927), Anablepsoides derhami (Fels&Huber 1985), Anablepsoides hartii

(Boulenger 1890), Anablepsoides holminae (Eigenmann 1909), Anablepsoides limoncochae

(Hoedeman 1962), Anablepsoides micropus (Steindachner 1863), Anablepsoides stagnatus

(Eigenmann 1909) and Anablepsoides urophthalmus (Günther 1866).

Amphibious behaviour has been reported in eight species of the family Fundulidae. These

North American species are widely distributed phylogenetically and occur in two deeply divergent genera, Lucania and Fundulus [Fig. 2.2(b)]. Only one species of Lucania, Lucania parva (Baird & Girard 1855) has been recorded to leave the water (Baylis, 1982). Within

Fundulus, seven amphibious species have been described in two subgenera (Ghedotti & Davis,

2013). In the subgenus Fundulus, amphibious species include Fundulus confluentus Goode &

Bean 1879, Fundulus grandis Baird & Girard 1853, F. heteroclitus, Fundulus majalis (Walbaum

1792) and Fundulus similis (Baird & Girard 1853). Within Zygonectes, Fundulus nottii (Agassiz

1854) and Fundulus luciae (Baird 1855) are known to emerse.

The large phylogenetic diversity of amphibious cyprinodontiforms raises the question whether this behaviour is ancestral or has evolved repeatedly. To date, there is no clear pattern.

The six families containing amphibious species are found on either side of the most basal division in the order (between the Aplocheiloidei and Cyprinodontoidei; Fig. 2.1), suggesting the most parsimonious explanation may be that the common cyprinodontiform ancestor was able to leave water. This hypothesis requires amphibiousness to have been repeatedly lost in many cyprinodontiform families, genera and species (Fig. 2.1). Furthermore, the relative rarity of amphibious taxa (only about 3% of cyprinodontiforms demonstrate the trait) suggests that the ability to emerse may not be the basal condition. Instead, the tendency of cyprinodontiforms to inhabit shallow, murky or ephemeral habitats may have exposed many species to the environmental conditions thought to drive the evolution of amphibious habits (Parenti, 1981;

Sayer & Davenport, 1991). Annual life cycles in cyprinodontiforms are thought to have evolved independently six times, probably as an alternative strategy for occupying the same habitat types that favour amphibiousness (Furness et al., 2015). Until more is known about the emersion tolerances and adaptive specializations of many more cyprinodontiform species, no firm conclusions are possible.

Within the Rivulidae, the evolutionary picture appears clearer. Amphibious habits are documented in four genera and in two subfamilies, including the basal Kryptolebiatinae [Fig.

2.2(a)], and anecdotes of emersion in Rivulus species are widespread (Scheel, 1990; Huber,

1992). Thus, it appears probable that amphibiousness is a trait that existed in the earliest rivulids.

Considering that Nothobranchiidae, the sister family to Rivulidae, also contains many amphibious species, this origin of amphibiousness might even predate the nothobranchiid–rivulid split. Why then are there so many fully aquatic species in these families? One possibility is that

22 jack-of-all-trades species able to survive in and out of water may be competitively disadvantaged in aquatic habitats. For example, both locomotion (Gibb et al., 2013) and respiration (Turko et al., 2012) in amphibious fishes appear to require morphological trade-offs that may result in these species being outperformed in water by fully aquatic fishes. Thus, the potential cost of being amphibious in environments where emersion is unnecessary may have driven the reversion to aquatic habits in many rivulids. Annual fishes within these groups may represent an extreme end of this continuum, in which all energy is allocated to rapid growth and reproduction.

Although further work is required to clarify the evolution of amphibious and annual natural histories in the Nothobranchiidae and Rivulidae, this system appears to have great potential for studying the life-history trade-offs required to colonize variable and ephemeral habitats (Edward

& Chapman, 2011).

In their report demonstrating air-breathing ability by emersed F. heteroclitus, Halpin &

Martin (1999) suggested that the common ancestor of all Fundulus species was amphibious. This conclusion was based on the identification of three amphibious species, F. heteroclitus, F. majalis and F. nottii, which are divided by the deepest fork in the phylogeny of Fundulus [Fig.

2.2(b); Ghedotti & Davis, 2013]. Five additional fundulid species have also been reported to emerse (Kushlan, 1973; Baylis, 1982; Graham, 1997), one in the sister genus to Fundulus (L. parva), which further supports this idea and suggests that emersion may be a basal trait in the family Fundulidae, rather than only present in Fundulus. As is seen in the Rivulidae, however, there are many fully aquatic species in the Fundulidae. Therefore, an alternative hypothesis is that amphibious habits have evolved repeatedly in this family. The common ancestor of the subgenus Fundulus (which contains five amphibious species) may have evolved the ability to emerse, while the more distantly related amphibious species F. nottii, F. luciae and L. parva may

23 represent two or three additional independent origins of emersion ability [Fig. 2.2(b)]. Overall, it appears probable that emersion tolerance is a trait ancestral to the family Fundulidae (Halpin &

Martin, 1999) because it is probably more difficult to evolve a complex ability, such as amphibiousness, than to lose it (Murphy & Collier, 1997).

The majority of amphibious cyprinodontiforms are found in the Rivulidae and

Fundulidae, but members of the Nothobranchiidae, Cyprinodontidae, Anablepidae and

Poeciliidae will also emerse. There are probably many additional cyprinodontiform species that emerse as part of their natural history, but there are as yet no formal reports of these behaviours.

The evolutionary origins of amphibious habits within this diverse group of fishes are unclear and merit additional study. Phylogenetically broad studies of both the environmental factors that cause amphibious cyprinodontiforms to emerse and the physiological mechanisms they use will be valuable for discovering common themes between species.

2.3 Behaviour and ecology

Amphibious fishes emerse for many reasons and exhibit a variety of behaviours while out of water. Sayer & Davenport (1991) reviewed the factors driving amphibious fishes to emerse and divided them broadly into abiotic factors (e.g. habitat drying and poor water quality) and biotic factors (e.g. competition, predation and feeding). Almost all of these factors can trigger emersion in cyprinodontiforms. One recurring reason for emersion in many amphibious fishes is reproduction (Sayer & Davenport, 1991; Graham, 1997). Terrestrial egg laying is common in both the Rivulidae and Fundulidae (Huber, 1992). Depositing eggs in air has many advantages for fishes, including increasing the availability of oxygen and offering protection from aquatic predators, but these developing embryos must overcome other challenges such as preventing

24 desiccation. Recent reviews (Podrabsky et al., 2010; Martin & Carter, 2013; Polacik &

Podrabsky, 2015) do an excellent job summarizing the current state of knowledge in the field.

2.3.1 Emersion stimuli

Habitat drying is perhaps the most common environmental factor that drives amphibious cyprinodontiforms to emerse. When the water evaporates from salt marshes or inland lagoons, C. dearborni is able to survive in thin films of water between pebbles or in algae for several days

(Kristensen, 1970). Similarly, J. floridae is capable of surviving in damp mud if water levels are low (Baber et al., 2002), and F. confluentus can tolerate at least 24 h out of water during dry periods in Floridian cypress and willow swamps (Kushlan, 1973). Instead of remaining during low tide, groups of F. majalis stranded in tide pools will travel several meters overland to return to the ocean (Mast, 1915). At the beginning of the dry season in the Amazon, the rivulid A. derhami will move overland to congregate in the deepest pools of water (Lüling, 1971). The closely related A. micropus uses a different strategy to tolerate the dry season, retreating to mud and leaf litter at the bottom of ponds where it can survive for at least 3 months (Pazin et al.,

2006). Leaf litter is also a common terrestrial refuge for K. marmoratus, and the amount of leaf litter in a habitat may be a significant predictor of abundances of this species in Florida (Huehner et al., 1985; Richards et al., 2011). During emersion in the field, K. marmoratus will also hide in damp mud, under coconuts, in discarded beer cans and even within termite tunnels in rotting logs

[Fig. 2.3(a); Taylor et al., 2008]. This log packing behaviour has been documented in both

U.S.A. and Belize populations. Many fish are able to occupy a single log; more than 100 individuals were recorded from a 1.5m long log (9 cm in diameter) at Peter Douglas Cay, Belize, and 57 fish were collected from a similarly sized log at Big Pine Key, Florida, U.S.A. (Taylor et

25 al., 2008). The interrupted dartfish Parioglossus interruptus Suzuki & Senou 1994, a perciform from Indonesia, has independently evolved a similar habit of hiding in teredinid tunnels in rotten mangrove logs during emersion (Hendy et al., 2013). The logs inhabited by P. interruptus were up to 6.5°C cooler than nearby tide pools due to evaporative cooling and may thus provide a thermal refuge for emersed fish; mangrove logs in the New World may play a similar role for K. marmoratus.

Poor water quality is another common trigger for emersion. Under conditions of aquatic hypoxia, many fishes use aquatic surface respiration to obtain oxygen from the thin and relatively oxygen-rich layer of water at the surface (Kramer & McClure, 1982). Some amphibious cyprinodontiforms go a step further and emerse in response to aquatic stressors. For example, F. heteroclitus may emerse under hypoxic conditions and then switch to aerial respiration (Halpin & Martin, 1999). Abel et al. (1987) were the first to test whether hypoxia

- caused emersion in K. marmoratus, but used a moderate level of dissolved oxygen (c. 2 mg O2 l

1 -1 ) and concluded that the fish were unresponsive. Extreme hypoxia (<0.4 mg O2 l ), however, stimulated emersion in K. marmoratus (Regan et al., 2011). High levels of hydrogen sulphide, common in the mangrove swamps typically inhabited by K. marmoratus (Taylor, 2012), also cause emersion (Abel et al., 1987). Receding water levels led to increased concentrations of hydrogen sulphide, and Abel et al. (1987) speculated that this is the cue used by K. marmoratus to leave mangrove pools and find new bodies of water. Other acute abiotic changes, such as fluctuations in temperature, salinity or pH have also been hypothesized to elicit emersion (Sayer

& Davenport, 1991). K. marmoratus will emerse when water temperatures decrease from 25 to

20°C (Huehner et al., 1985). Increased water temperatures (c. 36°C) have also recently been found to induce emersion by K. marmoratus, which may then be able to benefit from evaporative

26 cooling in air (D. Gibson, E. Sylvester, A. Turko, G. Tattersall & P. Wright, unpubl. data).

Additional data indicates that aquatic acidosis resulting from either hypercapnia or the addition of dilute HCl also stimulates emersion by K. marmoratus (Robertson et al. 2015). In mangrove forest pools and crab burrows inhabited by K. marmoratus, many water quality variables (e.g. oxygen, carbon dioxide, hydrogen sulphide, temperature, ammonia and pH) fluctuate on a daily basis (Ellison et al., 2012). A systematic study of the behavioural responses of fishes to the isolated effects of each of these potential stimuli, as well as their interactive effects on behaviour, is warranted. In addition, nothing is known about the potential interactions between abiotic and biotic factors. For example, would emersion in response to competition or aggression occur more readily in hypoxic or hydrogen sulphide-rich habitats? Finally, only the acute emersion responses of amphibious cyprinodontiforms have been investigated thus far. Future experiments should investigate if chronic but less severely degraded water quality also leads to increased rates of emersion.

Emersion can be used to escape competition, aggression and predation by other fishes.

Intraspecific aggression has been shown to cause emersion in K. marmoratus under laboratory conditions (Taylor, 1990), and these fish will also emerse in response to a human disturbance

(submerging a hand underwater) in crab burrows in the field (Taylor, 1988). The closely related

K. sepia will also emerse in response to disturbance by humans (Vermeulen & Hrbek, 2005), as will the more distantly related A. ahli (Scheel, 1990). African killifish in the Aphyosemion subgenus Diapteron (A. fulgens, A. cyanostictum and A. georgiae) have been seen leaping from water after losing an intraspecific aggressive contest; this behaviour occurs both in the wild and in aquaria (Brosset & Lachaise, 1995). Emersion in response to the threat of predation is commonly reported. Cynodonichthys brunneus escapes predation from species of Hoplias by

27 jumping out of the water (Abel et al., 1987), and A. hartii also emerse in response to a predator threat (Seghers, 1978). Goodyear (1970) reported that F. nottii will leap out of water and remain on the bank of a pond for several minutes to escape predation by largemouth bass Micropterus salmoides (Lacépède 1802). Similarly, G. affinis and L. parva will jump onto water lilies to escape predatory bluegill Lepomis macrochirus Rafinesque 1819 (Baylis, 1982). Emersion is, however, not a foolproof strategy. Baylis (1982) noted that small fish flipping across a lily pad are sometimes spotted from below by the predatory L. macrochirus and consumed lily and all.

In addition to leaving the water to avoid predation, some cyprinodontiforms emerse to capture prey. Huehner et al. (1985) reported that K. marmoratus emerse to capture termites on land. According to Davis et al. (1990), K. marmoratus are capable of pursuing and capturing prey on land by flipping or leaping, but always return to water to consume their meal. In captivity, K. marmoratus leap 11 cm in the air to catch terrestrial prey, an impressive feat for a fish that is typically 2–4 cm in length (Taylor, 1990). Impressive jumping ability is also exhibited by A. hartii, which can jump 14 cm vertically to capture ants or other insects (Seghers,

1978). Pronko et al. (2013) recently reported that K. marmoratus will employ three distinct locomotory behaviours, ‘pounces’, ‘squiggles’ and ‘launches’, to leave water and capture prey presented in a laboratory environment. Launches are driven by a swimming burst that propels the fish out of water, while pounces are a shorter-distance movement in which the fish curls into a C- shape and rapidly extends, sliding its ventral aspect over the substratum. Squiggles are oscillatory movements used to travel over land without breaking contact with the substratum. A similar behaviour to the launch of K. marmoratus is used by A. anableps in the laboratory to capture terrestrial insects near the water’s edge (Kushner et al., 2009). In the wild, A. anableps

28 are often found emersed for several hours at low tide during which time they feed on small invertebrates (Tee-Van, 1922; Zahl et al., 1977; Brenner & Krumme, 2007).

The ability to leave water and travel over land allows amphibious cyprinodontiforms to colonize new habitats unavailable to fully aquatic species. Several species, including F. majalis,

A. micropus and L. kirovskyi, have been reported to move over land into new aquatic habitats

(Mast, 1915; Espírito-Santo et al., 2013). These terrestrial sojourns can be brief, as in the case of an unidentified species of Poecilia observed briefly ‘flipping’ across the top of a sandbar to reach an upstream segment of a creek (Lefebvre & Spahn, 1987) or F. luciae, which moves overmud between shallow pools (Byrne, 1978). At the other end of the spectrum, A. hartii can cover 53 cm in a single jump and move 363 cm in 5 min, allowing these fish to disperse widely

(Seghers, 1978). Anablepsoides hartii frequently move between pools of water after rainfall, thus rapidly occupying open habitats. In Trinidad, A. hartii colonized a newly dug pit within 2 months, even though the pit was at least 20 m from the nearest temporary water body and 500 m from the nearest permanent stream (Jordan, 1923). To determine the ability of K. marmoratus to colonize new habitats, Taylor (1990) dug artificial holes that simulated burrows of the land crab

Cardisoma guanhumi, the typical habitat of this species. When the burrows were sampled 6 months later, K. marmoratus were found within them. These fish have also been observed in puddles that form in the middle of roadways after rainfall, suggesting that new or ephemeral habitats can be rapidly colonized (Davis et al., 1990). In a laboratory mesocosm, K. marmoratus moved over land between artificial burrows up to six times per day, providing further evidence for the ability of these fish to exploit new habitats (Taylor, 1990). Little is known about the movement patterns of individual fish in the wild, in large part, because performing mark– recapture studies of such small fish in the mangrove habitat is quite difficult.

While many cyprinodontiforms emerse to reproduce, disperse or avoid deteriorating environmental conditions, some species emerse even when there is no discernable stimulus.

Sayer & Davenport (1991) urge caution before concluding that emersions are voluntary, as subtle biological or abiotic factors may influence behavioural decisions. Therefore, while the emersion behaviours in the following cases may appear voluntary, the cues for emersion are best summarized as unknown. Tee-Van (1922) described apparently voluntary emersion behaviour in the rivulid A. stagnatus, which would often jump out of the water in holding aquariums and adhere to leaves or stones. One individual, lying alert, remained emersed for 9 days before returning to water. Similarly, A. beniensis is frequently inexplicably found emersed and on blades of grass or algae filaments above well-oxygenated water (Lüling, 1971). Aphyosemion franzwerneri remain emersed among tree roots overhanging small pools by day (Scheel, 1990).

Fundulopanchax gardneri has been observed emersed in aquaria on top of floating plants or if provided with a cradle of filter floss that sits just above the water (M. Davies, pers. comm.). The undescribed Aphyosemion sp. Lobaye, a close relative of A. elegans (Collier, 2006), will similarly emerse on top of floating vegetation [Fig. 2.3(b); D. Taylor, pers. comm.]. Kryptolebias marmoratus voluntarily leave water under both laboratory and field conditions. Turko et al.

(2011) video-recorded fish for 7 days in clean brackish water in the laboratory to minimize abiotic cues for emersion and analysed the time each spent voluntarily out of water [Fig. 2.3(c)].

Some individuals never emersed, but others spent up to 78% of the 7 day period out of water. In

Belize, a remotely operated video camera was placed inside a crab burrow to record emersion patterns of K. marmoratus under natural conditions [Video 2.S1]. Emersion by an unknown number of fish was common in this habitat, in 30 min of filming 11 instances of emersion were recorded, ranging in duration from 6 s to 7 min (mean ± s.e. duration of 2.8 ± 0.7 min). These

30 emersions were shorter than those recorded in the laboratory by Turko et al. (2011), which averaged 61 min in one population and 20 min in another. At several points in the video, the movement of predatory mangrove tree crabs Goniopsis cruentata appear to startle emersed K. marmoratus and triggered their return to the water, which may explain why the emersion durations were shorter in the field relative to undisturbed laboratory conditions. Interestingly, movements by another, larger species of detritivorous crab (C. guanhumi, the burrow-excavating species) did not startle emersed K. marmoratus in the same way. These emersions occurred without any clear biotic stimulus (e.g. no observable chasing by conspecifics, risk of predation or terrestrial feeding opportunities), although abiotic factors such as hypoxia or elevated hydrogen sulphide probably play a role. Perhaps voluntary emersion allows some species to exploit oxygen-rich air and minimize energy expenditure by not swimming and reducing ventilation of the gills.

2.3.2 Terrestrial locomotion

Locomotion in terrestrial environments is challenging for most fishes, which have body forms, support structures and sensory systems that have evolved to facilitate life in water (Sayer,

2005). Many fully aquatic fishes, such as M. salmoides, are only capable of producing a chaotic thrashing movement that results in vertical but not in horizontal displacement (the kinematics of terrestrial locomotion in fishes were recently reviewed by Gibb et al., 2013). Amphibious cyprinodontiforms, however, are able to navigate out of water in a controlled and directional manner despite their lack of obvious morphological specializations for terrestrial locomotion

(Ashley-Ross et al., 2014). Body orientation on a slope can be sensed by emersed G. affinis, which always roll to return to water if they are positioned perpendicularly to the incline, but

31 often jump or flip back to water if they are pointed up or downhill (Boumis et al., 2014). At low tide, F. majalis emerse from tide pools and jump in the direction of the ocean, even if this requires moving uphill or into a breeze (Mast, 1915). The navigational cues used by F. majalis are unknown, but Goodyear (1970) has demonstrated that F. nottii use the sun to navigate while emersed. In these experiments, fish were captured from four different locations around a pond, at the four cardinal points of the compass. The fish were then placed, without water, on a flat table and their direction of travel was recorded. Goodyear (1970) found that, on sunny days, F. nottii would overwhelmingly travel in the direction that would have led them to the water’s edge based on their location of capture from the pond, e.g. fish captured from the north shore travelled south, which would have returned them to the water had they been emersed on the northern bank. On overcast days, however, fish moved randomly. These results suggest that emersed F. nottii use the sun to navigate, and also that these fish may emerse frequently enough prior to being captured for experimentation to learn the local environment and optimal direction of travel to return to water. Kryptolebias marmoratus also appear capable of learning and navigating terrestrial landscapes. In the field, emersed K. marmoratus tend to jump in the direction of crab burrows when disturbed or accidentally dropped by a collector (Huehner et al., 1985). Local landscapes can also be learned under laboratory conditions (Taylor, 1990; Pronko et al., 2013).

Taylor (1990) built a simulated mangrove swamp aquarium that contained two replica crab burrows separated by 8 cm of emersed substratum in order to film a fish moving between the burrows. When a barrier was installed to force the fish to take a longer (23 cm) and indirect path to switch between burrows, the new route was learned within 24 h. A new barrier was then inserted which completely separated the two burrows except for a small hole (1 cm diameter) drilled at ground level; the fish learned the location of this hole within 5 days and used it to move

32 between burrows. When the small hole was subsequently plugged, the K. marmoratus would repeatedly orient towards the location of the former hole and try to cross to the other burrow.

Despite their ability to move overland to inhabit unoccupied pools of water, many amphibious rivulids are endemic to only small regions of South America (Huber, 1992). In contrast, K. marmoratus are found in mangrove habitats throughout the tropical west Atlantic, from southern U.S.A. to northern Brazil, although the species designation in Brazil is uncertain

(Costa, 2011b; Taylor, 2012). Why are K. marmoratus so widespread when other rivulid species are not? One possibility is that logs invaded by fish during periods of low water are spread throughout the western Atlantic Ocean by storms and ocean currents (Taylor et al., 2008; Taylor,

2012). Another possible scenario is the dispersal of embryos, which are highly adhesive and likely candidates for transport on floating mangrove leaves or detritus. While this type of waif dispersal is probably rare, as suggested by strong population structure among Belizean,

Bahamian and American populations, a few specimens of K. marmoratus genetically similar to

Belizean populations have been discovered in Florida (Tatarenkov et al., 2007). Log or debris transport aside, two other derived traits in K. marmoratus are also required for their observed wide distribution. Unlike most other rivulids, K. marmoratus are euryhaline (Costa, 1998) and capable of tolerating fresh water and salinities as high as 114 (King et al., 1989; LeBlanc et al.,

2010). This salinity tolerance not only allowed the colonization of brackish mangrove swamps, but also would enable K. marmoratus to survive exposure to seawater during dispersal voyages

(Costa et al., 2010; Taylor, 2012). Second, K. marmoratus is one of the only two (the other being the sister species Kryptolebias hermaphroditus Costa 2011) known self-fertilizing hermaphroditic vertebrates (Harrington, 1961; Tatarenkov et al., 2009; Costa et al., 2010; Costa,

2011b). The ability to self-fertilize would allow a single K. marmoratus to found a new population after long-range dispersal, and would also be advantageous even after shorter movements over land to unoccupied crab burrows or pools (Tatarenkov et al., 2009; Avise &

Tatarenkov, 2012).

Overall, amphibious cyprinodontiforms emerse to reproduce, escape poor or deteriorating habitats or to exploit new niches, and some also appear to emerse voluntarily. Emersions range in duration from seconds to months and may involve actively moving about on land or simply retreating to a damp crevice to wait for water to return. A suite of morphological and physiological adaptations permit these terrestrial sojourns, and these are reviewed in the next section.

2.4 Physiology

To survive both in and out of water, amphibious fishes require adaptations that facilitate the maintenance of water balance, gas exchange, nitrogenous waste excretion and ionoregulation

(Sayer, 2005). In fully aquatic fishes, the gills are the main site of these physiological processes, but maintaining gill functionality during emersion is problematic. Gill tissues must remain moist, but without the buoyant support of water the secondary lamellae of the gills may collapse and coalesce thus reducing their effective surface area. Many amphibious fishes have evolved specialized gill morphologies or air-breathing organs, but these strategies do not appear to be used by cyprinodontiforms. Instead, during emersion, the cutaneous epithelium appears to take on many of the roles typically assigned to the gills (Wright, 2012; Glover et al., 2013).

Unfortunately, reports of amphibious cyprinodontiform physiology are limited to studies of K. marmoratus and one investigation of F. heteroclitus, so while other amphibious species

34 conceivably also use their skin during emersion, the role of the cutaneous surface in these fishes is unknown.

The skin may be especially important to K. marmoratus during emersion, as the gills are remodelled during terrestrial acclimation and surface area is reduced. Ong et al. (2007) demonstrated that 1week of emersion induces the proliferation of a group of cells in the spaces between the secondary lamellae in K. marmoratus but recedes within 7 days of returning to water. This group of cells, the inter-lamellar cell mass, was also enlarged in K. marmoratus that voluntary emersed frequently over a 7 day recording period (Turko et al., 2011). Aquatic respiratory function was impaired in K. marmoratus with reduced gill surface area after they returned to water but before the inter-lamellar cell mass regressed (Turko et al., 2012). The cost imposed by an enlarged inter-lamellar cell mass in water suggests that gill remodelling provides some kind of benefit during emersion, although the nature of the benefit is not yet clear. The inter-lamellar cell mass may protect and support the delicate secondary lamellae from collapsing and coalescing on land when they are not supported by water, or the reduced gill surface area resulting from an enlarged cell mass may minimize water loss (Ong et al., 2007). Gill remodelling may also reduce the surface area available for parasite attachment (Nilsson et al.,

2012). Gill surface area is correlated with parasite loads in the gills of Carassius carassius (L.

1758) (Nilsson et al., 2012) and Liza ramada (Risso 1826) (Caltran & Silan, 1996), and the gills of wild K. marmoratus sometimes contain bacterial cysts (Ellison et al., 2011).

Only a few studies have investigated the morphological specializations that permit cutaneous respiration in cyprinodontiforms. Grizzle & Thiyagarajah (1987) described capillaries in K. marmoratus that reach within 1 μm of the epidermal surface. These capillaries are densest anteriorly and dorsally, but blood vessels are also present in the fins. Recent use of

35 immunohistochemical labelling of the protein CD31, a common marker for angiogenesis, has revealed that these surfaces receive increased blood flow during emersion (Fig. 2.4; Turko et al.,

2014). The fins, which comprise c. 40% of the total cutaneous surface area of K. marmoratus, also demonstrated significant angiogenesis in terrestrially acclimated fish (Cooper et al., 2012).

Unexpectedly, Turko et al. (2014) noticed that increased CD31 fluorescence was not limited to the cutaneous surface, blood flow to the epithelium lining the mouth and opercula also increased during emersion. Further experiments demonstrated that K. marmoratus gulp air approximately five times per hour over several days after emersion, suggesting that the bucco-opercular chamber and possibly the gills may supplement cutaneous respiration. The use of cutaneous capillaries for gas exchange has never been directly investigated in F. heteroclitus, but Connolly

(1925) demonstrated that the head and caudal fin of these fish are well supplied with blood.

Vasodilation of these cutaneous blood vessels occurred when aquatic F. heteroclitus were placed on a red background, turning the fish slightly pink and improving camouflage. Whether similar changes in blood flow occur in response to emersion remains an open question.

Aerial gas exchange rates in both K. marmoratus and F. heteroclitus have been measured in several studies. Halpin & Martin (1999) measured oxygen consumption by F. heteroclitus in and out of water. These studies found that emersed fish reduced oxygen consumption by c. 75% relative to their rate of consumption in water, but despite these low rates of oxygen consumption no oxygen debt was evident upon returning to water after 1 h of emersion. Excretion of carbon dioxide was probably not problematic for emersed F. heteroclitus either, as respiratory exchange ratios (carbon dioxide released: oxygen consumed) were determined to be between 0.7 and 1.0 depending on temperature. To date, the only published report of aerial oxygen consumption by

K. marmoratus suggests that fish decrease oxygen uptake after 1 day of emersion (Abel et al.,

1987). Carbon dioxide excretion by emersed K. marmoratus, however, occurs at rates similar to those in water immediately after emersion and at significantly higher rates (c. 40%) after 3 days of terrestrial acclimation (Ong et al., 2007). Similarly, the activities of several enzymes involved in amino-acid and mitochondrial oxidative metabolism are unchanged or increased during emersion (Frick & Wright, 2002). How are these contradictory results explained? Measuring oxygen consumption by very small animals, <100 mg in the case of K. marmoratus, in relatively oxygen-rich atmospheric air (compared with water) is notoriously difficult (Ong et al., 2007).

With the recent availability of optode technology that allows precise sensing of oxygen concentrations, it should now be possible to directly measure oxygen consumption in and out of water in K. marmoratus and other minute amphibious fishes.

While there is some information on gas exchange in emersed F. heteroclitus (Halpin &

Martin, 1999) and K. marmoratus (Ong et al., 2007), there are no data available on partial pressures within the blood. Carbon dioxide is relatively insoluble in air compared with water, and thus tends to accumulate in the blood of air-breathing animals (Rahn, 1966; Ultsch, 1987).

An increased partial pressure of carbon dioxide in the blood could potentially impair oxygen uptake by reducing both the oxygen binding affinity (via the Bohr effect) and carrying capacity

(via the Root effect) of haemoglobin, thus resulting in hypoxemia (Graham, 1997; Shartau &

Brauner, 2014). Considering that cyprinodontiforms have probably evolved under conditions of aquatic hypoxia, Turko et al. (2014) tested the hypothesis that K. marmoratus use a co-opted hypoxia response to maintain oxygen transport in the face of carbon dioxide accumulation during emersion. This hypothesis was soundly rejected, terrestrially acclimated fish increased haemoglobin-oxygen binding affinity, while fish acclimated to hypoxia in water increased the number of circulating red blood cells. Increased haemoglobin-oxygen binding affinity in

37 terrestrially acclimated K. marmoratus probably offsets the decrease in affinity resulting from the Bohr effect, maintaining the balance between oxygen uptake and oxygen delivery. If not hypoxia, what cue initiates a change in haemoglobin affinity in K. marmoratus? One possibility is that emersed fish respond to increased partial pressures of carbon dioxide in the blood.

Recent evidence suggests that the skin of K. marmoratus may be used for oxygen sensing as well as gas exchange. In fishes, innervated chemosensory cells called neuroepithelial cells are responsible for sensing oxygen concentrations and initiating cardiorespiratory responses to hypoxia (Zachar & Jonz, 2012; Jonz et al., 2015). Neuroepithelial cells are typically not only found in the gills of adult fishes, but are also present on the cutaneous surface of larval fishes before the gills are developed. The skin of adult K. marmoratus has recently been reported to contain an abundance of neuroepithelial cells (45% more cells in the skin than the gills) and these cells respond to aquatic hypoxia (Regan et al., 2011). Whether these cutaneous oxygen sensors are also used during emersion has yet to be investigated. The situation is paradoxical, as aerial environments are rich in oxygen and these conditions are thought to reduce the need for oxygen sensing. Acclimation to hyperoxia, for example, reduced the densities of neuroepithelial cells in zebrafish Danio rerio (Hamilton 1822) (Vulesevic et al., 2006). Future studies are required to determine the functions and benefits of cutaneous neuroepithelial cells during emersion.

In addition to being the main location for respiration, the cutaneous epithelium of K. marmoratus is used for nitrogenous waste excretion during emersion. Fishes typically excrete waste nitrogen as ammonia by diffusion across the gills, but this is difficult in the absence of water. Uniquely among amphibious fishes, K. marmoratus excrete a relatively large proportion

(40%) of nitrogenous waste as ammonia gas via volatilization (Frick & Wright, 2002; Wright,

2012). Emersed fish volatilize ammonia gas within 24 h of leaving water, and continue to use this strategy over at least 11 days of terrestrial acclimation (Litwiller et al., 2006). This strategy depends on the presence of ammonia-transporting Rhesus glycoproteins in the skin, as well as the ability for fish to finely regulate the pH of the skin surface (Hung et al., 2007; Wright

&Wood, 2009; Cooper et al., 2013). It is unknown whether other amphibious cyprinodontiforms also possess these specializations to permit ammonia volatilization during emersion, or whether alternative strategies are used, such as suppressing amino acid catabolism, detoxifying ammonia via urea or glutamine synthesis or enhancing cellular tolerance to high ammonia concentrations

(Ip et al., 2004).

Maintaining water and ion balance during air exposure is one of the greatest challenges for amphibious fishes (Sayer, 2005; Takei, 2015). Many species rely on a thick skin or mucous coat during emersion, but for fish such as K. marmoratus the multi-functional nature of the cutaneous epithelium may preclude these methods of retaining water (LeBlanc et al., 2010;

Turko et al., 2011). Nonetheless, emersed K. marmoratus are able to maintain water balance.

Terrestrially acclimated fish, which are thought to be incapable of feeding while out of water, lose no more mass than fasted fish in water and the proportional body water content may actually increase as fatty energy stores are utilized (Frick &Wright, 2002; Litwiller et al., 2006; LeBlanc et al., 2010). Embryos of F. heteroclitus that develop terrestrially downregulate aquaporin expression, possibly to conserve water (Tingaud-Sequeira et al., 2009). The ability of adult cyprinodontiforms to similarly regulate cutaneous water permeability at the protein level has yet to be investigated.

Ion balance is largely maintained during emersion in K. marmoratus by a large cutaneous population of ionocytes (LeBlanc et al., 2010). These cells are typically found in the gills of

39 adult fishes, but K. marmoratus possess approximately equal numbers in the skin and gills.

Cutaneous ionocytes are also known in some other cyprinodontiforms. Ionocytes are present in the opercular epithelium of F. heteroclitus and in the skin of Poecilia reticulata (Schwerdtfeger

& Bereiter-Hahn, 1978; Hiroi & McCormick, 2012; Glover et al., 2013). Given the diversity of amphibious cyprinodontiforms, it would not be surprising if cutaneous ionocytes are found in other species as well.

While many advances have been made in the understanding of the physiological adaptations used by amphibious cyprinodontiforms to tolerate life out of water, many questions remain unanswered. One avenue of research that appears particularly promising is to utilize phylogenetic and inter-population comparisons to understand the particular traits that influence emersion tolerance. For example, why can K. marmoratus and A. micropus survive for many months out of water, while congeners do not emerse at all? Or perhaps most species within the genera Kryptolebias and Anablepsoides are in fact amphibious, but natural history knowledge is lacking. Intraspecific comparisons across populations are also worthwhile. Life-history traits are known to vary among populations of A. hartii (Walsh & Reznick, 2010; Oufiero et al., 2011;

Walter et al., 2011), F. heteroclitus (Burnett et al., 2007; Whitehead et al., 2011) and K. marmoratus (Grageda et al., 2005; Earley et al., 2012). Determining if there is similar variation in amphibious ability and addressing the physiological mechanisms that underlie this variation are promising future directions for research.

In conclusion, the order Cyprinodontiformes contains a diversity of amphibious species and more are sure to be discovered. For example, 97 valid species of Rivulus were described 20 years ago by Huber (1992). Since then, at least 54 new species have been described within the various genera that formerly comprised Rivulus (Costa, 2011a). Behavioural observations and

40 natural history data are lacking for many of these species, but some will almost certainly prove to be amphibious. Is there some basal cyprinodontiform trait that has allowed so many species to invade land? Perhaps amphibiousness is a key evolutionary innovation that has enabled cyprinodontiforms to invade new ecological niches, as in the cichlids where the evolution of pharyngeal jaws is thought to underlie their dramatic radiation in Africa (Liem, 1973). The ability to leave water has certainly allowed cyprinodontiforms to take advantage of several ecological niches not available to fully aquatic fishes. Emersion can be used to exploit ephemeral bodies of water, reproduce in oxygen-rich terrestrial environments safe from aquatic predators, tolerate deteriorating water conditions, feed on terrestrial prey and avoid aggressive or predatory interactions in water. Understanding the various environmental factors that drive emersion in cyprinodontiforms may even provide insight into why the earliest tetrapods made the transition to a terrestrial existence.

2.5 References

Abel, D. C., Koenig, C. C. & Davis,W. P. (1987). Emersion in the mangrove forest fish Rivulus

marmoratus, a unique response to hydrogen sulfide. Environmental Biology of Fishes 18,

67–72.

Ashley-Ross, M. A., Perlman, B. M., Gibb, A. C. & Long, J. H. Jr. (2014). Jumping sans legs:

does elastic energy storage by the vertebral column power terrestrial jumps in bony fishes?

Zoology 117, 7–18. doi: 10.1016/j.zool.2013.10.005

Avise, J. C. & Tatarenkov, A. (2012). Allard’s argument versus Baker’s contention for the

adaptive significance of selfing in a hermaphroditic fish. Proceedings of the National

Academy of Sciences of the United States of America 109, 18862–18867.

doi:10.1073/pnas.1217202109

Baber, M. J., Childers, D. L., Babbitt, K. J. & Anderson, D. H. (2002). Controls on fish

distribution and abundance in temporary wetlands. Canadian Journal of Fisheries and

Aquatic Sciences 59, 1441–1450. doi: 10.1139/F02-116

Baylis, J. R. (1982). Unusual escape response by two cyprinodontiform fishes, and a bluegill

predator’s counter-strategy. Copeia 1982, 455–457.

Boumis, R. J., Ferry, L. A., Pace, C. M. & Gibb, A. C. (2014). Heads or tails: do stranded fish

(mosquitofish, Gambusia affinis) know where they are on a slope and how to return to the

water? PLoS One 9, e104569. doi: 10.1371/journal.pone.0104569

Brenner, M. & Krumme, U. (2007). Tidal migration and patterns in feeding of the four-eyed fish

Anableps anableps L. in a north Brazilian mangrove. Journal of Fish Biology 70, 406–427.

doi: 10.1111/j.1095-8649.2007.01313.x

Brosset, A. & Lachaise, D. (1995). Evolution as a lottery conflicting with evolution via sexual

selection in African rainforest-dwelling killifishes (Cyprinodontidae, Rivulinae, Diapteron).

Evolutionary Biology 28, 217–264.

Burnett, K. G., Bain, L. J., Baldwin, W. S., Callard, G. V., Cohen, S., Di Giulio, R. T., Evans, D.

H., Gómez-Chiarri, M., Hahn, M. E., Hoover, C. A., Karchner, S. I., Katoh, F., MacLatchy,

D. L., Marshall, W. S., Meyer, J. N., Nacci, D. E., Oleksiak, M. F., Rees, B. B., Singer, T.

D., Stegeman, J. J., Towle, D. W., Van Veld, P. A., Vogelbein, W. K., Whitehead, A., Winn,

R. N. & Crawford, D. L. (2007). Fundulus as the premier teleost model in environmental

biology: opportunities for new insights using genomics. Comparative Biochemistry and

Physiology D 2, 257–286.

Byrne, D. M. (1978). Life history of the spotfin killifish, Fundulus luciae (Pisces:

Cyprinodontidae) in Fox Creek Marsh, Virginia. Estuaries 1, 211–227. Caltran, H. & Silan,

P. (1996). Gill filaments of Liza ramada, a biotope for ectoparasites: surface area acquisition

using image analysis and growth models. Journal of Fish Biology 49, 1267–1279.

Clack, J. A. (2002). Gaining Ground: The Origin and Evolution of Tetrapods. Bloomington, IN:

Indiana University Press.

Collier, G. E. (2006). The genus Aphyosemion: taxonomic history and molecular phylogeny.

Journal of the American Killifish Association 39, 147–168.

Connolly, C. J. (1925). Adaptive changes in shades and color of Fundulus. Biological Bulletin

48, 56–77.

Cooper, C. A., Litwiller, S. L., Murrant, C. L. & Wright, P. A. (2012). Cutaneous vasoregulation

during short- and long-term aerial acclimation in the amphibious mangrove rivulus,

Kryptolebias marmoratus. Comparative Biochemistry and Physiology B 161, 268–274. doi:

10.1016/j.cbpb.2011.12.001

Cooper, C. A., Wilson, J. M. & Wright, P. A. (2013). Marine, freshwater and aerially acclimated

mangrove rivulus (Kryptolebias marmoratus) use different strategies for cutaneous ammonia

excretion. American Journal of Physiology. Regulatory, Integrative and Comparative

Physiology 304, 599–612. doi: 10.1152/ajpregu.00228.2012

Costa, W. J. E. M. (1998). Phylogeny and classification of Rivulidae revisited: origin and

evolution of annualism and miniaturization in rivulid fishes (Cyprinodontiformes:

Aplocheiloidei). Journal of Comparative Biology 3, 33–92.

Costa, W. J. E. M. (2011a). Phylogenetic position and taxonomic status of Anablepsoides,

Atlantirivulus, Cynodonichthys, Laimosemion and Melanorivulus (Cyprinodontiformes:

Rivulidae). Ichthyological Exploration of Freshwaters 22, 233–249.

Costa, W. J. E. M. (2011b). Identity of Rivulus ocellatus and a new name for a hermaphroditic

species of Kryptolebias from south-eastern Brazil (Cyprinodontiformes: Rivulidae).

Ichthyological Exploration of Freshwaters 22, 185–192.

Costa, W. J. E. M. (2012). The caudal skeleton of extant and fossil cyprinodontiform fishes

(Teleostei: Atherinomorpha): comparative morphology and delimitation of phylogenetic

characters. Vertebrate Zoology 62, 161–180.

Costa, W. J. E. M. (2013). Historical biogeography of aplocheiloid killifishes (Teleostei:

Cyprinodontiformes). Vertebrate Zoology 63, 139–154.

Costa, W. J. E. M., Lima, S.M. Q. & Bartolette, R. (2010). Androdioecy in Kryptolebias killifish

and the evolution of self-fertilizing hermaphroditism. Biological Journal of the Linnean

Society 99, 344–349.

Davis, W. P., Taylor, D. S.&Turner, B. J. (1990). Field observations on the ecology and habits of

the mangrove rivulus (Rivulus marmoratus) in Belize and Florida. Ichthyological

Exploration of Freshwaters 1, 123–134.

Earley, R. L., Hanninen, A. F., Fuller, A., Garcia, M. J. & Lee, E. A. (2012). Phenotypic

plasticity and integration in the mangrove rivulus (Kryptolebias marmoratus): a prospectus.

Integrative and Comparative Biology 52, 814–827. doi: 10.1093/icb/ics118

Edward, D. A. & Chapman, T. (2011). Mechanisms underlying reproductive trade-offs: costs of

reproduction. In Mechanisms of Life History Evolution (Flatt, T. & Heyland, A., eds), pp.

137–152. Toronto, ON: Oxford University Press.

Ellison, A., Cable, J. & Consuegra, S. (2011). Best of both worlds? Association between

outcrossing and parasite loads in a selfing fish. Evolution 65, 3021–3026.

doi:10.1111/j.1558-5646.2011.01354.x

Ellison, A., Wright, P., Taylor, D. S., Cooper, C., Regan, K., Currie, S. & Consuegra, S. (2012).

Environmental diel variation, parasite loads, and local population structuring of a mixed-

mating mangrove fish. Ecology and Evolution 2, 1682–1695. doi: 10.1002/ece3.289

Espírito-Santo, H. M. V., Rodríguez, M. A. & Zuanon, J. (2013). Reproductive strategies of

Amazonian stream fishes and their fine-scale use of habitat are ordered along a hydrological

gradient. Freshwater Biology 58, 2494–2504. doi: 10.1111/fwb.12225

Frick, N. T. & Wright, P. A. (2002). Nitrogen metabolism and excretion in the mangrove killifish

Rivulus marmoratus II. Significant ammonia volatilization in a teleost during air-exposure.

Journal of Experimental Biology 205, 91–100.

Furness, A. I., Reznick, D. N., Springer, M. S. & Meredith, R. W. (2015). Convergent evolution

of alternative developmental trajectories associated with diapause in African and South

American killifish. Proceedings of the Royal Society B 282, 20142189. doi:

10.1098/rspb.2014.2189

Ghedotti, M. J. & Davis, M. P. (2013). Phylogeny, classification, and evolution of salinity

tolerance of the North American topminnows and killifishes, family Fundulidae (Teleostei:

Cyprinodontiformes). Fieldiana Life and Earth Sciences 7, 1–65.

Gibb, A. C., Ashley-Ross, M. A. & Hsieh, S. T. (2013). Thrash, flip, or jump: the behavioral and

functional continuum of terrestrial locomotion in teleost fishes. Integrative and Comparative

Biology 53, 295–306. doi: 10.1093/icb/ict052

Glover, C. N., Bucking, C. & Wood, C. M. (2013). The skin of fish as a transport epithelium: a

review. Journal of Comparative Physiology B 183, 877–891. doi: 10.1007/s00360-013-

0761-4

Goodyear, C. P. (1970). Terrestrial and aquatic orientation in the starhead topminnow, Fundulus

nottii. Science 168, 603–605.

Gordon, M. S., Boetius, I., Evans, D. H., McCarthy, R. & Oglesby, L. C. (1969). Aspects of the

terrestrial life in amphibious fishes. I. Themudskipper, Periophthalmus sobrinus. Journal of

Experimental Biology 50, 141–149.

Grageda, M. V. C., Sakakura, Y., Minamimoto, M. & Hagiwara, A. (2005). Differences in life-

history traits in two clonal strains of the self-fertilizing fish Rivulus marmoratus.

Environmental Biology of Fishes 73, 427–436.

Graham, J. B. (1997). Air-Breathing Fishes: Evolution, Diversity, and Adaptation. San Diego,

CA: Academic Press.

Graham, J. B. & Lee, H. J. (2004). Breathing air in air: in what ways might extant amphibious

fish biology relate to prevailing concepts about early tetrapods, the evolution of vertebrate

air breathing, and the vertebrate land transition? Physiological and Biochemical Zoology 77,

720–731.

Grizzle, J. M. & Thiyagarajah, A. (1987). Skin histology of Rivulus ocellatus marmoratus:

apparent adaptation for aerial respiration. Copeia 1987, 237–240.

Halpin, P. M. & Martin, K. L. M. (1999). Aerial respiration in the salt marsh fish Fundulus

heteroclitus (Fundulidae). Copeia 1999, 743–748.

Harrington, R.W. Jr. (1961). Oviparous hermaphroditic fish with internal self-fertilization.

Science 134, 1749–1750.

Hendy, I.W., Eme, J., Dabruzzi, T. F., Nembhard, R. V., Cragg, S. M. & Bennett,W. A. (2013).

Dartfish use teredinid tunnels in fallenmangrove wood as a low-tide refuge. Marine Ecology

Progress Series 486, 237–245. doi: 10.3354/meps10370

Hiroi, J. & McCormick, S. D. (2012). New insights into gill ionocyte and ion transporter

function in euryhaline and diadromous fish. Respiratory Physiology and Neurobiology 184,

257–268. doi: 10.1016/j.resp. 2012.07.019

Huber, J. H. (1992). Review of Rivulus, Ecobiogeography, Relationships. Paris: Société

Française D’ichtyologie.

Huehner, M. K., Schramm, M. E. & Hens, M. D. (1985). Notes on the behavior and ecology of

the killifish Rivulus marmoratus Poey 1880 (Cyprinodontidae). Florida Scientist 48, 1–6.

Hung, C.Y. C., Tsui, K. N. T.,Wilson, J. M., Nawata, C. M.,Wood, C. M.&Wright, P.A. (2007).

Rhesus glycoprotein gene expression in the mangrove killifish Kryptolebias marmoratus

exposed to elevated environmental ammonia levels and air. Journal of Experimental Biology

210, 2419–2429.

Ip, Y. K., Chew, S. F. & Randall, D. J. (2004). Five tropical fishes, six different strategies to

defend against ammonia toxicity on land. Physiological and Biochemical Zoology 77, 768–

782.

Jonz,M. G., Zachar, P. C., Da Fonte, D. F. &Mierzwa, A. S. (2015). Peripheral chemoreceptors

in fish: a brief history and a look ahead. Comparative Biochemistry and Physiology A 186,

27–38. doi: 10.1016/j.cbpa.2014.09.002

Jordan, D. S. (1923). Habits of the Trinidad Guapin, Rivulus hartii (Boulenger). Copeia 1923,

69–70.

King, J. A. C., Abel, D. C. & DiBona, R. (1989). Effects of salinity on chloride cells in the

euryhaline cyprinodontid fish Rivulus marmoratus. Cell and Tissue Research 257, 367–377.

Kramer, D. L. & McClure, M. (1982). Aquatic surface respiration, a widespread adaptation to

hypoxia in tropical freshwater fishes. Environmental Biology of Fishes 7, 47–55.

Kristensen, I. (1970). Competition in three cyprinodont fish species in the Netherlands Antilles.

Studies on the Fauna of Curacao and Other Caribbean Islands 119, 82–101.

Kushlan, J. A. (1973). Differential responses to drought in two species of Fundulus. Copeia

1973, 808–809.

Kushner, S. A., Gibb, A. C., Arena, A. & Ferry-Graham, L. A. (2009). Four-eyed fish (Anableps

anableps) use the same jaw-opening movements to produce a distinct prey-capture behavior

across environments. Integrative and Comparative Biology 49, e258.

LeBlanc, D. M., Wood, C. M., Fudge, D. S. & Wright, P. A. (2010). A fish out of water: gill and

skin remodeling promotes osmo- and ionoregulation in the mangrove killifish Kryptolebias

marmoratus. Physiological and Biochemical Zoology 83, 932–949. doi: 10.1086/656307

Lefebvre, L. & Spahn, D. (1987). Gray kingbird predation on small fish (Poecilia sp.) crossing a

sandbar. Wilson Bulletin 99, 291–292.

Liem, K. F. (1973). Evolutionary strategies and morphological innovations: cichlid pharyngeal

jaws. Systematic Zoology 22, 425–441.

Litwiller, S. L., O’Donnell, M. J. & Wright, P. A. (2006). Rapid increase in the partial pressure

of NH3 on the cutaneous surface of air-exposed mangrove killifish, Rivulus marmoratus.

Journal of Experimental Biology 209, 1737–1745.

Lüling, K. H. (1971). Ökologische Beobachtungen und Untersuchungen am Biotop des Rivulus

beniensis (Pisces, Cyprinodontidae). Beitrage zur Neotropischen Fauna 6, 163–193.

Martin, K. L. & Carter, A. L. (2013). Brave new propagules: terrestrial embryos in anamniotic

eggs. Integrative and Comparative Biology 53, 233–247. doi: 10.1093/icb/ict018

Mast, S. O. (1915). The behavior of Fundulus, with especial reference to overland escape from

tide-pools and locomotion on land. Journal of Animal Behaviour 5, 341–350.

Murphy, W. J. & Collier, G. E. (1997). A molecular phylogeny for aplocheiloid fishes

(Atherinomorpha, Cyprinodontiformes): the role of vicariance and the origins of annualism.

Molecular Biology and Evolution 14, 790–799.

Nelson, J. S. (2006). Fishes of the world, 4th edn. Hoboken, NJ: John Wiley and Sons.

Nilsson, G. E., Dymowska, A. & Stecyk, J. A. W. (2012). New insights into the plasticity of gill

structure. Respiratory Physiology and Neurobiology 184, 214–222. doi: 10.1016/j.resp.

2012.07.012

Ong, K., Stevens, E. D. & Wright, P. A. (2007). Gill morphology of the mangrove killifish

(Kryptolebias marmoratus) is plastic and changes in response to terrestrial air exposure.

Journal of Experimental Biology 210, 1109–1115.

Oufiero, C. E., Walsh, M. R., Reznick, D. N. & Garland, T. Jr. (2011). Swimming performance

trade-offs across a gradient in community composition in Trinidadian killifish (Rivulus

hartii). Ecology 92, 170–179.

Parenti, L. R. (1981). A phylogenetic and biogeographic analysis of cyprinodontiform fishes

(Teleostei, Atherinomorpha). Bulletin of the American Museum of Natural History 168,

335–557.

Pazin, V. F. V., Magnusson, W. E., Zuanon, J. & Mendonca, F. P. (2006). Fish assemblages in

temporary ponds adjacent to “terra-firme” streams in Central Amazonia. Freshwater Biology

51, 1025–1037. doi: 10.1111/j.1365-2427.2006.01552.x

Podrabsky, J. E., Tingaud-Sequeira, A. & Cerda, J. (2010). Metabolic dormancy and responses to

environmental desiccation in fish embryos. In Dormancy and Resistance in Harsh

Environments (Lubzens, E., Cerda, J. & Clark, M., eds), pp. 203–226. Berlin: Springer

Verlag. doi: 10.1007/978-3-642-12422-8

Polacik, M. & Podrabsky, J. E. (2015). Temporary environments. In Extremophile Fishes

Ecology, Evolution, and Physiology of Teleosts in Extreme Environments (Riesch, R.,

Tobler, M. & Plath, M., eds), pp. 217–246. Berlin: Springer Verlag. doi: 10.1007/978-3-

319-13362-1

Pronko, A. J., Perlman, B. M. & Ashley-Ross, M. A. (2013). Launches, squiggles and pounces,

oh my! The water-land transition in mangrove rivulus (Kryptolebias marmoratus). Journal

of Experimental Biology 216, 3988–3995. doi: 10.1242/jeb.089961

Rahn, H. (1966). Aquatic gas exchange: theory. Respiration Physiology 1, 1–12.

Regan, K. S., Jonz, M. G. & Wright, P. A. (2011). Neuroepithelial cells and the hypoxia

response in the amphibious fish Kryptolebias marmoratus. Journal of Experimental Biology

214, 2560–2568. doi: 10.1242/jeb.056333

Richards, T. M., Krebs, J. M. & McIvor, C. C. (2011). Microhabitat associations of a semi-

terrestrial fish, Kryptolebias marmoratus (Poey 1880) in a mosquito-ditched mangrove

forest, west-central Florida. Journal of Experimental Marine Biology and Ecology 401, 48–

56.

Robertson, C. E., Turko, A. J., Jonz,M. G. & Wright, P. A. Hypercapnia and low pH induce

neuroepithelial cell proliferation and emersion behaviour in the amphibious fish

Kryptolebias marmoratus. Journal of Experimental Biology. doi: 10.1242/jeb.123133.

Sayer, M. D. J. (2005). Adaptations of amphibious fish for surviving life out of water. Fish and

Fisheries 6, 186–211. doi: 10.1111/j.1467-2979.2005.00193.x

Sayer, M. D. J. & Davenport, J. (1991). Amphibious fish: why do they leave water? Reviews in

Fish Biology and Fisheries 1, 159–181.

Scheel, J. J. (1990). Atlas of Killifishes of the Old World. Neptune City, NJ: TFH Publications.

Schwerdtfeger, W. K. & Bereiter-Hahn, J. (1978). Cell and tissue transient occurrence of

chloride cells in the abdominal epidermis of the guppy, Poecilia reticulata Peters, adapted to

sea water. Cell and Tissue Research 471, 463–471.Seghers, B. (1978). Feeding behavior and

terrestrial locomotion in the cyprinodontid fish, Rivulus hartii (Boulenger). Vereinigung für

Theoretische und Angerwandte Limnologie 20, 2055–2059.

Shartau, R. B. & Brauner, C. J. (2014). Acid–base regulation in bimodal breathers. Journal of

Fish Biology 84, 682–704.

Takei, Y. (2015). From aquatic to terrestrial life: evolution of the mechanisms for water

acquisition. Zoological Science 32, 1–7.

Tatarenkov, A., Gao, H., Mackiewicz, M., Taylor, D. S., Turner, B. J. & Avise, J. (2007). Strong

population structure despite evidence of recent migration in a selfing hermaphroditic

vertebrate, the mangrove killifish (Kryptolebias marmoratus). Molecular Ecology 16, 2701–

2711.

Tatarenkov, A., Lima, S. M. Q., Taylor, D. S. & Avise, J. C. (2009). Long-term retention of self-

fertilization in a fish clade. Proceedings of the National Academy of Sciences of the United

States of America 6, 14456–14459. doi: 10.1073/pnas.0907852106

Taylor, D. S. (1988). Observations on the ecology of the killifish Rivulus marmoratus Poey

(Cyprinodontidae) in an infrequently flooded mangrove swamp. Northeast Gulf Science 10,

63–68.

Taylor, D. S. (1990). Adaptive specializations of the cyprinodont fish Rivulus marmoratus.

Florida Scientist 53, 239–248.

Taylor, D. S. (2012). Twenty-four years in the mud: what have we learned about the natural

history and ecology of the mangrove rivulus, Kryptolebias marmoratus? Integrative and

Comparative Biology 52, 724–736. doi: 10.1093/icb/ics062

Taylor, D. S., Turner, B. J., Davis,W. P. & Chapman, B. B. (2008). A novel terrestrial fish

habitat inside emergent logs. American Naturalist 171, 263–266.

Tee-Van, J. (1922). The land minnow of Kartabo. Bulletin of the New York Zoological Society

25, 72–74.

Tingaud-Sequeira, A., Zapater, C., Chauvigné, F., Otero,D. & Cerdà, J. (2009).Adaptive

plasticity of killifish (Fundulus heteroclitus) embryos: dehydration-stimulated development

and differential aquaporin-3 expression. American Journal of Physiology. Regulatory,

Integrative and Comparative Physiology 296, 1041–1052. doi: 10.1152/ajpregu.91002.2008

Turko, A. J., Earley, R. L. & Wright, P. A. (2011). Behaviour drives morphology: voluntary

emersion patterns shape gill structure in genetically identical mangrove rivulus. Animal

Behaviour 82, 39–47.

Turko, A. J., Cooper, C. A. & Wright, P. A. (2012). Gill remodelling during terrestrial

acclimation reduces aquatic respiratory function of the amphibious fish Kryptolebias

marmoratus. Journal of Experimental Biology 215, 3973–3980. doi: 10.1242/jeb.074831

Turko, A. J., Robertson, C. E., Bianchini, K., Freeman, M. & Wright, P. A. (2014). The

amphibious fish Kryptolebias marmoratus uses different strategies to maintain oxygen

delivery during aquatic hypoxia and air exposure. Journal of Experimental Biology 217,

3988–3995. doi: 10.1242/jeb.110601

Ultsch, G. R. (1987). The potential role of hypercarbia in the transition from water-breathing to

air-breathing in vertebrates. Evolution 41, 442–445.

Vermeulen, F. B.M. & Hrbek, T. (2005). Kryptolebias sepia n. sp. (Actinopterygii:

Cyprinodontiformes: Rivulidae), a new killifish from the Tapanahony River drainage in

southeast Surinam. Zootaxa 928, 1–20.

Vulesevic, B., McNeill, B. & Perry, S. F. (2006). Chemoreceptor plasticity and respiratory

acclimation in the zebrafish Danio rerio. Journal of Experimental Biology 209, 1261–1273.

Walsh, M. R.&Reznick, D. N. (2010). Influence of the indirect effects of guppies on life-history

evolution in Rivulus hartii. Evolution 64, 1583–1593. doi: 10.1111/j.1558-

5646.2009.00922.x

Walter, R. P., Blum, M. J., Snider, S. B., Paterson, I. G., Bentzen, P., Lamphere, B. A. &

Gilliam, J. F. (2011). Isolation and differentiation of Rivulus hartii across Trinidad and

neighboring islands. Molecular Ecology 20, 601–618. doi: 10.1111/j.1365-

294X.2010.04968.x

Whitehead, A., Roach, J. L., Zhang, S. & Galvez, F. (2011). Genomic mechanisms of evolved

physiological plasticity in killifish distributed along an environmental salinity gradient.

Proceedings of the National Academy of Sciences of the United States of America 108,

6193–6198. doi: 10.1073/pnas.1017542108

Wright, P. A. (2012). Environmental physiology of the mangrove rivulus, Kryptolebias

marmoratus, a cutaneously breathing fish that survives for weeks out of water. Integrative

and Comparative Biology 52, 792–800. doi: 10.1093/icb/ics091

Wright, P. A. & Wood, C. M. (2009). A new paradigm for ammonia excretion in aquatic

animals: role of rhesus (Rh) glycoproteins. Journal of Experimental Biology 212, 2303–

2312. doi: 10.1242/jeb.023085

Zachar, P. C. & Jonz, M. G. (2012). Neuroepithelial cells of the gill and their role in oxygen

sensing. Respiratory Physiology and Neurobiology 184, 301–308. doi: 10.1016/j.resp.

2012.06.024

Zahl, P. A., McLaughlin, J. J. A. & Gomprecht, R. J. (1977). Visual versatility and feeding of the

four-eyed fishes, Anableps. Copeia 1977, 791–793.

Figures

Figure 2.1. Phylogenetic placement of cyprinodontiform families containing amphibious species

(after Costa, 2012). Families including amphibious species are in bold; the amphibious species are listed to the right along with the first report of emersion in that species.

Figure 2.2. Phylogenetic distribution of amphibious fishes within the cyprinodontiform families

(a) Rivulidae (after Costa, 2011a, 2013) and (b) Fundulidae (after Ghedotti & Davis, 2013).

Terminal clades containing amphibious species are in bold.

Figure 2.3. (a) In Long Caye, Belize, Kryptolebias marmoratus were found inside rotting logs, as first described by Taylor et al. (2008). (b) The undescribed Aphyosemion sp. Lobaye often leaves water, apparently voluntarily, and basks on top of floating vegetation in aquaria (photo credit: Dominique Taylor). (c) An individual K. marmoratus voluntarily emersed in a cup above water taken from its crab burrow in Belize.

Figure 2.4. Capillaries (green) extend between the scales of Kryptolebias marmoratus into the epidermis. This photograph of a sagittal section was taken through the dorsal surface of the head, but these capillaries are seen across the cutaneous epithelium. An antibody against the endothelial protein CD31 was used to label capillaries; for methods see Turko et al. (2014). C, capillaries; S, scales; M, muscle. Scale bar=100 μm.

Supplementary Material

Video 2.S1. Video S1 A video camera placed inside a crab burrow captured several emersions by Kryptolebias marmoratus in Long Caye, Belize (17°13.08′ N; 087°35.66′ W). Several attempts were often required before fish would remain emersed. Movements by the partially submerged blue land crab Cardisoma guanhumi, a detritivore, did not readily startle the emersed

K. marmoratus. Sudden movements by emersed mangrove tree crabs Goniopsis cruentata, a predatory species, caused emersed K. marmoratus to immediately flip back into the water. Video available online at: http://youtu.be/0nBaE5rTwlM/

CHAPTER 3: SKELETAL STIFFENING IN AN AMPHIBIOUS FISH OUT OF WATER IS A RESPONSE TO

INCREASED BODY WEIGHT

Published as: Turko, A. J., Kültz, D., Fudge, D., Croll, R. P., Smith, F. M., Stoyek, M. R. and

Wright, P. A. (2017). Skeletal stiffening in an amphibious fish out of water is a response to

increased body weight. Journal of Experimental Biology, 220, 3621-3631

Abstract

Terrestrial animals must support their bodies against gravity, while aquatic animals are effectively weightless because of buoyant support from water. Given this evolutionary history of minimal gravitational loading of fishes in water, it has been hypothesized that weight-responsive musculoskeletal systems evolved during the tetrapod invasion of land and are thus absent in fishes. Amphibious fishes, however, experience increased effective weight when out of water – are these fishes responsive to gravitational loading? Contrary to the tetrapod-origin hypothesis, we found that terrestrial acclimation reversibly increased gill arch stiffness (∼60% increase) in the amphibious fish Kryptolebias marmoratus when loaded normally by gravity, but not under simulated microgravity. Quantitative proteomics analysis revealed that this change in mechanical properties occurred via increased abundance of proteins responsible for bone mineralization in other fishes as well as in tetrapods. Type X collagen, associated with endochondral bone growth, increased in abundance almost nine-fold after terrestrial acclimation. Collagen isoforms known to promote extracellular matrix cross-linking and cause tissue stiffening, such as types IX and

XII collagen, also increased in abundance. Finally, more densely packed collagen fibrils in both gill arches and filaments were observed microscopically in terrestrially acclimated fish. Our results demonstrate that the mechanical properties of the fish musculoskeletal system can be fine-tuned in response to changes in effective body weight using biochemical pathways similar to those in mammals, suggesting that weight sensing is an ancestral vertebrate trait rather than a tetrapod innovation.

3.1 Introduction

The invasion of land by aquatic vertebrates poses a fundamental physical problem: animals that were once buoyant in water become effectively heavier in air (Denny, 1993).

Increased effective gravity in terrestrial environments is thought to have driven the evolution of robust and responsive musculoskeletal systems in tetrapods to provide postural support and improve locomotion (Clack, 2002). Tetrapod skeletons are also highly responsive to changes in mechanical load (Currey, 2002). In mammals, increased body weight or exercise causes gains in muscle mass and bone density while decreases in mechanical forces during bedrest or spaceflight have the opposite effect (Robling et al., 2006). Would fishes, with an evolutionary history of effective weightlessness in water, respond similarly to increased body weight (Horn, 2005;

Martin, 2007)? While most fishes never experience gravitational loading out of water, over 200 species of extant amphibious fishes experience increased effective weight during terrestrial sojourns (Ord and Cooke, 2016; Wright and Turko, 2016). If weight-induced skeletal plasticity evolved at the base of the tetrapod lineage, amphibious fishes should be unresponsive to gravitational loading and may thus be expected to possess constitutively stronger and stiffer support systems to provide reinforcement while out of water. Alternatively, if weight responsiveness is a fundamental characteristic of vertebrate bone or has convergently evolved in amphibious fishes, terrestrial acclimation should cause amphibious fishes to increase the strength and stiffness of skeletal support structures.

There is evidence that fish bones are responsive to extreme and dynamic mechanical loads, such as surgically implanted opercular springs (Atkins et al., 2015), vertebral lordosis

(Kranenbarg et al., 2005), hard versus soft diets (Meyer, 1987; Huysseune et al., 1994) or cantilevered billfish rostra (Atkins et al., 2014). The effect of gravitational loading in fishes has

62 not been studied, however. While mechanical loads caused by gravity are similar to those imparted by other, more routine sources (e.g. muscle contraction or crushing prey), they differ and may have been overlooked both because gravitational forces are relatively weak and because the nature of loading is static. For example, a common view has held that bone growth or resorption occurs only when mechanical loading of bone exceeds threshold levels, and the force of gravity would not be sufficient to surpass these thresholds (McBride and Silva, 2012).

However, recent evidence suggests that these thresholds do not exist (Sugiyama et al., 2012;

Christen et al., 2014). Furthermore, dynamic or cyclical mechanical forces are thought to be the predominant signals for bone growth and mineralization, and gravitational loading of inactive fish out of water (e.g. Turko et al., 2014) would be largely static and therefore may not be expected to cause a typical skeletal loading response (Turner, 1998; Yang et al., 2006).

The objective of this study was to determine whether there is a cellular response in the skeleton of amphibious fishes experiencing mechanical loading owing to increased effective gravity when out of water. These experiments allowed us to test a critical prediction made by the

‘weight-insensitive fish hypothesis’ – that amphibious fishes should not respond to increased effective body weight because weight-sensitive musculoskeletal systems evolved during the tetrapod invasion of land. In contrast, evidence that amphibious fishes respond to gravitational loading using similar biochemical pathways as tetrapods would be strong evidence that weight responsiveness is an ancestral character of vertebrate skeletal tissue. Mangrove rivulus

(Kryptolebias marmoratus) provided an ideal experimental model for our studies: these fish can survive terrestrially for several months (Taylor, 2012), have a sequenced genome (Kelley et al.,

2016) and predicted proteome, and are small enough to fit into a random positioning machine to experience simulated microgravity on Earth. By terrestrially acclimating fish while effectively

63 weightless, we could tease apart the effect of gravity from other confounding effects of terrestrial acclimation. We focused on weight-induced plasticity of the gill arches using mechanical testing, quantitative proteomics and histology. Gill arches were chosen because of their relatively simple skeletal support, provided by the long and slender ceratobranchial bone, and lack of involvement in locomotion or other behaviours that differ between aquatic and terrestrial environments.

Skeletal plasticity of the pectoral girdle in response to walking versus swimming has been observed in Polypterus senegalus, for example (Standen et al., 2014), but our aim was to separate the impact of altered effective weight from plasticity resulting from different patterns of musculoskeletal use (DiGirolamo et al., 2013).

3.2 Materials and Methods

3.2.1 Animals

Captive-reared mangrove rivulus, Kryptolebias marmoratus (Poey 1880), originating from Belize (lineage 50.91; Tatarenkov et al., 2010) were obtained from a breeding colony maintained at the University of Guelph (19.3±0.1 mm, 82.1±1.5 mg). Fish were housed under standard conditions described elsewhere (Frick and Wright, 2002). Zebrafish, Danio rerio

(Hamilton 1822), were obtained from the pet trade to provide comparison with fully aquatic fish of similar body size (20.8±0.8 mm, 93.4±9.6 mg). All experiments were approved by the

University of Guelph animal care committee (animal utilization protocols 2239 and 2478).

3.2.2 Terrestrial acclimation

Adult mangrove rivulus were terrestrially acclimated (25°C) on damp filter paper soaked with 15‰ water as described previously (Ong et al., 2007). For proteomics and transmission

64 electron microscopy (TEM) experiments, fish were terrestrially acclimated for 14 days and compared with control fish held in 15‰ brackish water. To determine the effect of terrestrial acclimation on the mechanical properties and collagen composition of gill arches, fish were divided into five treatment groups: control (15‰ water), 7 or 14 days of terrestrial exposure, or

14 days terrestrial exposure followed by recovery (14 days or 12 weeks) in water. Body length and mass were similar between treatment groups (P>0.05).

3.2.3 Mechanical testing

The force required to deform individual gill arches was measured using a custom tensile testing apparatus (Fudge et al., 2003). Briefly, gill arches dissected from the left side were separated by cutting through each epibranchial and hypobranchial bone (dorsal and ventral extremities). Separated gill arches were held in 15‰ brackish water with 0.1 mmol l-1 phenylmethylsulfonyl fluoride to minimize protein degradation prior to tensile testing; all tests were completed within ∼45 min of euthanasia. Individual gill arches were mounted using cyanoacrylate gel between a glass microbeam (force transducer) and a fine tungsten point epoxied to a micrometer (Fig. 3.1A). Retraction of the micrometer caused the glass rod to bend and the gill arch to stretch and deform. Tensile tests were filmed through a dissecting microscope

(MD200 microscope camera, AmScope, Irvine, CA; 2 frames s−1) and quantified using ImageJ

(Schneider et al., 2012).

Applied forces were calculated from the deflection of the glass microbeam using beam theory (Fudge et al., 2003). Extension of the gill arch (a combination of stretching and deformation) was calculated by taking the linear distance between each end of the arch in each video frame. Because of variation in cross-sectional size and shape along the length of the gill

65 arches, we were unable to meaningfully calculate traditional stiffness (standardized to cross- sectional area). Instead, effective ‘stiffness’ of the whole gill arch was calculated as the slope of the force–extension curve from 0 to 5% extension and is presented in newtons (N), or calculated as a spring constant from the slope of the force–length curves (N mm-1). All four gill arches were tested in the time-course experiment; only the first two gill arches were tested after microgravity acclimation as these were larger, easier to handle and all four gill arches responded to terrestrial acclimation in the same way (i.e. there was no significant treatment by arch interaction, P>0.05).

3.2.3 Microgravity simulation

Mangrove rivulus were acclimated to simulated microgravity (25°C, 12 h:12 h light:dark) for 7 days using a desktop random positioning machine (RPM; Airbus Defence and Space,

Leiden, Netherlands) that simulates microgravity by rotating animals randomly about two axes so that the mean vector force of gravity over time approaches zero (Borst and van Loon, 2009;

Herranz et al., 2013; Wuest et al., 2015). We assumed that the gill arches deformed more slowly than the gravitational vector changed, which would reduce the amount of strain that the arches experienced in the RPM compared with the terrestrially acclimated fish maintained at 1 g. In contrast, if the gill arches deformed faster than the rate of rotation by the random positioning machine, we would expect to see stiffening as the arches would continuously experience strain from gravity acting in an ever-shifting, random vector, assuming these deformations occurred slowly enough to be sensed (Ueki et al., 2008). The RPM was operated in ‘real random’ mode, with both direction and intervals of rotation randomized, and the maximum angular velocity=20 deg s−1, minimum angular velocity=8 deg s−1 and acceleration=30 deg s−2. An onboard accelerometer and proprietary software calculated that we maintained a simulated effective

66 gravity of 0.06 g throughout the experiment, slightly less than half of the gravity experienced on the moon (0.16 g; Hirt and Featherstone, 2012).

Microgravity-acclimated animals were housed under terrestrial conditions in standard six-well plates with a corrugated bottom lined with gauze soaked in brackish water. This allowed fish to move freely within a well without ‘tumbling’ during rotation, confirmed by video recording (Movie 1). Activity was quantified from 1 h recording periods at 1 h, 24 h, 3 days and

7 days as the proportion of video frames in which a body movement occurred. Control (1 g) fish, in water or air, were placed in identical six-well plates located beside the RPM to ensure these fish experienced a similar environment as the microgravity-acclimated fish. After 7 days, fish were euthanized with an overdose of MS222 and the mechanical properties of gill arches were measured.

3.2.4 Proteomics

Mangrove rivulus (control, n=6; 14 days terrestrial exposure, n=6) were euthanized with an overdose of MS222 and gill baskets were dissected, snap-frozen in liquid nitrogen and stored at −80°C until analysis. Protein extraction, protein assays and in-solution trypsin digestion were performed as reported previously (Kültz et al., 2013). Tryptic peptides from each sample (200 ng total) were injected with a nanoAcquity sample manager (Waters, Milford, MA, USA) and trapped for 1 min at 15 μl min−1 on a Symmetry trap column (Waters 186003514). They were then separated on a 1.7 μm particle size BEH C18 column (250 mm×75 μm, Waters 186003545) using a 125-min linear gradient ranging from 3% to 35% acetonitrile by reversed phase liquid chromatography using a nanoAcquity UPLC (Waters). Nano-electrospray ionization (nESI) was achieved by elution from a pico-emitter tip (New Objective FS360–20-10-D-20,Woburn, MA,

USA) into an nESI source fitted on an ImpactHD UHR-QTOF mass spectrometer (Bruker

Daltonics, Bremen, Germany). Batch-processing of samples was controlled with Hystar 3.2, and peak lists were generated with DataAnalysis 4.2 (Bruker Daltonics) using data-dependent acquisition as previously described (Kültz et al., 2015).

To identify gill proteins, raw data were imported into PEAKS Studio 8.0 (BSI, Waterloo,

Canada) followed by charge state deconvolution, deisotoping and peak list generation. PEAKS

8.0, Mascot 2.2.7 (Matrix Science, London, UK; version 2.2.07) and X! Tandem Cyclone

(www.thegpm.org/tandem/) search engines were used to identify proteins from MSMS spectra using trypsin as the enzyme, and maximum of one missed cleavage permitted, Cys carbamidomethylation, Met oxidation, Pro hydroxylation, N-terminal carbamylation and N- terminal acetylation as variable modifications. The precursor ion mass tolerance was set to 20 ppm (100 ppm for X! Tandem searches) and fragment ion mass tolerance was set to 0.02 amu.

The K. marmoratus proteome (38,516 sequences) was downloaded on 5 July 2016 fromNCBI and decoy entries were added for each sequence by PEAKS 8.0. The resulting database containing 77,032 total entries was used for all searches to determine protein false discovery rate

(FDR). Results from all three search engines were consolidated in Scaffold 4.4 (Proteome

Software Inc., Portland, OR, USA) with peptide identifications being accepted if FDR <0.1%.

Protein identifications were accepted FDR <1.0% and contained at least one identified peptide.

Proteins that contained similar peptides and could not be differentiated based on MS/MS analysis alone were grouped into clusters by the Scaffold software to satisfy the principles of parsimony.

Label-free quantitative profiling of peptide intensities and calculation of relative protein abundances in each sample was performed with PEAKS 8.0 using the following parameters:

FDR <1%, confident samples per peptide ≥4, mass error tolerance=30 ppm and retention time shift tolerance=3 min.

3.2.5 Histology

The collagen-specific dye Picrosirius Red (Electron Microscopy Sciences, Hatfield, PA,

USA) was used to determine the density and/or thickness of collagen fibres in sections of gill arches and filaments. Gill arches were fixed overnight (4°C) in 10% buffered formalin, decalcified (Cal-Ex, Fisher Scientific) for 1 h (room temperature) and double-embedded in 2% agar followed by paraffin wax to aid tissue orientation. Cross-sections (3 μm for gill arch analysis, 5 μm for filaments) were cut through the middle of the ceratobranchial bone of each gill arch, the main skeletal support structure. Tissues were stained with Picrosirius Red and photographed under circularly polarized light as previously described (Johnson et al., 2014).

Under polarized light, thin and/ or disorganized collagen appears green, while densely packed collagen appears red (Rich and Whittaker, 2005). Instead of separating collagen hues into arbitrary colour ‘bins’ as described previously (MacKenna et al., 1994; Rich and Whittaker,

2005; Johnson et al., 2014), we simply calculated the mean hue of collagen-staining pixels in each gill arch after normalizing the 256 colour histogram so that the lowest-intensity green was scored with a value of 1 (traditionally, hue value=127) while the most intense red pixels received a score of 155 (traditionally, hue value=229).

TEM was used to describe the structure of gill arch and filament extracellular matrix.

Gills were rapidly dissected from euthanized fish (control n=2, 14 days air exposure n=2) and immediately fixed in freshly prepared 2% glutaraldehyde/2% tannic acid in 0.1 mol l−1 cacodylate buffer (pH 7.4) as previously described (King et al., 1989). Gill arches were washed

(3×, 0.5 mol l−1 HEPES), postfixed in 1% osmium tetroxide (2 h), washed (3×, 0.5 mol l−1

HEPES), stained in 2% uranyl acetate (2 h) and then dehydrated through a graded ethanol series.

Samples were then infiltrated with a descending series of ethanol:LR White resin (Electron

Microscopy Sciences), put into capsules and cured (24 h, 60°C). Thin sections (100 nm) were mounted on mesh copper grids coated with a layer of Formvar/carbon (Electron Microscopy

Sciences) and post-stained with 2% uranyl acetate followed by lead citrate. Images were obtained using a Philips CM10 electron microscope (80 kV) equipped with an Olympus/SIS

Morada top mount 11MP CCD camera.

3.2.6 Calcium content of gill baskets

Gill baskets from five individuals were rapidly dissected from euthanized fish and pooled for a single sample. Samples were dried (65°C, 24 h), weighed and then digested with 3:1 trace metal grade HNO3: HCl for 3 h. Digested samples were heated (90°C, 2 h) and filtered, and calcium content was measured using flame atomic absorption spectrometry (Varian SpectrAA

220 double beam flame atomic absorption spectrometer, Agilent Technologies, Santa Clara, CA,

USA). Bone mineral density was approximated assuming all calcium was present in the form of hydroxyapatite [Ca5(PO4)3(OH)], the main mineral component of bone. Hydroxyapatite is

39.9% calcium by mass, thus bone mineral density was calculated by dividing the measured calcium content by this fraction (0.399).

3.2.7 Statistical analyses

Two-way repeated-measures ANOVAs were used to detect differences in mechanical properties and collagen density between gill arches and across treatment groups. In a few

70 instances, dissected gill arches were lost or contaminated with cyanoacrylate and excluded from all analyses. Data were natural-logarithm (ln) transformed when necessary to meet assumptions of normality and equal variance. Statistical significance for label-free quantitation is based on

PEAKSQ –log10P-values, which have been calculated using a previously developed algorithm that has been optimized for proteomics data (Cox and Mann, 2008). A significance threshold of – log10P- value ≥5 and fold change of ≥1.5 were applied. To clarify visualization of protein abundances, values were z-score transformed (mean protein abundance was subtracted from each individual value, and the result was divided by the standard deviation) to normally distribute abundances and enhance the dynamic range for each protein (Berrar et al., 2007).

3.3 Results

3.3.1 Mechanical properties of gill arches

Applying force to isolated gill arches caused both small changes in shape (i.e. straightening) and straining of the gill skeleton. Together, these deformations provided an overall measure of the lengthening of the gill arch. The amount of force required to extend the gill arches significantly increased after 7 days of acclimation out of water when calculated from both force–extension curves (P<0.001; Figs 3.1B and 3.2A) and force–length curves (P<0.001; Fig.

3.S1). No additional increase in gill arch stiffness was observed after a further 7 days of terrestrial acclimation. Increased gill arch stiffness persisted after a 14 day recovery period in water, but returned to control values after 12 weeks (Fig. 3.2A, Fig. 3.S1). The fourth gill arch was significantly stiffer than arches 1 or 2 (force–extension P=0.002; Fig. 3.2B; force–length

P<0.001; Fig. 3.S1), possibly because arch 4 was straighter at the outset. The stiffening we

71 observed in terrestrially acclimated fish was widespread across all four gill arches (interaction

P>0.05; Fig. 3.2A, Fig. 3.S1).

Terrestrially acclimated fish held under conditions of simulated microgravity showed no obvious ill effects, and activity was similar to that of terrestrially acclimated fish at 1 g (P>0.05;

Fig. 3.S2). Microgravity-acclimated fish showed no righting response when inverted, which could have negated the effect of the random rotations by allowing gravity to act on the gill arches in a consistent vector (Movie 1). Gill arches from terrestrially acclimated fish in simulated microgravity were of comparable stiffness to those from control fish in water (P>0.05; Fig. 3.2B,

Fig. 3.S3). Terrestrially acclimated fish placed at the base of the random positioning machine to control for effects of noise and vibration had significantly stiffer gill arches relative to fish in water, as observed in the time course experiment (force–extension P=0.001; Fig. 3.2B; force– length P=0.025; Fig. 3.S3).

To determine whether gill arches from the amphibious K. marmoratus are constitutively stiffer than gills in fully aquatic fishes, we tested the mechanical properties of gill arches from similarly sized D. rerio. Overall, gill arches of control K. marmoratus in water were significantly stiffer than those of D. rerio (force–extension P=0.002; Fig. 3.3; force–length P=0.006; Fig.

3.S4). There was no significant effect of gill arch when ‘stiffness’ was calculated from force– extension curves (P>0.05, interaction P>0.05), but the fourth gill arch of both species was significantly stiffer than the first using force–length calculations (P=0.012, interaction P>0.05).

3.3.2 Proteomic responses to terrestrial acclimation

Quantitative proteomics analysis of gill tissue identified 1095 proteins, 152 of which significantly changed in abundance after 14 days of terrestrial acclimation. Of these differentially

72 expressed proteins, 23 were structural components of the cytoskeleton or extracellular matrix

(Fig. 3.4). Many of the cytoskeletal proteins that increased in abundance are components of the microfilament network. These included actin, responsible for forming microfilaments, as well as several proteins that regulate the folding (T-complex) and bundling (plastin-3) of the actin cytoskeleton. Type II keratin, part of the intermediate filament network, also significantly increased in abundance after terrestrial acclimation (Fig. 3.4).

Collagen was particularly responsive to terrestrial acclimation; of the 34 collagen isoforms identified in the proteomics analysis, 71% increased in abundance (Table 3.S1), significantly more than expected by chance (Z-test for proportions, P=0.016). Abundances of both alpha subunits of type I collagen, the major organic structural component of bone, increased significantly (P<0.001), but only by ∼30% (relative to total protein abundance), below our a priori threshold for consideration. Considering the large absolute abundance of type I collagen in the gills (5.8±0.3% of total protein in control samples, 7.4±1.0% after 14 days terrestrial acclimation), this 30% increase nonetheless represents a considerable increase in the absolute number of collagen molecules present. Abundance of type X collagen, associated with bone formation, increased 8.8-fold (P<0.0001) after terrestrial acclimation (Fig. 3.4). Gills of terrestrially acclimated fish also showed increased abundance of collagen types IX and XII, isoforms that enhance cross-linking between extracellular matrix components (Fig. 3.4).

3.3.3 Morphological changes after terrestrial acclimation

We used polarized light imaging of Picrosirius-Red-stained cross-sections through the middle of the ceratobranchial bone to assess the relative density of collagen in the extracellular matrix. Collagen had a relatively green hue in control fish, indicating less densely packed

73 molecules, while in terrestrially acclimated fish the gill arches had a significantly redder hue, indicative of more densely packed collagen (P=0.004; Fig. 3.5A,B). After recovery in water, the hue of gill arch collagen was intermediate and not statistically different from either control or terrestrially acclimated fish (P>0.05). Across all treatments, the third and fourth gill arches were significantly redder in hue than the first and second arches (P=0.012, interaction P>0.05).

Consistent with both the proteomic data and polarized light imaging of cross-sections, TEM revealed densely packed collagen molecules at the margins of the ceratobranchial bones in terrestrially acclimated fish (Fig. 3.5C,D). Furthermore, a dark band of mineral was present in the ceratobranchial bone margin after terrestrial acclimation (Fig. 3.5D).

The gill filaments protrude perpendicularly from the gill arch and are supported by a cartilage rod produced by a central core of chondrocytes. Similar to the ceratobranchial bones,

Picrosirius-Red stained sections through the gill filaments were significantly redder in hue under polarized light after terrestrial acclimation (P<0.001; Fig. 3.6A). Electron micrographs of longitudinal sections through the filament-supporting cartilage rods showed denser aggregations of highly organized collagen in terrestrially acclimated fish (Fig. 3.6B,C), a pattern similar to what we observed in the gill arches.

Calcium content of whole gill baskets was very low (66.7± 1.5 mg g−1 dry mass; Fig.

3.7A), but these measurements were consistent with the low degree of mineralization observed in electron micrographs of the gill skeleton (Fig. 3.7B). If we assume that this calcium is present as hydroxyapatite, the major mineral component of bone, this is equal to a mineral content of 16.7±

0.4%. Gill basket calcium content did not change after terrestrial acclimation (P>0.05; Fig. 3.7A).

3.4 Discussion

Our results are contrary to the hypothesis that vertebrate weight sensitivity evolved with the tetrapod invasion of land. Like astronauts moving between Earth and outer space (Robling et al., 2006), we found that the gill skeleton of the amphibious fish K. marmoratus is stiffened in response to increased effective weight, probably via both the modification of existing tissues and the formation of new endochondral bone. Furthermore, many of the protein-level changes we observed in the gills of terrestrially acclimated K. marmoratus were similar to those associated with increased body weight in tetrapods, such as increased abundances of collagen isoforms associated with endochondral bone formation and cross-linking of the extracellular matrix. These results suggest a fundamentally conserved response to changes in body weight in vertebrates.

3.4.1 Weight responsiveness in an amphibious fish

Gill arches of terrestrially acclimated fish stiffened rapidly, within 1 week, and no further change in mechanical properties was observed after an additional week out of water. In most bone, desensitization to mechanical forces occurs within hours to days (Turner, 1998; Robling et al., 2001). Thus, it seems probable that extracellular matrix-synthesizing cells of the gill skeleton were desensitized to mechanical loading after 1 week of terrestrial acclimation. Alternatively, the stiffening that occurred within the first week may have been sufficient to support the increased weight of the gill arches, thus removing the stimulus for further plasticity (Robling et al., 2006).

The increased gill stiffness caused by terrestrial acclimation persisted after 2 weeks of recovery in water, but was reversible after 3 months. The extended time required to reverse the terrestrially induced response is consistent with the normal rate of protein turnover in fishes

(∼1% per day in trout at 15°C; Lyndon and Houlihan, 1998) and suggests either that stiffened

75 gill baskets do not impose a large and/or immediate cost or that K. marmoratus are not able to quickly decrease gill arch stiffness.

Stiffening of gill arches in terrestrially acclimated fish could have been caused by several factors other than the influence of increased effective weight. For example, altered branchial muscle function and/ or surface tension from droplets of residual water in the branchial chamber could impart forces not experienced in water (Vogel, 1984), or stiffening could result simply from dehydration of connective tissues (Avery and Bailey, 2008). By holding terrestrially acclimated fish under simulated microgravity, we explicitly tested the influence of bodyweight on gill mechanical properties. Simulated microgravity eliminated the stiffening response observed during normal terrestrial acclimation, strongly suggesting that gill arch stiffening during terrestrial acclimation is caused by increased effective body weight and not some other aspect of air exposure. Random positioning machines have been used successfully to mimic microgravity conditions in experiments using cultured cells (e.g. Kraft et al., 2000; Pardo et al.,

2005; Patel et al., 2007), plants (Kittang et al., 2013) and invertebrates (Ricci and Boschetti,

2003), but to our knowledge this is the first successful ground-based simulation of microgravity for an adult vertebrate. Amphibious fishes such as K. marmoratus may be a useful experimental model for future experiments investigating weight-induced musculoskeletal remodelling (Rea et al., 2016; Reilly and Franklin, 2016).

Adjustments to the morphology or mechanical properties of bones have widely been thought to require larger forces or strain rates than those imposed by gravitational loading (e.g.

Frost, 1990; Aiello et al., 2015). However, sheep bones experimentally loaded with just 0.3 g of force have been shown to increase bone density and the rate of bone formation, demonstrating that even very small loads can be important for determining the mechanical properties of tissues

(Rubin et al., 2001). Other studies in mammals have also shown that very low degrees of tissue strain are sufficient to cause skeletal remodelling (Sugiyama et al., 2012; Christen et al., 2014).

Our results provide further evidence that very small mechanical signals, in this case increased effective weight in the absence of buoyant support from water, can modify skeletal morphology.

The prevalent mammalian view is that osteocytes, cells embedded within the bone matrix, are responsible for mechanosensitive responses in bone (Robling et al., 2006; Bonewald, 2011;

Prideaux et al., 2016). In contrast, the bones of many teleosts including K. marmoratus lack osteocytes (Parenti, 1986; Shahar and Dean, 2013; Doherty et al., 2015). Without these mechanosensitive osteocytes, how can the ceratobranchial bones respond to increased effective body weight? In accordance with our results, several recent studies in other teleost fishes have demonstrated that anosteocytic bone can respond to changes in mechanical loading, suggesting that skeletal mechanosensitivity is more complicated than has been assumed (Witten and

Huysseune, 2009; Shahar and Dean, 2013; Witten and Hall, 2015).

3.4.2 How are gill arches stiffened?

We used a combination of quantitative proteomics and histology to understand the morphological and biochemical changes underlying gill arch stiffening in terrestrially acclimated fish. Overall, our results indicate that stiffening of the gill arch skeleton in response to terrestrial acclimation is probably caused by complementary changes to the extracellular matrix of the gill arch skeleton, including growth via the synthesis and mineralization of new collagen plus the modification of existing matrix by the formation of cross-links. The changes to the extracellular matrix that occurred in K. marmoratus are also well known to occur in mammals in response to changes in gravitational loading, such as during spaceflight (Humphrey et al., 2014).

Collagen is the main organic component of extracellular matrix and a major determinant of stiffness in bone and cartilage (Porter et al., 2006). Of the 1095 proteins identified in our proteomics analysis, type X collagen showed the largest increase in abundance of any protein. In mammals, type X collagen is a well-known marker for endochondral bone formation and ossification (Ricard-Blum et al., 2000; Shen, 2005). This protein is expressed by hypertrophic chondrocytes at the margins of growing endochondral bones, where it provides structural support and promotes mineralization (Jacenko et al., 1993; Rosati et al., 1994). Furthermore, experiments using cultured chondrocytes show that expression of type X collagen increases in response to mechanical strain (Yang et al., 2006). In fishes, type X collagen is similarly associated with bone formation (Renn and Winkler, 2010; Eames et al., 2012). Interestingly, in euteleost fishes, type X collagen is restricted to osteoblasts (Renn and Winkler, 2010), while in more basal fishes, including D. rerio and Lepisosteus oculatus, both chondrocytes and osteoblasts produce type X collagen (Eames et al., 2012). This suggests that common patterns of mechanically induced bone growth are found in all vertebrates, but there may be differences in the specific mesenchymal cells (chondrocytes versus osteoblasts) that are responsible for bone matrix deposition and mineralization in teleosts and mammals.

In addition to mineral density, cross-linking between collagen molecules in the extracellular matrix is an important determinant of the mechanical properties of bone (Oxlund et al., 1995; Banse et al., 2002; Avery and Bailey, 2008; Saito and Marumo 2010). Terrestrial acclimation of K. marmoratus increased the abundance of two collagen isoforms, types IX and

XII, that enhance cross-linking and probably contributed to the observed increase in tissue stiffness. Type IX collagen coats the surface of other collagen fibrils and acts to bind the fibrillary network, which enhances structural integrity (Hulmes, 2008; Wess, 2008). Type XII is

78 another cross-link forming collagen that is similarly localized to the surfaces of fibrils (Wess,

2008). This collagen isoform is upregulated by mechanical stress, regulates bone formation and is an important determinant of skeletal stiffness (Izu et al., 2011; Chiquet et al., 2014). Femurs of mutant mice lacking type XII collagen are less stiff than normal mice, and Picrosirius Red imaging of these bones indicated less organized and smaller collagen fibrils (Izu et al., 2011).

Similarly, our histological evidence suggests that collagen was more densely packed and organized in terrestrially acclimated K. marmoratus, possibly facilitated by increased cross- linking owing to upregulation of types IX and XII collagen. Forming cross-links to stabilize existing extracellular matrix would provide a rapid and energetically inexpensive way to increase tissue stiffness that may complement the synthesis of new matrix (Oxlund et al., 1995).

There was no change in overall gill calcium content in response to terrestrial acclimation, suggesting large-scale mineralization is not responsible for gill arch stiffening. However, we noticed in several electron micrographs that a dark band of mineral was deposited at the margin of the ceratobranchial bones after terrestrial acclimation. This amount of additional mineral would probably be too small to be detectable in samples of whole gill arches, but may nevertheless be an important mechanism to increase stiffness. We were also surprised to find very low calcium content in the gill skeleton of mangrove rivulus (∼17% dry mass), approximately half of what has been reported in other teleost fishes (Cohen et al., 2012). The low mineral content of K. marmoratus gills, however, is consistent with the relatively low forces required to deform the gill arches (Currey, 2003; Barak et al., 2013). Given the small and irregular nature of K. marmoratus gill arches, it was not possible to accurately determine

Young’s modulus – the material stiffness of a gill arch excluding effects of shape and size.

However, using our Picrosirius-Red-stained sections we measured a mean gill arch cross-

79 sectional area of 0.025±0.002 mm2, which allowed us to calculate an approximate Young’s modulus of 2–5 MPa. This value is closer to measurements of calcified cartilage in chondrichthyan vertebral centra (Porter et al., 2006) than to bones in other teleost fishes (Cohen et al., 2012). One possibility is that the skeletons of these tiny animals are adequately strong and stiff even with minimal mineralization owing to the relatively large cross-sectional area of the bones, which determines strength, compared with body size (Schmidt-Nielsen, 1984). The average mass of fish used in these experiments was 82±1 mg, at least 100 times smaller than other fishes in which bone mechanical properties and mineral density has been studied (Porter et al., 2006; Cohen et al., 2012).

Mechanosensitive responses to terrestrial acclimation may not be restricted to the gill arches of K. marmoratus, as we also saw evidence for structural remodelling at the level of the gill filaments and even within individual cells. Gill filaments are supported by cartilage rods

(Wilson and Laurent, 2002), and terrestrial acclimation caused increased collagen density and organization within this cartilage in a manner that mirrored our results in the gill arches.

Morphological adjustments at the level of the gill lamellae may also enhance structural support.

In previous experiments, we have found that terrestrially acclimated K. marmoratus enlarge a mass of cells that fills the space between gill lamellae (Ong et al., 2007; LeBlanc et al., 2010;

Turko et al., 2012). While the function of these cells has not been directly tested, one hypothesis is that this cell mass provides structural support out of water (Nilsson et al., 2012; Wright, 2012).

Gill lamellae may also be supported by stiffening of the cytoskeleton. Our proteomics analysis showed increased abundance of actin and other proteins that regulate the microfilament network, which is thought to be the major determinant of intra-cellular mechanical properties (Rotsch and

Radmacher, 2000) and may increase the stiffness of individual gill cells during terrestrial

80 acclimation. Gill lamellar support may also be enhanced by laminins, which increased in abundance almost threefold in our study and provide structure to basement membranes of epithelia (Aumailley and Smyth, 1998).

3.4.3 Evolution of weight sensing in vertebrates

Considering the evolutionary history of effective weightlessness of fishes in water, it is intuitively reasonable to hypothesize that these animals are unable to sense body weight (Anken and Rahmann, 1999; Rahmann and Anken, 2002) and have skeletal systems unresponsive to changes in effective gravity (Martin, 2007). The different supportive requirements for animals in water versus on land have certainly influenced overall skeletal size – bone mass is related isometrically to body size in buoyant aquatic animals, while in terrestrial animals the relationship is exponential because bone strength is a function of cross-sectional area rather than volume

(Reynolds, 1977; Schmidt-Nielsen, 1984). Furthermore, weight and use-responsive musculoskeletal systems are not ubiquitous in tetrapods. In mice and humans, for example, skeletal mechanosensitivity is heritable and controlled by several genetic loci (Tajima et al.,

2000; Kesavan et al., 2007; Kapur et al., 2010; Judex et al., 2013). Similarly, the musculoskeletal systems of many aestivating amphibians and hibernating mammals do not atrophy during months of inactivity (Doherty et al., 2015; Reilly and Franklin, 2016), suggesting the mechanosensitivity of support structures may be shaped by selection. However, contrary to the hypothesis that weight-sensing originated with the tetrapod invasion of land, we found that the branchial skeleton of K. marmoratus stiffened in response to increased effective gravity during terrestrial acclimation using biochemical pathways similar to those known in mammals. We also recently reported reversible hypertrophy of aerobic body muscle in terrestrially acclimated K.

81 marmoratus despite minimal activity, suggesting that weight-bearing responses are widespread throughout the musculoskeletal system (Brunt et al., 2016). Together, these data indicate that gravitational loading enhances body support in a teleost fish, despite an evolutionary history of effective weightlessness in water.

In the absence of time travel, it is impossible to know whether weight responsiveness in fishes represents the ancestral state. The ability of K. marmoratus to stiffen the gill skeleton when out of water may have evolved convergently with weight sensitivity in tetrapod bones to maintain structural integrity or prevent damage from weight bearing. More likely, considering the similarity of the proteomic response in the teleost K. marmoratus to the mechanisms used by tetrapods to remodel bone, the gravitational response we observed may indicate that the ability of bone-producing osteoblasts to respond to even slight changes in mechanical load is a fundamental feature of vertebrate skeletons (Shahar and Dean, 2013; Schilder, 2016).

Mechanical stresses owing to differences in tissue density would occur even in neutrally buoyant, effectively weightless fishes, possibly providing an adaptive benefit to the ability to fine tune the skeleton. In amphibious fishes, similar fine tuning of the skeleton would result in stiffer bones to provide body support, and if environmental conditions favoured highly terrestrial individuals, these plastic traits could become genetically assimilated in subsequent generations

(Pigliucci et al., 2006; Standen et al., 2014). In support of this idea, we found that the gill arches of water-acclimated mangrove rivulus were stiffer than those of comparably sized zebrafish, a fully aquatic species, indicating that phenotypic plasticity may complement constitutively stiffer gill arches in the amphibious mangrove rivulus. It has also been hypothesized that novel mechanical forces acting on skeletons, such as gravity on the bones of fish out of water, could give rise to evolutionary innovations in bone structure (Danos and Staab, 2010).

Mechanosensitive fish osteoblasts may have thus served as a valuable exaptation for producing a strong, stiff, weight-responsive skeleton during the tetrapod invasion of land (Khayyeri and

Prendergast, 2013; Doherty et al., 2015).

3.5 References

Aiello, B. R., Iriarte-Diaz, J., Blob, R. W., Butcher, M. T., Carrano, M. T., Espinoza, N. R.,

Main, R. P. and Ross, C. F. (2015). Bone strain magnitude is correlated with bone strain rate

in tetrapods: implications for models of mechanotransduction. Proc. R. Soc. B Biol. Sci.

282, 20150321.

Anken, R. H. and Rahmann, H. (1999). Effect of altered gravity on the neurobiology of fish.

Naturwissenschaften 86, 155-167.

Atkins, A., Dean, M. N., Habegger, M. L., Motta, P. J., Ofer, L., Repp, F., Shipov, A.,Weiner, S.,

Currey, J. D. and Shahar, R. (2014). Remodeling in bone without osteocytes: billfish

challenge bone structure–function paradigms. Proc. Natl Acad. Sci. USA 111, 16047-16052.

Atkins, A., Milgram, J., Weiner, S. and Shahar, R. (2015). The response of anosteocytic bone to

controlled loading. J. Exp. Biol. 218, 3559-3569.

Aumailley, M. and Smyth, N. (1998). The role of laminins in basement membrane function. J.

Anat. 193, 1-21.

Avery, N. C. and Bailey, A. J. (2008). Restraining cross-links responsible for the mechanical

properties of collagen fibers: natural and artificial. In Collagen Structure and Mechanics (ed.

P. Fratzl), pp. 81-110. New York: Springer.

Banse, X., Sims, T. J. and Bailey, A. J. (2002). Mechanical properties of adult vertebral

cancellous bone: correlation with collagen intermolecular cross-links. J. Bone Miner. Res.

17, 1621-1628.

Barak, M. M., Lieberman, D. E. and Hublin, J.-J. (2013). Of mice, rats and men: trabecular bone

architecture in mammals scales to body mass with negative allometry. J. Struct. Biol. 183,

123-131.

Berrar, D. P., Granzow, M. and Dubitzky, W. (2007). Introduction to genomic and proteomic

data analysis. In Fundamentals of Data Mining in Genomics and Proteomics (ed. W.

Dubitzky, M. Granzow and D. P. Berrar), pp. 1-38. New York: Springer.

Bonewald, L. F. (2011). The amazing osteocyte. J. Bone Miner. Res. 26, 229-238.

Borst, A. G. and van Loon, J. J. W. A. (2009). Technology and developments for the random

positioning machine, RPM. Microgravity Sci. Technol. 21, 287-292.

Brunt, E. M., Turko, A. J., Scott, G. R. and Wright, P. A. (2016). Amphibious fish jump better

on land after acclimation to a terrestrial environment. J. Exp. Biol. 219, 3204-3207.

Chiquet, M., Birk, D. E., Bonnemannc, C. G. and Koch, M. (2014). Collagen XII: protecting

bone and muscle integrity by organizing collagen fibrils. Int. J. Biochem. Cell Biol. 53, 51-

54.

Christen, P., Ito, K., Ellouz, R., Boutroy, S., Sornay-Rendu, E., Chapurlat, R. D. and van

Rietbergen, B. (2014). Bone remodelling in humans is load-driven but not lazy. Nat.

Commun. 5, 4855.

Clack, J. A. (2002). Gaining ground: The Origin and Evolution of Tetrapods. Bloomington, IN:

Indiana University Press.

Cohen, L., Dean, M., Shipov, A., Atkins, A., Monsonego-Ornan, E. and Shahar, R. (2012).

Comparison of structural, architectural and mechanical aspects of cellular and acellular bone

in two teleost fish. J. Exp. Biol. 215, 1983-1993.

Cox, J. and Mann, M. (2008). MaxQuant enables high peptide identification rates, individualized

ppb-range mass accuracies and proteome-wide protein quantification. Nature Biotechnol.

26, 1367-1372.

Currey, J. D. (2002). The Mechanical Adaptations of Bones. Princeton, NJ: Princeton University

Press.

Currey, J. D. (2003). The many adaptations of bone. J. Biomech. 36, 1487-1495.

Danos, N. and Staab, K. L. (2010). Can mechanical forces be responsible for novel bone

development and evolution in fishes? J. Appl. Ichthyol. 26, 156-161.

Denny, M. W. (1993). Air and Water: The Biology and Physics of Life’s Media. Princeton, NJ:

Princeton University Press.

DiGirolamo, D. J., Kiel, D. P. and Esser, K. A. (2013). Bone and skeletal muscle: neighbors with

close ties. J. Bone Miner. Res. 28, 1509-1518.

Doherty, A. H., Ghalambor,C.K. and Donahue, S.W. (2015). Evolutionary physiology of bone:

bone metabolism in changing environments. Physiology 30, 17-29.

Eames, B. F., Amores, A., Yan, Y.-L. and Postlethwait, J. H. (2012). Evolution of the osteoblast:

skeletogenesis in gar and zebrafish. BMC Evol. Biol. 12, 27.

Frick, N. T. and Wright, P. A. (2002). Nitrogen metabolism and excretion in the mangrove

killifish Rivulus marmoratus I. The influence of environmental salinity and external

ammonia. J. Exp. Biol. 205, 79-89.

Frost, H. M. (1990). Skeletal structural adaptations to mechanical usage (SATMU): 1.

Redefining Wolff’s law: the bone modeling problem. Anat. Rec. 226, 403-413.

Fudge, D. S., Gardner, K. H., Forsyth, V. T., Riekel, C. and Gosline, J. M. (2003). The

mechanical properties of hydrated intermediate filaments: insights from hagfish slime

threads. Biophys. J. 85, 2015-2027.

Herranz, R., Anken, R., Boonstra, J., Braun, M., Christianen, P. C. M., de Geest, M. Hauslage, J.,

Hilbig, R., Hill, R. J. A., Lebert, M. et al. (2013). Ground-based facilities for simulation of

microgravity: organism-specific recommendations for their use, and recommended

terminology. Astrobiology 13, 1-17.

Hirt, C. and Featherstone, W. E. (2012). A 1.5 km-resolution gravity field model of the Moon.

Earth Planet. Sci. Lett. 329-330, 22-30.

Horn, E. R. (2005). Gravity effects on life processes in aquatic animals. Adv. Space Biol. Med.

10, 247-301.

Hulmes, D. J. S. (2008). Collagen diversity, synthesis and assembly. In Collagen Structure and

Mechanics (ed. P. Fratzl), pp. 15-48. New York: Springer.

Humphrey, J. D., Dufresne, E. R. and Schwartz, M. A. (2014). Mechanotransduction and

extracellular matrix homeostasis. Nat. Rev. Mol. Cell Biol. 15, 802-812.

Huysseune, A., Sire, J.-Y. and Meunier, F. J. (1994). Comparative study of lower pharyngeal jaw

structure in two phenotypes of Astatoreochromis alluaudi (Teleostei: Cichlidae). J. Morphol.

221, 25-43.

Izu, Y., Sun, M., Zwolanek, D., Veit, G., Williams, V., Cha, B., Jepsen, K. J., Koch, M. and

Birk, D. E. (2011). Type XII collagen regulates osteoblast polarity and communication

during bone formation. J. Cell Biol. 193, 1115-1130.

Jacenko, O., LuValle, P. A. and Olsen, B. R. (1993). Spondylometaphyseal dysplasia in mice

carrying a dominant negative mutation in a matrix protein specific for cartilage-to-bone

transition. Nature 365, 56-61.

Johnson, A. C., Turko, A. J., Klaiman, J. M., Johnston, E. F. and Gillis, T. E. (2014). Cold

acclimation alters the connective tissue content of the zebrafish (Danio rerio) heart. J. Exp.

Biol. 217, 1868-1875.

Judex, S., Weidong, Z., Donahue, L. R. and Ozcivici, E. (2013). Genetic loci that control the loss

and regain of trabecular bone during unloading and reambulation. J. Bone Miner. Res. 28,

1537-1549.

Kapur, S., Amoui, M., Kesavan, C.,Wang, X., Mohan, S., Baylink, D. J. and Lau, K.-H. W.

(2010). Leptin receptor (Lepr) is a negative modulator of bone mechanosensitivity and

genetic variations in Lepr may contribute to the differential osteogenic response to

mechanical stimulation in the C57BL/6J and C3H/HeJ pair of mouse strains. J. Biol. Chem.

285, 37607-37618.

Kelley, J. L., Yee, M.-C., Brown, A. P., Richardson, R. R., Tatarenkov, A., Lee, C. C., Harkins,

T. T., Bustamante, C. D. and Earley, R. L. (2016). The genome of the self-fertilizing

mangrove rivulus fish, Kryptolebias marmoratus: a model for studying phenotypic plasticity

and adaptations to extreme environments. Genome Biol. Evol. 8, 2145-2154.

Kesavan, C., Baylink, D. J., Kapoor, S. and Mohan, S. (2007). Novel loci regulating bone

anabolic response to loading: expression QTL analysis in C57BL/6JXC3H/HeJ mice cross.

Bone 41, 223-230.

Khayyeri, H. and Prendergast, P. J. (2013). The emergence of mechanoregulated endochondral

ossification in evolution. J. Biomech. 46, 731-737.

King, J. A. C., Abel, D. C. and DiBona, D. R. (1989). Effects of salinity on chloride cells in the

euryhaline cyprinodontid fish Rivulus marmoratus. Cell Tissue Res. 257, 367-377.

Kittang, A.-I., Winge, P., van Loon, J. J. W. A., Bones, A. M. and Iversen, T.-H. (2013).

Adaptation response of Arabidopsis thaliana to random positioning. Adv. Space Res. 52,

1320-1331.

Kraft, T. F. B., van Loon, J. J. W. A. and Kiss, J. Z. (2000). Plastid position in Arabidopsis

columella cells is similar in microgravity and on a random-positioning machine. Planta 211,

415-422.

Kranenbarg, S., van Cleynenbreugel, T., Schipper, H. and van Leeuwen, J. (2005). Adaptive

bone formation in acellular vertebrae of sea bass (Dicentrarchus labrax L.). J. Exp. Biol.

208, 3493-3502.

Kultz, D., Li, J., Gardell, A. and Sacchi, R. (2013). Quantitative molecular phenotyping of gill

remodeling in a cichlid fish responding to salinity stress. Mol. Cell. Proteomics 12, 3962-

3975.

Kultz, D., Li, J., Sacchi, R., Morin, D., Buckpitt, A. and Van Winkle, L. (2015). Alterations in

the proteome of the respiratory tract in response to single and multiple exposures to

naphthalene. Proteomics 15, 2655-2668.

LeBlanc, D. M., Wood, C. M., Fudge, D. S. and Wright, P. A. (2010). A fish out of water: gill

and skin remodeling promotes osmo-and ionoregulation in the mangrove killifish

Kryptolebias marmoratus. Physiol. Biochem. Zool. 83, 932-949.

Lyndon, A. R. and Houlihan, D. F. (1998). Gill protein turnover: costs of adaptation. Comp.

Biochem. Physiol. 119, 27-34.

MacKenna, D. A., Omens, J. H., McCulloch, A. D. and Covell, J. W. (1994). Contribution of

collagen matrix to passive left ventricular mechanics in isolated rat hearts. Am. J. Physiol.

Heart Circ. Physiol. 266, 1007-1018.

Martin, R. B. (2007). The importance of mechanical loading in bone biology and medicine. J.

Musculoskelet. Neuronal Interact. 7, 48-53.

McBride, S. H. and Silva, M. J. (2012). Adaptive and injury response of bone to mechanical

loading. Bonekey Rep. 1, 192.

Meyer, A. (1987). Phenotypic plasticity and heterochrony in Cichlasoma managuense (Pisces,

Cichlidae) and their implications for speciation in cichlid fishes. Evolution 41, 1357-1369.

Nilsson, G. E., Dymowska, A. and Stecyk, J. A. W. (2012). New insights into the plasticity of

gill structure. Respir. Physiol. Neurobiol. 184, 214-222.

Ong, K. J., Stevens, E. D. and Wright, P. A. (2007). Gill morphology of the mangrove killifish

(Kryptolebias marmoratus) is plastic and changes in response to terrestrial air exposure. J.

Exp. Biol. 210, 1109-1115.

Ord, T. J. and Cooke, G. M. (2016). Repeated evolution of amphibious behavior in fish and its

implications for the colonization of novel environments. Evolution 70, 1747-1759.

Oxlund, H., Barckman, M., Ortoft, G. and Andreassen, T. T. (1995). Reduced concentrations of

collagen cross-links are associated with reduced strength of bone. Bone 17, 365-371.

Pardo, S. J., Patel, M. J., Sykes, M. C., Platt, M. O., Boyd, N. L., Sorescu, G. P., Xu, M., van

Loon, J. J. W. A., Wang, M. D. and Jo, H. (2005). Simulated microgravity using the

Random Positioning Machine inhibits differentiation and alters gene expression profiles of

2T3 preosteoblasts. Am. J. Physiol. Cell Physiol. 288, C1211-C1221.

Parenti, L. R. (1986). The phylogenetic significance of bone types in euteleost fishes. Zool. J.

Linn. Soc. 87, 37-51.

Patel, M. J., Liu, W., Sykes, M. C., Ward, N. E., Risin, S. A., Risin, D. and Jo, H. (2007).

Identification of mechanosensitive genes in osteoblasts by comparative microarray studies

using the rotating wall vessel and the random positioning machine. J. Cell. Biochem. 101,

587-599.

Pigliucci, M., Murren, C. J. and Schlichting, C. D. (2006). Phenotypic plasticity and evolution by

genetic assimilation. J. Exp. Biol. 209, 2362-2367.

Porter, M. E., Beltrán, J. L., Koob, T. J. and Summers, A. P. (2006). Material properties and

biochemical composition of mineralized vertebral cartilage in seven elasmobranch species

(Chondrichthyes). J. Exp. Biol. 209, 2920-2928.

Prideaux, M., Findlay, D. M. and Atkins, G. J. (2016). Osteocytes: the master cells in bone

remodelling. Curr. Opin. Pharmacol. 28, 24-30.

Rahmann, H. and Anken, R. H. (2002). Gravity related research with fishes- perspectives in

regard to the upcoming international space station, ISS. Adv. Space Res. 30, 697-710.

Rea, G., Cristofaro, F., Pani, G., Pascucci, B., Ghuge, S. A., Corsetto, P. A., Imbriani, M., Visai,

L. and Rizzo, A. M. (2016). Microgravity-driven remodeling of the proteome reveals

insights into molecular mechanisms and signal networks involved in response to the space

flight environment. J. Proteomics 137, 3-18.

Reilly, B. D. and Franklin, C. E. (2016). Prevention of muscle wasting and osteoporosis: the

value of examining novel animal models. J. Exp. Biol. 219, 2582-2595.

Renn, J. and Winkler, C. (2010). Characterization of collagen type 10a1 and osteocalcin in early

and mature osteoblasts during skeleton formation in medaka. J. Appl. Ichthyol. 26, 196-201.

Reynolds, W. W. (1977). Skeleton weight allometry in aquatic and terrestrial vertebrates.

Hydrobiologia 56, 35-37.

Ricard-Blum, S., Dublet, B. and van der Rest, M. (2000). Unconventional Collagens. Oxford:

Oxford University Press.

Ricci, C. and Boschetti, C. (2003). Bdelloid rotifers as model system to study developmental

biology in space. Adv. Space Biol. Med. 9, 25-39.

Rich, L. and Whittaker, P. (2005). Collagen and picrosirius red staining: a polarized light

assessment of fibrillar hue and spatial distribution. J. Morphol. Sci. 22, 97-104.

Robling, A. G., Burr, D. B. and Turner, C. H. (2001). Recovery periods restore

mechanosensitivity to dynamically loaded bone. J. Exp. Biol. 204, 3389-3399.

Robling, A. G., Castillo, A. B. and Turner, C. H. (2006). Biomechanical and molecular

regulation of bone remodeling. Annu. Rev. Biomed. Eng. 8, 455-498.

Rosati, R., Horan, G. S. B., Pinero, G. J., Garofalo, S., Keene, D. R., Horton,W. A., Vuorio, E.,

de Crombrugghe, B. and Behringer, R. R. (1994). Normal long bone growth and

development in type X collagen-null mice. Nature Genet. 8, 129-135.

Rotsch, C. and Radmacher, M. (2000). Drug-induced changes of cytoskeletal structure and

mechanics in fibroblasts: an atomic force microscopy study. Biophys. J. 78, 520-535.

Rubin, C., Turner, A. S., Bain, S., Mallinckrodt, C. and McLeod, K. (2001). Low mechanical

signals strengthen long bones. Nature 412, 603-604.

Saito, M. and Marumo, K. (2010). Collagen cross-links as a determinant of bone quality: a

possible explanation for bone fragility in aging, osteoporosis, and diabetes mellitus.

Osteoporos. Int. 21, 195-214.

Schilder, R. J. (2016). (How) do animals know how much they weigh? J. Exp. Biol. 219, 1275-

1282.

Schmidt-Nielsen, K. (1984). Scaling: Why is Animal Size so Important. Cambridge: Cambridge

University Press.

Schneider, C. A., Rasband,W. S. and Eliceiri, K.W. (2012). NIH Image to ImageJ: 25 years of

image analysis. Nat. Methods 9, 671-675.

Shahar, R. and Dean, M. N. (2013). The enigmas of bone without osteocytes. BoneKEy Rep. 2,

343.

Shen, G. (2005). The role of type X collagen in facilitating and regulating endochondral

ossification of articular cartilage. Orthod. Craniofacial. Res. 8, 11-17.

Standen, E. M., Du, T. Y. and Larsson, H. C. E. (2014). Developmental plasticity and the origin

of tetrapods. Nature 513, 54-58.

Sugiyama, T., Meakin, L. B., Browne,W. J., Galea, G. L., Price, J. S. and Lanyon, L. E. (2012).

Bones’ adaptive response to mechanical loading is essentially linear between the low strains

associated with disuse and the high strains associated with the lamellar/woven bone

transition. J. Bone Miner. Res. 27, 1784-1793.

Tajima, O., Ashizawa, N., Ishii, T., Amagai, H., Mashimo, T., Liu, L. J., Saitoh, S., Tokuyama,

K. and Suzuki, M. (2000). Interaction of the effects between vitamin D receptor

polymorphism and exercise training on bone metabolism. J. Appl. Physiol. 88, 1271-1276.

Tatarenkov, A., Ring, B. C., Elder, J. F., Bechler, D. L. and Avise, J. C. (2010). Genetic

composition of laboratory stocks of the self-fertilizing fish Kryptolebias marmoratus: a

valuable resource for experimental research. PLoS ONE 5, e12863.

Taylor, D. S. (2012). Twenty-four years in the mud: what have we learned about the natural

history and ecology of the mangrove rivulus, Kryptolebias marmoratus? Integr. Comp. Biol.

52, 724-736.

Turko, A. J., Cooper, C. A. and Wright, P. A. (2012). Gill remodelling during terrestrial

acclimation reduces aquatic respiratory function of the amphibious fish Kryptolebias

marmoratus. J. Exp. Biol. 215, 3973-3980.

Turko, A. J., Robertson, C. E., Bianchini, K., Freeman, M. and Wright, P. A. (2014). The

amphibious fish Kryptolebias marmoratus uses different strategies to maintain oxygen

delivery during aquatic hypoxia and air exposure. J. Exp. Biol. 217, 3988-3995.

Turner, C. H. (1998). Three rules for bone adaptation to mechanical stimuli. Bone 23, 399-407.

Ueki, M., Tanaka, N., Tanimoto, K., Nishio, C., Honda, K., Lin, Y.-Y., Tanne, Y., Ohkuma, S.,

Kamiya, T., Tanaka, E. et al. (2008). The effect of mechanical loading on the metabolism of

growth plate chondrocytes. Ann. Biomed. Eng. 36, 793-800.

Vogel, S. (1984). Life in Moving Fluids: The Biology of Flow. Princeton, NJ: Princeton

University Press.

Wess, T. J. (2008). Collagen fibrillar structure and hierarchies. In Collagen Structure and

Mechanics (ed. P. Fratzl), pp. 49-80. New York: Springer.

Wilson, J. M. and Laurent, P. (2002). Fish gill morphology: inside out. J. Exp. Zool. 293, 192-

213.

Witten, P. E. and Hall, B. K. (2015). Teleost skeletal plasticity: modulation, adaptation, and

remodelling. Copeia 103, 727-739.

Witten, P. E. and Huysseune, A. (2009). A comparative view on mechanisms and functions of

skeletal remodelling in teleost fish, with special emphasis on osteoclasts and their function.

Biol. Rev. 84, 315-346.

Wright, P. A. (2012). Environmental physiology of the mangrove rivulus, Kryptolebias

marmoratus, a cutaneously breathing fish that survives for weeks out of water. Integr.

Comp. Biol. 52, 792-800.

Wright, P. A. and Turko, A. J. (2016). Amphibious fishes: evolution and phenotypic plasticity. J.

Exp. Biol. 219, 2245-2259.

Wuest, S. L., Richard, S., Kopp, S., Grimm, D. and Egli, M. (2015). Simulated microgravity:

critical review on the use of random positioning machines for mammalian cell culture.

Biomed. Res. Int. 2015, 971474.

Yang, X., Vezeridis, P. S., Nicholas, B., Crisco, J. J., Moore, D. C. and Chen, Q. (2006).

Differential expression of type X collagen in a mechanically active 3-D chondrocyte culture

system: a quantitative study. J. Orthop. Surg. 1, 15.

Figures

Figure 3.1. Mechanical testing of mangrove rivulus gill arches. (A) Representative video frames, 8 s apart, showing the glass microbeam force transducer used to measure stiffness of isolated gill arches; the vertical dashed line indicates the starting position of the microbeam for comparison between frames. Scale bar, 1 mm. (B) Representative force–extension curves for one individual gill arch from a water-acclimated fish (grey circles) and a terrestrially acclimated fish

(black triangles) used to calculate mechanical properties. Gill extension was calculated as the change in length (ΔL) divided by the initial length (L0).

Figure 3.2. Increased gill stiffness after terrestrial acclimation is reversible in Kryptolebias marmoratus. (A) Force required to deform all four gill arches from the left side of control fish in water (n=15), after 7 days (n=13) or 14 days in air (n=13), or 14 days in air followed by 14 days

(n=12) or 12 weeks (n=12) recovery in water. (B) Force required to deform gill arches 1 and 2 in fish acclimated to water at 1 g (n=12), 7 days in air at 1 g (n=12) or 7 days in simulated microgravity (0.06 g; n=10). Different letters above treatment groups denote significant effects of treatment (two-way repeated-measures ANOVA and Holm–Sidak post hoc tests, P≤0.001).

Figure 3.3. Increased stiffness of K. marmoratus versus Danio rerio gill arches. Force required to deform gill arches from the left side of amphibious K. marmoratus acclimated to water (n=15) compared with those of the fully aquatic zebrafish (n=9). The K. marmoratus data presented here are the control values repeated from Fig. 2A for comparison. Different letters above treatment groups denote significant differences between species (two-way repeated- measures ANOVA and Holm–Sidak post hoc tests, P=0.002).

Figure 3.4. Relative abundances of structural proteins that significantly changed expression after 2 weeks of terrestrial acclimation. Each column represents a single individual and each row a unique protein (control n=6, 14 days in air n=6). Red rectangles represent proteins that increased expression after terrestrial acclimation and blue represents proteins that were less abundant. Relative protein abundances were z-score transformed to normally distribute values within each row and make it easier to visualize differences between treatments.

Figure 3.5. Increased density of collagen molecules in K. marmoratus gill arches after terrestrial acclimation. (A) Boxplot showing the mean hue value of collagen observed in ceratobranchial bone cross-sections of control fish in water (n=15), after 7 days (n=13) or 14 days in air (n=13), or 14 days in air followed by 14 days (n=12) or 12 weeks (n=12) recovery in water. Larger values indicate larger and more densely packed collagen fibrils. Different letters above treatment groups denote significant effects of treatment (two-way repeated-measures

ANOVA and Holm–Sidak post hoc tests, P=0.004). (B) Representative image of a control

Picrosirius-Red-stained gill arch under polarized light used to measure collagen hue. Scale bar,

100 μm. (C,D) Representative transmission electron micrograph of a ceratobranchial bone from a control fish in water (C) and from a fish acclimated terrestrially for 14 days (D). Note the dark band of mineralized bone (arrow) and the especially dense packing of collagen (c) beneath the osteoblast (o) along the outer margin of the bone (b) in the terrestrially acclimated fish. Scale bar, 1 μm.

100

Figure 3.6. Increased collagen density in mangrove rivulus gill filaments after terrestrial acclimation. (A) Boxplot showing the hue value of collagen measured in gill filament cross- sections; larger values indicate larger and more densely packed collagen fibrils (n=6 control and n=6 terrestrially acclimated fish; two-way repeated-measures ANOVA and Holm–Sidak post hoc tests P<0.001). (B,C) Representative transmission electron micrograph of a filament-supporting cartilaginous rod from a control fish in water (B) and from a fish acclimated terrestrially for 14 days (C). Note the diffuse and disorganized collagen (c) alongside the chondrocyte (ch) in the water-acclimated fish, which becomes more densely packed and organized after terrestrial acclimation. Scale bar, 1 μm.

101

Figure 3.7. Calcium content of K. marmoratus gill arches. (A) Boxplot showing the mean calcium content of gill baskets (n=4 samples for each time point; 5 gill baskets pooled per sample) from K. marmoratus acclimated to water (control) or over a time course of air exposure and recovery (one-way ANOVA P=0.045, but for all Holm–Sidak post hoc tests, P>adjusted critical P). (B) Transmission electron micrograph of a ceratobranchial bone cross-section of a control fish held in water, showing only a few dark mineralized bands (arrowheads). Diffuse collagen molecules (c) and osteoblasts (o) can be observed at the margin of the bone. Scale bar, 1

μm.

102

Supplementary Material

Table 3.S1. Abundances of all collagen isoforms identified in the gills of Kryptolebias marmoratus acclimated to water or 14d in air.

Relative Coverage NCBI Accession Protein annotation #Peptides #Unique abundance P value (%) (air/water) gi|1041063337 collagen alpha-2(I) chain 67 80 75 1.03 0.2041737945 gi|1041090756 collagen alpha-1(I) chain-like isoform X1 55 73 67 1.3 0.0000426580 gi|1041065690 collagen alpha-1(I) chain-like 50 70 66 1.33 0.0000012589 gi|1041073117 collagen alpha-1(II) chain isoform X2 12 13 10 1.09 0.0398107171 gi|1041074737 collagen alpha-1(II) chain-like isoform X2 4 5 2 0.89 0.0002398833 gi|1041075806 collagen alpha-2(IV) chain isoform X2 2 3 2 0.88 0.0028183829 gi|1041145053 collagen alpha-4(IV) chain-like 1 1 1 1.26 0.0000013804 gi|1041075808 collagen alpha-1(IV) chain 3 4 3 1.65 0.0001819701 gi|1041138366 collagen alpha-6(IV) chain-like 2 4 3 1.23 0.0034673685 gi|1041132849 collagen alpha-2(V) chain 8 12 9 0.99 0.7762471166 gi|1041125201 collagen alpha-1(V) chain-like isoform X2 5 9 6 1.2 0.0000012882 gi|1041106386 collagen alpha-1(VI) chain 33 33 32 1.24 0.0000676083 gi|1041145385 collagen alpha-3(VI) chain partial 27 56 55 1.2 0.0001096478 gi|1041106315 collagen alpha-2(VI) chain-like 33 37 33 1.18 0.4897788194 gi|1041145452 collagen alpha-3(VI) chain-like partial 23 39 37 1.1 0.0257039578 gi|1041113033 collagen alpha-2(VI) chain-like isoform X2 20 19 18 1.01 0.4677351413 gi|1041138699 collagen alpha-3(VI) chain-like partial 17 25 25 1.05 0.1348962883 gi|1041145951 collagen alpha-4(VI) chain-like partial 23 15 15 1.19 0.1621810097 gi|1041141506 collagen alpha-2(V) chain-like 12 12 11 1.13 0.0000933254 gi|1041063832 collagen alpha-2(IX) chain 5 3 1 0.61 0.0263026799 gi|1041119925 collagen alpha-1(IX) chain-like isoform X3 2 1 1 2.37 0.0000000000 gi|1041126807 collagen alpha-1(IX) chain-like 2 1 1 0.77 0.3801893963 gi|1041074889 collagen alpha-3(IX) chain isoform X1 2 2 1 0.89 0.0000000000 gi|1041062211 collagen alpha-1(X) chain-like 2 1 1 8.84 0.0000000000 gi|1041144154 collagen alpha-1(X) chain-like 3 1 1 1.69 0.0000000000 gi|1041118682 collagen alpha-1(XI) chain-like isoform X2 6 9 5 0.95 0.2398832919 gi|1041088438 collagen alpha-1(XI) chain-like 8 13 8 1.26 0.0011748976 gi|1041059509 collagen alpha-1(XI) chain-like 3 5 1 1.02 0.0000000000 gi|1041061459 collagen alpha-1(XII) chain isoform X2 25 72 65 0.74 0.0004365158 gi|1041140211 collagen alpha-1(XII) chain-like partial 17 19 18 0.98 0.1778279410 gi|1041146700 collagen alpha-1(XII) chain-like isoform X1 8 2 1 2.35 0.0000000000 gi|1041147096 collagen alpha-1(XII) chain-like 4 1 1 0.88 0.0056234133 gi|1041146058 collagen alpha-1(XII) chain-like partial 6 6 3 1.16 0.0025703958 gi|1041148375 collagen alpha-1(XII) chain-like partial 6 2 2 1.26 0.8511380382

103

Figure 3.S1. Increased gill stiffness (calculated as a spring constant from force/length curves) after terrestrial acclimation is reversible in K. marmoratus. Force relative to gill arch length required to deform all four gill arches from the left side of control fish in water (n=15), after 7d (n=13) or 14d in air (n=13), or 14d in air followed by 14d (n=12) or 12 weeks (n=12) recovery in water. Different letters above treatment groups denote significant effects of treatment

(two-way repeated-measures ANOVA and Holm-Sidak post hoc tests, P<0.001).

104

Figure 3.S2. Activity of mangrove rivulus in the simulated microgravity experiment.

Control fish in water (black line; n=6) reduced activity to levels comparable to terrestrially acclimated fish after 3 d of acclimation to a 6-well plate. Terrestrially-acclimated fish moved infrequently under both normal gravity conditions (dark grey line; n=6), and when exposed to simulated microgravity in the random positioning machine (light grey line; n=6). All fish were fasted for the duration of the experiment. Asterisks denote significant differences between water- acclimated and each terrestrially-acclimated group within a time point (repeated measures

ANOVA and Holm-Sidak post hoc tests, treatment*time P<0.05).

105

Figure 3.S3. Increased gill stiffness (calculated as a spring constant from force/length curves) after terrestrial acclimation in K. marmoratus does not occur in simulated microgravity. Force relative to gill arch length required to deform gill arches 1 and 2 in fish acclimated to water at 1g (n=12), 7d in air at 1g (n=12), or 7d in simulated microgravity (0.06g; n=10). Different letters above treatment groups denote significant effects of treatment (two-way repeated-measures ANOVA and Holm-Sidak post hoc tests, P=0.025).

106

Figure 3.S4. Increased gill stiffness (calculated as a spring constant from force/length curves) of K. marmoratus versus D. rerio gill arches. Force relative to gill arch length required to deform gill arches from the left side of amphibious K. marmoratus acclimated to water (n=15) compared to those of the fully aquatic zebrafish (n=9). The K. marmoratus data presented here are the control values repeated from Fig. 3.S1 for comparison. Different letters above treatment groups denote significant differences between species (two-way repeated-measures ANOVA and

Holm-Sidak post hoc tests, P=0.006).

107

Movie 3.1. Time lapse video of K. marmoratus during acclimation to microgravity in the random positioning machine. Images were captured every 2s and are played back at 60x speed.

Available online at: http://jeb.biologists.org/lookup/doi/10.1242/jeb.161638.supplemental

108

CHAPTER 4: GILL REMODELLING DURING TERRESTRIAL ACCLIMATION IN THE PRIMITIVE

AMPHIBIOUS FISH POLYPTERUS SENEGALUS

Submitted to the Journal of Morphology as: Turko, A. J., Maini, P., Wright, P. A. and Standen,

E. M. Gill remodelling during terrestrial acclimation in the primitive amphibious fish

Polypterus senegalus. Manuscript ID: JMOR-18-0145

109

Abstract

Fishes are effectively weightless in water due to the buoyant support of the environment, but amphibious fishes must cope with increased effective weight when on land. Delicate structures such as gills are especially vulnerable to collapse and loss of surface area out of water. We tested the “structural support” hypothesis that amphibious Polypterus senegalus solve this problem using phenotypically plastic changes that provide mechanical support and increase stiffness at the level of the gill lamellae, the filaments, and the whole arches. After 7 d in terrestrial conditions, enlargement of an inter-lamellar cell mass filled the water channels between gill lamellae, probably to provide structural support and/or reduce evaporative water loss. Similar gill remodelling has been described in several other diverse clades of fishes, suggesting this may be an ancestral character of all Actinopterygians. There was no change in the mechanical properties or collagen composition of filaments or arches after 7 days out of water, but 8 months of terrestrial acclimation caused a reduction in gill arch length and bone volume. These gill filament and arch results do not provide evidence for the “structural support” hypothesis, and instead suggest that P. senegalus reduce investment in the gills when on land. This strategy may save energy, but could have negative consequences for aquatic respiration when fish return to water.

These results are also strikingly similar to the evolutionary trend of gill loss that occurred during the tetrapod invasion of land, raising the possibility that genetic assimilation of gill plasticity was an underlying mechanism.

110

4.1 Introduction

The different physical properties of air and water make the invasion of land by aquatic animals one of the most dramatic transitions in the evolutionary history of vertebrates. Despite the challenges, over 20 clades of extant fishes have independently evolved amphibious lifestyles

(Ord & Cooke, 2016; Wright & Turko, 2016). Aquatic fishes are effectively weightless due to the buoyant support of the surrounding water, but when on land amphibious fishes must support their body weight against the force of gravity. The effects of gravity are especially problematic for fish gills, which collapse and coalesce in the absence of buoyant support from water

(Graham, 1997; Long & Gordon, 2004). Collapse of the gills reduces their effective surface area, potentially impairing branchial respiration, osmoregulation, and nitrogenous waste excretion

(Sayer, 2005; Wright, 2012).

In water-breathing fishes, the gill filaments and lamellae are typically long and thin to maximize surface area and minimize diffusion distances between water and blood. However, gill filaments and lamellae of amphibious fishes tend to be relatively short and stout, which probably serves to prevent collapse and coalescence out of water and thus maintains some degree of gill function (Graham, 1997; Hughes, 1984). Additional morphological adaptations are seen in some amphibious fishes. The gill filaments of the highly terrestrial mudskipper Periophthalmodon schlosseri are intricately branched, which is thought to prevent tight packing and coalescence out of water thus maintaining channels for air flow through the gills (Low, Lane, & Ip, 1988).

Furthermore, adjacent lamellae are fused at several points along their length which produces a collapse-resistant lattice-like structure. Similar lamellar fusions are present in the distantly related bowfin Amia calva, and probably serve the same function (Daxboeck, Barnard, &

Randall, 1981). In the climbing perch Anabas testudineus and other anabantoid fishes, both

111 filaments and lamellae are fused into an elaborate labyrinthine organ that is used to breathe air

(Munshi, 1968; Munshi, Olson, Ojha, & Ghosh, 1986). These structural adaptations are all constitutively expressed, and little is known about whether gill morphology in amphibious fishes can also be shaped by phenotypic plasticity (Wright & Turko, 2016).

Phenotypically plastic structural responses to terrestrial exposure in the gills have only been reported in a single amphibious fish species. In mangrove rivulus, Kryptolebias marmoratus, the gill arches are reversibly stiffened after 7 days out of water via increased collagen synthesis and bone mineralization (Turko et al., 2017). An increased density of collagen fibrils was also observed in the cartilage rods that support the gill filaments. Terrestrially-acclimated mangrove rivulus also reversibly enlarge an inter-lamellar cell mass (ILCM), epithelial tissue that fills the water channels between lamellae (LeBlanc, Wood, Fudge, & Wright, 2010; Ong, Stevens, &

Wright, 2007). Inter-lamellar channels are similarly plugged with mucous in the gills of African lungfish Protopterus annectens during aestivation out of water (Sturla, Paola, Carlo, Angela, &

Maria, 2002). These responses may provide structural support for the lamellae, or may limit evaporative water loss via reduction of functional gill surface area (Nilsson, Dymowska, &

Stecyk, 2012; Wright, 2012). Modification of gill surface area via an ILCM also occurs in several other groups of fully aquatic fishes, including cyprinids (Dhillon et al., 2013; Sollid, De

Angelis, Gundersen, & Nilsson, 2003; Tzaneva, Bailey, & Perry, 2011; Wu et al., 2017), catfishes (Phuong, Huong, Nyengaard, & Bayley, 2017), stickleback (Gibbons, McBryan, &

Schulte, 2018), and salmonids (Blair, Matheson, He, & Goss, 2016). Given this diversity, one possibility is that ILCM-based gill remodelling may be an ancestral trait of all fishes (Nilsson,

2007; Nilsson et al., 2012). However, no data exist for the most basal clades, which would provide a strong test of the ancestral origin hypothesis.

112

The objective of our study was to test whether the gills of the basal Actinopterygian

Polypterus senegalus are phenotypically plastic in response to terrestrial acclimation. The order

Polypteriformes is the sister group to all other ray-finned fishes, and contains several amphibious species including P. senegalus (Du, Larsson, & Standen, 2016). These fish possess lungs for air breathing, can survive out of water for 8 months, and the musculoskeletal system of the pectoral girdle is remodelled after terrestrial acclimation to improve locomotor performance (Du &

Standen, 2017; Standen, Du, & Larsson, 2014). To test the hypothesis that phenotypic plasticity is also important to support the gills out of water, we examined structural changes at the level of the gill lamellae, filaments, and arches. Specifically, we tested three predictions of the structural support hypothesis: terrestrial acclimation should cause (1) enlargement of an ILCM between lamellae, (2) increased stiffness and collagen content of gill filaments, and (3) increased stiffness, collagen content and bone volume of the gill arch skeleton.

4.2 Materials and Methods

4.2.1 Experimental Animals

Juvenile Polypterus senegalus (61.8 ± 1.0 mm, 1.18 ± 0.08 g, means ± SE) were obtained commercially (AQUAlity Tropical Fish Wholesale Inc., Mississauga, ON, Canada) and housed under conditions (recirculating filtered City of Ottawa tap water, 25.5 ± 1.1 °C, 12 h light : 12 h dark photoperiod) described previously (Standen et al., 2014). Fish were fed a high protein diet daily, and a 10% by volume water was made twice a week. Fish were held under fully aquatic conditions with no opportunity to leave water prior to experimentation. All experiments were approved by the University of Ottawa Animal Care Committee (University of Ottawa protocol

#BL-2069).

113

4.2.2 Experimental Protocol

Two series of experiments were conducted to assess plasticity of gill morphology over different timescales. First, we tested whether the gills of P. senegalus are flexible in response to relatively short term (7 days) terrestrial acclimation. We used histology to examine structural changes at all three levels of gill organization (lamellae, filaments, and arches), combined with mechanical testing of the gill filaments and arches. Second, to assess whether chronic acclimation (8 months) to terrestrial conditions influenced the size and structure of the gill arches, we capitalized on previously collected micro-computed tomography (micro-CT) data

(Standen et al. 2014).

4.2.3 Series I – Short-term terrestrial acclimation

Fish were held in water under standard laboratory conditions (control, n = 13), or terrestrially acclimated (n = 13) for 7 days as described previously (Standen et al., 2014).

Animals were held in flat-bottomed mesh baskets directly above filtered aquarium water, and humidity was maintained near 100% with an ultrasonic Exo Terra mini fogger. Live plants and small pieces of plastic pipe were included to provide cover and reduce stress.

4.2.3.1 Mechanical testing

Gill baskets were carefully dissected from freshly euthanized fish (250 mg/L buffered tricaine methanesulfonate) in aquarium water containing a protease inhibitor (0.1 mM phenylmethylsulfonyl fluoride) to minimize tissue degradation prior to testing. All tests were conducted within one hour of euthanasia. All four arches from the left side and three gill

114 filaments from the first gill arch on the right side were used for tensile testing. To isolate single filaments without damaging their attachment to the gill arch, the ceratobranchial bone was first cut into three sections (dorsal, middle, ventral). The distal halves of all but one of the filaments per section were then carefully trimmed away, leaving a single protruding filament. For mechanical testing, the tip of the intact filament and the attached ceratobranchial bone segment were used as attachment points. All four gill arches from the left side were used to test the stiffness of the ceratobranchial bones. The arches were carefully separated by cutting through the epibranchial and basibranchial bones, and soft tissue (gill filaments and overlying epithelial and connective tissue) was removed from each end to allow the ceratobranchial bones to be securely fastened to the testing apparatus.

Mechanical properties of gill filaments and arches were measured using a custom tensile testing apparatus described previously (Fudge, Gardner, Forsyth, Riekel, & Gosline, 2003; Turko et al., 2017). Briefly, gill tissue was mounted between a moveable micrometer and the tip of a glass microbeam with cyanoacrylate gel. The other end of the microbeam was held securely in place with epoxy resin. To conduct a trial, the micrometer was retracted which caused the microbeam to bend and the tissue to stretch and deform. All trials were video-recorded (Leica

M60 dissecting scope with an MC170 HD camera), and individual frames were subsequently analyzed in ImageJ (National Institutes of Health, Bethesda, MD, USA) to measure bending of the glass microbeam and extension of the gill tissue. Gill filaments were tested to failure, while gill arches were only tested until approximately 10% extension due to limitations of the testing apparatus.

The forces applied to the gill tissue were calculated using beam theory (Fudge et al., 2003), given the deflection and dimensions of the glass rod. Tissue extension was calculated as the

115 linear distance between the two attachment points, and thus includes both stretching and deformation. Variation in the size and shape of gill arches and filaments along their length made it impossible to calculate traditional material properties (i.e. standardized to cross-sectional area).

Thus, we instead calculated effective equivalents using the whole tissues and were careful to use animals of equivalent size in the control and terrestrially-acclimated treatments (standard length p = 0.47). Effective “stiffness” was calculated as the slope of the force-extension curve. For gill filaments, the entire curve (until failure) was used, while for gill arches only the initial portion of the curve (0-5% extension) was used as this section of the curve was always linear and best reflects the low forces applied by gravity. By stretching the gill filaments until they snapped, we were also able to measure the amount of force that caused failure (“strength”), maximum extension (“extensibility”), and work required to break the filament (“toughness”), which was calculated as the area under the force-extension curve (Kreplak & Fudge, 2006).

4.2.3.2 Histology

Immediately after mechanical tests were completed, tissues were fixed overnight in 10% neutral buffered formalin (4°C), decalcified for 1 h at room temperature (Cal-Ex, Fisher

Scientific), and stored in 70% ethanol. To aid tissue orientation, samples were double embedded: first in 2% agar under a dissecting microscope (Blewitt, Pogmore, & Talbot, 1982), and then routinely infiltrated with paraffin wax (Shandon Excelsior ES, Fisher Scientific). Cross-sections

(5 µm) were first taken through the middle of each ceratobranchial for collagen analysis. The paraffin was then melted and tissues re-oriented so that lamellar morphology could be examined in sagittal sections.

116

The density and thickness of collagen fibres in the gill arches and filaments was measured with picrosirius red dye (Electron Microscopy Sciences, Hatfield, PA, USA) as described previously (Johnson, Turko, Klaiman, Johnston, & Gillis, 2014; Turko et al., 2017). Briefly, sections were photographed under circularly polarized light so that collagen appears as a spectrum of green (diffuse, disorganized fibres) to red (densely packed fibres) while other tissues are blue-black (supporting information Figure 4.S1; Rich & Whittaker, 2005). Photographs were imported into ImageJ, separated into hue-saturation-brightness components, and the mean hue value was measured after normalizing the 256 colour histogram so that the weakest green had a value of 1 while the deepest red was equal to 155.

To quantify changes in lamellar length and ILCM coverage, sagittal sections were routinely stained with hematoxylin and eosin. Five lamellae per arch were randomly selected, and the length of each lamella (µm) and coverage by the adjacent ILCM (%) were measured (LeBlanc et al., 2010, Ong et al., 2007; Turko, Earley, & Wright, 2011).

4.2.4 Series II – Long-term terrestrial acclimation

To test whether chronic acclimation (8 months) to terrestrial conditions caused plastic changes to the gill arches of P. senegalus, we re-analyzed micro-CT scans originally used to examine the pectoral girdle (Standen et al. 2014). Experimental fish (control n = 6, terrestrial n =

16) were fixed in 4% paraformaldehyde and stored in 70% ethanol before being scanned in a

SkyScan 1172 micro-CT scanner. The raw tomography projection images were reconstructed into cross-sectional images with NRecon construction software (SkyScan), and gill arches were manually segmented in Amira (FEI Visualization Sciences Group). The length of each gill arch was measured by inserting six evenly-distributed reference markers along the length of the arch

117 in the centre of the cross-sectional area, and summing the total distance between these markers.

Curvature was calculated by dividing these lengths by the straight length (i.e. tip-to-tip) of each arch. Total bone volume of each gill arch was estimated as the sum of all voxels constituting the skeleton in micro-CT scans.

4.2.5 Statistical analysis

To test for phenotypically plastic responses to terrestrial acclimation, we compared the height of the gill lamellae, ILCM coverage, and mechanical properties of the gill filaments using

Student’s t-tests. The mechanical properties and collagen density of gill arches was compared with two-way repeated measures ANOVA to also test for differences among gill arches. To account for differences in body size after the long-term acclimation experiment, we used repeated measures ANCOVA to test for changes in gill arch length (covariate: body mass) and bone volume (covariate: arch length). Curvature was unaffected by body size (p = 0.18) and thus was analyzed using two-way repeated measures ANOVA. Data were natural logarithm (ln) transformed when necessary to meet assumptions of normality and equal variance, and

Greenhouse-Geisser corrections were used when ANCOVA sphericity assumptions were violated. RStudio (https://www.rstudio.com/) was used for all analyses (critical α=0.05).

Throughout the text, values are given as means ± SE.

4.3 Results

4.3.1 Short-term terrestrial acclimation

In control P. senegalus, the water channels between gill lamellae were distinct but approximately half-filled by an ILCM (Figure 4.1a). After 7 d of terrestrial acclimation, the

118

ILCM had enlarged by 30% and covered a significantly greater proportion of the gill lamellae (p

= 0.01, Figure 4.1b,c). Lamellar length was unaffected by terrestrial acclimation (p = 0.89,

Figure 1d).

Tensile testing of gill filaments produced force-extension curves that were relatively linear (Figure 4.2a). There were no obvious weak points within the filaments, as failure occurred abruptly and at variable locations along the length of the filament. Effective “stiffness”, the amount of force required to extend the filament, did not change after terrestrial acclimation (p =

0.995, Figure 4.2b). The amount of force required to break the filament was similarly unaffected

(p = 0.80, Figure 4.2c). Filaments were relatively elastic, and could be extended 36.7 ± 2.8% before failure. Terrestrial acclimation did not affect the extensibility (p = 0.92, Figure 4.2d), nor the amount of work required to reach failure (p = 0.36, Figure 4.2e). The hue of collagen in picrosirius red-stained sections through the filaments did not change after terrestrial acclimation

(p = 0.38, Figure 4.2f).

Force-extension curves from gill arches were linear at extensions below 5% (Figure

4.3a). Effective “stiffness” was the same across all four gill arches (p = 0.66), and did not change after terrestrial acclimation (p = 0.90, interaction p = 0.13, Figure 4.3b). Similarly, there was no difference in the hue of picrosirius red-stained collagen among gill arches (p = 0.46) or between conditions (p = 0.72, interaction p = 0.51, Figure 4.3c).

4.3.2 Long-term terrestrial acclimation

At the end of the 8-month acclimation period, P. senegalus raised under terrestrial conditions were significantly shorter (p < 0.001) and weighed less (p = 0.018) than control fish reared in aquatic conditions. Condition factor (Fulton’s K) was statistically similar between

119 groups (p = 0.13). The ceratobranchial bones of all four gill arches were clearly distinguishable in micro-CT scans (Figure 4.4a). After controlling for the significant effect of body mass using

ANCOVA (p < 0.001), all four gill arches were significantly shorter in terrestrially acclimated fish than controls held in water (p = 0.007, Figure 4.4b). The first gill arch was longest and the arches significantly decreased in size moving posteriorly (p < 0.001, Figure 4.4b). There was no significant treatment by gill arch interaction (p = 0.66). The volume of the gill arches (total number of voxels) was significantly related to arch length (p < 0.001). Therefore, to test whether terrestrial acclimation influenced bone volume irrespective of length, we used arch length as a covariate in an ANCOVA. After controlling for the effect of arch length, there was not a consistent effect of terrestrial acclimation on all four gill arches (interaction p < 0.001, Figure

4.4c). The volume of the first (Tukey’s post-hoc p = 0.005) and fourth (p = 0.01) gill arches was significantly smaller after terrestrial acclimation, while arches two and three were unaffected (p

= 0.99 and 0.69, respectively). Curvature of the gill arches was not affected by terrestrial acclimation (p = 0.34), but varied among arches (p = 0.004, interaction p = 0.93, Figure 4.4d).

4.4 Discussion

In the absence of buoyant support from the environment, we hypothesized that P. senegalus induce phenotypically plastic structural modifications of the gills to prevent collapse when out of water. In support of this hypothesis, we found that acclimation to terrestrial conditions caused enlargement of an ILCM that could prevent the lamellae from coalescing and/or reduce evaporative water loss. In contrast, we found no evidence for the prediction that the skeleton would grow and ossify to resist the force of gravity. There were no changes in the mechanical properties of the gill filaments or arches in fish held on land for 7 days. Furthermore,

120

8 months of terrestrial acclimation resulted in shorter gill arches and further decreases in bone volume. During long-term emersion, P. senegalus appear to be allocating resources away from gill tissue that is presumably sub- or non-functional out of water, given the presence of lungs for air-breathing. This example of phenotypic plasticity mirrors the general trend towards gradually smaller gill arches seen in the fossil record during the tetrapod invasion of land. It is possible that gill loss in these early tetrapods occurred via genetic assimilation of an initially plastic response.

4.4.1 The inter-lamellar cell mass

Terrestrial acclimation caused significant enlargement of the ILCM in P. senegalus.

Considering the gill lamellae of P. senegalus lack any obvious morphological adaptations to resist collapse out of water, such as distal connections present in some mudskippers and bowfin, the ILCM may play an important supportive role and maintain separation between adjacent lamellae. Only one other species of amphibious fish, K. marmoratus, is known to enlarge an

ILCM after terrestrial acclimation. Gill remodelling in K. marmoratus occurs over similar time scales as in P. senegalus (one week), and histologically the ILCM appears very similar in both species (Ong et al., 2007).

The ILCMs of aquatic control P. senegalus were relatively well-developed, covering more than half the length of the lamellae. In other words, the gill surface area of these fish was substantially smaller than it could be if the lamellae were fully exposed. Large gill surface area enhances respiratory gas exchange, but also imposes costs such as increased rates of ion loss in freshwater or increased chances for parasite attachment (Gonzalez, 2011; Gonzalez &

McDonald, 1992; Nilsson, 2007). Thus, by maintaining a relatively low gill surface area in water, P. senegalus can potentially reduce the costs of large gills without decreased respiratory

121 performance given the presence of lungs. Gas exchange across the gills versus lungs has never been measured in P. senegalus, but in the closely related reedfish Erpetoichthys calabaricus the gills account for about 30% of oxygen uptake, while the lungs and skin account for 40% and

30% respectively (Sacca & Burggren, 1982). Gill surface area is similarly reduced in several other air-breathing fishes with alternative air-breathing organs. For example, gill lamellae of the air-breathing striped catfish Pangasianodon hypothalamuss (Phuong et al., 2017) and armoured catfish Hypostomus aff. pyreneusi (Scott et al., 2017) are covered by an extensive ILCM under normoxic conditions.

Many fully aquatic fishes also reversibly modify gill surface area (see Introduction). Did gill remodelling in these phylogenetically diverse taxa evolve repeatedly, or is this an ancestral trait shared by all fishes? Our discovery of a modifiable ILCM in P. senegalus, a member of the most basal Actinopterygian clade, supports the hypothesis that gill remodelling is an ancient trait shared by all ray-finned fishes. However, we cannot rule out the possibility that gill remodelling has evolved repeatedly in fishes. To test this convergence hypothesis, further investigations of the mechanisms regulating ILCM development and the specific gill cell types that undergo proliferation in the ILCM of different fish species are required.

4.4.2 Plasticity of gill filaments and arches

We found no evidence for phenotypically plastic changes enhancing structural support to the gill filaments or arches of P. senegalus acclimated to terrestrial conditions for one week.

Mechanical properties of both tissues were similar in fish acclimated to water or land, and there was no indication of collagen remodelling in the extracellular matrix. Surprisingly, after eight months of terrestrial acclimation, gill arches were shorter relative to aquatic control fish, and

122 normalized bone volume was further reduced in two of the four arches. Curvature of the arches did not change, indicating that the observed reduction in length was not due to an overall change in shape. These results are contrary to the predictions made by the “structural support” hypothesis. Instead, P. senegalus appear to reduce investment in gill tissue that is presumably non-functional out of water, which may decrease energetic costs of growth and/or minimize the weight and supportive requirements of the gill basket.

The gill reductions strategy used by P. senegalus in this study contrasts strongly with our earlier work in K. marmoratus, which increase collagen deposition, mineralization, and stiffness of the gill arches after terrestrial acclimation (Turko et al. 2017). Thus, there does not appear to be a generalizable skeletal response to life out of water that is shared by independently-evolved groups of amphibious fishes. However, this result is consistent with other animal studies that have demonstrated substantial inter-specific variation in responsiveness to mechanical forces and gravitational loading (Reilly & Franklin, 2016). In addition to approximately 400 million years of phylogenetic divergence between mangrove rivulus and P. senegalus (Near et al., 2012), differences in the mode of aerial respiration (lung in P. senegalus versus skin and bucco- opercular cavity in K. marmoratus (Cooper, Litwiller, Murrant, & Wright, 2012; Turko,

Robertson, Bianchini, Freeman, & Wright, 2014)), and bone type (osteocytic in P. senegalus versus anosteocytic in K. marmoratus (Atkins, Milgram, Weiner, & Shahar, 2015)) may influence the gill tissue response to gravitational load.

If P. senegalus respond to terrestrial acclimation by minimizing energetic investment in the gills, it makes sense that no changes in material properties were observed within the first week. Degradation and removal of bone and extracellular matrix is a relatively slow process that typically takes several weeks to become measurable in mammalian studies (e.g. Warner et al.

123

2006, Gross, Poliachik, Prasad, & Bain, 2010). Lower metabolic rates in fishes versus mammals would presumably slow the process further. In K. marmoratus, stiffening of gill arches that occurred within 7 d of terrestrial acclimation was not reversible after 14 d back in water; stiffness eventually returned to control values after 16 weeks (Turko et al. 2017). Rather than resulting from active degradation of the gill skeleton, the smaller ceratobranchial bones we measured in P. senegalus after long term terrestrial acclimation probably reflect a reduced branchial growth rate relative to that of the rest of the body.

Bone growth is thought to require dynamic cycles of mechanical loading to stimulate collagen production and mineralization by osteoblasts (DiGirolamo, Kiel, & Esser, 2013; Fiaz,

Van Leeuwen, & Kranenbarg, 2010). Mechanical forces during feeding trigger bone growth in cichlid jaws (Witten & Hall, 2015), and swim training similarly increases bone formation and mineralization of Atlantic salmon Salmo salar (Totland et al., 2011) and zebrafish Danio rerio

(Fiaz et al., 2012; Suniaga et al., 2018). In water-breathing fishes, the gill arches and filaments would be subject to forces from flowing water, muscular contractions, and distortion caused by movement of the mouth and opercula during the ventilatory cycle. Our results suggest that this dynamic mechanical stimulation may be required for normal gill development in P. senegalus.

Mechanical loading by gravity in fishes out of water, on the other hand, would be largely static and may not promote bone development. There is even some evidence that static loading can decrease bone and cartilage growth, resulting in shorter bones (Cancel, Grimard, Thuillard-

Crisinel, Moldovan, & Villemure, 2009; Grodzinsky, Levenston, Jin, & Frank, 2000; Villemure et al., 2005). If this is the case in P. senegalus, differences in the static loads experienced among the gill arches may explain why we found differential responses in bone volume. After long term terrestrial acclimation, the first and fourth gill arches were significantly less dense than in aquatic

124 controls, while no difference was observed in arches two or three. The second and third arches are comprised of smaller ceratobranchial and enlarged hypobranchial bones connected end-to- end by cartilage that may act as a hinge, allowing these arches to deform and thus reduce compressive loading by gravity (see Figure 4a). On the other hand, the first and fourth arches are supported almost entirely by a single ceratobranchial bone and lack substantial cartilage content to allow deformation. Thus, these arches may have also experienced the most static load, which in turn may have caused the reduced bone volume that we measured.

4.4.3 Phenotypic plasticity and evolution

Phenotypic plasticity of extant amphibious fishes has been proposed as a model for studying the tetrapod invasion of land (Graham & Lee, 2004: Standen et al., 2014), because plastic changes in phenotype can lead to evolutionary change via genetic assimilation (Pigliucci,

Murren & Schlichting, 2006; West-Eberhard, 2003). This process occurs through inter- generational reductions in the environmental responsiveness of a given trait, and can occur regardless of whether the plasticity is adaptive or non-adaptive (Ghalambor, McKay, Carrol, &

Reznick, 2007). Polypteriform fishes such as P. senegalus are especially useful models in this regard, as these fish represent the most ancestral ray-finned fish lineage and bear a large resemblance to hypothesized early tetrapods (Standen et al., 2014). Both inhabit(ed) shallow swamps that periodically become hypoxic or dry up, requiring air-breathing or amphibious behaviour for survival (Du et al., 2016; Graham et al., 2014). Similarly, P. senegalus and the earliest tetrapods were both probably bimodal breathers, using both gills and lungs. During the tetrapod invasion of land, the major evolutionary trend at the level of the gills was reduction and eventual loss (Graham & Lee, 2004). For example, the largely aquatic Acanthostega had

125 filament-bearing gills that resemble those of P. senegalus (Coates & Clack, 1991), but the more terrestrial Ichthyostega was without gill arches entirely (Clack, 2002). The mechanism by which gills were lost by early tetrapods is unknown. However, our data showing a phenotypically plastic reduction of the gill arches after terrestrial acclimation in P. senegalus suggests that genetic assimilation may have played a key role. As early tetrapods spent more time out of water, heritable changes to the regulatory pathways controlling plasticity of gill arches may have reduced the scope of environmental sensitivity, leading to the constitutive expression of ever- smaller gills. Such a process would also necessitate the shifting of physiological processes that occur primarily across the gills of fishes, such as respiration, osmoregulation, and nitrogenous waste excretion, to other organs including the lungs and kidneys (Graham & Lee, 2004).

4.5 References

Atkins, A., Milgram, J., Weiner, S., & Shahar, R. (2015). The response of anosteocytic bone to

controlled loading. Journal of Experimental Biology, 218, 3559-3569.

Blair, S. D., Matheson, D., He, Y., & Goss, G. G. (2016). Reduced salinity tolerance in the

Arctic grayling (Thymallus arcticus) is associated with rapid development of a gill

interlamellar cell mass: implications of high-saline spills on native freshwater salmonids.

Conservation Physiology, 4, cow010.

Blewitt, E. S., Pogmore, T., & Talbot, I. C. (1982). Double embedding in agar/paraffin wax as an

aid to orientation of mucosal biopsies. Journal of Clinical Pathology, 35, 365.

Cancel, M., Grimard, G., Thuillard-Crisinel, D., Moldovan, F., & Villemure, I. (2009). Effects of

in vivo static compressive loading on aggrecan and type II and X collagens in the rat growth

plate extracellular matrix. Bone, 44, 306-315.

126

Clack, J. A. (2002). Gaining ground: the origin and evolution of tetrapods. Bloomington,

Indiana: Indiana University Press.

Coates, M. I., &. Clack, J. A. (1991). Fish-like gills and breathing in the earliest known tetrapod.

Nature 352, 234–236.

Cooper, C. A., Litwiller, S. L., Murrant, C. L., & Wright, P. A. (2012). Cutaneous vasoregulation

during short- and long-term aerial acclimation in the amphibious mangrove rivulus,

Kryptolebias marmoratus. Comparative Biochemistry and Physiology B, 161, 268-274.

Daxboeck, C., Barnard, D. K., & Randall, D. J. (1981). Functional morphology of the gills of the

bowfin, Amia calva L., with special reference to their significance during air exposure.

Respiration Physiology, 43, 349-364.

Dhillon, R. S., Yao, L., Matey, V., Chen, B.-J., Zhang, A.-J., Cao, Z.-D., Fu, S.-J., Brauner, C.

J.,Wang, Y. S., & Richards, J. G. (2013). Interspecific differences in hypoxia-induced gill

remodeling in carp. Physiological and Biochemical Zoology, 86, 727-739.

DiGirolamo, D. J., Kiel, D. P., & Esser, K. A. (2013). Bone and skeletal muscle: neighbors with

close ties. Journal of Bone and Mineral Research, 28, 1509-1518.

Du, T. Y., & Standen, E. M. (2017). Phenotypic plasticity of muscle fiber type in the pectoral

fins of Polypterus senegalus reared in a terrestrial environment. Journal of Experimental

Biology, 220, 3406-3410.

Du, T. Y., Larsson, H. C. E., & Standen, E. M. (2016). Observations of terrestrial locomotion in

wild Polypterus senegalus from Lake Albert, Uganda. African Journal of Aquatic Science, 41,

67-71.

127

Fiaz, A. W., Van Leeuwen, J. L., & Kranenbarg, S. (2010). Phenotypic plasticity and mechano‐

transduction in the teleost skeleton. Journal of Applied Ichthyology, 26, 289-293.

Fiaz, A. W., Léon-Kloosterziel, K. M., Gort, G., Schulte-Merker, S., van Leeuwen, J. L., &

Kranenbarg, S. (2012). Swim-training changes the spatio-temporal dynamics of

skeletogenesis in zebrafish larvae (Danio rerio). PLoS ONE 7, e34072.

Fudge, D. S., Gardner, K. H., Forsyth, V. T., Riekel, C., & Gosline, J. M. (2003). The

mechanical properties of hydrated intermediate filaments: insights from hagfish slime threads.

Biophysical Journal, 85, 2015-2027.

Ghalambor, C. K., McKay, J. K., Carroll, S. P., & Reznick, D. N. (2007). Adaptive versus

nonadaptive phenotypic plasticity and the potential for contemporary adaptation in new

environments. Functional Ecology, 21, 394–407.

Gibbons, T. C., McBryan, T. L., & Schulte, P. M. (2018). Interactive effects of salinity and

temperature acclimation on gill morphology and gene expression in threespine stickleback.

Comparative Biochemistry and Physiology A, 221, 55-62.

Gonzalez, R. J. (2011) The osmorespiratory compromise. In A. P. Farrell (Ed.), Encyclopaedia

of fish physiology: From genome to environment. (pp. 1389-1394). San Diego: Elsevier.

Gonzalez, R. J., & McDonald, D.G. (1992). The relationship between oxygen consumption and

ion loss in a freshwater fish. Journal of Experimental Biology, 163, 317–332.

Graham, J. B. (1997). Air-breathing fishes: evolution, diversity and adaptation. San Diego:

Academic Press.

Graham, J. B., & Lee, H. J. (2004). Breathing air in air: in what ways might extant amphibious

fish biology relate to prevailing concepts about early tetrapods, the evolution of vertebrate air

128

breathing, and the vertebrate land transition? Physiological and Biochemical Zoology, 77,

720-731.

Graham, J. B., Wegner, N. C., Miller, L. A., Jew, C. J., Chin Lai, N., Berquist, R. M., Frank, L.

R., & Long, J. A. (2014). Spiracular air breathing in polypterid fishes and its implications for

aerial respiration in stem tetrapods. Nature Communications 5, 3022.

Grodzinsky, A. J., Levenston, M. E., Jin, M., & Frank, E. H. (2000). Cartilage tissue remodeling

in response to mechanical forces. Annual Review of Biomedical Engineering, 2, 691-713.

Gross, T. S., Poliachik, S. L., Prasad, J., & Bain, S. D. (2010). The effect of muscle dysfunction

on bone mass and morphology. Journal of Musculoskeletal Neuronal Interactions, 10, 25-34.

Hughes, G. M. (1984). General anatomy of the gills. In W. S. Hoar, & D. J. Randall (Eds.), Fish

Physiology Vol 10A. (pp. 1-72). New York: Academic Press.

Johnson, A. C., Turko, A. J., Klaiman, J. M., Johnston, E. F., & Gillis, T. E. (2014). Cold

acclimation alters the connective tissue content of the zebrafish (Danio rerio) heart. Journal

of Experimental Biology, 217, 1868-1875.

Kreplak, L., & Fudge, D. (2006). Biomechanical properties of intermediate filaments: from

tissues to single filaments and back. BioEssays, 29, 26–35.

LeBlanc, D. M., Wood, C. M., Fudge, D. S., & Wright, P. A. (2010). A fish out of water: gill and

skin remodeling promotes osmo-and ionoregulation in the mangrove killifish Kryptolebias

marmoratus. Physiological and Biochemical Zoology, 83, 932-949.

Long, J. A., & Gordon, M. S. (2004). The greatest step in vertebrate history: A paleobiological

review of the fish-tetrapod transition. Physiological and Biochemical Zoology, 77, 700–719.

129

Low, W. P., Lane, D. J. W., & Ip, Y. K. (1988). A comparative study of terrestrial adaptations of

the gills in three mudskippers: Periophthalmus chrysospilos, Boleophthalmus boddaerti, and

Periophthalmodon schlosseri. Biological Bulletin, 175, 434-438.

Munshi, J. S. D. (1968). The accessory respiratory organs of Anabas testudineus (Bloch)

(Anabantidae, Pisces). Proceedings of the Linnean Society of London, 179, 107-126.

Munshi, J. S. D., Olson, K. R., Ojha, J., & Ghosh, T. K. (1986). Morphology and vascular

anatomy of the accessory respiratory organs of the air‐breathing climbing perch, Anabas

testudineus (Bloch). American Journal of Anatomy, 176, 321-331.

Near, T. J., Eytan, R. I., Dornburg, A., Kuhn, K. L., Moore, J. A., Davis, M. P., Wainwright, P.

C., Friedman, M., & Smith, W. L. (2012). Resolution of ray-finned fish phylogeny and timing

of diversification. Proceedings of the National Academy of Sciences USA, 109, 13698-13703.

Nilsson, G. E. (2007). Gill remodeling in fish – a new fashion or an ancient secret? Journal of

Experimental Biology, 210, 2403-2409.

Nilsson, G. E., Dymowska, A., & Stecyk, J. A. W. (2012). New insights into the plasticity of gill

structure. Respiratory Physiology and Neurobiology, 184, 214-222.

Ong, K. J., Stevens, E. D., & Wright, P. A. (2007). Gill morphology of the mangrove killifish

(Kryptolebias marmoratus) is plastic and changes in response to terrestrial air exposure.

Journal of Experimental Biology, 210, 1109-1115.

Ord, T. J., & Cooke, G. M. (2016). Repeated evolution of amphibious behavior in fish and its

implications for the colonization of novel environments. Evolution 70, 1747-1759.

130

Phuong, L. M., Huong, D. T. T., Nyengaard, J. R., & Bayley, M. (2017). Gill remodelling and

growth rate of striped catfish Pangasianodon hypophthalmus under impacts of hypoxia and

temperature. Comparative Biochemistry and Physiology A, 203, 288-296.

Pigliucci, M., Murren, C. J., & Schlichting, C. D. (2006). Phenotypic plasticity and evolution by

genetic assimilation. Journal of Experimental Biology, 209, 2362-2367.

Reilly, B. D., & Franklin, C. E. (2016). Prevention of muscle wasting and osteoporosis: the value

of examining novel animal models. Journal of Experimental Biology, 219, 2582-2595.

Robling, A. G., Duijvelaar, K. M., Geevers, J. V., Ohashi, N., & Turner, C. H. (2001).

Modulation of appositional and longitudinal bone growth in the rat ulna by applied static and

dynamic force. Bone, 29, 105-113.

Sacca, R., & Burggren, W. (1982). Oxygen uptake in air and water in the air-breathing reedfish

Calamoichthys calabaricus: role of skin, gills and lungs. Journal of Experimental Biology, 97,

179-186.

Sayer, M. D. J. (2005). Adaptations of amphibious fish for surviving life out of water. Fish and

Fisheries, 6, 186-211.

Scott, G. R., Matey, V., Mendoza, J. -A., Gilmour, K. M., Perry, S. F., Almeida-Val, V. M. F., &

Val, A. L. (2017). Air breathing and aquatic gas exchange during hypoxia in armoured

catfish. Journal of Comparative Physiology B, 187, 117-133.

Sollid, J., De Angelis, P., Gundersen, K., &Nilsson, G. E. (2003). Hypoxia induces adaptive and

reversible gross morphological changes in crucian carp gills. Journal of Experimental

Biology, 206, 3667-3673.

131

Standen, E. M., Du, T. Y., & Larsson, H. C. E. (2014). Developmental plasticity and the origin

of tetrapods. Nature, 513, 54-58.

Sturla, M., Paola, P., Carlo, G., Angela, M. M., & Maria, U. B. (2002). Effects of induced

aestivation in Protopterus annectens: A histomorphological study. Journal of Experimental

Zoology, 292, 26–31.

Suniaga, S., Rolvien, T., vom Scheidt, A., Fiedler, I. A. K., Bale, H. A., Huysseune, A., Witten,

P. E., Amling, M. & Busse, B. (2018). Increased mechanical loading through controlled

swimming exercise induces bone formation and mineralization in adult zebrafish. Scientific

Reports, 8, 3646.

Totland, G. K., Fjelldal, P. G., Kryvi, H., Løkka, G., Wargelius, A., Sagstad, A., Hansen, T. &

Grotmol, S. (2011). Sustained swimming increases the mineral content and osteocyte density

of salmon vertebral bone. Journal of Anatomy, 219, 490–501.

Turko, A. J., Earley, R. L., & Wright, P. A. (2011). Behaviour drives morphology: voluntary

emersion patterns shape gill structure in genetically identical mangrove rivulus. Animal

Behaviour, 82, 39-47.

Turko, A. J., Robertson, C. E., Bianchini, K., Freeman, M., & Wright, P. A. (2014). The

amphibious fish Kryptolebias marmoratus uses different strategies to maintain oxygen

delivery during aquatic hypoxia and air exposure. Journal of Experimental Biology, 217,

3988-3995.

Turko, A. J., Kültz, D., Fudge, D., Croll, R. P., Smith, F. M., Stoyek, M. R., & Wright, P. A.

(2017). Skeletal stiffening in an amphibious fish out of water is a response to increased body

weight. Journal of Experimental Biology, 220, 3621-3631.

132

Tzaneva, V., Bailey, S., & Perry, S. F. (2011). The interactive effects of hypoxemia, hyperoxia,

and temperature on the gill morphology of goldfish (Carassius auratus). American Journal of

Physiology, 300, R1344–R1351.

Villemure, I., Chung, M. A., Seck, C. S., Kimm, M. H., Matyas, J. R., & Duncan, N. A. (2005).

Static compressive loading reduces the mRNA expression of type ii and x collagen in rat

growth-plate chondrocytes during postnatal growth. Connective Tissue Research, 46, 211–

219.

Warner, S. E., Sanford, D. A., Becker, B. A., Bain, S. D., Srinivasan, S., & Gross, T. S. (2006).

Botox induced muscle paralysis rapidly degrades bone. Bone, 38, 257-264.

West-Eberhard, M. J. (2003). Developmental plasticity and evolution. Oxford: Oxford

University Press.

Witten, P. E., & Hall, B. K. (2015). Teleost skeletal plasticity: modulation, adaptation, and

remodelling. Copeia, 103, 727-739.

Wright, P. A. (2012). Environmental physiology of the mangrove rivulus, Kryptolebias

marmoratus, a cutaneously breathing fish that survives for weeks out of water. Integrative

and Comparative Biology, 52, 792-800.

Wright, P. A., & Turko, A. J. (2016). Amphibious fishes: evolution and phenotypic plasticity.

Journal of Experimental Biology, 219, 2245-2259.

Wu, C. -B., Liu, Z. -Y., Li, F. -G., Chen, J., Jiang, X. -Y., & Zou, S. -M. (2017). Gill remodeling

in response to hypoxia and temperature occurs in the hypoxia sensitive blunt snout bream

(Megalobrama amblycephala). Aquaculture, 479, 479-486.

133

Figures

Figure 4.1. Representative light micrographs of Polypterus senegalus gills stained with hematoxylin and eosin. Water channels between the gill lamellae of were moderately filled with an inter-lamellar cell mass (ILCM; outlined) under control conditions (a), and largely filled after

7 d of terrestrial acclimation (b). Scale bar = 50 µm. (c) Coverage of the gill lamellae by the adjacent ILCM. (d) Mean length of measured lamellae. The asterisk denotes a significant increase in terrestrially acclimated fish (p = 0.01).

134

Figure 4.2. Mechanical properties of Polypterus senegalus gill filaments after short-term (7 d) terrestrial acclimation. (a) Representative force-extension curve of a single filament from a control fish used to calculate mechanical properties. (b) Effective “stiffness” was calculated as the slope of the curve. (c) Force at failure and (d) extension at failure were the largest values of applied force and extension that were measured in each trial. (e) Effective “toughness” was calculated as the area under the curve. (f) Hue values of collagen in longitudinal filament sections stained with picrosirius red. Mechanical properties and collagen hue were not affected by terrestrial acclimation (all p > 0.35).

135

Figure 4.3. Mechanical properties of Polypterus senegalus gill arches after short-term (7 d) terrestrial acclimation. (a) Representative force-extension curve of a single gill arch from a control fish. Extension of the gill arch was calculated as the change in length divided by initial length, and effective “stiffness” was calculated as the slope of the curve. (b) Effective “stiffness” of gill arches from fish in water (n = 14) or after terrestrial acclimation (n = 13). (c) Hue values of collagen in ceratobranchial cross-sections stained with picrosirius red. Mechanical properties and collagen hue did not differ among gill arches or between treatments (all p > 0.45).

136

Figure 4.4. Morphology of Polypterus senegalus gill arches after long-term (8 month) terrestrial acclimation. (a) Representative micro-CT scan of the ceratobranchial (c) and hypobranchial (h) bones of left-hand arches 1 through 4, plus the basibranchial bone (b). Neither the hypobranchials of arches 1 and 4, nor the cartilage connections between bones were visible in the micro-CT scans. Anterior is up, scale bar is 1 mm. (b) Length of gill arches of control fish in water and after long-term terrestrial acclimation measured from micro-CT scans. The presented values are mass-corrected to the mean body mass (5.65 g) using ANCOVA for visualization. (c)

Bone volume of gill arches calculated as the number of voxels of bone present in micro-CT scans. Presented values are standardized to the average length of each gill arch using ANCOVA, and thus do not simply reflect differences in overall gill size. (d) Curvature of the gill arches.

Different letters indicate significant differences between gill arches, and asterisks denote significant differences between control and treatment groups (p < 0.05).

137

Supplementary Material

Figure 4.S1. Representative image of a picrosirius red stained cross-section through the first gill arch, under polarized light, of a control Polypterus senegalus reared in water. Collagen appears green through red, while other tissue types and background appear blue-black. Hue from polarized light images was used to estimate the fibre size and density of collagen molecules.

Scale bar = 100 µm.

138

CHAPTER 5: FISH RESPIRATORY FUNCTION DEPENDS ON GILL FILAMENT CALCIFICATION

Prepared in the style of Proceedings of the Royal Society B

Authors: Andy J. Turko, Bianca Cisternino, and Patricia A. Wright.

139

Abstract

The structure and function of fish gills is closely linked to lifestyle and environmental conditions. Factors that influence gill surface area are well-studied, but much less is known about how the mechanical properties of gill tissues determine function. In several groups of fishes, including killifishes (Cyprinodontiformes) and some other percomorphs, the bases of the gill filaments are surrounded by a calcified “sheath” of unknown purpose. We tested two functional hypotheses: (1) calcified gill filaments enhance water flow through the gill basket, improving aquatic respiratory function, and (2) considering many killifishes are amphibious, calcification could provide support for gills out of water. We surveyed >100 species of killifishes and found that generally hypoxia tolerant clades (e.g. Poecilid livebearers) had the most calcified gill filaments, in support of the respiratory function hypothesis. There was no difference in calcification between amphibious and fully aquatic species. Two additional lines of evidence support the respiratory function hypothesis. First, acclimation to hypoxia or warm temperatures in the laboratory (to increase respiratory demands) increased calcium deposition on the gill filaments of the model killifish Kryptolebias marmoratus. Second, when calcification was gently removed (15% EDTA solution) branchial resistance to water flow decreased, suggesting calcification of the gill filaments maintains the integrity of the branchial basket. These results show that respiratory function is determined by the mechanical properties of the gill basket in addition to surface area, which is important for understanding how fishes cope with aquatic hypoxia and climate change.

140

5.1 Introduction

Gills are multi-functional organs that provide the primary site of exchange between the fish and the external environment. The size and structure of gills varies among species, usually reflecting differences in habitat or lifestyle. Athletic or hypoxia-tolerant fishes tend to have large gills relative to their body mass, while sluggish fishes have less gill surface area (e.g. Hughes

1984, Palzenberger and Pohla 1992). Similarly, hypoxia tolerance and aerobic exercise capacity are positively correlated with gill surface area (Mandic et al. 2009, Crans et al 2015, Borowiec et al. 2016). The importance of gill size to the respiratory capacity of fishes has even led to the recent hypothesis that the adaptive responses of fishes to climate change are ultimately limited by gill surface area (Pauly and Cheung 2017, 2018, but see Lefevre et al. 2017, 2018). Despite this focus on surface area, respiratory function probably also depends on other factors including ventilation and the biomechanics of water flow over the gills, but the importance of these aspects of gill structure and function are not well understood.

The branchial apparatus of all ray-finned fishes consists of four pairs of branchial arches, each bearing two rows of perpendicularly oriented filaments (Hughes 1984, Wilson and Laurent

2002). The tips of the filaments from adjacent arches are positioned in close proximity, which forces the respiratory water flow between gill lamellae that protrude along the length of the filaments. These lamellae are the main site of exchange between the fish and environment due to very short water-blood diffusion distances (Hughes 1984, Wilson and Laurent 2002).

Gill filaments are typically supported by a cartilage rod that joins proximately with the gill arch and runs the length of the filament. However, a few species have specialized support structures. In amphibious mudskippers such as Periophthalmodon schlosseri, the gill filaments are branched at several points along their length, which is thought to prevent coalescence and

141 maintain channels for air movement when these fish move onto land and the gills no longer receive buoyant support from water (Low et al. 1988). Ram-ventilating tunas have calcified filaments that are fused with neighbouring filaments several times along their length to maintain orientation in high water flows (Wegner et al. 2012). A calcified “sheath” surrounding the base of the cartilage filament rod has also been reported in a few other species including stickleback and some cyprinids (Muir and Kendall 1968, Conway and Mayden 2009). The calcified portion of the filament is clearly stiffer than the uncalcified cartilage, but the functional significance of this stiffness is unknown.

We tested two hypotheses to explain the function of calcified gill filaments in fishes.

First, calcification may be important for maintaining position of the gills in the ventilatory water current, allowing the filament tips from one gill arch to remain interwoven with the filaments from the adjacent arch. This branchial “sieve” arrangement is essential to force inspired water to pass between the lamellae. Otherwise, if the filaments of neighbouring arches bend and separate in the ventilatory current, inspired water will follow the path of least resistance and escape between the gill arches bypassing the lamellae (Hughes 1984). If calcification provides this type of filament support, it could be especially important under high flow conditions (e.g. hyperventilation during hypoxia). We also tested the hypothesis that calcification of the gill filaments is a structural adaptation to support the gills of amphibious fishes out of water. In preliminary investigations, we discovered calcification in the gills of several amphibious killifishes that regularly leave water (Livingston et al. 2018). Calcified gill filaments may prevent filament collapse and coalescence in the absence of other morphological adaptations such as the branching structure described in mudskippers (Low et al. 1988).

142

To understand how widespread gill filament calcification is in fishes, we used a phylogenetic comparative approach and investigated gill structure in over 100 species. We focused on the killifishes (Cyprinodontiformes), as these fishes occupy a wide variety of habitats and ecological niches that may select for different gill characteristics. Some species inhabit flowing and well oxygenated water, while others typically experience stagnant and hypoxic conditions (McKinsey and Chapman 1998, Johnson and Bagley 2011, Tobler and Plath 2011).

Furthermore, amphibious lifestyles have evolved independently several times in this group

(Turko and Wright 2015). If gill filament calcification improves respiratory function, prevalence of this trait should be associated with active or hypoxia-tolerant species. Alternatively, amphibious species should have more calcified filaments if this is an adaptation to provide mechanical support out of water. We also investigated whether gill filament calcification is a phenotypically plastic trait in fishes acclimated to increased respiratory demand or terrestrial conditions. Finally, we measured the resistance of the gill basket to water flow with and without the presence of filament calcification. If calcification prevents inspired water from escaping between the tips of the filaments, then gill resistance should decrease in response to experimental decalcification of the filaments.

5.2 Methods

5.2.1 Experimental animals

To investigate patterns of gill filament calcification in killifishes and their relatives, we opportunistically acquired specimens (N = 106 species, 1-15 individuals/species) available through the pet trade, during fieldwork, or that were generously donated by other researchers.

Species examined, sample sizes, and acquisition information are provided in Table S1. While a

143 comprehensive survey of the ~30,000 fishes was beyond the scope of our present study, we included some representatives from closely related orders, including silversides (Atheriniformes) and medaka (Beloniformes), as well as more distantly related groups (e.g. Centrarchidae,

Cichlidae, Cyprinidae, Salmonidae) to assess the prevalence of filament calcification in phylogenetically diverse fishes.

We conducted a series of laboratory experiments on the model cyprinodontiform species

Kryptolebias marmoratus (strain 50.91; Tatarenkov et al. 2010) and Poecilia wingei to study the functional role of gill filament calcification. Fishes were acquired from a breeding colony maintained at the Hagen Aqualab, University of Guelph and raised under constant conditions (12 h:12 h light: dark cycle, 25°C) in 15‰ brackish water (K. marmoratus) or freshwater (P. wingei) and fed freshly hatched Artemia nauplii three times per week.

5.2.2 Staining and clearing

To visualize the gill skeleton, freshly euthanized fish were lightly fixed (2% paraformaldehyde in phosphate buffered saline, pH 8.0) for 2h at room temperature, and then stored in 70% ethanol (4°C) prior to staining. A few samples (see Table S1) donated by colleagues were placed directly in 70-95% ethanol when paraformaldehyde was unavailable; pilot experiments using K. marmoratus showed this method of tissue storage did not affect staining results. Gill baskets were dissected and stained with an acid-free mixture of Alizarin red

(calcium stain, 0.1%) and Alcian blue (cartilage stain, 2%) in 70% ethanol (3-5 d at room temperature) as described in detail elsewhere (Walker and Kimmel 2007). Note that this method of staining (acid-free) differs from conventional acid differentiation of Alcian blue staining, as the acid reduces or eliminates calcium staining in the filaments. Samples were cleared in a

144 descending series of potassium hydroxide in glycerol (Walker and Kimmel 2007), left-side arches were mounted in 70% glycerol, and brightfield photographs were taken with a Nikon

Eclipse 90i microscope. The calcified and total length of 6 filaments (3 afferent, 3 efferent) at the midpoint of the first gill arch were measured in ImageJ. The proportion of filament that was calcified did not depend on filament position (i.e. afferent vs. efferent, paired t-test P=0.89), so we simply used the mean of the six measurements for each individual in subsequent analyses.

5.2.3 Phylogenetic analysis

A phylogeny of the fishes we studied was manually assembled in RStudio using the R package phytools (Revell 2012). We used the recently published Cyprinodontiform tree from

Reznick et al (2017) as a backbone. Additional phylogenetic information was obtained from the following sources: Atheriniformes (Campanella et al. 2015), Nothobranchiidae (Collier et al.

2009, Dorn et al. 2014), Rivulidae (Costa 2006, Costa 2010, Furness et al. 2015), Fundulidae

(Ghedotti and Davis 2013, Rodgers et al. 2018), Cyprinodontidae (Echelle et al. 2005). When divergence time estimates within the order Cyprinodontiformes differed between phylogenies, they were proportionately re-scaled to match Reznick et al (2017). Divergence time estimates for more distantly related orders were obtained from Betancur-R et al. (2017).

Ancestral state reconstruction was performed using the fastAnc function in phytools to investigate patterns of gill filament calcification among species. Many statistical tests have been created to measure phylogenetic signal, a measure of whether inter-specific trait variation is determined by phylogenetic position (versus the null assumption that traits are randomly distributed among species), each providing slightly different information. To test for an overall signal in the phylogeny, we used five of the most common test statistics (Abouheif’s Cmean,

145

Moran’s I, Blomberg’s K and K*, Pagel’s λ) using the phyloSignal function of the phylosignal package (Keck et al. 2016). We investigated local indicators of spatial association to determine the specific taxa that exhibited phylogenetic signal (significant Moran’s Ii) using the lipaMoran function. To focus on clade-level patterns and to minimize the effects of terminal branch length, we based the proximity matrix on the number of nodes in the phylogeny (argument prox.phylo =

“nNodes”). We also compared Brownian and Ornstein–Uhlenbeck models of trait evolution using Akaike's information criterion.

We compared the extent of gill filament calcification in aquatic versus amphibious killifishes using phylogenetic generalized least squares (Revell 2012) under Brownian, Ornstein-

Uhlenbeck, and null models of trait evolution.

5.2.4 Plasticity of filament calcification

To investigate the environmental responsiveness of gill filament calcification, K. marmoratus were acclimated (28 d) to terrestrial conditions (n = 18, aquatic control n = 17), progressive aquatic hypoxia (10.6 [start]- 4.3 [finish] kPa O2; n = 15, normoxic control n = 16), warm water (30°C, n = 12, control = 25°C, n = 11) or fresh water versus sea water (35‰, each n

= 26), as described previously (Rossi et al. 2018). Terrestrially acclimated fish were held on a moist filter paper substrate (Ong et al. 2007). All other fish were held in mesh-bottomed holding containers (120 mL) suspended in water. Oxygen saturation was decreased by 2 kPa every 7 d and was maintained by an automated control system (Oxy-reg, Loligo Systems, Tjele, Denmark) that displaced O2 by bubbling N2 gas. To test the generality of plastic responses, we also acclimated guppies P. wingei to the same aquatic hypoxia (n = 14, control n =13) and elevated temperatures (each n = 10); this species would not survive terrestrial or sea water acclimation. To

146 facilitate comparison, both species were held in fresh water for these experiments. Fish were fed

Artemia nauplii three times per week during the acclimation period except for fish in the terrestrial acclimation experiment. Kryptolebias marmoratus are unable to feed while out of water, so in this experiment the aquatic control group was similarly starved.

Immediately after the acclimation period, fish were bathed in a 0.2% calcein solution

(Fisher Scientific; pH 7.0, 20 min) to label newly deposited calcium (Petrie et al. 2014). Fish were then gently transferred to three baths of clean water (5 mins each) to remove excess calcein and euthanized with an overdose of tricaine methanesulfonate (MS222; 500 mg L-1). Gills were quickly dissected, right-side arches mounted in 50% glycerol, and photographed using the FITC channel of a Nikon Eclipse 90i microscope. All samples were photographed using identical microscope and camera settings. Arches from the left side were placed in Alcian/Alizarin double stain, and the extent of filament calcification was determined as described above. Fluorescent intensity and the proportion of calcification of six filaments at the midpoint of the arch were compared within each species and experiment using Student’s t-tests.

5.2.5 Gill basket resistance

We measured the resistance of the gill basket with and without calcified gill filaments to test whether stiff filaments maintain the flow of ventilatory water over the lamellae. In each trial

(n = 5), a polyethelene catheter (PE-160) was inserted into the mouth of a heavily anesthetized

(300 mg L-1 MS222) K. marmoratus and held in place with cyanoacrylate gel. The catheter was attached to a peristaltic pump that provided pulsatile water flow (15‰ brackish water, 300 mg L-

1 MS222, pH = 8.1) simulating natural ventilation. An in-line pressure transducer (BP stopcock transducer MLT0670, AD Instruments, Colorado Springs, CO, USA) located approximately 10

147 cm upstream of the fish was allowed continuous pressure recording. Fish were immobilized in a plastic trough and submerged in an overflowing bath to maintain head pressure from the surrounding water. Given the constant water flow in this setup, pressure was directly proportional to resistance of the gills. After a 2 h stabilization period (35 ventilations min-1), pressure was recorded for 5 - 6 minutes at each of five ventilatory frequencies (Vf = 35, 80, 115,

-1 150, 185 min ) that represent the normal range of Vf from resting to maximal in this species

(Turko et al. 2012). Pressure at each Vf was measured in triplicate; preliminary experiments indicated that these measurements were consistent for more than 6 h. The flow of brackish water across the gills was then replaced with a 15% ethylenediaminetetraacetic acid solution (300 mg

-1 L MS222, pH 8.1, Vf = 35) for 90 min to gently remove the calcified sheath from the gill filaments. Then, the flow through the catheter was switched back to the original water, and pressure was again measured at each of the five levels of Vf in triplicate. At the end of the experiment, gills were removed and stained as described above to verify that the filament calcification had been removed. Data were analyzed with a two-way repeated measures ANOVA to test for effects of Vf and decalcification on gill resistance.

5.3 Results

Every killifish species that we examined had calcification at the filament base. This calcification always resembled a “sheath” that surrounded the base of the supportive cartilage rod, and calcification never extended into the middle of the cartilage (figure 5.1). In K. marmoratus, filaments positioned in the middle of the gill arch tended to be proportionately more calcified than filaments located at either edge (figure 5.2). Similar calcification was observed in almost every other Percomorph fish studied (but not in the Centrarchiform

148

Ambloplites rupestris), but calcification was absent in specimens representing more distantly related families including the Cyprinidae and Salmonidae (figure 5.3). The proportion of filament calcification scaled weakly and inconsistently with body size (species * mass P<0.001,

R2=0.026, species * length P<0.001, R2=0.064). The weakness of this relationship suggests that simply standardizing the proportion of calcification to total filament length accounts for almost all variation due to body size, thus we did not include body size as a covariate in subsequent analyses. There was no difference in the degree of filament calcification between aquatic and amphibious killifishes (t-test P=0.739, all pgls models P>0.26; figure 5.3B). Each measure of overall phylogenetic signal was significant (Cmean, I, K, K*, λ; all P<0.0001), indicating that the extent of calcification (% of filament length that was calcified) was distributed non-randomly throughout the phylogeny. Most of the variation tended to be among clades rather than individual species (Blomberg’s K = 1.021). The local indicator analysis (Moran’s Ii) showed a pattern of significantly more calcification in Poecilids and Cyprinodon pupfishes, and reduced calcification in annual Nothobranchius killifishes (figure 5.3). The families Aplocheilidae, Cyprinodontidae, and Poeciliidae tended to have the most calcified gill filaments, while the Fundulidae and

Nothobranchiidae were least (figure 5.3C). Generally, the distribution of calcification across the phylogeny appeared to follow a Brownian pattern (λ = 0.839). However, the Ornstein–

Uhlenbeck model of evolution provided a better fit for the pattern of calcification than a simple

Brownian model (ΔAICc = 40).

Fluorescent intensity of calcein staining in the gill filaments of K. marmoratus significantly increased in response to air exposure, seawater acclimation, hypoxia, and a 5°C temperature increase (all P<0.05; figure 5.4). There was no increase in fluorescent intensity of P. wingei gill filaments after hypoxic or warm acclimation however (P>0.11, figure 5.4D,E). The

149 proportion of the filament that was calcified did not change in response to any acclimation condition (all P>0.05, figure 5.S1).

Pressure (proportional to resistance) within the branchial chamber significantly decreased after decalcification at all values of Vf except the highest Vf tested (two-way repeated measures

ANOVA, interaction P<0.05; Holm-Sidak post-hoc P<0.05; figure 5.5a).

5.4 Discussion

Our study provides three independent lines of evidence that support the hypothesis that gill filament calcification is a mechanism to improve gill function. First, we found a strong phylogenetic signal within the killifishes such that the most pronounced calcification occurred in clades (Poecilid mollies and Cyprinodon pupfishes) generally known to survive in hot and hypoxic habitats that demand strategies for increased oxygen uptake. Second, increased respiratory demand during hypoxic and warm acclimation caused increased calcium deposition in the gill filaments of K. marmoratus. Third, experimental removal of the calcified sheath from the gill filaments decreased the resistance of the gills to water flow across a wide range of ventilation frequencies. This result indicates that this calcification plays an important role maintaining the integrity of the gill basket, forcing inspired water through the narrow inter- lamellar channels where gas exchange can occur most efficiently. Evidence for the alternative hypothesis, that gill filament calcification supports the gills of amphibious killifishes out of water, was equivocal. Gill filaments of amphibious species were calcified to the same extent as fully aquatic relatives, suggesting that there is no selective advantage to calcification in species that leave water. Terrestrial acclimation triggered calcium deposition on the calcified portion of the filaments, however, indicating that there may be an advantage to increased stiffness.

150

5.4.1 Phylogenetic patterns of filament calcification

Calcified gill filaments appear to be widespread within the Percomorpha (sensu

Betancur-R et al. 2017), which would imply that approximately 40% of the over 30,000 fish species possess this trait. In addition to the Cyprinodontiformes, we sampled a few species within the Cichliformes and Gobiiformes and found calcification present in the gill filaments.

Calcified gill filaments have also been reported in Scombriformes (Wegner et al. 2012), and

Gasterosteiformes (Muir and Kendall 1968, Conway and Mayden 2009). However, we found one percomorph species with no filament calcification - the Centrarchiform Ambloplites rupestris.

Further sampling effort is needed to test whether this represents a loss of filament calcification in this species or if a more intensive survey would reveal some variation within the Percomorpha.

Two families within the order Cypriniformes, which diverged from the percomorphs about 250 million years ago (Betancur-R et al. 2017), have also been found to have a calcified sheath surrounding the cartilage rod of the gill filaments. However, the morphology of this calcification is subtly different from that in the percomorph fishes. In the Cypriniformes, the v-shaped base and the lower portion of the cartilage rod is calcified (Conway and Mayden 2009), while in the percomorphs the calcification begins above the v-shaped base. These subtle differences suggest that this trait evolved convergently in these two clades.

The extent of gill filament calcification was highly dependent on phylogeny. Within the order Cyprinodontiformes, the families Cyprinodontidae and Poeciliidae tended to have the most calcified gill filaments. These fishes are generally restricted to tropical and sub-tropical regions of the Americas, and often inhabit warm water low in dissolved oxygen (Meffe and Snelson

1989, McKinsey and Chapman 1998, Tobler and Plath 2011, Dzul-Caamal et al. 2012). Such an

151 association between these types of habitats and highly calcified gill filaments is consistent with the hypothesis that the calcification is a mechanism for enhanced respiratory function. Therefore, it was surprising that we found a relatively low extent of filament calcification in the family

Nothobranchiidae, especially in the annual genus Nothobranchius. These annual fishes typically inhabit temporary pools in African grasslands, and in response have evolved extremely short adult lifespans and desiccation-tolerant embryonic stages (Vrtilek et al. 2018). Given the presumably high metabolic demands of this accelerated pace of life, and the often-hypoxic nature of these temporary pools (Podrabsky et al. 2015), a relatively high degree of gill filament calcification would be expected in these fishes. Furthermore, annual lifecycles also evolved convergently several times in South American rivulid killifishes (Furness et al. 2015), and these species showed average levels of filament calcification. Thus, the extent of gill filament calcification in a given species is probably determined by a much more complex set of environmental and natural history variables than simply oxygen supply and demand. It is also possible that the phylogenetic signal we observed is the result of random evolutionary processes and is thus non-adaptive (Revell et al. 2008). However, considering that selection seems to maintain other biomechanical aspects of gill structure, such as lamellar spacing, at functional optima (Park et al. 2014), we think it is unlikely that the phylogenetic signal we observed is simply the result of neutral processes.

5.4.2 Calcified filaments improve respiratory function

Efficient branchial respiration requires inspired water to pass between the gill lamellae, where the diffusion distance to the blood is shortest (Hughes 1984). Forcing water through this path requires the tips of neighbouring gill filaments to be in very close proximity, otherwise the

152 ventilatory water current will “escape” between the filament tips through the path of least resistance without offloading oxygen to the blood (Hughes 1972, Strother 2013). This path between the filaments has been called an “anatomical dead space” (Hughes 1984) or a “non- respiratory shunt” (Strother 2013), and is thought to be most problematic for fishes at high rates of ventilation. For example, increased gill ventilation can cause decreased branchial resistance

(Hughes and Shelton 1958¸ Hughes and Saunders 1970), less efficient oxygen uptake (Saunders

1962, Hughes and Umezawa 1968), and the separation of adjacent gill filaments (Strother 2013).

Our data strongly support the hypothesis that gill filament calcification is a structural mechanism to minimize the loss of water through the non-respiratory shunt. When we gently removed filament calcification with EDTA, we measured a decrease in the resistance of the gills to water flow at all but the highest Vf we studied. We assume that the reduced stiffness of the decalcified filaments caused them to bend in the ventilatory current, thus allowing water to flow through this relatively low resistance pathway (see schematic figure 5.5b). At the highest Vf, however, it seems likely that even calcified filaments were not sufficiently stiff to prevent water from escaping through the non-respiratory shunt. However, we found that acclimation to conditions of increased respiratory demand caused calcium deposition on the filaments of K. marmoratus. This additional calcification would presumably increase filament stiffness and may minimize the loss of inspired water through the non-respiratory shunt, thus increasing respiratory performance. In support of this hypothesis, we previously found that hypoxia-acclimated K. marmoratus could maintain rates of oxygen uptake with much lower ventilation volumes than control fish under a wide range of environmental oxygen tensions (Turko et al. 2012). Calcium deposition was not observed in P. wingei acclimated to hypoxia or increased temperatures, however. One possibility is that these poecilids were not as challenged by the acclimation

153 conditions, or were able to increase oxygen uptake using behavioural strategies such as aquatic surface respiration (Kramer and McClure 1982), and thus did not require increased gill ventilation. Calcification of bone and cartilage is usually controlled at the tissue level (Robling et al. 2006), thus the cue for calcium deposition in the gill filaments is probably increased mechanical loading caused by higher ventilation rates.

In addition to receiving structural support from the cartilage rod, filament position can be actively controlled by muscle. Relatively large adductor muscles connect every filament to the gill arch (Bijtel 1949, Hughes 1984). Contractions of these muscles serve to move the filaments of a single gill arch closer together, thus increasing the width of the non-respiratory shunt between adjacent arches. This is an important mechanism for clearing the gills of particulate matter via “coughing”, but would also temporarily decrease respiratory gas exchange (Hughes

1984). Abductor muscles, which would position the filaments to close the non-respiratory shunt, are typically very small or absent and the passive mechanical properties of the filaments are thought to hold their position in the water flow (Bijtel 1949). Thus, stiffening of the cartilage rod via calcification would provide a simple mechanism to maintain the position of filaments during respiration without requiring muscular support.

5.4.3 Filament calcification in amphibious killifishes

Evidence for the hypothesis that calcified filaments are an adaptation to increase structural support in the gills of amphibious killifishes was equivocal. There was no evolutionary pattern indicating higher proportions of calcification in amphibious versus aquatic species, and the presence of calcified filaments in the sister orders to the Cyprinodontiformes indicates that filament calcification did not evolve in response to amphibious lifestyles. However, our results

154 show that the filaments in the middle of the gill arch were the most calcified, and these would also experience the most gravitational load out of water due to their horizontal orientation and cantilevered attachment to the gill arch. Furthermore, calcein staining revealed calcium deposition on the gill filaments of terrestrially-acclimated K. marmoratus. This result is consistent with our previous report of enhanced gill structural support in fish out of water through increased collagen density in gill filaments (Turko et al. 2017 [Chapter 3]). Thus, while filament calcification probably did not initially evolve in response to amphibiousness, this mechanism may have been co-opted by amphibious killifishes to increase gill support on land.

5.4.4 Perspectives

Our results provide strong evidence that calcification of the gill filaments improves aquatic respiratory function. If this is the case, why do some fishes (e.g. Nothobranchius) have minimal calcification, and other species (e.g. salmonids) have none? The multi-functional nature of gills means that their structure and function is the product of a series of trade-offs (Hughes

1984, Wilson and Laurent, 2002). Stiff, calcified filaments may be too fragile, have negative consequences for other aspects of gill function, be unsuitable for some environments (e.g. may increase clogging by suspended sediment), or be energetically costly to build and maintain.

Furthermore, our phylogenetic analysis suggested that an OU model of evolution better described variation in calcification than a simple Brownian model, which is often interpreted as an indication of evolutionary constraint (but see Revell et al. 2008, Losos 2008) that would therefore imply more calcification is not always better. Different amounts of calcification may also reflect differences in ventilatory mechanisms (e.g. ram ventilation vs buccal pumping), which seems to be the main driver of variation in inter-lamellar spacing (Park et al. 2014).

155

In many relatively basal ray-finned fishes (e.g. salmonids), the two rows of gill filaments on an arch are connected by a membranous inter-branchial septum that is thought to provide support (Hughes 1984, Wilson and Laurent 2002). The evolutionary trend in derived fishes is towards a reduction or loss of this septum (Olson 2000), and no septa were visible in our killifish gill samples. One possibility is that calcification evolved to replace the supportive function of the septum, although it is not clear whether this would provide a functional benefit or simply be an alternative and equally effective method of support.

Ultimately, our results highlight the importance of the mechanical properties of gill filaments when assessing respiratory performance. Up to now the research focus has been on gill surface area as a functional correlate of respiratory capacity in fishes (e.g. Pauly and Cheung

2017, Lefevre et al. 2017), with less consideration of the importance of the path of water flow through the gill basket. Variation in filament stiffness, both within and between species, may be an additional feature that determines the tolerance of fishes to harsh environmental conditions or a changing climate.

156

5.5 References

Betancur-R, R, Wiley EO, Arriata G, Acero A, Bailly N, Miya M, Lecointre G, Orti G. 2017

Phylogenetic classification of bony fishes. BMC Evol. Biol. 17, 162.

Bijtel JH. 1949 The structure and mechanism of movements of the gill filaments in Teleostei.

Arch. Neerl. Zool. 8, 267-288.

Borowiec BG, Crans KD, Khajali F, Pranckevicius NA, Young A, Scott GR. 2016 Interspecific

and environment-induced variation in hypoxia tolerance in sunfish. Comp. Biochem.

Physiol. A Mol. Integr. Physiol. 198, 59-71.

Campanella D, Hughes LC, Unmack PJ, Bloom DD, Piller KR, Orti G. 2015 Multi-locus fossil-

calibrated phylogeny of Atheriniformes (Teleostei, Ovalentaria). Mol. Phylogenet. Evol. 86,

8–23.

Collier GE, Murphy WJ, Espinoza M. 2009 Phylogeography of the genus Epiplatys

(Aplocheiloidea: Cyprinodontiformes). Mol. Phylogenet. Evol. 50, 190–196.

Conway KW and Mayden RL. (2009). Gill-filament ossifications: a possible morphological

synapomorphy uniting the families Balitoridae and Cobitidae (Ostariophysi: Cypriniformes).

J. Fish Biol. 75, 2839-2844.

Costa WJEM. 2006 Descriptive morphology and phylogenetic relationships among species of the

Neotropical annual killifish genera Nematolebias and Simpsonichthys (Cyprinodontiformes:

Aplocheiloidei: Rivulidae). Neotrop. Ichthyol. 4, 1-26.

Costa WJEM. 2010 Historical biogeography of cynolebiasine annual killifishes inferred from

dispersal–vicariance analysis. J. Biogeogr. 37, 1995–2004.

157

Crans KD, Pranckevicius NA, Scott GR. 2015 Physiological tradeoffs may underlie the

evolution of hypoxia tolerance and exercise performance in sunfish (Centrarchidae). J. Exp.

Biol. 218, 3264-3275.

Dorn A, Musilova Z, Platzer M, Reichwald K, Cllerino A. 2014 The strange case of East African

annual fishes: aridification correlates with diversification for a savannah aquatic group?

BMC Evol. Biol. 14, 210.

Dzul-Caamal R, Olivares-Rubio HF, Medina-Segura CG, Vega-Lopez A. 2012 Endangered

Mexican Fish Under Special Protection: Diagnosis of Habitat Fragmentation, Protection,

and Future – A Review. In Endangered species: Habitat, protection and ecological

significance (ed. ML Lucas-Borja), pp. 109-130. New York, NY: Nova Science Publishers.

Echelle AA, Carson EW, Echelle AF, Van Den Bussche RA, Dowling TE, Meyer A. 2005

Historical Biogeography of the New-World Pupfish Genus Cyprinodon (Teleostei:

Cyprinodontidae). Copeia 2005, 320-339.

Furness AI, Reznick DN, Springer MS, Meredith RW. 2015 Convergent evolution of alternative

developmental trajectories associated with diapause in African and South American killifish.

Proc. R. Soc. B 282, 20142189.

Ghedotti, MJ, Davis MP. 2013 Phylogeny, classification, and evolution of salinity tolerance of

the North American topminnows and killifishes, family Fundulidae (Teleostei:

Cyprinodontiformes). Fieldiana, Life earth sci. 7, 1-65.

Hughes GM. 1972 Morphometrics of fish gills. Respir. Physiol. 14, 1-25.

Hughes GM. 1984. General anatomy of the gills. In Fish Physiology vol. 10A (eds. WS Hoar and

DJ Randall), pp. 1-72. San Diego, CA: Academic Press.

158

Hughes GM, Saunders RL. 1970 Responses of the respiratory pumps to hypoxia in the rainbow

trout (Salmo gairdneri). J. Exp. Biol. 53, 529-545.

Hughes GM, Shelton, G. 1958. The mechanism of gill ventilation in three freshwater teleosts. J.

Exp. Biol. 35, 807-823.

Hughes GM, Umezawa SI. 1968 On respiration in the dragonet Callionymus lyra L. J. Exp. Biol.

49, 565-582.

Johnson JB, Bagley JC. 2011. Ecological drivers of life-history divergence. In Ecology and

Evolution of Poeciliid Fishes (eds. JP Evans, A Pilastro and I Schlupp), pp. 38-49. Chicago,

IL: University of Chicago Press.

Keck F, Rimet F, Bouchez A, Franc A. 2016 phylosignal: an R package to measure, test, and

explore the phylogenetic signal. Ecol. Evol. 6, 2774-2780.

Kramer, D. L. & McClure, M. (1982). Aquatic surface respiration, a widespread adaptation to

hypoxia in tropical freshwater fishes. Environ. Biol. Fish. 7, 47–55.

Lefevre S, McKenzie DJ and Nilsson GE. (2017). Models projecting the fate of fish populations

under climate change need to be based on valid physiological mechanisms. Glob. Change

Biol. 23, 3449-3459.

Lefevre S, McKenzie DJ and Nilsson GE. (2018). In modelling effects of global warming,

invalid assumptions lead to unrealistic projections. Glob. Change Biol. 24, 553-556.

Livingston, M. D., Bhargav, V. V., Turko, A. J., Wilson, J. M. and Wright, P. A. (2018).

Widespread use of emersion and cutaneous ammonia excretion in Aplocheiloid killifishes.

Proc. R. Soc. B, 20181496.

159

Losos JB. 2011 Phylogenetic niche conservatism, phylogenetic signal and the relationship

between phylogenetic relatedness and ecological similarity among species. Ecol. Lett. 11,

995-1007.

Low WP, Lane DJW, Ip YK. 1988. A comparative study of terrestrial adaptations of the gills in

three mudskippers – Periophthalmus chrysospilos, Boleophthalmus boddaerti, and

Periophthalmodon schlosseri. Biol. Bull. 175, 434-438.

Mandic M, Todgham AE, Richards JG. 2009 Mechanisms and evolution of hypoxia tolerance in

fish. Proc. R. Soc. B 276, 735-744.

McKinsey DM, Chapman LJ. 1998. Dissolved oxygen and fish distribution in a Florida spring.

Environ. Biol. Fish. 53, 211-223.

Meffe, G. K. and Snelson, F. F. 1989. Ecology and evolution of livebearing fishes (Poeciliidae).

Englewood Cliffs, NJ: Prentice Hall.

Muir BS and Kendall JI. 1968. Structural modifications in the gills of tunas and some other

oceanic fishes. Copeia 1968, 388-398.

Olson, K. R. 2000. Respiratory system. In The Laboratory Fish (ed. GK Ostrander). pp. 357-368.

San Diego, CA: Academic Press.

Ong KJ, Stevens ED, Wright PA. 2007 Gill morphology of the mangrove killifish (Kryptolebias

marmoratus) is plastic and changes in response to terrestrial air exposure. J. Exp. Biol. 210,

1109–1115.

Palzenberger M, Pohla H. 1992. Gill surface area of water-breathing freshwater fish. Rev. Fish

Biol. Fisher. 2, 187-216.

160

Park K, Kim W, Kim H-Y. 2014 Optimal lamellar arrangement in fish gills. Proc. Natl. Acad.

Sci. USA, 111, 8067-8070.

Pauly D and Cheung WWL. 2018a. Sound physiological knowledge and principles in modeling

shrinking of fishes under climate change. Glob. Change Biol. 24, e15-e26.

Pauly D and Cheung WWL. 2018b. On confusing cause and effect in the oxygen limitation of

fish. Glob. Change Biol. accepted manuscript https://doi.org/10.1111/gcb.14383

Petrie TA, Strand NS, Tsung-Yang C, Rabinowitz JS and Moon RT. 2014. Macrophages

modulate adult zebrafish tail fin regeneration. Development 141, 2581-2591.

Podrabsky JE, Riggs CL, Wagner JT. 2015 Tolerance of environmental stress. In Annual Fishes:

Life History Strategy, Diversity, and Evolution (ed. G Garcia, B Berois, R de Sa). pp. 159-

184. Boca Raton, FL: CRC Taylor & Francis.

Revell LJ. 2012 phytools: an R package for phylogenetic comparative biology (and other things).

Methods Ecol. Evol. 3, 217-223.

Revell LJ, Harmon LJ, Collar DC. 2008. Phylogenetic signal, evolutionary process, and rate.

Syst. Biol. 57, 591-601.

Reznick DN, Furness AI, Meredith RW, Springer MS. 2017 The origin and biogeographic

diversification of fishes in the family Poeciliidae. PLoS ONE 12, e0172546.

Rodgers R, Roach JL, Reid NM, Whitehead A, Duvernell DD. 2018 Phylogenomic analysis of

Fundulidae (Teleostei: Cyprinodotiformes) using RNA-sequencing data. Mol. Phylogenet.

Evol. 121, 150–157.

Robling, A. G., Castillo, A. B. and Turner, C. H. 2006 Biomechanical and molecular regulation

of bone remodeling. Annu. Rev. Biomed. Eng. 8, 455-498.

161

Rossi GS, Turko AJ, Wright PA. 2018 Oxygen drives skeletal muscle remodeling in an

amphibious fish out of water. J. Exp. Biol. 221, jeb180257.

Saunders R. 1962. The irrigation of the gills in fishes: II. Efficiency of oxygen uptake in relation

to respiratory flow activity and concentrations of oxygen and carbon dioxde. Can. J. Zool.

40, 817-862.

Strother JA. 2013 Hydrodynamic resistance and flow patterns in the gills of a tilapine fish. J.

Exp. Biol. 216, 2595-2606.

Tatarenkov A, Ring BC, Elder JF, Bechler DL, Avise JC. 2010 Genetic composition of

laboratory stocks of the self-fertilizing fish Kryptolebias marmoratus: a valuable resource

for experimental research. PLoS ONE 5, e12863.

Tobler M and Plath M. 2011. Living in extreme environments. In Ecology and Evolution of

Poeciliid Fishes (eds. JP Evans, A Pilastro and I Schlupp), pp. 120-127. Chicago, IL:

University of Chicago Press.

Turko AJ and Wright PA 2015. Evolution, ecology and physiology of amphibious killifishes

(Cyprinodontiformes). J. Fish Biol. 87, 815-835.

Turko AJ, Cooper CA, Wright PA. 2012 Gill remodelling during terrestrial acclimation reduces

aquatic respiratory function of the amphibious fish Kryptolebias marmoratus. J. Exp. Biol.

215, 3973-3980.

Turko AJ, Kültz D, Fudge D, Croll RP, Smith FM, Stoyek MR, Wright PA. 2017. Skeletal

stiffening in an amphibious fish out of water is a response to increased body weight. J. Exp.

Biol. 220, 3621-3631.

162

Vrtílek, M., Žák, J., Pšenicka, M. and Reichard, M. (2018). Extremely rapid maturation of a wild

African annual fish. Curr. Biol. 28, R822-R824.

Walker MB and Kimmel CB. 2007. A two-color acid-free cartilage and bone stain for zebrafish

larvae. Biotech. Histochem. 82, 23-28.

Wegner NC, Sepulveda CA, Aalbers SA, Graham JB. 2012 Structural adaptations for ram

ventilation: Gill fusions in scombrids and billfishes. J. Morph. 274, 108-120.

Wilson JM and Laurent, P. 2002. Fish gill morphology: inside out. J. Exp. Zool. 293, 192-213.

163

Figures

Figure 5.1. Representative gill filaments stained with Alizarin red and Alcian blue, each demonstrating a calcified sheath surrounding the base of the supportive cartilage rod. Blue branches on the phylogeny indicate fully aquatic species, green branches indicate amphibious fishes.

164

165

Figure 5.2. (A) The degree of filament calcification in Kryptolebias marmoratus depends on the position along the gill arch. Filaments at the dorsal and ventral ends are less calcified than those in the middle. Gray band indicated the 95% confidence interval, error bars represent s.e.m. (B)

Representative cleared and stained K. marmoratus gill arch. Scale bar = 0.5 mm.

166

Figure 5.3. (A) Ancestral state reconstruction of gill filament calcification in the

Cyprinodontiformes and several outgroup taxa. Relatively blue branches indicate no or minimal calcification, while red indicates extreme calcification. The dotplot above the phylogeny shows mean values of calcification for each species. Red dots represent significant local indicators of spatial association (Moran’s Ii, P<0.05), indicating a cluster of species with a high degree of similarity in calcification values compared to the overall phylogeny. Branch lengths indicate divergence times; length of coloured scale bar = 50 million years. (B) There was no difference in the extent of filament calcification between fully aquatic and amphibious killifishes (P>0.05).

167

Figure 5.4. (A) Representative calcein-labelled gill of Kryptolebias marmoratus. Fluorescent intensity proportional to the amount of recently deposited calcium. Scale bar = 0.5 mm. (B)

Calcein fluorescence in K. marmoratus acclimated to water versus air, or (C) fresh versus 35‰ seawater for 28d. (D) Calcein fluorescence in K. marmoratus and Poecilia wingei acclimated to hypoxia for 28d, or (E) to 25°C versus 30°C for 28d. Fluorescent intensity is presented in arbitrary units (a.u.) that have been normalized to a control value of 1.0 within each experiment.

* indicates P<0.05, ** P<0.01.

168

Figure 5.5. (A) Resistance of the gill basket in Kryptolebias marmoratus before (control) and after decalcification at five ventilatory frequencies (Vf) maintained with a peristaltic pump. Light gray lines are values of individual fish, black lines show means ± SE. * indicates P<0.05, ns; not significant. (B) Schematic diagram of a longitudinal section through a fish head showing the hypothesized pattern of water flow over control gills with calcification at the base of the filaments (top; solid lines) or escaping between the tips of gills with decalcified filaments

(bottom; dashed lines).

169

Supplementary Material

Table 5.S1. Species, sample sizes (n), calcification (proportion of filament length), standard error (SE), and acquisition information for all fishes used for phylogenetic analysis.

species n calcification SE source Amatitlania nigrofasciata 4 0.2738 0.0192 aquarium hobby Ambloplites rupestris 1 0.1133 - researcher donation Ameca splendens 1 0.4886 - aquarium hobby Anablepsoides hartii 4 0.4416 0.0173 aquarium hobby Aphyolebias rubrocaudatus 3 0.3430 0.0401 aquarium hobby Aphyosemion australe 5 0.2746 0.0253 aquarium hobby Aphyosemion bivittatum 1 0.2253 - aquarium hobby Aphyosemion celiae 3 0.3006 0.0126 aquarium hobby Aphyosemion fulgens 1 0.3379 - researcher donation Aphyosemion gabunense 2 0.3344 0.0206 researcher donation Aphyosemion primigenium 5 0.2033 0.0198 researcher donation Aphyosemion striatum 5 0.4051 0.0058 researcher donation Aplocheilus lineatus 5 0.4102 0.0176 aquarium hobby Aplocheilus panchax 1 0.5237 - aquarium hobby Ataeniobius toweri 4 0.4216 0.0154 aquarium hobby Austrofundulus leohoignei 3 0.3129 0.0656 aquarium hobby Austrofundulus limnaeus 6 0.3319 0.0269 researcher donation Chapalichthys pardalis 5 0.3328 0.0159 aquarium hobby Cynodonichthys hildebrandi 7 0.5086 0.0221 aquarium hobby Cynodonichthys tenuis 3 0.4377 0.0011 aquarium hobby Cynolebias itapicuruensis 6 0.2474 0.0222 aquarium hobby Cyprinodon albivelis 2 0.4960 0.0103 researcher donation Cyprinodon nazas 1 0.5767 - researcher donation Cyprinodon tularosa 1 0.5367 - researcher donation Danio rerio 4 0.0000 0.0000 researcher donation Dermogenys pusilla 6 0.3694 0.0245 aquarium hobby Epiplatys chaperi 4 0.3793 0.0203 aquarium hobby Epiplatys dageti 2 0.6814 0.3347 aquarium hobby Epiplatys togolensis 2 0.4261 0.0127 aquarium hobby Floridichthys polyommus 8 0.3408 0.0567 wild caught Fundulopanchax gardneri 5 0.2961 0.0248 aquarium hobby Fundulopanchax gresensi 3 0.3351 0.0056 aquarium hobby Fundulopanchax walkeri 2 0.3480 0.0864 aquarium hobby Fundulus confluentus 3 0.3747 0.0169 researcher donation Fundulus grandis 2 0.2036 0.0184 researcher donation Fundulus heteroclitus 6 0.4316 0.0299 researcher donation Fundulus rathbuni 1 0.3460 - researcher donation

170

Fundulus similis 2 0.2806 0.0107 researcher donation Gambusia affinis 3 0.3606 0.0367 aquarium hobby Gambusia yucatana 15 0.4314 0.0129 wild caught Girardinus microdactylus 3 0.5133 0.0227 aquarium hobby Gnatholebias zonatus 4 0.3357 0.0164 researcher donation Goodea gracilis 5 0.2919 0.0612 aquarium hobby Heterandria formosa 2 0.4335 0.1646 aquarium hobby Hypsolebias antenori 1 0.3768 - aquarium hobby Hypsolebias sertanejo 2 0.2985 0.0415 aquarium hobby Ilyodon furcidens 1 0.4274 - aquarium hobby Ilyodon whitei 7 0.3063 0.0363 aquarium hobby Ilyodon xantusi 4 0.3447 0.0362 aquarium hobby Jenynsia onca 5 0.3129 0.0067 aquarium hobby Jordanella floridae 3 0.4421 0.0418 aquarium hobby Kryptolebias marmoratus 5 0.4290 0.0529 laboratory stock Laimosemion xiphidius 4 0.3896 0.0519 aquarium hobby Limia nigrofasciata 1 0.3842 - aquarium hobby Limia perugiae 11 0.5730 0.0136 aquarium hobby Lucania goodei 3 0.3253 0.0281 researcher donation Lucania parva 3 0.3043 0.0177 researcher donation Melanotaenia splendida 6 0.2658 0.0135 aquarium hobby Menidia menidia 5 0.4125 0.0093 wild caught Micropoecilia parae 4 0.4123 0.0066 researcher donation Nematolebias papilliferus 6 0.3533 0.0215 aquarium hobby Nothobranchius brieni 6 0.2220 0.0125 aquarium hobby Nothobranchius cardinalis 6 0.1866 0.0268 aquarium hobby Nothobranchius furzeri 1 0.3826 - aquarium hobby Nothobranchius fuscotaneatus 2 0.3433 0.0183 aquarium hobby Nothobranchius guentheri 1 0.2154 - researcher donation Nothobranchius hassoni 1 0.3510 - researcher donation Nothobranchius jubbi 6 0.2960 0.0192 researcher donation Nothobranchius kadleci 10 0.2613 0.0063 aquarium hobby Nothobranchius kafuensis 6 0.1971 0.0313 aquarium hobby Nothobranchius kilomberoensis 3 0.2678 0.0133 aquarium hobby Nothobranchius milvertzi 2 0.2522 0.0600 aquarium hobby Nothobranchius polli 2 0.2023 0.0359 aquarium hobby Nothobranchius rachovii 6 0.2698 0.0092 aquarium hobby Nothobranchius rubroreticulatus 6 0.2700 0.0256 aquarium hobby Nothobranchius wattersi 5 0.1735 0.0164 aquarium hobby Oncorhynchus mykiss 6 0.0000 0.0000 researcher donation Oryzias dancena 1 0.5417 - aquarium hobby Oryzias woworae 5 0.4561 0.0176 aquarium hobby Oxyeleotris marmorata 1 0.3253 - researcher donation Pelvicachromis pulcher 1 0.2621 - aquarium hobby Poecilia gillii 1 0.5630 - researcher donation

171

Poecilia latipinna 1 0.6923 - aquarium hobby Poecilia mexicana 1 0.5827 - aquarium hobby Poecilia orri 12 0.5543 0.0178 wild caught Poecilia reticulata 6 0.4776 0.0198 aquarium hobby Poecilia sphenops 6 0.6759 0.0120 aquarium hobby Poecilia wingei 6 0.4637 0.0137 laboratory stock Poeciliopsis gracilis 1 0.4556 - researcher donation Poeciliopsis infans 1 0.4773 - researcher donation Poeciliopsis prolifica 3 0.4728 0.0152 researcher donation Poeciliopsis retropinna 2 0.3741 0.0005 researcher donation Poeciliopsis turneri 1 0.5506 - researcher donation Poropanchax normani 2 0.4374 0.0648 aquarium hobby Pterolebias longipinnis 2 0.5276 0.0495 aquarium hobby Rachovia brevis 1 0.0454 - aquarium hobby Rivulus cylindraceus 5 0.3854 0.0251 aquarium hobby Simpsonichthys margaritatus 4 0.2548 0.0277 aquarium hobby Simpsonichthys suzarti 2 0.3263 0.0181 aquarium hobby Skiffia lermae 2 0.5370 0.0186 aquarium hobby Strongylura notata 1 0.1459 - wild caught Xenotoca eiseni 4 0.4113 0.0154 aquarium hobby Xiphophorus alvarezi 8 0.3669 0.0269 aquarium hobby Xiphophorus hellerii 6 0.3719 0.0188 aquarium hobby Xiphophorus maculatus 6 0.4038 0.0173 aquarium hobby Xiphophorus mayae 6 0.3465 0.0092 aquarium hobby Xiphophorus nezahualcoyotl 6 0.3874 0.0174 aquarium hobby

172

Figure 5.S1. Gill filament calcification in Kryptolebias marmoratus and Poecilia wingei after

28d of acclimation to (A) water versus air, (B) aquatic normoxia or hypoxia, (C) fresh versus

35‰ seawater, or (D) 25°C versus 30°C.

173

CHAPTER 6: PROLONGED SURVIVAL OUT OF WATER IS LINKED TO A GENERALLY SLOW PACE

OF LIFE IN A SELFING AMPHIBIOUS FISH

Prepared in the style of Functional Ecology

Authors: Andy J. Turko, Justine E. Doherty, Irene Yin-Liao, Kelly Levesque, Perryn Kruth, Joe

Holden, Ryan L. Earley, and Patricia A. Wright.

174

Abstract

1. Metabolic rate and life history traits vary widely both among and within species,

reflecting trade-offs in the allocation of energy and resources, but the proximate and

ultimate causes for this variation are not well understood. We tested the hypothesis that

environmental heterogeneity mediates pace of life trade-offs, using three isogenic strains

of the amphibious fish Kryptolebias marmoratus that vary in the amount of time each can

survive out of water without feeding. These fish are one of two self-fertilizing

hermaphroditic vertebrates, resulting in isogenic and effectively “clonal” strains which

enable repeated experimentation with the same genotype.

2. Consistent with pace of life theory, the strain that could survive longest generally

exhibited a “slow” phenotype characterized by the lowest metabolic rate, largest scope

for metabolic depression, slowest consumption of energy stores, and least investment in

reproduction under standard conditions. Growth rates were fastest in the otherwise

“slow” strain, however.

3. We tested for fitness trade-offs between the “fast” and “slow” strains using microcosm

experiments (12 months), in which fish were held with either constant water availability

or fluctuating conditions where water was absent for half of the experiment. Overall, the

“slow” strain grew larger and was in better condition, and under conditions of low water

availability the “slow” strain produced more embryos. However, the “fast” strain had

larger adult population sizes under both conditions.

4. Our results show that genetically based differences in the pace of life of amphibious fish

determine how long these animals can survive out of water. Limited water availability

175

favoured the “slow” phenotype, as predicted, but there was no detectable cost under

control conditions. Thus, differences in pace of life may among genotypes may reflect a

conditionally neutral instead of antagonistic trade-off.

5. Genetically divergent “fast” and “slow” wild populations of K. marmoratus have been

found in habitats with high or low water availability, respectively. Together with our

laboratory data, these findings support the idea that environmental heterogeneity can be

an important factor that drives differences in metabolic rate and pace of life.

176

6.1 Introduction

Life history and metabolic rate are often conceptually linked to form pace of life syndromes, in which organisms span a continuum of “fast” (fast growth, early age at maturity, high metabolic rate) to “slow” (slow growth, delayed maturity, low metabolic rate) lifestyles

(Arnqvist, Stojkovic, Ronn & Immonen, 2017; Auer, Dick, Metcalfe & Reznick, 2018; Reale et al., 2010; Ricklefs & Wikelski, 2002). Pace of life varies substantially both between and within species, and a major goal of physiological ecology is to understand the mechanistic causes and ultimate consequences of this variation (e.g. Burton, Killen, Armstrong & Metcalfe, 2011;

Careau & Garland, 2012; Stearns, 1992). Variation between fast and slow lifestyles is thought to reflect trade-offs in energy allocation, for example to growth versus reproduction. Energy allocation depends first on energy acquisition and resource availability. Spatial or temporal environmental heterogeneity in resource availability is thought to be a key factor that produces and maintains variation in pace of life (Reid, Armstrong & Metcalfe, 2011). However, the direction and mechanistic relationships between environmental conditions, life history, and metabolic rate remain unclear (Burton et al., 2011; Koons, Metcalf & Tuljapurkar, 2008).

To understand the mechanisms responsible for generating and maintaining variation in pace of life, it is useful to examine organisms that inhabit extreme or variable environments (Passow,

Arias-Rodriguez, & Tobler, 2017). For example, amphibious fishes experience extreme and abrupt changes in the physical environmental when moving between water and land (Dejours,

1988; Sayer, 2005). Capturing and consuming prey is problematic for most amphibious fishes on land, as the relatively low density of air makes suction feeding difficult (Heiss, Aerts & Van

Wassenbergh, 2018). Thus, most amphibious fishes must rely on internal energy stores when out of water, although there are a few exceptions (Michel, Heiss, Aerts & Van Wassenbergh, 2015;

177

Van Wassenbergh et al., 2006). The ability to deeply depress metabolic rate and aestivate is thought to be a key factor that enables prolonged survival out of water in amphibious fishes such as the African lungfish Protopterus aethiopicus, which reduce rates of O2 consumption by about

80% on land (Smith, 1931, Delaney, Lahiri, & Fishman, 1974). However, many other species do not aestivate, and the mechanisms that underlie the wide variation in emersion tolerance are not well understood. In these non-aestivating species, a relatively low overall metabolic rate would conserve resources and could enable prolonged survival out of water. Thus, environmental variation in water availability may be a key factor that causes and maintains intra-specific variation in metabolism and life history traits of amphibious fishes, but this hypothesis has not been tested.

Our objective was to understand the proximate and ultimate reasons why some isogenic laboratory strains of K. marmoratus can survive out of water longer than others. First, we tested the hypothesis that the amount of time mangrove rivulus can spend out of water is limited by their pace of life. This hypothesis predicts that emersion-tolerant fish will have relatively slow metabolic and growth rates, reduced consumption of energy stores (glycogen, lipid, and protein), and low levels of activity and reproductive output. We then tested the hypothesis that the costs and benefits of different paces of life ultimately depend on environmental conditions – i.e. there is an antagonistic trade-off between emersion tolerance and aquatic performance. Specifically, we evaluated the prediction that a relatively “fast” lifestyle would be favoured when food and water were constantly available, while a “slow” pace of life would be favoured in a fluctuating environment that frequently lacked water.

To test these hypotheses, we studied the amphibious mangrove rivulus Kryptolebias marmoratus. This fish is one of only two known self-fertilizing hermaphroditic vertebrates

178

(Avise and Tatarenkov, 2015; Harrington, 1961) - the other is the sister species K. hermaphroditus (Costa, 2011). This “selfing” reproductive system allows for the production of large numbers of isogenic, effectively clonal individuals and ultimately enables repeated experiments on the same genotype (Earley, Hanninen, Fuller, Garcia & Lee, 2012; Tatarenkov,

Ring, Elder, Bechler & Avise, 2010; Turko, Earley & Wright, 2011). Mangrove rivulus are capable of surviving more than 66 d out of water (Taylor, 1990), and survive the dry season in leaf litter or packed nose-to-tail in tunnels within rotting mangrove logs (Taylor, Turner, Davis &

Chapman, 2008). There is no evidence that K. marmoratus aestivates when on land (Ong,

Stevens & Wright, 2007; Wright, 2012), but they are largely inactive (Turko, Robertson,

Bianchini, Freeman & Wright, 2014; Turko et al., 2017) and do not eat (Pronko, Perlman &

Ashley-Ross, 2013; Wells, Turko & Wright, 2015). We first measured energy use and oxygen uptake in isogenic strains acclimated to terrestrial conditions for 21 d. Then, fish from “fast” and

“slow” strains were reared together for 12 months in microcosms in which water was either always present, or was absent for random periods totaling 6 months, and the number, condition, and reproductive output of fish was measured.

6.2 Materials and Methods

6.2.1 Animals

Kryptolebias marmoratus hermaphrodites were raised individually in the Hagen Aqualab,

University of Guelph, in 120 mL plastic holding cups. Fish from the isogenic strains 50.91

(“Belize”, from Twin Cayes, Papa Gabriel, Belize), SLC8E (“Florida”, from St. Lucie County,

Florida, USA), and HON11 (“Honduras”, from Bay Islands, Utila, Honduras) were used in these experiments (Tatarenkov, Ring, Elder, Bechler & Avise, 2015). Fish were kept at 25°C, 15‰,

179 with 12 h: 12 h light: dark cycle and were fed Artemia nauplii 3 times per week (Frick & Wright,

2002). Fish were fed to satiation for 3 consecutive days immediately prior to experimentation but were not fed for the duration of the experiments (21 d). In the microcosm experiments fish were fed live Artemia nauplii three times per week, but food was not added to fluctuating microcosms when water was not present. Water samples taken immediately prior to feeding contained live

Artemia that remained from the previous feeding period, indicating that food was available ad libitum.

6.2.2 Emersion tolerance

Emersion tolerance was assessed as described previously (Wells et al., 2015). Size-matched fish (n = 20/strain) were terrestrially acclimated on moist filter paper (Ong, Stevens & Wright,

2007), and survival was monitored at least once per day. For ethical reasons, the experiment was terminated when 20% of each strain remained; these fish were euthanized with an overdose of tricaine methanesulfonate (MS222; 500 mg L-1).

6.2.3 Metabolic rate

To test the hypothesis that emersion tolerance is conferred by low metabolic rate, O2 consumption (n = 8-10/strain) was compared in water (control) and after air exposure using intermittent flow respirometry (Loligo Systems WITROX 4, Tjele, Denmark) as described elsewhere (Sutton, Turko, McLaughlin & Wright, 2018; Livingston, Bhargav, Turko, Wilson &

Wright, 2018) with the following modifications. Aquatic O2 consumption was measured in triplicate (3 h chamber acclimation, 12-15 min recordings, 10 min flushing periods), then the water was drained from the chambers and aerial O2 consumption was measured once per day for

180

7 d at the same time (12:00-16:00) to minimize the effect of diurnal metabolic rhythms (Rodela

& Wright, 2006). Fresh humidified air was introduced into the chambers between measurements.

Fish were weighed before and after the experiment and the average mass was used to standardize

O2 uptake. To measure O2 uptake after three weeks in air, an initial O2 consumption rate in water was first determined for a separate group of fish as described above (n = 10/strain). Fish were then air-exposed on filter paper for 21 d and the measurement was repeated.

6.2.4 Energy reserves and consumption

To test whether the amount of energy reserves and/or rate of energy use was related to emersion tolerance, we measured activity, overall body condition, and body composition. To determine activity during emersion, fish (n = 8/strain) were photographed every 5 s for 1 h at six time points (1 h, and 1, 3, 7, 14, 21 d out of water). Activity was quantified by determining the proportion of photos in which fish moved (Turko et al., 2014). Fulton’s condition factor (K), a general index of body condition (Froese, 2006) was calculated for control fish in water (n =

6/strain), and in a separate group of fish (n = 12/strain) that were terrestrially acclimated for 21 d.

To measure body composition, independent groups of fish were required because measurements of energy reserves were terminal and the small size of mangrove rivulus precluded measuring multiple energy reserves in the same sample. Fish held in water (control) or air (21 d) were euthanized with an overdose of MS222, blotted dry, weighed, and snap frozen in liquid nitrogen

(for glycogen and protein determination) or dried (48 h at 50°C) for lipid analysis and measurement of water content. Glycogen content of whole fish (n = 8-9/strain) was measured enzymatically (Bergmeyer, Berndt, Schmidt & Stork, 1974). Whole body lipid stores (n = 6-

12/strain) were measured by chloroform extraction (Junior & Peixoto, 2013). Crude protein (n =

181

6-10/strain) was measured using the Kjeldahl method (AOAC 1995). Tecator Kjeltec digestion and distillation units (Foss, Eden Prairie, MN, USA) were used for protein analysis and the percentage of total nitrogen was determined based on a dry matter basis (%N × 6.25) (Bureau et al., 2000). Total body water content (n = 6-12/strain) was calculated by subtracting wet body mass from dry body mass and dividing by the wet mass.

To calculate energy use during emersion, lipid, glycogen, and protein utilization were each calculated by subtracting the average mass-specific energy stores (mg g-1) in terrestrially- acclimated fish (Eterr) from those of the control fish (Econ), accounting for changes in body water content (W) and the average change in body mass (ΔBM), according to the formula:

퐸푛푒푟푔푦 푐표푛푠푢푚푒푑 = (퐸푐표푛)(1 − 푊푐표푛) − (퐸푡푒푟푟)(1 − 푊푡푒푟푟)(훥퐵푀)

All data used for these calculations is provided in Table 6.S1. Total energy use was calculated by adding the energy contained in the consumed glycogen (17 kJ g-1), lipid (37 kJ g-1), and protein

(17 kJ g-1). Overall standard deviations were calculated using standard methods of error propagation, and effective degrees of freedom were calculated using the Welch-Satterthwaite equation (JCGM, 2008). Using these values, we compared energy use among the isogenic lineages with one-way ANOVAs and post hoc Holm–Sidak tests. Data were ln transformed when necessary to meet assumptions of normal distribution and equal variance.

6.2.5 Life history traits

Routinely collected records from our K. marmoratus breeding colony were used to compare overall embryo production, clutch size, and age at first reproduction among the strains. To assess cumulative reproductive output, we only used data from fish that hatched within one year of those used for experiments, released at least one embryo, were never used for any experiments,

182 and survived for over 18 months (Belize n = 42, Florida n = 27, Honduras n = 20). Age at first reproduction was determined for a larger subset of fish that simply released an embryo prior to use in any experiments (Belize n = 90, Florida n= 72, Honduras n = 90).

6.2.6 Microcosms

We used a 12-month microcosm experiment to test the hypothesis that there is an environmentally mediated trade-off between emersion intolerant/high metabolism and emersion tolerant/low metabolism phenotypes. Belize and Honduras fish (n = 3/strain) were placed into each microcosm at the start of the experiment. Fish were size-matched to minimize performance differences in aggressive/competitive interactions between the strains (Earley & Hsu, 2008), which resulted in Belize fish being slightly older (434 ± 9.4 versus 325 ± 17 days old) but of similar mass (Figure 6.S1) at the beginning of the experiment. All fish were sexually mature and had released at least one embryo in the laboratory colony before being placed in a microcosm to standardize reproductive status. In control microcosms (n = 10) we maintained constant water levels, while fluctuating microcosms (n = 10) were randomly drained and refilled (every 1-3 weeks) such that water was absent for half of the experiment. Microcosms were constructed from

10 L plastic boxes filled half-way with 15‰ water. The bottom was covered with soft filtration media to provide a moist substrate in the fluctuating treatment when water was absent, and 20 pieces of plastic pipe (3 cm length, 1.5 cm diameter) and three green acrylic yarn mops were added to provide shelter. Emersion periods never exceeded 3 weeks as our emersion tolerance data showed that 100% of fish survived this length of time, and the goal of this experiment was explicitly to test whether sub-lethal effects could mediate a trade-off between “fast” and “slow” fish. Water levels were altered (fluctuating) or water was refreshed (control) via permanently- installed plastic tubing under the filter media to minimize disturbance.

183

After the 12-month experiment, all adult fish were euthanized (MS222), photographed (for length measurements), weighed, and a piece of caudal fin tissue was fixed in DNA preservative

(0.25 M ethylenediaminetetraacetic acid, 20% dimethyl sulfoxide, NaCl saturated, pH = 7.5;

Seutin, White & Boag, 1991) for determination of genetic identity. Gonads and liver were dissected and weighed to assess condition. Gills were also removed to test whether gill surface area was related to pace of life differences. The number and length of gill filaments was measured in whole mounts of the left-side arches. Sex of each fish was assessed based on external morphology (Scarsella, Gresham & Earley, 2018) and appearance of ovarian tissue in the gonads. All embryos were collected and fixed in DNA preservative (see above) for genetic identification. No larvae or juvenile fish were found.

The genetic strain (i.e. Belize or Honduras) of each adult and embryo was determined using previously described protocols and microsatellite markers (Mackiewicz et al. 2006; Tatarenkov et al., 2010). Genomic DNA was extracted and purified using a commercially available kit according to manufacturer’s instructions (GeneJET DNA Purification Kit, Fisher Scientific).

Microsatellite “R18” from Mackiewicz et al. (2006) was used to differentiate between strains, as this is one of the three most divergent loci between Belize and Honduran fish and amplified more consistently than the other most divergent loci (R3 and R34). PCR products were run on an acrylamide gel (5%, 3000V for 2 h at 55°C), and were manually scored for strain identity. We were able to identify 100% of the adult fish, however, some embryos (20 of 81 from control, 209 of 979 from fluctuating microcosms) could not be confidently assigned to either strain, probably due to low DNA content. These unidentified embryos were excluded from statistical analysis.

184

6.3 Results

Survival out of water was significantly longer in the Honduras strain relative to the Belize and Florida strains (Kaplan-Meier log-rank statistic = 11.198, p = 0.004; Figure 6.1). The

Honduras strain had consistently lower (by 30-50%) rates of O2 consumption in both water and air relative to Belize and Florida strains over 7 d (two-way ANOVA, F2,24 = 9.96, p < 0.001;

Figure 6.2A) and 21 d in air (two-way ANOVA, F2,27 = 4.81, p = 0.016; Figure 6.2B). Overall,

O2 consumption increased in all strains after 2 and 3 d in air compared to other timepoints (two- way ANOVA, F2,24 = 2.96, p = 0.004, interaction p = 0.20; Figure 6.2A), but was on average

44% lower after 21 days in air (two-way ANOVA, F2,27 = 18.48, p < 0.001, interaction p = 0.75;

Figure 6.2B).

There was no difference in initial body condition among the three strains, but terrestrial acclimation for 21 d resulted in significantly lower condition factor of Belize and Florida, but not

Honduras, fish (two-way ANOVA F2,47 = 3.35, interaction p = 0.024; Figure 6.3). Similarly, in paired measurements of different individuals, Honduras fish lost significantly less body mass than other strains after 21 d out of water (ANOVA F2,27 = 6.92, p = 0.004, Figure 6.S3). All three strains in water had similar levels of glycogen, lipid, and protein stores at the beginning of the experiment (two-way ANOVA, all p > 0.05; Figure 6.S2). After 21 d out of water, there was no significant difference among strains in lipid use (ANOVA, F2,83 = 0.92, p = 0.40; Figure 6.4A), but Florida fish used more glycogen than the other strains (ANOVA, F2,191 = 6.72, p = 0.002;

Figure 6.4B). Florida fish also consumed the most protein, and Honduras fish used the least

(ANOVA, F2,779 = 47.60, p < 0.001; Figure Fig 6.4C). Overall, Honduras fish metabolized less energy over 21 d on land compared to Belize and Florida fish (ANOVA, F2,65 = 5.21, p = 0.008;

Figure 6.4D). Water content was significantly elevated in all strains after 21 days on land (two-

185 way ANOVA, F1,53 = 75.31, p <0.001, interaction p = 0.60; Figure 6.S2D). Activity generally decreased over the course of 21 d out of water, but differed among the strains only at the 1 h time point (two-way ANOVA F2,16 = 5.67, interaction p = 0.001; Figure 6.S4).

Belize fish produced more embryos (Kruskal-Wallis ANOVA on ranks H2 = 10.61, p =

0.005; Figure 6.5A) and had larger mean clutches (Kruskal-Wallis ANOVA on ranks H2 = 20.12, p < 0.001; Figure 6.5B) over the first 18 months of life, while age at first reproduction tended to be earlier (Kruskal-Wallis ANOVA on ranks H2 = 6.91, p = 0.032; Figure 6.5C). Honduras fish were heavier (ANCOVA F1,287 = 163.75, p < 0.0001; Figure 6.S1A) and longer (ANCOVA F1,287

= 83.60, p < 0.0001; Figure 6.S1B) than Belize fish at a given age, indicating higher a growth rate.

At the end of the microcosm experiment, there were no differences in the number of adult fish per strain (two-way RM ANOVA, F1,18 = 3.01, p = 0.10) or in each treatment (F1,18 = 2.14, p

= 0.16, interaction p = 0.78; Figure 6.6A). Honduras fish were longer (two-way ANOVA, F1,110

= 49.95, p < 0.001; Figure 6.6B), heavier (two-way ANOVA, F1,110 = 80.40, p < 0.001; Figure

6.6C), and in better condition overall (two-way ANOVA, F1,110 = 36.24, p < 0.001; Figures 6.6D,

6.S5A). Gonado-somatic index was also significantly higher in Honduras fish (two-way

ANOVA, F1,110 = 93.76, p < 0.001; Figures 6.6E, 6.S5B), but there was no difference in hepato- somatic index (two-way ANOVA, F1,110 = 1.4, p = 0.23; Figures 6.6F, 6.S5C). Fish from control microcosms were longer (two-way ANOVA, F1,110 = 168.04, p < 0.001) heavier (two-way

ANOVA, F1,110 = 214.96, p < 0.001), and in better condition (Fulton’s K: two-way ANOVA,

F1,110 = 42.02, p < 0.001; GSI F1,110 = 7.77, p = 0.006) than fish from fluctuating microcosms, but there was no difference in HSI (F1,110 = 3.70, p = 0.057; Figures 6.6, 6.S5). After controlling for body length, total gill filment length was not different between strains (ANCOVA, F1,82 = 1.22, p

186

= 0.27; Figure 6.S6) or microcosm conditions (ANCOVA, F1,82 = 0.85, p = 0.36; Figure 6.S6).

There were almost 11-fold more embryos recovered from fluctuating (89.8 ± 7.6) versus control

(8.1 ± 1.4) microcosms (Mann-Whitney, U = 0, p < 0.001). After accounting for the number adult fish in each microcosm, embryo quantity was dependent on a significant strain by treatment interaction (ANCOVA, F1,32 = 4.36, p = 0.045). Honduras fish produced significantly more embryos than Belize fish under fluctuating conditions (Tukey, p = 0.021; Figure 6.7), but there was no difference between strains under control conditions (p = 0.98). Two embryos collected from a single fluctuating microcosm were heterozygous at the microsatellite locus we used for identification, indicating outcrossing between the two strains. No adult males were found in the microcosm that contained the heterozygous embryos. Seven adult males were present at the end of the experiment, each in a different microcosm (5 control, 2 fluctuating).

6.4 Discussion

Using a self-fertilizing amphibious fish, we were able to directly relate survival out of water to genetically-based metabolic and life history phenotypes. In support of the proximate hypothesis that the overall pace of life determines the length of time non-aestivating amphibious fish can survive out of water, we found that emersion tolerance was negatively correlated with metabolic rate and the rate of fuel use, but not the size of initial energy reserves. The emersion- tolerant Honduras strain also produced fewer offspring, due to an increased age of first reproduction and smaller average clutch size, consistent with a slow pace of life. We then used long term microcosms to test the ultimate hypothesis that a sublethal trade-off between emersion tolerance and aquatic performance mediates the relative advantages of different paces of life.

Contrary to the prediction made by this hypothesis, “slow” Honduras fish did not outcompete

187

relatively “fast” Belize fish (in terms of adult population size) under conditions of low water

availability. However, Honduras fish produced more embryos than Belize fish in the water-

limited condition, in support of the trade-off hypothesis. Furthermore, Honduras fish tended to be

larger and in better condition regardless of water availability, suggesting an overall advantage to

low metabolic rate under our experimental conditions.

6.4.1 Metabolism and emersion tolerance

The rate of O2 consumption in Honduras fish was consistently lower than the other strains in both water and air, suggesting that this strain has an inherently slow metabolism that is genetically based. Furthermore, the scope of metabolic depression after 21 d in air in Honduras fish was larger (58% reduction) than either the Belize (-31%) or Florida (-44%) strains, which probably allowed Honduras fish to conserve protein stores. Previous studies have suggested that

K. marmoratus maintain or increase metabolic rate for several days after moving from water to land, but these experiments did not examine fish that were out of water for longer than 5 days

(Ong et al., 2007, Livingston et al., 2018). While probably helpful for survival during prolonged emersion, the scope of metabolic depression we found in K. marmoratus is smaller than has been measured in classically aestivating amphibious fishes such as P. aethiopicus, Synbranchus marmoratus, and Lepidogalaxias salamandroides that reduce O2 consumption by 65-80% during months-long aestivation in mud (Guppy & Withers, 1999). Interestingly, the control rate of O2 consumption in Honduras fish was almost identical to the depressed rate in the other two strains.

One possibility is that the constitutively low metabolic rate of the Honduras strain evolved via the genetic assimilation of metabolic plasticity, which would be expected if these fish were found in habitats that regularly dried (Pigliucci et al., 2006; Lande, 2009). More research is required to

188 understand the mechanism by which Honduras fish achieve low metabolic rate (e.g. increased efficiency or reduced expenditure), especially given the high growth rate of this strain. Overall, however, constitutive expression of a low baseline metabolic rate, combined with the ability to further reduce metabolism during extended periods without water, probably allows the Honduras strain to prolong survival out of water by conserving energy reserves.

There was a small but consistent increase in O2 consumption after 2-3 d out of water across

all three strains we investigated, despite the fact that fish remained largely motionless. Consistent

with this finding, previous work found increased rates of CO2 excretion in mangrove rivulus over

5 d of terrestrial acclimation (Ong et al., 2007). This transient increase in metabolic rate may

reflect the energetic cost of mounting phenotypically plastic responses during air exposure, such

as gill remodelling (Ong et al., 2007), cutaneous angiogenesis (Cooper, Litwiller, Murrant &

Wright, 2012; Turko et al., 2014), and enlargement of cutaneous ionocytes (LeBlanc, Wood,

Fudge & Wright, 2010).

The pattern of energy use varied among strains in a manner consistent with the differences in

whole animal O2 consumption. In independent experiments, Honduras fish lost the least amount

of body mass relative to the other two strains and showed no change in condition factor after 21

d out of water. Protein catabolism was also lowest in Honduras fish. Generally, teleosts use

protein and some lipids as fuel sources when food is not restricted, but glycogen and lipids are

more important during starvation (Jobling, 1994; Moyes & West, 1995). However, African

lungfish defend glycogen stores during aestivation, perhaps to facilitate rapid recovery when

routine metabolism must be restored (Frick, Bystriansky, Ip, Chew & Ballantyne, 2008). After

21 days of emersion, all three strains of mangrove rivulus we tested had consumed a large

fraction of their glycogen (~80% of initial) and lipid (~61%) stores, but only some protein (17-

189

29%, depending on strain). This pattern resembles that of a typical starving teleost, rather than an aestivating lungfish. Presumably, the Florida and Belize fish, with higher metabolic rates, began to metabolise protein earlier in the emersion period than Honduras fish, resulting in greater consumption at the 21 d timepoint. In Honduras fish, larger protein reserves after 21 d of air exposure, in addition to fueling continued emersion, may also help preserve locomotor ability when these fishes return to water, similar to inactive hibernating mammals that protect protein stores to minimize impairment of skeletal muscle function when they emerge from winter dens

(e.g. Hindle et al., 2015).

There were no differences in initial energy stores among the Belize, Florida, and Honduras strains. One possible explanation is that there may be costs to carrying large energy stores, such as increased attractiveness to predators (Jensen et al., 2012) or decreased locomotory performance (Gibb, Ashley-Ross & Hsieh, 2013). Alternatively, fish in the wild may detect environmental cues that indicate the onset of the dry season and respond by increasing internal energy stores (Griffiths & Kirkwood, 1995; Schultz & Conover, 1997). In our laboratory setting, however, such anticipatory feeding would not have been possible as the fish were given no signals that emersion was imminent.

6.4.2 Emersion tolerance trade-offs

Pace of life theory makes conflicting predictions about the direction of correlations between metabolic rate, growth, and reproductive investment (Burton et al., 2011). According to the acquisition model, a fast metabolism allows more resources to be acquired, leading to faster growth and increased reproductive investment (Mathot & Dingemanse, 2015). On the other hand, the allocation or compensation model suggests that limited resources are split between

190 various physiological processes via trade-offs; negative relationships between each of metabolic rate, growth, and reproduction are thus expected (Burton et al., 2011). We found that Honduras fish, with the lowest metabolic rate, also produced fewer embryos but grew faster than Belize fish under standard laboratory rearing conditions, suggesting a trade-off consistent with the allocation model. Furthermore, Honduras fish were much larger than Belize fish in both microcosm conditions. Although faster growth is typically correlated with high metabolic rates in animals (Allen, Rosenfeld & Richards, 2016), the negative relationship between growth and metabolic rate we found in K. marmoratus has also been found in some other fishes (Alvarez &

Nicieza, 2005; Norin and Malte, 2011).

Our results support, in part, the hypothesis that differences in metabolic rate between mangrove rivulus strains are ultimately caused and maintained by a trade-off between emersion tolerance and aquatic performance. This hypothesis predicts that the “slow” Honduras fish would have higher fitness than Belize fish in fluctuating conditions when water was periodically unavailable, and this was indeed the case by several metrics. Honduras fish were larger, in better condition, and produced more embryos than Belize fish in the fluctuating condition. However,

Honduras fish were also larger in the control microcosms and there was no difference between strains in embryo production, in contrast to the prediction that Honduras fish would have lower fitness in constant aquatic conditions. However, our results are consistent with many other studies of local adaptation, which often find that phenotypes which are advantageous under some conditions have neutral consequences in other environments, especially when environmental heterogeneity is generally high (Bono, Smith Jr., Pfennig & Burch, 2017).

It is not clear whether “fast” and “slow” phenotypes have effectively neutral fitness consequences under fully aquatic conditions, or whether there are situations that would favour

191 the “fast” Belize phenotype. One limitation to our microcosm experiment was that fish were unable to disperse. The natural habitat of mangrove rivulus and other rivuline killifish typically consists of a mosaic of small, intermittent pools (e.g. Furness et al. 2018; Sutton et al., 2018). A fast pace of life is often correlated with high activity (Gangloff et al., 2017; Rádai, Kiss & Barta,

2017), and perhaps Belize fish are more likely to leave water and disperse to unoccupied habitats. Embryo production by Belize fish was highest of any strain in our laboratory colony, typical of a dispersal phenotype. Furthermore, if embryo production in the microcosm experiment is standardized to gonad mass, rather than the number of adults, the Belize strain actually produced almost twice as many embryos per gram of gonad compared to Honduras fish.

We also found nearly 9-fold more embryos in fluctuating vs. control conditions, despite the larger overall size of control fish. We think it is unlikely that this disparity reflects differences in reproductive output, but instead is the result of cannibalism under aquatic conditions (Wells et al., 2015). Fluctuating microcosms were sampled after one week without water, so embryos deposited during this time could not have been consumed. Thus, Belize fish may live a traditionally “fast” pace of life that favours reproduction over growth, but the inability to escape competition/embryo cannibalism from the larger Honduras fish in our microcosms nullified the advantages of this life history strategy. High metabolic rates are often also correlated with aggression (Reale et al., 2010), and thus the fast Belize phenotype may be superior competitors in situations of relative food scarcity – our conditions of ad libitum food provisioning would have failed to detect such a benefit to increased metabolic rate however.

Relative investment in organ size and function is one mechanism that can mediate trade-offs between metabolic rate and performance (Allen et al., 2016; Careau & Garland, 2012; White &

Kearney, 2013). We found no evidence for such a trade-off in mangrove rivulus. Honduras fish

192 had the largest gonads relative to body mass, while there was no difference in the size of the liver or gills compared to Belize fish. However, gonads of Belize fish in the microcosm experiment produced twice as many embryos as Honduras fish when standardized to gonad mass, suggesting increased rates of protein synthesis and other metabolic activity that may have contributed to elevated whole-animal metabolic rate.

The androdioecious mating system of mangrove rivulus is thought to provide these fish a mechanism to benefit from both self-fertilization and outcrossing, depending on the context

(Avise and Tatarenkov, 2015; Ellison, Cable & Consuegra, 2011). This “best of both worlds” hypothesis predicts that animals in relatively suitable habitats should self-fertilize to maximize the genetic inheritance by the offspring, while those in less suitable habitats should opt for outcrossing to increase the genetic variation of progeny so that some offspring will be better suited to the environmental conditions. In our experiments, this could be reflected as an increased number of male Belize fish and higher rate of outcrossing in fluctuating microcosms, while in control conditions more Honduras males and outcrossing would be predicted (due to environmental mismatching). We found no evidence consistent with these predictions. Only 7 of

114 adult fish were male, and 2 of 979 identified embryos were the result of outcrossing between strains. One possibility is that despite being without water for 6 of 12 months, the fluctuating microcosms were not sufficiently stressful for the benefits of outcrossing to outweigh the cost of reduced genetic inheritance. Alternatively, the mating system of mangrove rivulus may not be flexible within adults, and could instead rely on developmental plasticity or stochastic epigenetic effects that act on early life stages. More research is needed to understand the origins and maintenance of androdioecy in this species.

193

We found two embryos that were heterozygous at the microsatellite locus we studied, indicating that these were the result of outcrossing. The current view is that males are required for outcrossing (Avise & Tatarenkov, 2015; Furness, Tatarenkov, & Avise, 2015), but no males were found in the microcosm that contained these outcrossed embryos. One possibility is that a male(s) was present when the embryos were fertilized, but subsequently died and decomposed before the microcosms were sampled. Alternatively, these embryos resulted from outcrossing between a hermaphroditic individual of each strain. If so, these two embryos would represent a hermaphrodite-driven outcrossing rate of ~0.2% given that we genotyped 831 embryos. Such a low rate is consistent with the results of Furness et al. (2015), who found no sign of outcrossing in a sample of 173 embryos.

6.4.3 Conclusions and perspective

Our findings show that pace of life was associated with emersion tolerance in K. marmoratus out of water, which allowed the “slow” Honduras strain to survive out of water for an average of

12 days longer than the relatively “fast” Belize strain. Furthermore, Honduras fish had a lower metabolic rate under control aquatic conditions, indicating genetic divergence between strains that could possibly be the result of genetic assimilation of metabolic plasticity (Pigliucci et al.,

2006; Lande, 2009). These results are consistent with other work showing generally “slow” lifestyles in extremophile fishes (e.g. Passow et al., 2017). We did not detect an obvious cost to emersion tolerance in our microcosm experiment, consistent with other studies of local adaptation that have often found conditional neutrality of variable traits instead of antagonistic trade-offs (Bono et al., 2017), but we also did not push environmental extremes to the limit, and life history trade-offs are often only revealed under extreme environmental conditions (Lemaître

194 et al., 2015). Furthermore, the mechanism(s) driving variation in metabolic rate are probably more nuanced than just water availability, as many other factors (e.g. temperature, food availability, crowding) in the wild will often covary with water availability.

Recently, we discovered genetically divergent wild populations of K. marmoratus occupying abiotically distinct habitats (high versus low water availability), and the phenotypes of these populations matched those predicted by pace of life theory (Turko et al., 2018). The population that inhabits an ephemeral pond (no water during the dry season) tended to be much larger, in better body condition (Fulton’s condition factor), and had metabolic rates that were about 30% lower than fish from a nearby site with much higher water availability. In combination with our laboratory data, these findings support the idea that environmental heterogeneity can be an important factor that drives differences in metabolic rate and pace of life.

195

6.5 References

Allen, D., Rosenfeld, J. & Richards, J. (2016). Physiological basis of metabolic trade-offs

between growth and performance among different strains of rainbow trout. Canadian Journal

of Fisheries and Aquatic Science, 73, 1493-1506.

Alvarez, D. & Nicieza, A. G. (2005). Is metabolic rate a reliable predictor of growth and survival

of brown trout (Salmo trutta) in the wild? Canadian Journal of Fisheries and Aquatic

Science, 62, 643-649.

Arnqvist, G., Stojković, B., Rönn, J. L. & Immonen, E. (2017). The pace-of-life: A sex-specific

link between metabolic rate and life history in bean beetles. Functional Ecology, 31, 2299-

2309.

Bergmeyer, H. U., Berndt, E., Schmidt, F., & Stork, H. (1974). D-Glucose: determination with

hexokinase and glucose-6-phosphate dehydrogenase. In Methods of enzymatic analysis, 2nd

Ed (ed HU Bergmeyer) pp. 1196–1201. New York, NY: Academic Press, Inc.

Bicego, K. C., Barros, R. C. H., & Branco, L. G. S. (2007). Physiology of temperature

regulation: comparative aspects. Comparative Biochemistry and Physiology, 147, 616-639.

Bono, L. M., Smith Jr., L. B., Pfennig, D. W., & Burch, C. L. (2017). The emergence of

performance trade-offs during local adaptation: insights from experimental evolution.

Molecular Ecology, 26, 1720-1733.

Booth, D. J., & Keast, J. A. (1986). Growth energy partitioning by juvenile bluegill sunfish,

Lepomis macrochirus Rafinesque. Journal of Fish Biology, 28, 37–45.

Bureau, D. P., Harris, A. M., Bevan, D. J., Simmons, L. A., Azevedo, P. A., & Cho, C. Y.

(2000). Feather meals and meat and bone meals from different origins as protein sources in

rainbow trout (Oncorhynchus mykiss) diets. Aquaculture, 181, 281-291.

196

Burton, T., Killen, S. S., Armstrong, J. D., & Metcalfe, N. B. (2011). What causes intraspecific

variation in resting metabolic rate and what are its ecological consequences? Proceedings of

the Royal Society B, 278, 3465-3473.

Careau, V., & Garland, T. (2012). Performance, personality, and energetics: correlation,

causation, and mechanism. Physiological and Biochemical Zoology, 85, 543–571.

Careau, V, Thomas, D., Pelletier, F., Turki, L., Landry, F., Garant, D., & Réale, D. (2011).

Genetic correlation between resting metabolic rate and exploratory behaviour in deer mice

(Peromyscus maniculatus). Journal of Evolutionary Biology, 24, 2153-2163.

Cooper, H. M., Herbin, M., & Nevo, E. (1993). Visual system of a naturally microphthalmic

mammal: The blind mole rat, Spalax ehrenbergi. Journal of Comparative Neurology, 328,

313-350.

Costa, W. J. E. M. (2011). Identity of Rivulus ocellatus and a new name for a hermaphroditic

species of Kryptolebias from south-eastern Brazil (Cyprinodontiformes: Rivulidae).

Ichthyological Explorations of Freshwaters, 22, 185–192.

Delaney, R. G., Lahiri, S., & Fishman, A. P. (1974). Aestivation in the African lungfish

Protopterus aethiopicus: cardiovascular and respiratory functions. Journal of Experimental

Biology, 61, 111-128.

Dykhuizen, D. (1978). Selection for tryptophan auxotrophy of Escherichia coli in glucose

limited chemostats as a test of the energy conservation hypothesis. Evolution, 32, 125- 150.

Earley, R. L., & Hsu, Y. (2008). Reciprocity between endocrine state and contest behavior in the

killifish, Kryptolebias marmoratus. Hormones and Behaviour, 53, 442-451.

197

Earley, R. L., Hanninen, A. F., Fuller, A., Garcia, M. J., & Lee, E. A. (2012). Phenotypic

plasticity and integration in the mangrove rivulus (Kryptolebias marmoratus): A

prospectus. Integrative and Comparative Biology, 52, 814-827.

Eduardo, J., Bicudo, P. W., & Johansen, K. (1979). Respiratory gas exchange in the

airbreathing fish, Synbranchus marmoratus. Environmental Biology of Fishes, 4, 55-64.

Fishman, A. P., Pack, A. I., DeLaney, R. G., & Galante, R. J. (1986). Estivation in Protopterus.

Journal of Morphology, 190, 237-248.

Frick, N. T., & Wright, P. A. (2002). Nitrogen metabolism and excretion in the mangrove

killifish Rivulus marmoratus II. Significant ammonia volatilization in a teleost during air-

exposure. Journal of Experimental Biology, 205, 91-100.

Froese, R. (2006). Cube law, condition factor and weight–length relationships: history, meta-

analysis and recommendations. Journal of Applied Ichthyology, 22, 241–253.

Gangloff, E. J., Vleck, D., & Bronikowski, A. M. (2015). Developmental and immediate thermal

environments shape energetic trade-offs, growth efficiency and metabolic rate in divergent

life-history ecotypes of the garter snake, Thamnophia elegans. Physiological and Biochemical

Zoology, 88, 550-563.

Gangloff, E. J., Chow, M., Leos-Barajas, V., Hynes, S., Hobbs, B., & Sparkman, A. M. (2017).

Integrating behaviour into the pace-of-life continuum: Divergent levels of activity and

information gathering in fast- and slow-living snakes. Behavioral Processes, 142, 156-163.

Gibb, A. C., Ashley-Ross, M. A., & Hsieh, S. T. (2013). Thrash, flip, or jump: The behavioral

and functional continuum of terrestrial locomotion in teleost fishes. Integrative and

Comparative Biology, 53, 295-306.

198

Gordon, M. S., Ng, W. W., & Yip, A. Y. (1978). Aspects of the physiology of terrestrial life in

amphibious fishes. III. The Chinese mudskipper Periophthalmus cantonensis. Journal of

Experimental Biology, 72, 57-75.

Gordon, M. S., Boetius, I., Evans, D. H., McCarthy, R., & Oglesby, L. C. (1969). Aspects of the

physiology of terrestrial life in amphibious fishes. I. The mudskipper Periophthalmus

sobrinus. Journal of Experimental Biology, 50, 141-149.

Graham, J. B. (1997). Air-Breathing Fishes: Evolution, Diversity, and Adaptation. San Diego,

CA: Academic Press.

Graham, J. B., Jones, C. B., & Rubinoff, I. (1985). Behavioural, physiological and ecological

aspects of the amphibious life of the pearl blenny Entomacrodus nigracans gill. Journal of

Experimental Marine Biology and Ecology, 89, 255-268.

Griffiths, D., & Kirkwood, R. C. (1995). Seasonal variation in growth, mortality, and fat stores

of roach and perch in Lough Neagh, Northern Ireland. Journal of Fish Biology, 47, 537–554.

Guppy, M. & Withers, P. (1999). Metabolic depression in animals: physiological perspectives

and biochemical generalizations. Biological Reviews, 74, 1-40.

Harrington, R. W. (1961). Oviparous hermaphroditic fish with internal self-fertilization. Science,

134, 1749-1750.

Harshman, L. G., Hoffmann, A. A., & Clark, A. G. (1999). Selection for starvation resistance in

Drosophila melanogaster: physiological correlates, enzyme activities and multiple stress

responses. Journal of Evolutionary Biology, 12, 370-379.

Heiss, E., Aerts, P., & Van Wassenbergh, S. (2018). Aquatic–terrestrial transitions of feeding

systems in vertebrates: a mechanical perspective. Journal of Experimental Biology, 221,

jeb154427.

199

Hindle, A. G., Otis, J. P., Epperson, L. E., Hornberger, T. A., Goodman, C. A., Carey, H. V., &

Martin, S. L. (2015). Prioritization of skeletal muscle growth for emergence from

hibernation. Journal of Experimental Biology, 218, 276-284.

Jensen, K., Mayntz, D., Toft, S., Clissold, F. J., Hunt, J., Raubenheimer, D., & Simpson, S. J.

(2012). Optimal foraging for specific nutrients in predatory beetles. Proceedings of the Royal

Society B, 279, 2212-2218.

Jobling, M. (1994). Fish bioenergetics. London, ENG: Chapman and Hall.

Joint Committee for Guides in Metrology (JCGM) (2008). Evaluation of measurement data –

Guide to the expression of uncertainty in measurement. Retrieved from:

www.bipm.org/utils/common/documents/jcgm/JCGM_100_2008_E.pdf

Junior, R. S. L., & Peixoto, P. E. C. (2013). Males of the dragonfly Diastatops obscura fight

according to predictions from game theory models. Animal Behaviour, 85, 663-669.

Lande, R. (2009). Adaptation to an extraordinary environment by evolution of phenotypic

plasticity and genetic assimilation. Journal of Evolutionary Biology, 22, 1435-1446.

Lauff, R. F., & Wood, C. H. (1996). Respiratory gas exchange, nitrogenous waste excretion, and

fuel usage during aerobic swimming in juvenile rainbow trout. Journal of Comparative

Physiology B, 166, 501-509.

LeBlanc, D. M., Wood, C. M., Fudge, D. S., & Wright, P. A. (2010). A fish out of water: gill and

skin remodeling promotes osmo-and ionoregulation in the mangrove killifish Kryptolebias

marmoratus. Physiological and Biochemical Zoology, 83, 932-949.

Lemaître, J.- F., Berger, V., Bonenfant, C., Douhard, M., Gamelon, M., Plard, F., & Gaillard, J.-

M. (2015). Early-late life trade-offs and the evolution of ageing in the wild. Proceedings of

the Royal Society B, 282, 20150209.

200

Litwiller, S. L., O'Donnell, M. J., & Wright, P. A. (2006). Rapid increase in the partial pressure

of NH3 on the cutaneous surface of air-exposed mangrove killifish, Rivulus

marmoratus. Journal of Experimental Biology, 209, 1737-1745.

Livingston, M. D., Bhargav, V. V., Turko, A. J., Wilson, J. M., &Wright, P. A. (2018).

Widespread use of emersion and cutaneous ammonia excretion in Aplocheiloid killifishes.

Proceedings of the Royal Society B, 20181496.

Mackiewicz, M., Tatarenkov, A., Perry, A., Martin, J. R., Elder, J. F., Bechler, D. L., & Avise, J.

C. (2006). Microsatellite documentation of outcrossing between inbred laboratory strains of

the self-fertilizing mangrove killifish (Kryptolebias marmoratus). Journal of Heredity, 97,

508–513.

Martin, K. L. (2014). Theme and variations: amphibious air-breathing intertidal fishes. Journal

of Fish Biology, 84, 577-602.

Mathot, K. J., & Dingemanse, N. J. (2015). Energetics and behavior: Unrequited needs and new

directions. Trends in Ecology and Evolution, 30, 199-206.

McCue, M. D. (2010). Starvation physiology: Reviewing different strategies animals use to

survive a common challenge. Comparative Biochemistry and Physiology A, 156, 1-18.

Michel, K. B., Heiss, E., Aerts, P., & Van Wassenbergh, S. (2015). A fish that uses its

hydrodynamic tongue to feed on land. Proceedings of the Royal Society B, 282, 1805.

Moyes, C. D., & West, T. G. (1995). Exercise metabolism of fish. In: Metabolic Biochemistry.

Biochemistry and Molecular Biology of Fish, vol. 4 (eds PW Hochachka, TP Mommsen)

Amsterdam, NL: Elsevier Science.

201

Mueller, P., & Diamond, J. (2001). Metabolic rate and environmental productivity: well-

provisioned animals evolved to run and idle fast. Proceedings of the National Academy of

Sciences, 98, 2550-2554.

Norin, T. and Malte, H. (2011). Repeatability of standard metabolic rate, active metabolic rate

and aerobic scope in young brown trout during a period of moderate food availability. Journal

of Experimental Biology, 214, 1668-1675.

Ong, K. J., Stevens, E. D., & Wright, P. A. (2007). Gill morphology of the mangrove killifish

(Kryptolebias marmoratus) is plastic and changes in response to terrestrial air exposure.

Journal of Experimental Biology, 210, 1109-1115.

Passow, C. N., Arias-Rodriguez, L., & Tobler, M. (2017). Convergent evolution of reduced

energy demands in extremophile fish. PLoS ONE, 12, e0186935.

Pigliucci, M., Murren, C. J., & Schlichting, C. D. (2006). Phenotypic plasticity and evolution by

genetic assimilation. Journal of Experimental Biology, 209, 2362-2367.

Pronko, A. J., Perlman, B. M., & Ashley-Ross, M. A. (2013). Launches, squiggles and pounces,

oh my! The water-land transition in mangrove rivulus (Kryptolebias marmoratus). Journal of

Experimental Biology, 216, 3988-3995.

Rádai, Z., Kiss, B. & Barta, Z. (2017). Pace of life and behaviour: rapid development is linked

with increased activity and voracity in the wolf spider Pardosa agrestis. Animal Behaviour,

126, 145-151.

Réale, D., Garant, D., Humphries, M. M., Bergeron, P., Careau, V., & Montiglio, P. O. (2010).

Personality and the emergence of the pace-of-life syndrome concept at the population level.

Philosophical Transactions of the Royal Society B, 365, 4051-4063.

202

Reid, D., Armstrong, J. D., & Metcalfe, N. B. (2011). Estimated standard metabolic rate interacts

with territory quality and density to determine the growth rates of juvenile Atlantic salmon.

Functional Ecology, 25, 1360-1367.

Ricklefs, R. E., & Wikelski, M. (2002). The physiology/life-history nexus. Trends in Ecology

and Evolution, 17, 462-468.

Rios, F. S., Moraes, G., Oba, E. T., Fernandes, M. N., Donatti, L., Kalinin, A. L., & Rantin, F. T.

(2006). Mobilization and recovery of energy stores in traira, Hoplias malabaricus Bloch

(Teleostei, Erythrinidae) during long-term starvation and after re-feeding. Journal of

Comparative Physiology B, 176, 721-728.

Rodela, T. M., & Wright, P. A. (2006). Metabolic and neuroendocrine effects on diurnal urea

excretion in the mangrove killifish Rivulus marmoratus. Journal of Experimental Biology,

209, 2704-2712.

Rozemeijer, M. J. C., & Plaut, I. (1993). Regulation of nitrogen excretion of the amphibious

Blenniidae Alticus kirki (Guenther, 1868) during emersion and immersion. Comparative

Biochemistry and Physiology A, 104, 57-62.

Schultz, E. T., & Conover, D. O. (1997). Latitudinal differences in somatic energy storage:

adaptive responses to seasonality in an estuarine fish (Atherinidae: Menidia menidia).

Oecologia, 109, 516–529.

Smith, H. W. (1931). Observations on the African lungfish, Protopterus aethiopicus, and on

evolution from water to land environments. Ecology, 12, 164-181.

Tatarenkov, A., Ring, B. C., Elder, J. F., Bechler, D. L., & Avise, J. C. (2010). Genetic

composition of laboratory stocks of the self-fertilizing fish Kryptolebias marmoratus: a

valuable resource for experimental research. PloS One, 5, e12863.

203

Tatarenkov, A., Earley, R. L., Perlman, B. M., Taylor, D. S., Turner, B. J., & Avise, J. C. (2015).

Genetic subdivision and variation in selfing rates among Central American populations of the

mangrove rivulus, Kryptolebias marmoratus. Journal of Heredity, 106, 276-284.

Taylor, D. S. (1990). Adaptive specializations of the cyprinodont fish Rivulus marmoratus.

Florida Scientist, 53, 239-248.

Taylor, D. S., Turner, B. J., Davis, W. P., & Chapman, B. B. (2008). A novel terrestrial fish

habitat inside emergent logs. American Naturalist, 171, 263-266.

Turbill, C., Bieber, C., & Ruf, T. (2011). Hibernation is associated with increased survival and

the evolution of slow life histories among mammals. Proceedings of the Royal Society B, 278,

3355-3363.

Turko, A. J., Earley, R. L., & Wright, P. A. (2011). Behaviour drives morphology: voluntary

emersion patterns shape gill structure in genetically identical mangrove rivulus. Animal

Behaviour, 82, 39-47.

Turko, A. J., Robertson, C. E., Bianchini, K., Freeman, M., & Wright, P. A. (2014). The

amphibious fish Kryptolebias marmoratus uses different strategies to maintain oxygen

delivery during aquatic hypoxia and air exposure. Journal of Experimental Biology, 217,

3988-3995.

Van Dijk, P. L. M., Staaks, G., & Hardewig, I. (2002). The effect of fasting and refeeding on

temperature preference, activity and growth of roach, Rutilus rutilus. Oecologia, 130, 496-

504.

Van Waarde, A. (1983). Aerobic and anaerobic ammonia production by fish. Comparative

Biochemistry and Physiology B, 74, 675–684.

204

Van Wassenbergh, S., Herrel, A., Adriaens, D., Huysentruyt, F., Devaere, S., & Aerts, P. (2006).

A catfish that can strike its prey on land. Nature, 440, 881.

Van Wassenbergh, S. (2013). Kinematics of terrestrial capture of prey by the eel-catfish

Channallabes apus. Integrative and Comparative Biology, 53, 258-268.

Walton, M. J., & Cowey, C. B. (1982). Aspects of intermediary metabolism in salmonid

fish. Comparative Biochemistry and Physiology B, 73, 59-79.

Wang, T., Hung, C. C. Y., & Randall, D. J. (2006). The comparative physiology of food

deprivation: From feast to famine. Annual Review of Physiology, 68, 223-251.

Wells, M. W., Turko, A. J., & Wright, P. A. (2015). Fish embryos on land: terrestrial embryo

deposition enhances development in the amphibious fish Kryptolebias marmoratus. Journal

of Experimental Biology, 218, 3249-3256.

White, C. R., & Kearney, M. R. (2013). Determinants of inter-specific variation in basal

metabolic rate. Journal of Comparative Physiology B, 183, 1-26.

Wilkie, M. P., Morgan, T. P., Galvez, F., Smith, R. W., Kajimura, M., Ip, Y. K., & Wood, C. M.

(2007). The African lungfish (Protopterus dolloi): Ionoregulation and osmoregulation in a

fish out of water. Physiological and Biochemical Zoology, 80, 99-111.

Williams, J. B., Miller, R. A., Harper, J. M., & Wiersma, P. (2010). Functional linkages for the

pace of life, life-history and environment in birds. Integrative and Comparative Biology, 50,

855-868.

Wright, P. A. (2012). Environmental physiology of the mangrove rivulus, Kryptolebias

marmoratus, a cutaneously breathing fish that survives for weeks out of water. Integrative

and Comparative Biology, 52, 792-800.

205

Wright, P. A., & Turko A. J. (2016). Amphibious fishes: Evolution and phenotypic plasticity

Journal of Experimental Biology, 219, 2245-2259.

206

Figures

Figure 6.1. Survival (proportion) in different strains (Belize, Florida, Honduras) of Kryptolebias marmoratus out of water. Different letters represent significant differences between strains (p <

0.05).

207

Figure 6.2. Mean ± SE rates of O2 consumption in different strains (Belize, Florida, Honduras) of Kryptolebias marmoratus (A) over 7d of terrestrial acclimation, and (B) after 21 d out of water. Asterisks denote a significant overall difference between the Honduran strain versus the other two strains (p < 0.05), and the daggers indicate significant overall differences in metabolic rate compared to the value in water (p < 0.05).

208

Figure 6.3. Fulton’s condition factor in different strains (Belize, Florida, Honduras) of

Kryptolebias marmoratus in water under normal conditions (control) and after 21 d out of water

(terrestrial). Asterisks represent significant differences between the control and air-acclimated groups within a strain (interaction p < 0.05).

209

Figure 6.4. Mean ± SE rates of fuel use after 21 d of terrestrial acclimation in 3 strains (Belize,

Florida, Honduras) of Kryptolebias marmoratus. (A) lipid use, (B) glycogen use, (C) protein use, and (D) total energy consumed. Data are presented relative to wet mass. Different letters represent significant differences (p < 0.05) between strains.

210

Figure 6.5. Reproductive measures of 3 strains (Belize, Florida, Honduras) of Kryptolebias marmoratus over the first 18 months of life. (A) total embryo production, (B) mean clutch size, and (C) age at first reproduction. Different letters represent significant differences (p < 0.05) between strains.

211

Figure 6.6. Number, body size and condition of Belize and Honduras strains of Kryptolebias marmoratus after 12 months in fully aquatic (control) or periodically drained (fluctuating) microcosms. (A) number of adult fish collected, (B) standard length, (C) wet mass, (D) Fulton’s condition factor, (E) gonado-somatic index (GSI), and (F) hepato-somatic index (HSI). Asterisks denote a significant overall difference between strains (p < 0.05), and different letters indicate significant differences between control and fluctuating microcosms (p < 0.05).

212

Figure 6.7. Number of embryos recovered from microcosms after 12 months. More embryos were released by the Honduras versus Belize strain under fluctuating conditions (p < 0.05).

213

Supplementary Material

Table 6.S1. Body composition data used to calculate energy use in three distinct genetic lineages of Kryptolebias marmoratus after 21 d of terrestrial acclimation.

lineage variable treatment mean SD n -1 glycogen (mg g dry mass) control 61.687 21.35 9 terrestrial 19.445 5.64 6 -1 lipid (mg g dry mass) control 109.711 25.84 6 terrestrial 49.481 28.55 11 50.91 -1 protein (mg g dry mass) control 598.168 19.98 10 (Belize) terrestrial 602.463 27.05 9 body water (proportion) control 0.762 0.009 6 terrestrial 0.785 0.013 12 Δ body mass (final intial-1) n/a 0.819 0.09 10 -1 glycogen (mg g dry mass) control 74.838 36.45 8 terrestrial 17.315 11.25 9 -1 lipid (mg g dry mass) control 129.411 18.36 6 terrestrial 89.090 26.84 12 SLC8E -1 protein (mg g dry mass) control 598.132 18.68 6 (Florida) terrestrial 626.655 26.45 7 body water (proportion) control 0.751 0.011 6 terrestrial 0.779 0.010 12 Δ body mass (final intial-1) n/a 0.765 0.12 10 -1 glycogen (mg g dry mass) control 60.320 21.30 9 terrestrial 14.627 2.10 9 -1 lipid (mg g dry mass) control 100.960 17.49 6 terrestrial 46.069 14.04 12 HON11 -1 protein (mg g dry mass) control 605.562 19.01 10 (Honduras) terrestrial 607.166 21.29 10 body water (proportion) control 0.771 0.005 6 terrestrial 0.792 0.004 12 Δ body mass (final intial-1) n/a 0.908 0.05 10

214

Figure 6.S1. Body mass (A) and standard length (B) as a function of age in two strains of

Kryptolebias marmoratus. Honduras fish were significantly heavier (p < 0.0001) and longer (p <

0.0001) than Belize fish at a given age.

215

Figure 6.S2. Energy and water content in 3 strains (Belize, Florida, Honduras) of Kryptolebias marmoratus under normal housing conditions (control) and after 21 d terrestrial acclimation. (A) lipids, (B) glycogen, (C) crude protein, and (D) water content. All values are relative to wet mass. Asterisks denote significant differences (p < 0.05) after terrestrial acclimation, and different letters indicate significant overall differences among strains (p < 0.05).

216

Figure 6.S3. Change in body mass of 3 strains (Belize, Florida, Honduras) of Kryptolebias marmoratus after 21 d terrestrial acclimation. Different letters represent significant differences (p

< 0.05) between strains.

217

Figure 6.S4. Mean ± SE time spent moving (%) over a 1 h interval in different strains (Belize,

Florida, Honduras) of Kryptolebias marmoratus over 21 d of terrestrial acclimation. Different letters indicate significant differences among strains at the 1 h time point (p < 0.05). Within the

Honduras strain, activity at 1 h is significantly higher than the 7 and 14 d timepoints (p < 0.05).

Within the Florida strain, activity at 1 h is significantly higher than all other timepoints (p <

0.05). There were no differences in the Belize strain over time (p > 0.05).

218

Figure 6.S5. Body mass (A), gonad mass (B), and liver mass (C) as a function of body length in

Belize and Honduras strains of Kryptolebias marmoratus after 12 months in fully aquatic

(control) or periodically drained (fluctuating) microcosms. Body mass was significantly greater in Honduras than Belize fish (p < 0.05), and under control versus fluctuating conditions (p <

0.05, interaction p > 0.05). Gonad mass was higher in Honduras than Belize fish (p < 0.05), and not affected by treatment (p > 0.05). Liver mass was not affected by strain or treatment (both p >

0.05).

219

Figure 6.S6. Calculated total length of gill filaments in Belize and Honduras strains of

Kryptolebias marmoratus after 12 months in fully aquatic (control) or periodically drained

(fluctuating) microcosms. There were no significant effects of strain or environment (p > 0.05).

220

CHAPTER 7: GENERAL DISCUSSION

221

7.1 Major findings

The central goal of my thesis was to investigate how amphibious fishes cope with the physical challenges of moving between aquatic and terrestrial environments. To date, the amphibious fish literature has largely focused on constitutive adaptations to physiology and morphology that enable life on land – i.e. asking how amphibious and fully aquatic fishes differ.

Much less attention has been given to the importance of intra-specific (e.g. local adaptation) and intra-individual variation (phenotypic plasticity) for determining performance out of water

(Wright and Turko, 2016). In this thesis, I examined how amphibious fishes use phenotypic plasticity (Chapters 3, 4, 5) and constitutively expressed, genetically based traits that vary both inter- (Chapter 5) and intra-specifically (Chapter 6) to cope with life in and out of water. My thesis has made substantial contributions to our understanding of how amphibious fishes have solved the physical challenges of living out of water, and has also provided important evidence that the mechanical properties of gill tissue are important determinants of aquatic function.

In Chapter 2 (Turko and Wright, 2015), I synthesized the literature on the evolution, ecology and physiology of the 34 species of amphibious killifishes known at the time (and provided two new reports of amphibious species). Since then, I have contributed to two additional projects that have discovered amphibious behaviour in three more aplocheiloid killifishes (Livingston et al., 2018) and two poecilids (Rossi, Tunnah, Martin, Turko, Taylor,

Currie, Wright, in review). In Chapter 2, I also developed two hypotheses to explain why there seem to be so many species of amphibious killifishes. First, apparently widespread amphibious behaviour could be the result of convergent evolution. Many amphibious killifishes inhabit stagnant and hypoxic aquatic habitats, widely considered to be strong driving forces for the evolution of terrestrially (Graham, 1997). Alternatively, amphibiousness in the common killifish

222 ancestor may have acted as a key innovation (Heard and Hauser, 1995), enabling an adaptive radiation that resulted in the >1000 extant killifishes. Of course, a “hybrid” evolutionary history is also possible, in which amphibiousness arose independently in phylogenetically distant clades, and some of these subsequently diverged into several amphibious species. Chapter 2 also established the killifishes as a good model system for investigating physiological adaptations to amphibious life. I later used this approach to investigate whether amphibious lifestyles affect gill morphology (Chapter 5). I also facilitated the use of this approach in a recent project, led by two undergraduates, that discovered that several amphibious killifishes can volatilize ammonia gas while out of water, probably by retaining cutaneous expression of larval ammonia-transporting

Rh proteins into adulthood (Livingston et al., 2018).

In Chapters 3, 4, 5, and 6, I investigated gill morphological plasticity of Kryptolebias marmoratus and Polypterus senegalus out of water. The responses of these fishes to terrestrial acclimation were surprisingly divergent. In K. marmoratus, short-term terrestrial acclimation (1 week) caused stiffening of the gill arches, driven by increased effective gravity on land, and was linked to increased abundance of collagen and other proteins associated with bone growth (Turko et al., 2017 [Chapter 3]). Collagen density also increased in the gill filaments (Turko et al., 2017

[Chapter 3]), and calcification of the filament sheath also intensified after 1 week in air (Chapter

5). Repeated, longer-term emersions had no effect on filament length or number (Chapter 6). In contrast, short-term terrestrial acclimation of P. senegalus did not affect the mechanical properties or collagen content of the gill arches or filaments but long-term acclimation caused significant reductions in arch length and volume (Chapter 5). Thus, there does not appear to be a generalizable response of gill tissue to terrestrial acclimation in amphibious fishes. This was a surprising result because in Chapter 3 I discovered that the proteomic response of K. marmoratus

223 to terrestrial acclimation resembled the tetrapod pattern of bone growth, and thus concluded that the common ancestor of all vertebrates probably possessed a weigh-responsive skeleton.

Reduced gill arches in terrestrially-acclimated P. senegalus are not obviously consistent with this view. Perhaps P. senegalus sensed increased weight of the gill arches on land and responded by lightening the load carried by the skull. Alternatively, mechanical stresses imparted by respiratory water flow may be an important promotor of gill development, and these forces were missing in fish on land (Witten et al., 2005; Gunter et al., 2013). Regardless, the skeletons of both P. senegalus and K. marmoratus are mechanically responsive, but the direction of plastic change is divergent.

In Chapter 5, I discovered a calcified sheath surrounding the base of the gill filaments in

K. marmoratus. Amphibious fishes are commonly thought to possess gills with short and stiff filaments to prevent collapse when out of water (Graham, 1997). It therefore seemed intuitive that the calcification in K. marmoratus filaments was another adaptation to provide this sort of mechanical support. I surveyed >100 species of killifishes and found that there was no difference in calcification between amphibious and fully aquatic species, contrary to the terrestrial support hypothesis. I was surprised to find that calcified filaments were present in all the killifishes I investigated. It is not clear why this trait seems to have been largely overlooked in other studies, but one possibility is that conventional Alcian blue staining, by far the most common technique for studying skeletal anatomy, uses low pH to selectively stain cartilage but may also dissolve fine skeletal elements (Walker and Kimmel, 2007). Instead, I used Mg2+ differentiation of Alcian blue staining, which does not damage calcified structures. While probably not an adaptation for life on land, I found strong support for the alternative hypothesis that calcified gill filaments are important for aquatic respiration. Increased respiratory demands caused calcium deposition on

224 the filaments of K. marmoratus, suggesting a functional benefit. Furthermore, removal of the calcified sheath reduced the resistance of the gill basket to water flow, suggesting respiratory water was bypassing the lamellar exchange surfaces. By demonstrating that the mechanical properties of the gill filaments may influence respiratory function in a wide variety of fishes, this chapter makes an important contribution to understanding the function of the gill that extends beyond amphibious fishes. Conventional wisdom has been that the most important aspect of gill morphology is surface area (e.g. Lefevre et al., 2017; Pauly and Cheung, 2018), but Chapter 5 demonstrates that the impact of morphology on the flow of water through the gill basket may also be critical.

In Chapter 6, I took a broader view of the ecological and evolutionary factors that could promote or constrain amphibious lifestyles in fishes. Specifically, I tested the hypothesis that long-term emersion tolerance would be enabled by low metabolic rates that allow for conservation of energy stores, but that there would be a cost to this “slow” lifestyle under fully aquatic conditions. Using the self-fertilizing K. marmoratus system, I found that genetically based low metabolic rate can increase survival time out of water by an average of almost two weeks. Interestingly, I also found that the strain with the lowest metabolic rate had the highest growth rate. This has only been observed rarely before in fishes (e.g. Alvarez and Nicieza, 2005;

Norin and Malte, 2011). I also found that emersion tolerant “slow” fish produced more embryos than “fast” fish in conditions of low water availability, consistent with the hypothesis that environmental conditions mediate a fitness trade-off between these phenotypes. “Slow” fish were also larger regardless of water availability, but population sizes were smaller. These results generally support the hypothesis that there is a trade-off between “fast” and “slow” life histories, but the ultimate explanation for pace of life variation in amphibious fishes is probably more

225 complex than simply water availability. Pace of life history studies have tended to focus on trade-offs in relation to single environmental traits, for example predation (e.g. Auer et al.,

2018), food availability (e.g. Finstad et al., 2007), or as I have done here with water availability

(Chapter 6). Natural systems vary in more than one dimension, however, and future pace of life studies should consider that trade-offs may result from complex interactions of environmental variables.

7.2 Thesis limitations

In this thesis, I pioneered several techniques (e.g. glass microbeam testing of isolated gill arches, simulated microgravity acclimation of adult fish, gill filament resistance measurements) that I combined with other established methods (e.g. histology, proteomics, micro-CT morphometrics, phylogenetic analysis, respirometry). This diversity of methods allowed me to integrate biochemical, tissue-level, and whole-animal responses of amphibious fishes to provide new insights into the strategies these animals use to survive out of water. However, there are limitations to the types of data that can be collected and conclusions that can be drawn from each of these methods. The major limitations are considered below.

A major goal of this thesis was to understand how the mechanical properties of the gills respond to environmental conditions and influence function. Accurately measuring these mechanical properties was challenging and there are some caveats to the methods I used. The fishes used in this thesis were very small (K. marmoratus ~0.1 g [Chapter 3]; P. senegalus ~1-2 g [Chapter 4]), and so were the gills (~1-3 mm long). To measure the stiffness of these tissues, I used a custom testing apparatus modified from Fudge et al. (2003). This method used the bending of a glass microbeam to calculate how much force was applied to an isolated gill arch in

226 tension. Compressive tests may have been better approximations of the forces imposed by gravity in live fish, but these were not possible given the small tissues. However, the slope of a force-extension curve is typically linear through the origin of the plot (i.e. in the region between slight compression and slight tension), so my measurements of stiffness and deformation in tension should be similar to those in compression. In these tests, I used hand-made glass microbeams, which allowed me to make force transducers of the appropriate size considering the mechanical properties of the gill arches. However, despite careful examination under the microscope, there were certainly small imperfections or departures from roundness in the glass beams. To account for this, the same beam in the same orientation was used for every test within a trial. For calculations of stiffness, I also assumed the material properties of glass from previously published work (Fudge et al., 2003). Thus, while the microbeam apparatus allowed me to make careful relative measures of arch properties between treatments, the absolute stiffness values could be slight over- or under-estimates.

The small size of gills used in this thesis also necessitated working with whole arches, which limited my ability to standardize my measurements of material properties. In typical investigations of bone stiffness, samples are trimmed to a standardized size and shape, which allows applied forces to be expressed as pressure (i.e. standardized to cross-sectional area;

Vogel, 2003). In my experiments (Chapters 3, 4), I instead measured the forces required to elongate whole arches, which included both deformation and strain components. In these experiments, I was not able to differentiate the effects of hypertrophy (more bone and connective tissue) from true remodelling (re-organization of existing tissue). There was also some variation in cross-sectional shape and area over the length of the gill arches I tested. If the cue for the stiffening I observed was local bone strain, the narrowest/least dense portions of the gill arches

227 would be expected to exhibit the largest response to terrestrial acclimation. I was unable to test this prediction using my testing apparatus, but this would be a promising direction for future research.

Experimentally manipulating body weight is difficult, and all currently available techniques have drawbacks. In Chapter 3, I used a random positioning machine to simulate microgravity conditions. In these experiments, the effective body weight of K. marmoratus was not actually reduced, but rather the gravitational vector was averaged to near-zero over time. If fish could detect and respond to changes in gravity faster than the random positioning machine rotated, then the physiological and biochemical responses I measured might not accurately reflect weightless conditions. Another approach is to use spaceflight to achieve a true reduction in body weight. However, besides the high cost of these studies, launch and landing conditions expose experimental animals to acutely high g forces, which may cause physiological changes that can be challenging to disentangle from the effects of microgravity (Vandenbrink and Kiss,

2016; Huang et al., 2018). Nonetheless, experiments on air-exposed amphibious fish in orbit would be an exciting direction for future research.

In Chapter 5, I assembled a phylogeny of cyprinodontiform fishes using several published trees, and subsequently used this phylogeny to estimate ancestral states of gill filament calcification and compare amphibious versus aquatic species. While the phylogenetic relationships of cyprinodontiforms has not changed substantially in the last decade, there have been major reorganizations in the last 30 years (e.g. Huber, 1992; Costa, 2011) and the status of several clades is still considered to be in flux (e.g. Furness et al., 2015; Furness et al., 2018).

Uncertainty regarding the topology of phylogenetic trees can reduce the accuracy of phylogenetically informed analyses and ancestral state reconstructions (Duchene and Lanfear,

228

2015), although the effect of uncertainty is often minor (Hanson-Smith et al., 2010). In my analysis, most inter-specific variation in gill filament calcification occurred at the genus or family level, rather than between closely related species (Chapter 5). Thus, phylogenetic uncertainty with respect to intra-generic relationships should not have a major impact on my conclusions, and the family-level relationships within the Cyprinodontiformes are well established. Furthermore, my comparison of amphibious and aquatic killifishes is limited by a lack of information regarding the natural behaviour of many species (Turko and Wright, 2015

[Chapter 2]). To account for these limitations, I analyzed the gill filament calcification with and without the effects of phylogeny, and with and without inclusion of several species of killifishes in which amphibiousness has never been confirmed in the literature but seems probable based on habitat information, body shape, and behaviour of close relatives. Altering these parameters had no qualitative effect on the outcome of the analysis, and it seems clear that gill filament calcification is not influenced by amphibious versus aquatic lifestyles.

To measure gill filament calcification in over 100 species of killifishes, I needed to be opportunistic regarding sampling. This meant that fishes were acquired from the aquarium trade, other researchers, or the wild. If there is a large scope for plasticity of filament calcification, and different species I studied were housed under different conditions, this could limit the accuracy of my phylogenetic analysis. Therefore, whenever possible, I spawned these fishes in captivity at the Hagen Aqualab, University of Guelph to standardize rearing conditions and minimize effects of phenotypic plasticity, but this was not possible in all cases and would not account for trans- generational effects. Most of the other fishes I studied were obtained from the pet trade, and it is reasonable to assume that rearing conditions in commercial facilities would be relatively consistent and amenable (e.g. well oxygenated, low in nitrogenous waste). Thus, it seems

229 unlikely that there would be strong plastic effects that would affect species differentially.

Furthermore, in acclimation experiments (28 d) using two species on either side of the deepest division in the Cyprinodontiformes, K. marmoratus and Poecilia wingeii, I found no evidence for plasticity in the extent filament calcification (Chapter 5). These experiments used adult fishes, however, and future work should examine the effect of early life environments (developmental plasticity) on filament calcification. In other fishes, other aspects of the filaments such as length and spacing can be affected by environmental oxygen availability (Chapman et al., 2000; Blank and Burggren, 2014; Borowiec et al., 2015).

I experimentally decalcified the gill filaments of K. marmoratus to test the hypothesis that calcification maintains the position of the filaments in the ventilatory current. I chose to use the calcium chelator ethylenediaminetetraacetic acid (EDTA), as this is considered a gentle method of tissue decalcification (Callis and Sterchi, 1998). However, EDTA also binds divalent metallic ions other than Ca2+ and can interfere with overall attachment between cells and the extracellular matrix (Crapo et al., 2011). Thus, while I was careful to use the minimum amount of EDTA required to only decalcify the gill filaments (gill arches were confirmed to remain calcified), it is possible that there were other off-target effects to the gill basket that could affect material properties. I also attempted to test the “filament position” hypothesis by experimentally increasing filament stiffness via formalin-induced extracellular matrix crosslinking, which is predicted to increase gill basket resistance. However, introduction of formalin into the buccal cavity caused immediate mucous production that dramatically increased gill resistance and swamped any signal from stiffer filaments.

In Chapter 6, I was only able to identify a single isogenic K. marmoratus lineage with a low metabolic rate, which limits the strengths of my conclusions in that study. Repeating the

230 finding of increased survival time out of water in a replicate “slow” lineage would provide compelling evidence that there is a functional link between those traits. In hindsight, stocking the microcosms with adult fish of each lineage rather than earlier life stages might have also reduced the likelihood of detecting a trade-off between “fast” and “slow” phenotypes, as life- history traits are often more correlated with juvenile than adult fitness (Promislow and Harvey,

1990). Repeating this experiment with embryos or newly hatched larvae would therefore provide a good additional test of the trade-off hypothesis. The design of the microcosm experiments also meant that I could not track individuals over time, which limited my ability to estimate some important life-history variables including growth, survival, and lifetime reproductive success.

However, the main objective of this experiment was to measure fitness metrics at the genotype level, and this was successfully accomplished.

7.3 Future directions

In the work presented in this thesis, I successfully answered many of the questions that I had initially posed. However, several exciting new research avenues have emerged. For example, my thesis has shown that changes in effective gravity can drive phenotypically plastic changes to gill morphology, but the direction and sensitivity of the response is species-specific. It would therefore be valuable to extend these studies into other clades of both amphibious and fully aquatic fishes. There are over 200 species of amphibious fishes, and amphibiousness has probably evolved more than 20 times in extant species, providing a broad range of taxa for future research (Graham, 1997; Wright and Turko, 2016). These amphibious fishes span a range of lifestyles (e.g. frequent active emersers, low tide “remainers”, long-term aestivators; Martin,

2014), and comparing responses to gravitational loading among these functional groups would

231 be a good place to start looking for broad patterns. In Chapter 3, I concluded that mechanical responsiveness of fish skeletons was probably an ancestral trait and argued that fully aquatic fish would also benefit from the ability to fine-tune skeletal properties in response to changes in mechanical load (Turko et al., 2017). To test this hypothesis, it would be valuable to study the response of fully aquatic fishes to increased effective weight. An excellent first step was made by

Atkins et al. (2015), who used a spring attached the opercular bone of tilapia to test the response to focused mechanical loading. Another approach would be to acclimate aquatic fishes to hypergravity. Centrifugation of fishes to increase gravitational loading has been attempted in a few studies (e.g. Anken et al. 2001, Anken et al. 2002), but the response of musculoskeletal tissues has not been investigated. In preliminary experiments (collaboration with G. Rossi), K. marmoratus survived in a centrifuge for two weeks at ~6.5 g, suggesting that this approach may be feasible for future studies.

Amphibious fish gills are substantially remodelled during terrestrial acclimation, but the functional significance of these changes is unknown. In K. marmoratus, the gill arches stiffened and the filaments increased in collagen density (Turko et al., 2017 [Chapter 3]) and calcification

(Chapter 5). It is possible that increased stiffness of gill tissue reduces the collapse and coalescence of gill filaments out of water, maintaining some functional surface area. When on land, K. marmoratus “gulp” air to ventilate the bucco-pharyngeal cavity (Turko et al., 2014).

Thus, if increased gill stiffness maintains air flow through the gills, there could be a respiratory benefit as has been suggested in the mudskipper Periophthalmodon schlosseri (Kok et al., 1998).

A direct test of this hypothesis would be valuable. If increased stiffness is desirable for life out of water in highly amphibious K. marmoratus, it is puzzling why the overall extent of filament calcification was minor compared to other killifishes (Chapter 5). Furthermore, the gill lamellae

232 are largely covered by an inter-lamellar cell mass during air exposure, which reduces surface area (Ong et al., 2007). An alternative hypothesis is that the increased mechanical support out of water is simply a local cellular response to deformation and does not confer a specific advantage under these conditions, but rather reflects an overall adaptive advantage to mechanically responsive skeletal systems under other conditions.

When out of water, both K. marmoratus (Ong et al., 2007) and P. senegalus (Chapter 4) fill the inter-lamellar spaces with a cell mass (Nilsson et al., 2012). This remodelling reduces gill surface area, but the function has not been tested. One possibility is that the inter-lamellar cell mass provides structural support out of water, preventing contact and possible fusion of adjacent lamellae (Ong et al., 2012; Wright, 2012). If a technique could be developed to prevent enlargement of the inter-lamellar cell mass in air (e.g. pharmacologically or using gene editing), histological examination for damage or lamellar fusion in gills of air-exposed fish would provide a strong test of this hypothesis. Alternatively, reduced gill surface area may function to minimize evaporative water loss. Future work should measure rates of evaporation from gills with well- versus poorly-developed inter-lamellar cell masses. Yet another possibility is that the size of the inter-lamellar cell mass is regulated by environmental oxygen availability, and thus the abundance of oxygen in air may cause enlargement as a regulatory byproduct without providing a functional benefit. Under aquatic conditions, the inter-lamellar cell mass is known to be highly responsive to changes in oxygen availability - hypoxia causes a reduction of the inter-lamellar cell mass in K. marmoratus (Turko et al., 2012), and some other species (reviewed by Nilsson et al., 2012), while aquatic hyperoxia causes enlargement in Carassius auratus (Tzaneva et al.,

2011). This hypothesis predicts that fish acclimated to aerial hypoxia (e.g. Turko et al., 2014)

233 would not enlarge the inter-lamellar cell mass, while aquatic hyperoxia (e.g. Rossi et al., 2018) should cause enlargement.

In Chapter 5, I found strong evidence that calcification of the gill filaments is an important structural feature that probably improves respiratory function in water-breathing fishes. However, there are many interesting avenues for follow-up research. I used resistance of the gill basket as a proxy for water flow patterns, but particle image velocimetry experiments to measure water flow through calcified and decalcified gill filaments would provide a stronger test of the prediction that more water should escape through the filament tips of decalcified gills (e.g.

Strother, 2013). Cannulated fish heads (e.g. Wright and Perry, 1989) could also be used to measure oxygen extraction efficiency of gills with, and without, calcified filaments. Alternative approaches to EDTA decalcification would also be helpful, given the possible off-target effects described above. If the developmental pathway that causes filament calcification was discovered, perhaps a genetic knock-down without calcified filaments could be tested. Standing individual variation or environmentally induced changes in the degree of filament calcification could also be used to test whether there is a positive relationship with respiratory variables such as oxygen extraction efficiency, but this would assume that other gill traits do not co-vary with filament calcification. Finally, studies of wild fishes inhabiting different environmental conditions (e.g. normoxia vs. hypoxia) or laboratory tests of developmental plasticity that find the extent of filament calcification to be associated with respiratory demand would provide persuasive evidence that relatively stiff filaments provide a respiratory advantage.

234

7.4 Conclusions

In this thesis, I integrated physiological, ecological, and evolutionary approaches to investigate how amphibious fishes cope with the challenges of living out of water. I discovered that K. marmoratus responds to increased effective body weight by stiffening the skeleton using the same molecular mechanisms as mammals, rejecting the common view that dynamic, gravity- responsive skeletons evolved during the tetrapod invasion of land. The gill skeleton of P. senegalus was also highly plastic but instead of becoming more robust as in K. marmoratus, the bones became shorter and smaller. While it is not clear why the responses of these two species differ, my work demonstrates that amphibious fish skeletons overall have a large scope for phenotypic plasticity. Genetic assimilation of plastic traits is thought to be an important mechanism for driving evolutionary change (e.g. Pigliucci et al., 2006, Ehrenreich and Pfenning,

2016). During the tetrapod invasion of land, some bones became more robust while others were reduced (Clack, 2002). The large scope for weight-induced plasticity I describe in my thesis shows that it is possible for these evolutionary trends in tetrapods to have been facilitated by genetic assimilation. I also discovered that calcification of gill filaments is a constitutively expressed trait that might be used by more than half of the ~30,000 ray-finned fishes to increase respiratory performance. This finding raises many new questions about the mechanisms used by fishes to maximize oxygen uptake, and may be especially important in the modern era of climate change and other anthropogenic disturbance of aquatic habitats. Finally, I discovered that survival out of water is enhanced by genetically-based low intrinsic metabolic rates and the ability to suppress metabolism in an amphibious fish. A fundamental goal of physiological ecology is understanding the factors that create and maintain phenotypic trait variation (e.g.

Burton et al., 2011). My study makes an important contribution by demonstrating that conditions

235 of limited resource availability can favour an overall “slow” pace of life, and that low intrinsic metabolic rate may evolve via genetic assimilation of metabolic rate plasticity.

7.5 References

Alvarez, D. and Nicieza, A. G. (2005). Is metabolic rate a reliable predictor of growth and

survival of brown trout (Salmo trutta) in the wild? Canadian Journal of Fisheries and

Aquatic Science, 62, 643-649.

Anken, R. H., Ibsch, M., Breuer, J. and Rahmann, H. (2001). Effect of hypergravity on the Ca/Sr

composition of developing otoliths of larval cichlid fish (Oreochromis mossambicus).

Comparative Biochemistry and Physiology. Part A, Molecular & Integrative Physiology,

128, 369–77.

Anken, R. H., Werner, K. and Rahmann, H. (2002). Effects of hypergravity on the development

of cell number and asymmetry in fish brain nuclei. Advances in Space Research, 30, 849–

853.

Atkins, A., Milgram, J., Weiner, S. and Shahar, R. (2015). The response of anosteocytic bone to

controlled loading. Journal of Experimental Biology, 218, 3559-3569.

Auer, S. K., Dick, C. A., Metcalfe, N. B. and Reznick, D. N. (2018). Metabolic rate evolves

rapidly and in parallel with the pace of life history. Nature Communications, 9, 14.

Blank, T. and Burggren, W. (2014). Hypoxia-induced developmental plasticity of the gills and

air-breathing organ of Trichopodus trichopterus. Journal of Fish Biology, 84, 808–826.

Borowiec, B. G., Darcy, K. L., Gillette, D. M. and Scott, G. R. (2015). Distinct physiological

strategies are used to cope with constant hypoxia and intermittent hypoxia in killifish

(Fundulus heteroclitus). Journal of Experimental Biology, 218, 1198-1211.

236

Burton, T., Killen, S. S., Armstrong, J. D. and Metcalfe, N. B. (2011). What causes intraspecific

variation in resting metabolic rate and what are its ecological consequences? Proceedings of

the Royal Society B, 278, 3465-3473.

Callis, G and Sterchi, D. (1998). Decalcification of bone: literature review and practical study of

various decalcifying agents. methods, and their effects on bone histology. Journal of

Histotechnology, 21, 49-58.

Chapman, L. J., Galis, F. and Shinn, J. (2000). Phenotypic plasticity and the possible role of

genetic assimilation: Hypoxia‐induced trade‐offs in the morphological traits of an African

cichlid. Ecology Letters, 3, 387-393.

Clack, J. A. (2002). Gaining ground: the origin and evolution of tetrapods. Bloomington,

Indiana: Indiana University Press.

Costa, W. J. E. M. (2011). Phylogenetic position and taxonomic status of Anablepsoides,

Atlantirivulus, Cynodonichthys, Laimosemion and Melanorivulus (Cyprinodontiformes:

Rivulidae). Ichthyological Exploration of Freshwaters, 22, 233–249.

Crapo, P. M., Gilbert, T. W. and Badylak, S. F. (2011). An overview of tissue and whole organ

decellularization processes. Biomaterials, 32, 3233-3243.

Duchêne, S. and Lanfear, R. (2015). Phylogenetic uncertainty can bias the number of

evolutionary transitions estimated from ancestral state reconstruction methods. Journal of

Experimental Zoology B, 324, 517-524.

Ehrenreich, I. M. and Pfennig, D. W. (2016). Genetic assimilation: a review of its potential

proximate causes and evolutionary consequences. Annals of Botany, 117, 769-779.

Finstad, A. G., Forseth, T., Ugedal, O. and Naesje, T. F. (2007). Metabolic rate, behaviour and

winter performance in juvenile Atlantic salmon. Functional Ecology, 21, 905-912.

237

Fudge, D. S., Gardner, K. H., Forsyth, V. T., Riekel, C. and Gosline, J. M. (2003). The

mechanical properties of hydrated intermediate filaments: insights from hagfish slime

threads. Biophysical Journal, 85, 2015-2027.

Furness, A. I., Reznick, D. N., Springer, M. S. and Meredith, R. W. (2015). Convergent

evolution of alternative developmental trajectories associated with diapause in African and

South American killifish. Proceedings of the Royal Society B, 282, 20142189.

Furness, A. I., Reznick, D. N., Tatarenkov, A. and Avise, J. C. (2018). The evolution of diapause

in Rivulus (Laimosemion). Zoological Journal of the Linnean Society, zly021.

Graham, J. B. (1997). Air-breathing Fishes: Evolution, Diversity, and Adaptation. San Diego,

CA: Academic Press.

Gunter, H. M., Fan, S., Xiong, F., Franchini, P., Fruciano, C. and Meyer, A. (2013). Shaping

development through mechanical strain: the transcriptional basis of diet‐induced phenotypic

plasticity in a cichlid fish. Molecular Ecology, 22, 4516-4531.

Hanson-Smith, V., Kolaczkowski, B. and Thornton, J. W. (2010). Robustness of ancestral

sequence reconstruction to phylogenetic uncertainty. Molecular Biology and Evolution, 27,

1988–1999.

Heard, S. B. and Hauser, D. L. (1995). Key evolutionary innovations and their ecological

mechanisms. Historical Biology, 10, 151-173.

Huang, B., Li, D.-G., Huang, Y. and Liu, C.-T. (2018). Effects of spaceflight and simulated

microgravity on microbial growth and secondary metabolism. Military Medical Research, 5,

18.

Huber, J. H. (1992). Review of Rivulus, Ecobiogeography, Relationships. Paris: Société

Française D’ichtyologie.

238

Kok, W. K., Lim, C. B., Lam, T. J. and Ip, Y. K. (1998). The mudskipper Periophthalmodon

schlosseri respires more efficiently on land than in water and vice versa for Boleophthalmus

boddaerti. Journal of Experimental Zoology, 280, 86-90.

Lefevre, S., McKenzie, D. J. and Nilsson, G. E. (2017). Models projecting the fate of fish

populations under climate change need to be based on valid physiological mechanisms.

Global Change Biology, 23, 3449-3459.

Livingston, M. D., Bhargav, V. V., Turko, A. J., Wilson, J. M. and Wright, P. A. (2018).

Widespread use of emersion and cutaneous ammonia excretion in Aplocheiloid killifishes.

Proceedings of the Royal Society B, 20181496.

Martin, K. L. (2014). Theme and variations: amphibious air-breathing intertidal fishes. Journal of

Fish Biology, 84, 577-602.

Nilsson, G. E., Dymowska, A. and Stecyk, J. A. W. (2012). New insights into the plasticity of

gill structure. Respiration Physiology and Neurobiology, 184, 214-222.

Norin, T. and Malte, H. (2011). Repeatability of standard metabolic rate, active metabolic rate

and aerobic scope in young brown trout during a period of moderate food availability.

Journal of Experimental Biology, 214, 1668-1675.

Ong, K. J., Stevens, E. D. and Wright, P. A. (2007). Gill morphology of the mangrove killifish

(Kryptolebias marmoratus) is plastic and changes in response to terrestrial air exposure.

Journal of Experimental Biology, 210, 1109-1115.

Pauly, D. and Cheung, W. W. L. (2018). Sound physiological knowledge and principles in

modeling shrinking of fishes under climate change. Global Change Biology, 24, e15-e26.

Pigliucci, M., Murren, C. J. and Schlichting, C. D. (2006). Phenotypic plasticity and evolution by

genetic assimilation. Journal of Experimental Biology, 209, 2362-2367.

239

Promislow, D. E. L. and Harvey, P. H. (1990). Living fast and dying young: A comparative

analysis of life‐history variation among mammals. Journal of Zoology, 220, 417-437.

Rossi, G. S., Turko, A. J. and Wright, P. A. (2018). Oxygen drives skeletal muscle remodeling in

an amphibious fish out of water. Journal of Experimental Biology, 221, jeb180257.

Strother, J. A. (2013). Hydrodynamic resistance and flow patterns in the gills of a tilapine fish.

Journal of Experimental Biology, 216, 2595-2606.

Turko, A. J. and Wright, P. A. (2015). Evolution, ecology and physiology of amphibious

killifishes (Cyprinodontiformes). Journal of Fish Biology, 87, 815-835.

Turko, A. J., Cooper, C. A. and Wright, P. A. (2012). Gill remodelling during terrestrial

acclimation reduces aquatic respiratory function of the amphibious fish Kryptolebias

marmoratus. Journal of Experimental Biology, 215, 3973-3980.

Turko, A. J., Robertson, C. E., Bianchini, K., Freeman, M. and Wright, P. A. (2014). The

amphibious fish Kryptolebias marmoratus uses different strategies to maintain oxygen

delivery during aquatic hypoxia and air exposure. Journal of Experimental Biology, 217,

3988-3995.

Turko, A. J., Kültz, D., Fudge, D., Croll, R. P., Smith, F. M., Stoyek, M. R. and Wright, P. A.

(2017). Skeletal stiffening in an amphibious fish out of water is a response to increased body

weight. Journal of Experimental Biology, 220, 3621-3631.

Tzaneva, V., Bailey, S. and Perry, S. F. (2011). The interactive effects of hypoxemia, hyperoxia,

and temperature on the gill morphology of goldfish (Carassius auratus). American Journal

of Physiology, 300, R1344-R1351.

Vandenbrink, J. P. and Kiss, J. K. (2016). Space, the final frontier: A critical review of recent

experiments performed in microgravity. Plant Science, 243, 115-119.

240

Vogel, S. (2003). Comparative Biomechanics: Life's Physical World. Princeton, NJ: Princeton

University Press.

Walker, M. B. and Kimmel, C. B. (2007). A two-color acid-free cartilage and bone stain for

zebrafish larvae. Biotechnic and Histochemistry, 82, 23-28.

Witten, P. E., Gil-Martens, L., Hall, B. K., Huysseune, A. and Obach, A. (2005). Compressed

vertebrae in Atlantic salmon Salmo salar: evidence for metaplastic chondrogenesis as a

skeletogenic response late in ontogeny. Diseases of Aquatic Organisms, 20, 237-246.

Wright, P. A. (2012). Environmental physiology of the mangrove rivulus, Kryptolebias

marmoratus, a cutaneously breathing fish that survives for weeks out of water. Integrative

and Comparative Biology, 52, 792-800.

Wright, P. A. and Perry, S. F. (1989). A critical analysis of gas transferin vivo and in a blood-

perfused trout head preparation. Fish Physiology and Biochemistry, 6, 297-308.

Wright, P. A. and Turko, A. J. (2016). Amphibious fishes: evolution and phenotypic plasticity.

Journal of Experimental Biology, 219, 2245-2259.

241