1 Sex Determination and Interspecies Hybridization in Zebrafish Danio

Sex determination and interspecies hybridization in zebrafish Danio rerio and pearl danio

D. albolineatus

Dissertation

Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy

in the Graduate School of The Ohio State University

Thomas Allin Delomas, M.S.

Graduate Program in Environment and Natural Resources

The Ohio State University

2018

Dissertation Committee

Konrad Dabrowski, Ph.D., Advisor

Macdonald Wick, Ph.D.

Joseph Ottobre, Ph.D.

Copyrighted by

Thomas Allin Delomas

2018

Abstract

Sustainable management of fisheries and improvement of aquaculture production depends on an increased scientific understanding of fish physiology, nutrition, genetics, and ecology. With over 33,000 described fish species, and hundreds of these species being commercially fished or farmed, it is impractical to develop scientific resources and thoroughly investigate the biology of each species. One solution to this problem is the utilization of model organisms. The zebrafish Danio rerio is a widely used model organism in the larger experimental biology community. However, several areas of research need to be addressed for its utility to increase, particularly for fisheries and aquaculture research. First, rearing methods need to be improved, with an emphasis on larval and early juvenile stages. The sex determination system is controversial, but has been suggested to be polygenic. Finally, interspecies hybridization, which is a key tool in genetic improvement for aquaculture species, has not been thoroughly explored in the

Danio genus.

We present a series of studies addressing these areas of research in order to increase the utility of the zebrafish model system with an emphasis on applications to fisheries and aquaculture research. First, we designed and evaluated a rearing method utilizing a novel set of environmental parameters (3 parts per thousand salinity, high densities of live food, algal turbidity, 24L:0D photoperiod) from 5 to 21 days post-

ii fertilization (dpf) that led to rapid growth rates (mean ± SD lengths of 19.4 ± 1.0 mm and

30.4 ± 1.5 mm at 21 and 42 dpf, respectively) and high fertility (232 ± 124 oocytes/female at 66 ± 3 dpf) (chapter 2). Next, we evaluated this protocol at temperatures close to the lower thermal limit for embryonic development (23°C) and observed no significant decrease in survival compared to a control group kept at optimum temperature (28.5°C) (chapter 3).

We then utilized this rearing protocol to perform a series of investigations into the zebrafish sex determination system. First, we produced triploid zebrafish and confirmed previous studies that found triploid zebrafish to be all male. We then treated triploid zebrafish with estradiol (100ng/L) from 5 to 28 dpf, and found that both treated and untreated triploids were all male. Untreated diploid siblings were also all male while treated diploid siblings were 11% male. This demonstrates that triploidy acts downstream of estradiol to induce male development (chapter 4).

Induced gynogenesis (inheritance of only maternal chromosomes) is a frequently used technique for investigating sex determination in fish species. Previous studies on gynogenesis in zebrafish reported inconsistent results and utilize irradiated zebrafish spermatozoa to induce embryonic development in zebrafish oocytes. This leaves open the possibility that rare, incompletely irradiated spermatozoa may cause gynogenetic progeny groups to be contaminated by biparental offspring. To address this methodological issue, we demonstrated that UV-irradiated common carp Cyprinus carpio spermatozoa activated embryonic development in zebrafish oocytes and that hybrids between zebrafish and common carp were inviable (chapter 5). We then used UV-irradiated common carp

iii spermatozoa to induce gynogenesis in zebrafish. Out of 52 adult gynogens, only one was female. This is consistent with zebrafish having a polygenic sex determination system where inbreeding causes male development. Male gynogens and their biparental siblings were outcrossed to the same group of unrelated females. The families sired by gynogen males were more likely to be female biased than families sired by biparental males. This suggests that inbreeding induces male development through recessive and/or overdominant male-determining alleles (chapter 6).

There are no studies investigating the sex determining systems of other Danio species, making it impossible to evaluate the evolution of sex determination in Danio. To address this gap, we investigated the sex determination system of pearl danio Danio albolineatus. We first performed a full-factorial mating and found that sex ratio varied between families from 5 to 100% male. Six breeding pairs were crossed twice and the sex ratios were not significantly different between the first and second crossings. Heritability was estimated at 0.89 (mean of posterior distribution) with a 95% credibility interval of

0.44 – 1.40. Together, these observations demonstrate that pearl danio has a polygenic sex determination system, and suggests that polygenic sex determination may be conserved between zebrafish and pearl danio. We then performed gynogenesis with pearl danio and the resulting gynogenetic families were strongly male-biased (91 ± 8% male).

This suggests that inbreeding also induces male development in pearl danio (chapter 7).

After observing the relationship between inbreeding and male development in zebrafish and pearl danio, we developed a hypothesis for why this may be an evolutionarily

iv adaptive trait: male-biased sex ratios improve parental fitness under conditions of inbreeding through male-specific dispersal and investment in mate searching (chapter 8).

Given the importance of interspecies hybridization to genetic improvement of aquaculture species, we explored whether zebrafish x pearl danio hybrids could serve as a model system for this technique. We first investigated the viability of both reciprocal crosses of zebrafish and pearl danio. Strongly asymmetric viability was observed between the two crosses, with zebrafish female x pearl danio male hybrids being viable and zebrafish male x pearl danio female hybrids being inviable past embryonic development.

This is an example of “Darwin’s corollary to Haldane’s rule”, but the molecular mechanisms responsible for this phenomenon have not been empirically investigated, presumably due to the lack of a suitable model system. As such, we propose that this hybrid could serve as a suitable model for future mechanistic studies (chapter 9).

We then investigated genetic influences on uniformity of growth in zebrafish x pearl danio hybrids. We found a recessive allele in zebrafish that caused hybrid offspring of zebrafish females homozygous for this allele to have dramatically higher uniformity of growth compared to those of other zebrafish females (mean coefficient of variation in length of 6 ± 1% compared to 21 ± 4% at 21 dpf) (chapter 10). Identification of the gene responsible for this effect and its mechanism of action could lead to similar genes being found in commercially important aquaculture species.

This series of studies both advances the zebrafish model system for fisheries and aquaculture research and demonstrates use of the model system to address key issues in evolutionary biology and aquaculture genetics.

Dedication

To Danielle

Acknowledgments

I would first like to thank my advisor, Konrad Dabrowski, for guiding me through this series of investigations. His advice and insight has been invaluable. I would also like to thank the members of my advisory committee, Macdonald Wick, Christine Beattie, and Joseph Ottobre for providing additional guidance and alternative perspectives on these studies. I would like to extend my sincere appreciation to Boris Gomelsky, who previously supervised my Master’s thesis research. The knowledge I gained while studying with him has provided the foundation for all of my work since.

I would like to thank my fellow students at Ohio State University, both in the

Aquaculture Laboratory and in the university as a whole, for frequent assistance and offering advice. Specific thanks are due to Mackenzie Miller, Megan Kemski, John

Grayson, and Kevin Fisher.

Finally, I would like to thank my family: my grandparents, Anthony and Barbara

Spellman, Valeria and Jim Ratliff, as well as my parents Mark and Clare, for encouraging scientific pursuits throughout my childhood as well as their constant love and support. I would like to extend my utmost gratitude and appreciation to my wife, Danielle, for her love, support, and encouragement throughout my time at Ohio State University.

vii

Vita

December 14, 1991 Born, Lexington, Kentucky, United States of America

2010 High School, The Gatton Academy of Mathematics and Science in

Kentucky, Western Kentucky University

2012 B.S. Molecular, Cellular and Developmental Biology, University

of Washington

2012 – 2013 Laboratory Technician, Department of Biosystems and

Agricultural Engineering, University of Kentucky

2013 Analytical Scientist, KD Analytical, Inc.

2013 – 2015 Graduate Research Assistant, Division of Aquaculture, Kentucky

State University

2015 M.S. Aquaculture / Aquatic Sciences, Kentucky State University

2015 – 2016 FAES Environmental Graduate Research Fellow, Ohio

Agricultural Research and Development Center, The Ohio State

University

2016 – 2017 Graduate Teaching Assistant, School of Environment and Natural

Resources, The Ohio State University

2018 Presidential Fellow, The Ohio State University

viii

Publications

Delomas, T.A. and Dabrowski, K. 2018. Larval rearing of zebrafish at suboptimal temperatures. Journal of Thermal Biology. In press.

Warner, J.L., Gomelsky, B., Delomas, T.A., Kramer, A.G, and Novelo, N.D. 2018. Reproductive ability of second generation (F2) ornamental (koi) carp (Cyprinus carpio L.) x goldfish (Carassius auratus L.) hybrids and characteristics of their offspring. Aquaculture Research. In press.

Delomas, T.A. and Dabrowski, K. 2017. The importance of controlling genetic variation - remarks on 'Appropriate rearing density in domesticated zebrafish to avoid masculinization: links with the stress response'. Journal of Experimental Biology. 220:4078. doi: 10.1242/jeb.164293

Delomas, T.A., Gomelsky, B., Anil, A., Schneider, K.J., and Warner, J.L. 2017. Spontaneous polyploidy, gynogenesis, and androgenesis in second generation (F2) koi x goldfish hybrids. Journal of Fish Biology. 90:80 – 92. doi:10.1111/jfb.13157

Delomas, T.A. and Dabrowski, K. 2016. Zebrafish embryonic development is induced by carp sperm. Biology Letters. 12:20160628. doi: 10.1098/rsbl.2016.0628.

Gomelsky B., Delomas, T.A., and Warner, J.L. 2016. Ploidy variation and viability of aneuploid ornamental koi carp obtained by crossing triploid females with diploid males. North American Journal of Aquaculture, 78:218 – 223.

Gomelsky, B., Warner, J.L., Delomas, T.A., and Nason, L.A. 2016. Inheritance of Red Eyes in Ornamental Koi Carp. North American Journal of Aquaculture, 78:251 – 258.

Gomelsky, B., Delomas, T. A., Schneider, K. J., Anil, A., and Warner, J. L. 2015. Inheritance of Sparkling Scales (Ginrin) Trait in Ornamental Koi Carp. North American Journal of Aquaculture, 77:313 – 317.

Gomelsky, B., Schneider, K. J., Anil, A., and Delomas, T. A. 2015. Gonad development in triploid ornamental koi common carp and results of crossing triploid females with diploid males. North American Journal of Aquaculture, 77:96 – 101.

Fields of Study

Major Field: Environment and Natural Resources

Table of Contents

Abstract ...... ii Dedication ...... vi Acknowledgments...... vii Vita ...... viii Table of Contents ...... x List of Tables ...... xiv List of Figures ...... xvii CHAPTER 1. INTRODUCTION ...... 1 Rearing methods ...... 3 Sex determination ...... 4 Hybridization ...... 10 CHAPTER 2. CHARACTERIZATION OF A METHOD FOR RAPID LARVAL AND JUVENILE ZEBRAFISH GROWTH AND ITS CONSEQUENCES ON REPRODUCTION ...... 13 Abstract ...... 13 Introduction ...... 14 Methods...... 17 Broodstock care and spawning ...... 17 Rearing during 5 – 21dpf ...... 18 Rearing after 21 dpf ...... 19 Sex identification ...... 19 Fertility testing ...... 20 Statistical analysis ...... 20 Results ...... 21 Stocking density, growth, and survival to 21 dpf ...... 21 Growth and survival during 22 – 42 dpf ...... 21 x

Sex ratio and fertility ...... 22 Discussion ...... 22 CHAPTER 3. LARVAL REARING OF ZEBRAFISH AT SUBOPTIMAL TEMPERATURES ...... 32 Abstract ...... 32 Introduction ...... 34 Materials and Methods ...... 36 Broodstock care and reproduction ...... 36 Rearing larvae and juveniles ...... 37 Sex identification and fertility testing ...... 38 Statistical analysis ...... 39 Results ...... 39 Discussion ...... 40 CHAPTER 4. WHY ARE TRIPLOID ZEBRAFISH ALL MALE? ...... 46 Abstract ...... 46 Introduction ...... 48 Materials and methods ...... 50 Experimental overview ...... 50 Fish and fish husbandry ...... 52 Breeding and induction of triploidy ...... 52 Confirmation of triploidy ...... 53 Estradiol (E2) treatment ...... 53 Sex identification and histology...... 54 Fertility testing ...... 55 Statistical analysis ...... 55 Results ...... 56 Experiment one ...... 56 Experiment two ...... 58 Discussion ...... 59 CHAPTER 5. COMMON CARP SPERM INDUCES ZEBRAFISH EMBRYONIC DEVELOPMENT ...... 71 Abstract ...... 71 Introduction ...... 72 xi

Methods...... 73 Fish ...... 73 Collection of Gametes and Generation of Haploids ...... 74 Flow Cytometry ...... 74 Results ...... 75 Discussion ...... 77 CHAPTER 6. EFFECTS OF HOMOZYGOSITY ON SEX DETERMINATION IN ZEBRAFISH ...... 87 Abstract ...... 87 Introduction ...... 89 Materials and Methods ...... 92 Broodstock care and gamete collection ...... 92 Gynogenesis ...... 92 Fish rearing ...... 93 Sex identification ...... 94 Effect of shock on sex ratio ...... 94 Fertility testing and outcrossing of meiotic gynogens ...... 95 Statistical analysis ...... 96 Results ...... 97 Discussion ...... 100 CHAPTER 7. POLYGENIC SEX DETERMINATION IN PEARL DANIO ...... 112 Abstract ...... 112 Introduction ...... 114 Methods...... 117 Fish and fish husbandry ...... 117 Mating design...... 118 Sex identification ...... 118 Gynogenesis ...... 118 Statistical analysis ...... 120 Results ...... 120 Full-factorial mating ...... 120 Gynogenesis ...... 121 Discussion ...... 122 xii

CHAPTER 8. WHY DOES INBREEDING CAUSE MALE BIASED SEX RATIOS? 129 Abstract ...... 129 Introduction ...... 130 Discussion ...... 130 CHAPTER 9. ASYMMETRIC VIABILITY IN RECIPROCAL CROSSES OF ZEBRAFISH AND PEARL DANIO...... 136 Abstract ...... 136 Introduction ...... 137 Materials and Methods ...... 138 Experimental overview ...... 138 Fish and broodstock husbandry ...... 139 Natural crosses ...... 139 In vitro fertilization ...... 139 Embryo incubation and larval rearing...... 139 Flow cytometry ...... 140 Results ...... 141 Discussion ...... 142 CHAPTER 10. MONOGENIC CONTROL OF UNIFORMITY OF GROWTH IN ZEBRAFISH X PEARL DANIO HYBRIDS ...... 146 Abstract ...... 146 Introduction ...... 148 Materials and methods ...... 149 Experimental overview ...... 149 Broodstock and breeding ...... 150 Larval and juvenile husbandry ...... 151 Oocyte and larval measurements ...... 151 Statistical analysis ...... 152 Results ...... 153 Discussion ...... 156 CHAPTER 11. CONCLUSION...... 166 REFERENCES ...... 169

xiii

List of Tables

Table 2.1 Mean ± SD weight, length and specific growth rate (SGR) and survival (5 – 21 and 21 – 42 dpf) in progenies with stocking density of 12 – 22 fish/L (n = 8 progenies)

*initial weight of 0.25 mg based on bulk weighing of 25 larvae at 5 dpf ...... 27

Table 2.2 Size and fertility of fish at 62 – 72 dpf (n = 9 breeding pairs) raised at a stocking density of 12 – 22 fish/L from 11 – 21dpf ...... 28

Table 3.1. Survival and growth of fish raised at standard and low temperatures during larval and early juvenile stages. All values are given as tank mean ± SD. *calculated with an initial weight of 0.25 mg based on bulk weighing of 25 larvae at 5 days post- fertilization (dpf) ...... 43

Table 3.2 Fertility of fish raised at standard and low temperatures during larval and early juvenile stages. All values are given as mean ± SD. dpf, days post-fertilization ...... 44

Table 4.1 Sex of diploid and triploid zebrafish offspring of three different breeding pairs.

...... 63

Table 4.2 Fertility of diploid and triploid zebrafish males tested through natural spawning or by in vitro fertilization with sperm collected by dissection and maceration of the testes.

Dpf is days post-fertilization. *Two males yielded three embryos each (total of six), and all six embryos survived to one dpf ...... 64

xiv

Table 4.3 Sex of adult diploid and triploid zebrafish exposed to either 0 ng/L or 100 ng/L estradiol from 5 to 28 days post-fertilization...... 65

Table 5.1 Ploidy of fish produced by separate crosses with wild-type females (n = 6). .. 80

Table 5.2 Ploidy of fish produced in propagations of one female that exhibited a high frequency of spontaneous diploidization of maternal chromosomes (SDM). *Diploids composed from 0.5 – 34.5% of the progeny...... 81

Table 5.3 Frequency of spontaneous diploidization of maternal chromosomes (SDM) in separate propagations of one “high frequency SDM” female and other females (n = 6 females) with UV irradiated common carp sperm...... 82

Table 6.1 Arithmetic mean ± SD percent survival of gynogenetic and control progeny groups. Age ‘1 – 5’ is the percentage of embryos (4 – 8 cells) surviving to 5 days post- fertilization (dpf). Survival at 1dpf is the percentage of embryos (4 – 8 cells) that survived to 1 dpf. Survival at later ages is given as the percent surviving from the previous age listed. Different letters within the same column indicate significant differences between groups...... 106

Table 6.2 Fertility of EP and LH gynogen males during natural spawning trials in comparison to control biparental siblings. Values are given as arithmetic mean ± SD.

Different letters within the same column indicate significant differences between groups.

...... 107

Table 6.3 Sex ratio of families sired by EP gynogen and control males. Dam group refers to the groups of pooled oocytes used to produce the sires. Bias was assessed by performing a Chi-square goodness of fit test against a 1 : 1 ratio...... 108

Table 7.1 Survival and sex segregation (number male : number female) in repeated crosses (separated by 3 – 4 weeks) of the same dam and sire pearl danio. The columns marked ‘5 dpf’ give survival from 2 – 4 cells to 5 days post-fertilization (dpf). No significant differences in sex ratio were found between repetitions of the same breeding pair (Chi-square test, p > 0.10)...... 125

Table 7.2 Survival of gynogenetic groups resulting from different shocks and survival of unshocked biparental siblings fertilized on the same date. Columns marked 1 dpf give survival from 4 – 8 cells to 1 dpf. *Both these gynogenetic groups were produced from the same batch of ooctyes, and so they share a biparental control. Mpa: minutes post- activation; dpf: days post-activation ...... 126

Table 7.3 Sex of meiotic gynogen offspring of six pearl danio dams. Dam estimated breeding values (EBV) were estimated by Bayesian techniques from a full-factorial mating of all six dams with four sires. Larger EBVs correspond to more male-biased additive genetic effects. *Only three gynogens survived from this dam and all were male.

...... 127

Table 10.1 Segregation of uniform and variable phenotypes in families from generations

F1 and F2. See Figure 10.2 for the pedigree of dams and sires. *significantly different with p < 0.05, binomial test ...... 160

Table 10.2 Variation in size of oocytes and hybrid larve produced by uniform and variable females (n = 3 females of each phenotype from family 1, 30 oocytes and 20 larvae per female) *significantly different with p < 0.05, t-test ...... 161

xvi

List of Figures

Figure 2.1 Relationship between stocking density from 11 – 21 dpf and mean individual weight at 21 dpf. The three different strains used were an AB x TL hybrid line (AB/TL),

Casper, and offspring of fish obtained from a local pet shop (Local)...... 29

Figure 2.2 Relationship between stocking density from 11 – 21 dpf and survival from initial stocking to 21 dpf. The three different strains used were an AB x TL hybrid line

(AB/TL), Casper, and offspring of fish obtained from a local pet shop (Local)...... 30

Figure 2.3. Relationship between stocking density from 11-21 dpf and adult sex ratio.

The three different strains used were an AB x TL hybrid line (AB/TL), Casper, and offspring of fish obtained from a local pet shop (Local)...... 31

Figure 3.1 Water temperature and mean weight of low temperature (LT) and control (CO) fish. Error bars represent standard deviation of mean weight between tanks...... 45

Figure 4.1 Survival of triploid and diploid progeny groups (n = 3 groups per treatment) from fertilization (2 – 4 cells) to 70 days post-fertilization (dpf). Error bars represent standard deviation. The dotted line represents the beginning of exogenous feeding...... 66

Figure 4.2 Mean individual weight and length of diploid and triploid zebrafish at 21

(juveniles) and 70 (males and females) days post-fertilization (dpf) (n = 3 groups per treatment, 10 fish per group). Error bars represent standard deviation. *statistically significant difference (Student’s t, p < 0.01). Two comparisons were statistically tested: xvii triploid juveniles vs diploid juveniles at 21 dpf and triploid males vs diploid males at 70 dpf...... 67

Figure 4.3 Histological sections of testes from adult (A) diploid and (B) triploid males.

The lumen in the testis from the diploid fish contains spermatozoa while none are visible in the lumen of the testis from the triploid fish. *Lumen of spermatogenic tubules...... 68

Figure 4.4 Histological sections of developing gonads from diploid and triploid zebrafish.

(A) An undifferentiated gonad from a diploid zebrafish at 21 dpf. (B) A juvenile ovary containing stage I oocytes from a diploid zebrafish at 21 dpf. (C) An undifferentiated gonad from a triploid zebrafish at 21 dpf. (D) A juvenile ovary containing stage I oocytes from a triploid zebrafish at 21 dpf. (E) An ovary containing stage I and II oocytes from a diploid zebrafish at 34 dpf. (F) A testis containing spermatocytes from a diploid zebrafish at 34 dpf. (G) A testis containing spermatocytes from a triploid zebrafish at 34 dpf. (H) A testis containing a stage I oocyte and spermatocytes from a triploid zebrafish at 34 dpf. I,

II, and S indicate stage I oocytes, stage II oocytes, and spermatocytes, respectively. Scale bars represent 50μm...... 69

Figure 4.5 Mean weight and length of diploid (D, DE2) and triploid (T, TE2) zebrafish exposed to either 0 ng/L (D, T) or 100 ng/L E2 (DE2, TE2) at 28 days post-fertilization.

Error bars represent standard deviation. Different letters indicate statistically significant differences between groups in both weight and length (ANOVA, p < 0.01)...... 70

Figure 5.1 Average percent of embryos surviving in propagations of wild-type females

(not high frequency SDM) from the first mitotic division (2-cells). Error bars represent

xviii standard deviation and only one direction is shown to avoid displaying overlapping bars.

...... 83

Figure 5.2 Development of A. haploid zebrafish, B. zebrafish x carp hybrids, C. diploid zebrafish...... 84

Figure 5.3 Variety of embryo phenotypes seen in zebrafish x carp hybrids and haploid zebrafish...... 85

Figure 5.4 Histograms displaying flow cytometric analysis of cells labeled with propidium iodide to measure nuclear DNA content. The rightmost peak on all three graphs is the rainbow trout RBC internal standard. A. haploid zebrafish, B. diploid zebrafish x carp hybrids, C. diploid zebrafish...... 86

Figure 6.1 Arithmetic mean ± SD length of gynogenetic and control fish. Different letters at the same time point indicate significant differences between groups. Dpf, days post- fertilization ...... 109

Figure 6.2 Histological sections of testes (scale bars represent 50 µm; * designates examples of the lumen of spermatogenic tubules). A. Sterile gynogen (LH) male with few spermatozoa in the lumen B. Fertile gynogen (LH) male with densely packed spermatozoa in the lumen C. Fertile control male with densely packed spermatozoa in the lumen...... 110

xix biased or unbiased by testing the sex ratio against the theoretical 1 : 1 ratio with a Chi- square goodness-of-fit test (p<0.05)...... 111

Figure 7.1 Distribution of sex ratio in 24 pearl danio families...... 128

Figure 9.1 Embryonic development of hybrids between zebrafish and pearl danio. A.

Pearl danio female x zebrafish male B. Zebrafish female x pearl danio male. Arrow indicates pericardial edema at 1 day post-fertilization (dpf)...... 144

Figure 9.2 Survival of hybrids during embryonic development (0 – 5 days post- fertilization, dpf) and larval and early juvenile stages (5 – 21 dpf). Error bars represent standard deviation and are shown only in one direction at time periods where they overlap. Data points at 1 dpf are shown offset to avoid overlapping. Crosses are listed as dam x sire...... 145

Figure 10.1 Example of variation in size of hybrids at 21 days post fertilization from a variable dam (left) and a uniform dam (right) ...... 162

Figure 10.2 Pedigree of broodstock and phenotypes of offspring. Dams and sires of zebrafish offspring are labeled with an identification code corresponding to Table 10.1.

...... 163

Figure 10.3 Coefficient of variation (CV) of length of hybrid offspring from five females in the P generation. Ten fish from each family were measured at 21 days post-fertilization to determine CV...... 164

xx points (8% and 13%) observed for CV of length between families from uniform females and variable females. Cutoff points for CV of weight (not shown on graph) were 27% and

37%...... 165

xxi

CHAPTER 1. INTRODUCTION

Sustainable management of fisheries and improvement of aquaculture production depends on advances in ichthyology. With over 33,000 described fish species (Froese and

Pauly, 2017), it is impractical to develop scientific resources and thoroughly investigate the biology of each species. It is also not practical to narrow our study to only those species that are most relevant to fisheries and aquaculture, as aquaculture alone includes over 300 species (Shelton and Rothbard, 2006). Additionally, many species of importance to fisheries managers and aquaculturists have long generation times, making studies that take course over the entire lifecycle or are multigenerational prohibitively resource intensive.

One solution to this problem is the utilization of model organisms. The zebrafish

Danio rerio has a long history of use in ichthyological studies (Creaser, 1934; Schmidt,

1930). Zebrafish are relatively small, with adults typically 250 – 800 mg in weight, can have generation times as short as 43 days (Dabrowski and Miller, 2018) and individual zebrafish females routinely produce 200 – 400 oocytes per clutch (approximately 0.5 oocytes/mg body weight, chapter 2). This small size, short life cycle, and high fecundity compared to other model vertebrates (mice, rats) have allowed the zebrafish to develop into a widely used model organism for the larger biological community over the past several decades (Holtzman et al., 2016; Lieschke and Currie, 2007; Spitsbergen and Kent,

2003). This increased utilization has allowed for the development of numerous scientific resources, most notably an annotated genome (Howe et al., 2013). The same qualities of small size and short generation time can allow zebrafish to become a model organism for fisheries and aquaculture research as they offer distinct advantages over species that are directly relevant to fisheries and aquaculture. The vast majority of commercially captured or farmed species reach larger sizes (over 1 kg, therefore requiring larger rearing environments) and have drastically longer generation times (multiple years). The zebrafish is, therefore, ideally positioned to become an effective model organism for the fields of ichthyology, fisheries management, and aquaculture (Dahm and Geisler, 2006;

Ribas and Piferrer, 2014).

Before the zebrafish can be fully utilized by fisheries and aquaculture scientists, there are several areas of research that need to be addressed. First, rearing methods for zebrafish vary widely among laboratories and deliver slow growth in comparison to other freshwater fish species (Dabrowski and Miller, 2018; Singleman and Holtzman, 2014).

Second, the sex determination system of zebrafish is unclear, but several studies have suggested that it is polygenic in nature (Liew and Orban, 2014). Finally, while interspecies hybridization is often used for genetic improvement in aquaculture, there is no data available on growth traits of interspecies Danio hybrids. In the following series of studies, we aimed to address these gaps in the literature and facilitate expanded use of the zebrafish model system to inform ichthyology, fisheries management and aquaculture.

Rearing methods

Standardized and effective rearing methods are of utmost importance for model species, as they simplify the comparison of experimental results among laboratories. This is acknowledged in the zebrafish community (Watts et al., 2016, 2012), but current rearing methods for zebrafish vary widely between laboratories and deliver inconsistent and slow growth compared to other freshwater fish species (Dabrowski and Miller, 2018;

Singleman and Holtzman, 2014). Dabrowski and Miller (2018) described a novel protocol for rearing zebrafish that resulted in rapid growth and the shortest generation time on record. These authors utilized a combination of environmental parameters

(continuous light, algal turbidity, high densities of live food, low salinity) throughout the life cycle to maximize food availability, and thereby maximize food consumption and growth. To improve rearing methods for the zebrafish model system and particularly to develop rearing methods with sufficient efficacy for multigenerational and quantitative genetic experiments to be performed, we designed and evaluated a practical rearing protocol that utilizes the conditions described by Dabrowski and Miller (2018) during the larval and early juvenile stages, then transitions to conventional rearing practices

(freshwater, dry diet, light/dark cycle, no added turbidity).

In chapter 2, we describe the evaluation of our protocol and test the hypothesis: improvement of live food availability, continuous light, and algal turbidity during the first

21 days post-fertilization will result in fast growth, earlier maturation, and high fertility.

In chapter 3, we utilize this protocol at temperatures close to the lower limit for zebrafish embryonic development (23°C) in order to determine if this protocol could allow for

3 efficient screening for temperature-sensitive mutations acting after embryonic development. We test the hypothesis: the set of environmental conditions for larval rearing described in chapter 2 will allow zebrafish to be raised at low temperatures with high survival rates.

Sex determination

The evolution of sexual reproduction is proposed to be explained by multiple arguments, including: the production of offspring with increased genetic variation

(Williams, 1975), the spreading of novel beneficial mutations in a population (Fisher,

1930; Morgan, 1913; Muller, 1932; Weismann, 1889), the removal of deleterious mutations (Kondrashov, 1988, 1982; Muller, 1964), and the cycling of beneficial allele combinations (Jaenike, 1978; Sturtevant and Mather, 1938). While many sexually reproducing species exhibit hermaphroditism, the majority of vertebrates are gonochoristic – individuals have either the male or the female phenotype. Gonochoristic species require developing embryos or juveniles to make a binary choice, termed sex determination, between these distinct phenotypes, and several different mechanisms have evolved to regulate this process.

In many species, the choice between male and female phenotypes is genetically determined. For example, in therian mammals (excluding select rodent species), the choice is controlled by inheritance of the Y chromosome, with males being heterogametic

(XY) and females being homogametic (XX). Specifically, the Sry gene located on the Y chromosome is responsible for inducing male development in therian mammals

(Koopman et al., 1991). In monotremes, sex is also controlled by inheritance of specific chromosomes, but multiple sex chromosomes that segregate as a group are observed. The number and actions of sex determining genes on these chromosomes is unknown

(Grützner et al., 2004). Birds also have a genetic sex determination system, with males being homogametic (noted as ZZ) and females being heterogametic (noted as WZ)

(Clinton and Haines, 1999). Many reptiles, such as snakes, also exhibit genetic sex determination (Ohno, 1967) while others, such as alligators and multiple turtle species, display temperature-dependent sex determination, with exposure to different temperatures during development leading to either male or female development (Bull,

1980).

As observed for reptiles, gonochoristic fish species also have a variety of sex determining mechanisms. In some fish species, such as pejerry Odontesthes regia, sex is determined by the environmental temperature during a particular period of juvenile development (Strüssmann et al., 1996). There are also examples of cichlid species where sex determination is regulated by a combination of temperature and pH (Roemer and

Beisenherz, 1996). However, most fish species studied have been shown to have genetic sex determination (Devlin and Nagahama, 2002). In these species, there are multiple examples of both male heterogamety (Bongers et al., 1999; Chourrout and Quillet, 1982;

Komen et al., 1991) and female heterogamety (Dabrowski et al., 2000; Glennon et al.,

2012). The primary genes responsible for sex determination have been identified in four fish species (all with male heterogamety), two species of medaka, Oryzias latipes

(Matsuda et al., 2002) and O. luzonensis (Myosho et al., 2012), tiger pufferfish Takifugu

5 rubripes (Kamiya et al., 2012) and rainbow trout Oncorhynchus mykiss (Yano et al.,

2012), although the sex determining gene in rainbow trout, sdY, has been proposed to function as the sex determining gene for the majority of salmonids (Yano et al., 2013).

There are also several examples of fish species with polygenic sex determination systems (Kallman, 1973; Vandeputte et al., 2007). As outlined by Bull (1983), there are two general types of polygenic systems: 1) systems with a small number of genes whose alleles have dominant and recessive relationships with each other and 2) systems where a large number of alleles at multiple genes appear to affect sex in an additive manner

(termed “multifactorial” and “polyfactorial”, respectively by Bull (1983), although this terminology was never used extensively by other authors). The first type of system

(multifactorial) is found in some rodents (Fredga et al., 1976; Herbst et al., 1978;

Veyrunes et al., 2010) and in the platyfish Xiphophorus maculatus (Kallman, 1973). In these species, the coexistence of three different alleles of the same gene has been observed: X, Y, and W. X is a recessive female determining allele, Y is a male determining allele that is dominant to X, and W is a female determining allele that is dominant to Y.

The second type of system (polyfactorial) has been described theoretically

(Bulmer and Bull, 1982) and has been proposed for a select number of species, including apple snails Pomacea canaliculate (Yusa, 2007) and European sea bass Dicentrarchus labrax (Vandeputte et al., 2007). In these species, sex is best modeled as a threshold trait with a continuous, normally distributed underlying liability (Bulmer and Bull, 1982;

Wright, 1934). This model implies that the liability of an individual is determined by a

6 large number of genes with small additive effects, commonly called the “infinitesimal model” (Barton et al., 2017; Fisher, 1918; Mendel, 1865).

It has been proposed that zebrafish has a polygenic sex determination system

(Anderson et al., 2012; Bradley et al., 2011; Liew et al., 2012), although damagingly high temperatures (34 – 37°C) (Uchida et al., 2004) have been shown to influence the sex ratio. Evidence for polygenic sex determination comes mainly from Liew et al. (2012) who showed, using both laboratory strains and a local strain, that, 1) sex ratios in zebrafish families ranged from 5 to 97% male with a median of 51% 2) individual breeding pairs repeatedly produced families with highly correlated sex ratios, and 3) sex ratio responded to selection. Additionally, mapping studies utilizing hybrids between a laboratory strain and two less-domesticated strains have identified multiple different quantitative trait loci (QTL) that are associated with sex (Anderson et al., 2012; Bradley et al., 2011).

Alternatively, one mapping study claims that while sex is polygenic in laboratory strains of zebrafish, wild zebrafish have a system of female heterogamety. This claim is based on mapping experiments that identified a single QTL associated with sex in multiple less-domesticated strains but no QTL in laboratory strains (Wilson et al., 2014).

However, one study reports that, under identical rearing conditions, variable sex ratios were observed in different families of zebrafish from the WIK strain (one of the less- domesticated strains suggested to have female heterogamety by Wilson et al. 2014) and hybrids between WIK and a strain recently isolated from the wild (Brown et al., 2012).

This observation questions the suggestion of Wilson et al. (2014), as a system of female heterogamety would predict a constant 1:1 sex ratio in all families.

To address this uncertainty in the literature, we performed a series of experiments investigating the sex determination system of zebrafish. We first addressed the effect of induced triploidy on sex determination in zebrafish. It has been reported that triploid zebrafish are all-male (Feitsma et al., 2007; Kavumpurath and Pandian, 1990; Mizgireuv et al., 2004), which could indicate that gene dosage plays an important role in the sex determination system. In chapter 4, we investigated the effect of triploidy on zebrafish gonad development and test the following hypotheses: 1) triploid zebrafish are all male,

2) gonad development in triploid zebrafish follows the same morphological progression as in diploid zebrafish, 3) triploidy acts upstream of estradiol in the sex determination pathway.

Gynogenesis (inheritance of only maternal chromosomes) is a frequently used method for examining the sex determination systems of fish species, and while it has been previously performed in zebrafish (Hörstgen-Schwark, 1993; Pelegri and Schulte-

Merker, 1998; Streisinger et al., 1981), the results reported by different groups are inconsistent. As part of inducing gynogenesis, these previous studies utilize irradiated zebrafish spermatozoa to activate embryonic development in oocytes. This leaves open the possibility for contamination of gynogenetic progeny groups by biparental offspring arising from rare, incompletely irradiated spermatozoa. We first address this methodological shortcoming in chapter 5, where we test the hypothesis: UV-irradiated common carp Cyprinus carpio spermatozoa activates haploid embryonic development in

8 zebrafish oocytes. Then, in chapter 6 we describe performing gynogenesis with zebrafish and tested the following hypotheses, both of which would be consistent with either female heterogamety or a polygenic sex determination system: 1) gynogenetic zebrafish progeny groups are mixed sex and biased towards females, 2) a proportion of gynogenetic zebrafish females produce offspring that are all-female when crossed to a biparental male.

There are currently no descriptions of the sex determining systems of other Danio species in the scientific literature, making it impossible to evaluate the evolution of the sex determining systems in Danio. This is a particularly important avenue of study, as theoretical work predicts that if sexually antagonistic alleles exist (alleles that benefit one sex but not the other), then polygenic sex determination is evolutionarily unstable (Rice,

1986). To address this gap in the literature, we then investigated the sex determination system of the closely related (McCluskey and Postlethwait, 2015) pearl danio Danio albolineatus. In chapter 7, we describe performing a full-factorial mating and gynogenesis with pearl danio and tested the following hypotheses, both of which would be consistent with female heterogamety: 1) different pearl danio breeding pairs always give offspring with a 1:1 sex ratio, 2) heritability of sex in pearl danio is 0, 3) gynogenetic pearl danio progeny groups are mixed sex and biased towards females.

Our results, described in both chapters 6 and 7, suggest that inbreeding causes sex ratios to be male-biased in both zebrafish and pearl danio. This phenomenon has been previously reported in zebrafish, and one hypothesis for why this may be an evolutionarily adaptive trait has been suggested: small size, which can be caused by

9 inbreeding depression, is less detrimental to male fitness than female fitness in zebrafish

(Brown et al., 2012; Lawrence et al., 2008). In chapter 8, we discuss this hypothesis and the literature describing the relationship between body size and male zebrafish reproductive success. We then suggest an alternative evolutionary hypothesis for this phenomenon: male-biased sex ratios improve parental fitness under conditions of inbreeding through male-specific dispersal and investment in mate searching.

Hybridization

Interspecies hybridization has long been used as a means to study the phenomenon of speciation (Darwin, 1859). Speciation is generally thought to begin with the separation of a population into two distinct mating groups, either by a geographic barrier (allopatric speciation) or reproductive isolation due to non-geographic barriers, such as behavioral differences (sympatric speciation). Additionally, “hybrid speciation” is thought to be possible, where hybrids occur between two distinct species and evolve into a separate species.

As the reproductively isolated groups genetically diverge, two categories of barriers to hybridization begin to develop that results in continuation of the speciation process: prezygotic and postzygotic barriers. Prezygotic barriers act to prevent fertilization and can include morphological differences in genitalia, as well as behavioral and geographical barriers. Postzygotic barriers act after fertilization and prevent the hybrid zygote from developing into a fertile adult. Examples of postzygotic barriers include inviability of hybrid embryos and sterility of hybrid adults.

One phenomenon that has been observed during experiments with interspecies hybrids is that in some hybrids, the reciprocal crosses will have asymmetric viability

(Bolnick and Near, 2005). This phenomenon has become known as “Darwin’s corollary to Haldane’s rule”, or simply “Darwin’s corollary” (Turelli and Moyle, 2007). Darwin’s corollary has been proposed to be the result of Dobzhansky-Muller incompatibilities between nuclear alleles in one species and uniparentally inherited traits in the other species, such as maternally deposited transcripts and the mitochondrial genome (Bolnick et al., 2008; Brandvain et al., 2014; Turelli and Moyle, 2007). Despite this proposed hypothesis, there has been no exploration of the molecular basis for this phenomenon, likely due to the lack of an appropriate model system. If asymmetric viability is present in an interspecies Danio hybrid, it could be an appropriate model system for investigations into Darwin’s corollary. In chapter 9, we describe the viability of both reciprocal crosses between zebrafish and pearl danio and test the hypothesis: reciprocal crosses of zebrafish and pearl danio have identical viability.

In addition to the use of interspecies hybridization for experimental biology, interspecies hybridization has been used in aquaculture to combine the beneficial traits of two parental species and exploit heterosis that sometimes results from hybridization

(Dunham, 2011). Examples of interspecies hybrids that are commercially farmed include hybrid striped bass (White bass Morone chrysops x striped bass Morone saxatilis)

(Bayless, 1968; Bishop, 1968; Bosworth et al., 1997; Garber and Sullivan, 2006), blue catfish Ictalurs furcatus x channel catfish Ictalurus punctatus (Argue et al., 2003;

Dunham et al., 1983; Li et al., 2014; Yant et al., 1975), and common carp x crucian carp

Carrasius auratus (Cherfas et al., 1994; Delomas et al., 2017; Liu et al., 2004, 2001).

Genetic improvement of interspecies hybrids is challenging in comparison to that of pure species, as the direct evaluation of breeding candidates requires using them to produce hybrid offspring (reciprocal recurrent selection). The phenotype of the breeding candidate itself is only of limited use in predicting the phenotype of its hybrid offspring because heterosis results from complex epistatic, dominant, and overdominant interactions (Jiang et al., 2017; Li et al., 2008). New methods are needed to facilitate the genetic improvement of parental lines for the production of interspecies aquaculture hybrids.

Interspecies Danio hybrids could allow these methods to be efficiently developed prior to evaluation in production species. We demonstrate this possibility in chapter 10 by testing the following hypothesis: uniformity of growth in zebrafish x pearl danio hybrids is not genetically regulated.

CHAPTER 2. CHARACTERIZATION OF A METHOD FOR RAPID LARVAL AND JUVENILE ZEBRAFISH GROWTH AND ITS CONSEQUENCES ON REPRODUCTION

Abstract

Despite widespread use of zebrafish as a model organism for finfish aquaculture, genetics, and molecular biology, husbandry methods and nutrition have not been standardized. We characterize a protocol for rearing zebrafish using a recently developed combination of environmental parameters that yield rapid growth rates: high densities of live food organisms (marine rotifers and brine shrimp nauplii), low salinity, continuous light, and algal turbidity during larval and early juvenile stages (5 – 21 days post- fertilization, dpf) followed by a transition to conventional rearing methods (dark/light photoperiod, freshwater) to accelerate maturation. We assessed a range of larval fish densities in a stagnant water system with continuous ad libitum feeding until 21 dpf and periodic partial water changes. Fish stocking densities of 12 – 22 fish/L gave maximum growth rate and growth rate steadily declined as stocking density increased past this threshold. Survival rates from 5 – 21 and 21 – 42 dpf were 87 ± 9 and 98 ± 3%, respectively. Mean ± SD weight and length were 64 ± 9 mg and 19.4 ± 1.0 mm at 21 dpf and 245 ± 28 mg and 30.4 ± 1.5 mm at 42 dpf. Fertility was assessed at 66 ± 3 dpf. Mean clutch size was 232 ± 124 oocytes/female and fertilization rate was 72 ± 16%. This protocol will significantly increase productivity in zebrafish facilities.

Introduction

Zebrafish have become a beneficial model organism for the production aquaculture and biological research communities (Dahm and Geisler, 2006; Ribas and

Piferrer, 2014; Ulloa et al., 2011). However, husbandry techniques and nutrition, particularly during the larval and juvenile stages, deliver inconsistent and slow growth compared to other freshwater species (see review in Dabrowski and Miller, 2018). This low growth rate of zebrafish is recognized in the field (Singleman and Holtzman, 2014), and there are many calls for the development and adoption of standardized nutrition and rearing protocols (Watts et al., 2016, 2012). The health of laboratory animals is both an animal welfare and a scientific concern, as undernourished, and consequently unhealthy, animals can severely bias experimental results. Use of the zebrafish model system continues to expand across fields, such as aquaculture (Chengfei et al., 2017; Pereira et al., 2017; Yossa et al., 2015), making the confounding effects of subpar nutrition increasingly questionable.

Currently, larval zebrafish are fed at the start of exogenous feeding a commercial dry diet, along with live feeds, such as brine shrimp (Artemia) nauplii, Protozoa

(Paramecium sp.), and marine (Brachionus plicatilis) or freshwater (B. calyciflorus) rotifers. Facilities that initially feed larvae with paramecium or rotifers typically transition to feeding Artemia nauplii after fish are large enough to readily consume the newly hatched nauplii. These different feeding regimes result in variable growth of larval fish, which demonstrates the significant impact of diet acceptance and nutritional quality

14 on larval growth, health, and survival. This also suggests that the choice of larval diet can limit the productivity of a zebrafish facility.

Goolish et al. (1999) compared the performance of larval zebrafish fed various

(formulated) dry diets to a live diet (paramecium and Artemia) for 14 days and demonstrated that a live diet results in significantly higher survival and growth. Similar studies in other cyprinids, the same family as zebrafish, concluded that cyprinid larvae grow best when fed live food (Abi-Ayad and Kestemont, 1994; Dabrowski, 1984;

Rottmann et al., 1991). However, some species, such as common carp Cyprinus carpio, have demonstrated excellent growth and survival on specific formulated feeds (Charlon and Bergot, 1984). The low growth rate of larvae fed dry diets is likely due both to the low attractiveness and palatability of dry diets, but also the poor utilization of nutrients from dry diets in stomachless fish (reviewed by Dabrowski et al. 2010). It can be concluded that, currently, for optimal growth and survival, a live diet must be used for zebrafish larvae throughout metamorphosis to juveniles.

Artemia and marine rotifers are the most widely used live feeds in aquaculture due to their availability and known culture methods. Both of these species are native to marine or inland saline environments (Triantaphyllidis et al., 1998) and do not survive for extended periods of time in freshwater. However, both survive at salinities as low as 1 – 3 parts per thousand (ppt) (Conte et al., 1972; Walker, 1981). Culture of zebrafish at 3 ppt allows rotifers and Artemia nauplii to stay alive for more than 24 hours, continue to be available in the water column, and thereby enables the zebrafish to feed continuously

(Dabrowski and Miller, 2018).

Zebrafish only incidentally consume food in the dark (Carrillo and McHenry,

2016; McElligott and O’malley, 2005), and so it is likely that continuous light increases growth by allowing the fish to forage constantly. Continuous light has been previously shown to increase larval survival and growth in other fish species (Barahona-Fernandes,

1979; Duray and Kohno, 1988).

Clay and algae turbidity have been shown to increase growth in both freshwater and marine fish larvae (Attramadal et al., 2012; McEntire et al., 2015; Naas et al., 1992;

Rieger and Summerfelt, 1997). The mechanisms responsible for increased growth are thought to be increased contrast of the prey organisms and decreased aggression between larvae. In the case of algal turbidity, it is also documented that algae acts as a food source for the rotifers and Artemia, preventing a decline in nutritional quality as they live in the fish tank (Reitan et al., 1997).

When used in combination, high densities of live food organisms, low salinity, continuous light, and algae turbidity were recently shown to increase growth rates and decrease time to maturity in zebrafish (Dabrowski and Miller, 2018). These authors utilized the above conditions throughout the entire lifecycle and achieved the fastest growth rates for zebrafish (250 ± 29 mg for males and 430 ± 5 mg for females at 42 days post-fertilization, dpf) and the shortest generation time (43 days) to date. However, the protocol described by Dabrowski and Miller (2018) is resource intensive as it involves the maintenance of these environmental conditions and the feeding of exclusively live food throughout the life cycle. We aimed to characterize a rearing protocol that takes advantage of the combination of growth-promoting environmental conditions during the

16 larval and early juvenile stages but used conventional rearing techniques once juveniles were large enough to transition to a formulated, commercial diet. Due to its use as a model organism, reproductive quality is of utmost importance to zebrafish husbandry. As such, we characterized not only the growth rate of fish reared with this new protocol, but also their reproductive quality.

Methods

Broodstock care and spawning

Broodstock zebrafish (2 – 10 months old, 300 – 800 mg weight) were kept in a recirculating system under a 13L:11D photoperiod and maintained at a temperature of 28

± 1°C. Broodstock were derived from an AB x TL hybrid line, Casper (Carolina

Biological, Burlington, NC), and fish from a local pet shop (local). No pattern of differences in the growth rate of different strains was evident (Figure 2.1) and so all strains were analyzed together. Broodstock were fed Artemia nauplii and dry pellets

(Otohime B2, Reed Mariculture, Campbell, CA).

Twenty-five progenies were produced by individual spawning pairs bred in mesh- bottomed breeding boxes. One male and one female zebrafish were placed in a breeding box the night prior to spawning, but kept separate by a transparent divider. The divider was removed the morning of spawning, approximately 15 minutes after the lights turned on. Embryos were collected and incubated in mesh bottom baskets suspended in a recirculating system (Pentair Aquatic Ecosystems, Apopka, FL) maintained at 28.5 ±

0.5°C. At 5 dpf, larvae were stocked into static water tanks for further rearing.

Rearing during 5 – 21dpf

Rearing from 5 to 21 dpf was performed similar to Dabrowski and Miller (2018).

Individual progenies were stocked into static water containers filled with 4 L of 3 ppt saline water (Instant Ocean, Blacksburg, VA) and placed in a water bath. The number of larvae varied among progenies, and initial stocking density varied as a result from 10 to

150 fish/L. Gentle aeration was provided through an airstone in each container and temperature was maintained at 28.5 ± 0.5°C. Fish were kept under a 24L:0D photoperiod.

Turbidity in the tanks was maintained at 10 – 20 NTU by periodic addition of Nanno

3600 Nannochloropsis algae paste (Reed Mariculture, Campbell, CA). For the first five days of feeding (5 – 10 dpf), marine rotifers Brachionus plicatilis were maintained in the fish tanks at a concentration of approximately 200 rotifers/mL. Rotifers were added to the tank as needed (1 – 3 times/day) to maintain constant density. Maintaining a high concentration of rotifers allowed ad libitum feeding independent of fish density. Rotifers were produced in our laboratory being fed Nannochloropsis algae paste and brewers yeast. At 11 dpf, the volume of water in the containers was increased to 6 L, thereby decreasing the density of fish by 1/3. A lower water volume was used for the first 5 days to maintain rotifers at a high density. From 6 – 21 dpf, fish were fed newly hatched (less than 24 hours since hatching) Artemia nauplii (Szyper, 1989), that were maintained in the fish tanks at a concentration of approximately 10 nauplii/mL. Nauplii were hatched at

28°C and then maintained at 18°C prior to being added to the tank as needed (1 – 4 times/day) to maintain the concentration. Every three days, 2/3 of the water volume was replaced with new 3 ppt saline water. This rate of water exchange was based on

18 observations of water quality (ammonia, dissolved oxygen, pH) by Dabrowski and Miller

(2018). Dissolved oxygen and un-ionized ammonia were maintained at acceptable levels for zebrafish (4 – 6 mg/L and > 0.01 mg/L, respectively).

Rearing after 21 dpf

At 21 dpf fish were counted and 10 fish from each progeny were randomly sampled, anesthetized with MS-222 (100 mg/L), measured (weight and total length), and then returned to their respective tanks. Up to sixty (in one progeny, fewer than 60 fish remained, and so all were transferred) randomly sampled fish from each of 22 (out of 25) randomly chosen progenies were transferred to a freshwater recirculating system (Thoren

Aquatics, Hazelton, PA) with a 13L:11D photoperiod and temperature maintained at 28 ±

1°C. Fish were stocked at a density of 20 fish/L, separated by their original progeny groups, and fed Artemia nauplii supplemented with dry feed (Otohime B1/B2, Reed

Mariculture, Campbell, CA). At 42 dpf, all fish were counted and 10 fish from each progeny were randomly sampled and measured (weight and total length). Sixteen randomly chosen progenies were raised past 42 dpf. Fish density was decreased to 1.6 fish/L and fish were fed dry feed (Otohime B2) supplemented with Artemia nauplii. Fish were raised until all were identifiable as male or female (55 – 70 dpf).

Sex identification

Sex of adult fish was assessed by visual observation of body shape and pelvic fin coloration (Parichy et al., 2009). When these observations were not conclusive, gentle pressure was applied to the abdomen and it was observed whether the fish expelled oocytes or sperm (Liew et al., 2012).

Fertility testing

Three spawning pairs (siblings) from each of three progenies were randomly chosen for testing (n = 9 pairs total). Spawning was performed as described above for broodstock fish. Fish that did not spawn on the first attempt were set up to spawn a second time on the following day. All pairs successfully spawned on either the first or second attempt. Fertilization rate (2 – 4 cells), 1 dpf viability (number alive at 1 dpf/number fertilized), and male and female post-spawn weight and length were recorded.

Statistical analysis

All biometric values are given as mean ± SD. Specific growth rate (SGR) was

ln 푓𝑖푛푎푙 푤푒𝑖푔ℎ푡−ln𝑖푛𝑖푡𝑖푎푙 푤푒𝑖푔ℎ푡 calculated as 100 ∗ and was calculated from the preceding 푑푎푦푠 weighing. The arcsine transformation (arcsine of the square root) was applied to sex ratio observations (proportion of fish in a progeny that were male). A two-tailed, one-sample t- test was performed on the transformed data to determine if the mean proportion of fish that were male was significantly different from 0.50. Pearson’s r was calculated to assess the correlation between weight at 21 dpf and weight at 42 dpf. A t-test was performed to determine if Pearson’s r was significantly different from zero. All statistical analyses were performed in R statistical software ver 3.2.5 (R Core Team, 2017).

Results

Stocking density, growth, and survival to 21 dpf

Stocking density of fish and mean individual weight at 21 dpf showed an inverse relationship (Figure 2.1). There was no apparent relationship between strain and growth.

A rapid decrease in mean individual weight occurred with stocking densities above approximately 22 fish/L after 11 dpf. The optimum stocking density for fish grown in this method was judged to be 12 – 22 fish/L based on weight at 21 dpf. For progenies stocked in this range, mean individual weight and length at 21 dpf was 64 ± 9 mg and 19.4 ± 1 mm, respectively (Table 2.1). There was no evidence for a relationship between stocking density and survival (Figure 2.2). Mean survival for fish stocked at 12 – 22 fish/L was 87

± 9% (Table 2.1). A small proportion of fish was observed to develop edema during the first 21 dpf. The percentage of fish in each progeny that developed edema was low but had high variation, with mean of 2 ± 6% and was not correlated with mean weight or survival.

Growth and survival during 22 – 42 dpf

For progenies stocked at 12-22 fish/L from 11 – 21 dpf, mean individual weight and length at 42 dpf was 245 ± 28 mg and 30.4 ± 1.5 mm, respectively. Mean survival in these progenies from 22 to 42 dpf was 98 ± 3% (Table 2.1). There was a strong linear relationship between mean individual weight at 21 dpf and 42 dpf (Pearson’s r = 0.73, p

< 0.001).

Sex ratio and fertility

Progenies became fully sexually dimorphic between 50 and 60 dpf. The mean percentage of fish that were male, 58 ± 24%, was not significantly different from 50% (t

= 1.22, p = 0.24). Average age at fertility testing was 66 ± 3 dpf. As is typical for zebrafish, male weight and length were smaller than these in females. All nine breeding pairs successfully spawned on the first or second attempt. Mean fertilization rate was 72

± 16% and mean clutch size was 232 ± 124 oocytes (Table 2.2). Mean embryo survival from 2 – 4 cells to 1 dpf was 97 ± 4%.

Discussion

Maintaining the environmental conditions identified by Dabrowski and Miller

(2018) throughout the life cycle is resource intensive, and it still must be determined if all zebrafish are capable of reaching the final stages of reproductive maturity after being exposed to continuous light for 42 – 47 days. In the present study, we demonstrate a practical protocol for rapid growth and full fertility that utilizes these conditions during the larval and early juvenile stages. Continuous light, live food, low salinity, and turbidity resulted in rapid growth from 5 to 21 dpf across a variety of zebrafish genetic backgrounds (AB/TL, Casper, and a local strain). We identified 12 – 22 fish/L as the density range that allowed for rapid growth. This density is similar to larval stocking densities reported as standard protocol (fewer than 30 fish/L) by seven out of eight surveyed globally distributed zebrafish facilities (Castranova et al., 2011).

Rearing zebrafish with continuous access to appropriately sized live food that is maintained in the water column during the larval stages has been shown to yield high survival and growth rates. Best et al. (2010) demonstrated that initial feeding with marine rotifers and maintaining low levels of salinity (5 ppt), followed by a transition to Artemia and a freshwater environment at 12 dpf resulted in mean length of approximately 13 mm at 23 dpf. Aoyama et al. (2015) demonstrated that initial feeding with freshwater rotifers followed by a transition to Artemia resulted in a mean length of approximately 16 mm at

23 dpf. Carvalho et al. (2006) utilized only Artemia nauplii beginning at 5 dpf, and at 26 dpf, fish had a mean length of 14.3 mm. Similar to the results obtained by Dabrowski and

Miller (2018), we demonstrate, in this study, that maintaining low salinity conditions (3 ppt), continuous light, turbidity, and feeding marine rotifers and Artemia resulted in faster growth, with a mean length of 19.4  1 mm at 21 dpf (Table 2.1).

The small percentage of fish that developed edema during the first 21 dpf, could be due to individual progeny’s deficiencies in osmoregulation exacerbated by the salinity

(3 ppt). The relatively high variability among families suggests a genetic influence on susceptibility to this condition. Since the percentage of fish with this pathology was so low, it is not likely to be an impediment in adoption of this method. Additionally, for

“weaker” strains of fish (such as knock-out mutants) that may be more susceptible, the benefits of salinity (continuous access to live food) likely outweigh any negative effects.

Lawrence et al. (2012) reported that by using the protocol developed by Best et al.

(2010), mean individual weight and length at 57 dpf was 220 – 250 mg and 28 – 31 mm, respectively. In the current study, fish reached the same size (245 ± 28 mg and 30.4 ± 1.5

23 mm) at 42 dpf. More recently, Uusi-Heikkila et al. (2016) utilized a commonly practiced rearing protocol, ad libitum dry feed and Artemia nauplii, in addition to five generations of selection for large body size, and obtained fish with lengths of 10 – 17 mm at 50 dpf.

This demonstrates that improvements in growth rate and productivity can be achieved with this new protocol. Additionally, it can be suggested that the large increase in growth rate we observed under improved conditions calls into question the results of previous studies on undernourished (slow-growing) fish.

There was a strong correlation between mean weight at 21 dpf and mean weight at

42 dpf. It can be concluded that slower growth due to high densities during the larval and early juvenile stages cannot be fully compensated for by growth later in life. Therefore, adherence to effective larval rearing protocols is necessary to maximize zebrafish growth and health.

While rapid growth in a laboratory model is desirable, superior growth has to be accompanied by fertility. Aoyama et al. (2015) reported a mean clutch size of 46 oocytes/female and 116 oocytes/female at 63 dpf and 71 dpf, respectively. Lawrence et al. (2012) reported that mean clutch size at 65 dpf and 71 dpf was 52 oocytes/female and

109 oocytes/female. Fish reared with the current protocol of a comparable age (66 ± 3 dpf) had mean clutch size of 232 ± 124 oocytes/female. The difference in clutch size between the current study and Aoyama et al. (2015) is likely due to the larger size of females reared with the current method (24 – 26 mm compared to 35 – 39 mm).

Lawrence et al. (2012) did not report the size of spawned females, but the mean fish size

(males and females combined) at the time of spawning was 30 – 34 mm and 270 – 360

24 mg. This is smaller than the females spawned in the current study, suggesting that the difference in mean clutch size compared to the current study can also be explained by a difference in female size and not a difference in relative fecundity. The mean clutch size reported for adult females (older than 126 dpf) in fertility trials across eight globally distributed zebrafish facilities was around 130 oocytes/female (Castranova et al., 2011).

The size of females at spawning was not reported in this study, although the females used were older than the fish tested in the current study. It is therefore unclear if the larger mean clutch size of females reared with the current method is due to differences in size or relative fecundity. This comparison demonstrates that the current method is an improvement over methods currently used by zebrafish facilities around the world. The present study also reinforces the conclusions of McMenamin et al. (2016) that age is not a suitable measure of development in post-embryonic zebrafish.

Mean fertilization rates (assessed at 1 dpf) reported by Aoyama et al. (2015) for fish at 63 dpf and 71 dpf were 85%. Mean fertilization rates reported by Lawrence et al.

(2012) for fish at 65 dpf was 66%. Fertilization rates in the current study (72 ± 16%, assessed at 2 – 4 cell stage) and survival of embryos from fertilization to 1 dpf (97 ± 4%) are comparable, demonstrating that there are no large differences in the quality of gametes produced by fish reared using these methods.

The sex ratio of zebrafish progenies is highly variable due to the polygenic nature of the zebrafish sex determination system, but on the population level, it is centered around a balanced (1:1) ratio (Liew et al., 2012). The mean sex ratio of progenies produced with this proposed method (58 ± 24% male) was not significantly different

25 from a balanced sex ratio (50% male) demonstrating that this method did not exert a detectable environmental influence (fish density, growth rate) on sex determination. It has been suggested that high rearing densities may induce male development in select zebrafish families (significant genotype by environment interaction) by causing chronic stress (Delomas and Dabrowski, 2017a; Liew et al., 2012; Ribas et al., 2017a, 2017b).

We did not observe a relationship between density and sex ratio in our study (Figure 2.3), demonstrating that our estimate of mean sex ratio was not biased by density-induced masculinization. The density treatment was not applied during the entire sex determination period (up to 23 mm fish length) (Maack et al., 2003; Maack and Segner,

2003). Additionally, the density range we identify as optimum (12 – 22 fish/L) is well below the density (40 fish/L) found to cause some families to be male biased (Ribas et al., 2017b), although survival under these conditions was previously reported to be below

50% (Ribas et al., 2017a).

The described protocol produces faster growth and equal or increased fertility compared to currently described methods in the literature. The effectiveness of this protocol is largely due to the application of continuous light, live food, low salinity, and turbidity (Dabrowski and Miller, 2018) until 21 dpf. This study provides a template for further refinement of zebrafish rearing techniques. Further optimization of this protocol is likely to yield even faster growth and earlier maturation than obtained in this study.

Age (dpf) 21 42

Weight (mg) 64 ± 9 245 ± 28

Length (mm) 19.4 ± 1 30.4 ± 1.5

SGR (% day-1) 32.1 ± 2.9* 7.7 ± 1.6

Survival (%) 87 ± 9 98 ± 3

Table 2.1 Mean ± SD weight, length and specific growth rate (SGR) and survival (5 – 21 and 21 – 42 dpf) in progenies with stocking density of 12 – 22 fish/L (n = 8 progenies)

*initial weight of 0.25 mg based on bulk weighing of 25 larvae at 5 dpf

Parameter Mean ± SD

Weight (mg) Male 358 ± 30

Female 476 ± 45

Length (mm) Male 35.6 ± 0.7

Female 37.1 ± 1.2

Age at spawning (dpf) 66 ± 3

Fertilization rate (% 2-4 cells) 72 ± 16

Embryo survival at 1 dpf (%) 97 ± 4

Clutch size (oocytes/female) 232 ± 124

Relative fecundity (oocytes/mg body weight) 0.49 ± 0.26

Table 2.2 Size and fertility of fish at 62 – 72 dpf (n = 9 breeding pairs) raised at a stocking density of 12 – 22 fish/L from 11 – 21dpf

Figure 2.1 Relationship between stocking density from 11 – 21 dpf and mean individual weight at 21 dpf. The three different strains used were an AB x TL hybrid line (AB/TL),

Casper, and offspring of fish obtained from a local pet shop (Local).

Figure 2.2 Relationship between stocking density from 11 – 21 dpf and survival from initial stocking to 21 dpf. The three different strains used were an AB x TL hybrid line

(AB/TL), Casper, and offspring of fish obtained from a local pet shop (Local).

Figure 2.3. Relationship between stocking density from 11-21 dpf and adult sex ratio.

The three different strains used were an AB x TL hybrid line (AB/TL), Casper, and offspring of fish obtained from a local pet shop (Local).

CHAPTER 3. LARVAL REARING OF ZEBRAFISH AT SUBOPTIMAL TEMPERATURES

Abstract

Temperature-sensitive mutants have been widely utilized in single-cell and invertebrate model systems, particularly to study the function of essential genes. Few temperature- sensitive mutants have been identified in zebrafish, likely due to the difficulty of raising zebrafish at low temperatures. We describe a novel rearing protocol that allows rapid growth of larval and juvenile zebrafish at 23°C compared to previous data in the literature. Embryos collected from four breeding pairs were maintained at 28.5 ± 0.5°C until 5 days post-fertilization (dpf) – the onset of exogenous feeding. Larvae were then divided to six tanks and three tanks were cooled to 23 ± 0.2°C. Fish were fed a live diet

(marine rotifers Brachionus plicatilis and Artemia nauplii) and maintained under a set of environmental parameters shown to increase growth rate: continuous light, low salinity

(3ppt), and algal turbidity. Mean total length and weight of fish at 21dpf were 12.7 ±

0.3mm and 20.5 ± 1.5 mg for the 23°C treatment and 18.5 ± 0.4mm and 67.3 ± 3.4mg for the 28.5°C control. By 35 dpf, the fish raised at 23°C had reached a mean length and weight of 18.9 ± 0.7mm and 76.4 ± 6.7mg, approximately the size control fish reached at

21 dpf. At 35 dpf, water temperature was raised to 28°C and fish were reared to maturity

(75 dpf) under standard conditions (freshwater, 13L:11D photoperiod, dry diet, no added algal turbidity). Sex ratio and fertility were assessed and compared between temperature 32 groups. There were no significant differences in sex ratio, fertilization rate, embryo viability at 1 dpf, clutch size, or relative fecundity. This rearing protocol will allow for efficient utilization of temperature-sensitive mutations in the zebrafish model system.1

1 Chapter 3 has been accepted for publication in the Journal of Thermal Biology as Delomas, T.A., Dabrowski, K., 2018. Larval rearing of zebrafish at suboptimal temperatures. Journal of Thermal Biology. TAD designed, performed, and interpreted the results of the experiment under supervision of KD. TAD was the primary author and drafted the manuscript. Both authors contributed to the editing process. 33

Introduction

Zebrafish were the first vertebrate species to be used in a large-scale genetic screen. These screens identified loss-of-function mutants for multiple genes involved in embryonic development (Driever et al., 1996; Haffter et al., 1996), and success has continued with screens for complex traits, such as behavioral phenotypes (Chiu et al.,

2016). The functions of many essential genes are difficult to analyze in loss-of-function mutants due to mortality of the embryo. In other model systems, such as yeast, this has been overcome by screening for conditional mutants, in which the mutated gene is functional under one set of environmental conditions but not fully functional under other conditions.

The most frequent examples of conditional mutants are temperature-sensitive mutants. Screens for temperature-sensitive yeast mutants have most notably identified essential genes involved in cell-cycle control (Hartwell et al., 1970; Nurse et al., 1976) and currently there are temperature-sensitive mutant strains available for close to half of all essential genes in Saccharomyces cerevisiae (Li et al., 2011). Temperature-sensitive mutants have also been used to study development throughout the life cycle and cell- cycle control in Caenorhabditis elegans (Hirsh and Vanderslice, 1976; O’Connell et al.,

1998).

Zebrafish can tolerate (based on loss of equilibrium) water temperatures as low as

6.2 and 10.6°C when acclimated to 20 and 30°C, respectively (Cortemeglia and

Beitinger, 2005) which allows temperature shifts to be used as an experimental tool in this vertebrate model system (López-Olmeda and Sánchez-Vázquez, 2011). A small

34 number of temperature-sensitive mutants have been found in zebrafish. Temperature- sensitive mutations affecting fin regeneration have been identified (Johnson and Weston,

1995; Nechiporuk et al., 2003) with a permissive temperature of 25°C, slightly below the optimum temperature for growth of zebrafish (28°C), and a restrictive temperature of

33°C. Additionally, one temperature-sensitive mutation in nodal-related 2 (ndr2), also known as cyclops, a nodal-related signaling factor involved in floor-plate specification, was found (Tian et al., 2003). This allowed precise determination of the timing of ndr2 action during embryonic development through temperature shift experiments.

Temperature shift experiments were not performed on juvenile fish to investigate the function of ndr2 after the completion of embryonic development, likely due to the difficulty of raising zebrafish at the permissive temperature (22°C).

Unlike yeast, there has not been widespread screening for temperature-sensitive mutations of essential genes in zebrafish. This is presumably due to the difficulty of raising zebrafish at low temperatures. Even at optimum temperatures, there is wide variation among laboratories in growth and survival of zebrafish (Lawrence, 2011).

Recently, a new method of rearing zebrafish that results in high survival and faster growth compared to traditional protocols was developed (chapter 2). This method utilizes continuous access to live food, low salinity, algal turbidity, and continuous light during the larval and early juvenile stages to increase growth rate. The low salinity allows marine food organisms to stay alive for 24 – 48 hours (Conte et al., 1972; Walker, 1981), continuous light allows continuous foraging, as zebrafish do not forage in the dark

(Carrillo and McHenry, 2016; McElligott and O’malley, 2005), and the algal turbidity

35 increases contrast of the live food organisms, decreases aggressive interactions between larvae, and acts as a food source for the marine live food (McEntire et al., 2015; Naas et al., 1992; Reitan et al., 1997; Rieger and Summerfelt, 1997).

We assessed whether applying this novel rearing method at temperatures close to the minimum for growth would result in (1) high survival, (2) acceptable growth rate and metamorphosis to juveniles, and (3) sexual maturation after returning to optimum temperatures.

Materials and Methods

Broodstock care and reproduction

Broodstock zebrafish were from an AB/TL hybrid line and their care was the same as that described in chapter 2. Four breeding pairs were naturally spawned on the same day. Breeding and embryo incubation was performed as described in chapter 2. At 5 days post-fertilization (dpf), swim-up larvae were collected and combined. Larvae were distributed randomly into six static water tanks with a water volume of 4 L and 120 larvae per tank (30 fish /L). Three tanks were maintained at 28.5 ± 0.5°C (control group,

CO) while three tanks were passively cooled to 23 ± 0.5°C (low temperature group, LT).

The LT group was maintained at 23°C, as this was reported to be the lower thermal limit for embryonic development in zebrafish (Schirone and Gross, 1968) and therefore is the lowest temperature that could be applied from fertilization to adulthood (egg to egg) in a screen for temperature-sensitive mutations. However, more recent work has shown that zebrafish embryos can survive temperatures as low as 22°C (Scott and Johnston, 2012).

Rearing larvae and juveniles

Fish were raised based on the method described in chapter 2. During the larval and early juvenile stages, fish were maintained in static water containers at low salinity (3 ppt), moderate turbidity (3 – 10 NTU) maintained with Nannochloropsis algae paste

(Nanno 3600, Reed Mariculture, Campbell, CA), and 24L:0D photoperiod. The CO group was fed marine rotifers (Brachionus plicatilis) until 10 dpf, when the water volume in the tank was increase to 6 L (20 fish/L) and the diet was switched to Artemia nauplii.

The CO group was maintained under these conditions until 21 dpf. At 21 dpf, 10 – 20 fish from each tank (both CO and LT groups) were measured and the CO group was stocked in a freshwater recirculating system at a density of 2 fish/L and with a 13L:11D photoperiod. Diet for the CO group was transitioned to dry feed (Otohime B1/B2) supplemented with Artemia nauplii. The CO group was maintained under these conditions for the remainder of the experiment.

As fish development and metamorphosis is dependent upon growth (McMenamin et al., 2016), we standardized changes in husbandry based on size and not time (age). The

LT group was maintained under the larval and early juvenile conditions (low salinity, turbidity, continuous light) until 35 dpf, when they had attained approximately the same size as the CO group at 21 dpf (Table 3.1). Additionally, the LT group was fed marine rotifers until 13 dpf, when they had attained the same mean length as the CO group had at

10 dpf (8.7 mm, n = 10 fish), and then fed Artemia nauplii until 35 dpf. At 35 dpf, 20 fish from each of the LT tanks were measured and the water temperature was raised to 28.5 ±

0.5°C over 8 hours. Fish were then stocked in a freshwater recirculating system at a

37 density of 2 fish/L and with a 13L:11D photoperiod. Diet was transitioned to dry feed

(Otohime B1/B2) supplemented with Artemia nauplii. The LT group was maintained under these conditions for the remainder of the experiment.

Sex identification and fertility testing

Once fish were large enough to determine their sex by examining external morphology (Parichy et al., 2009) (62 dpf for the CO group and 75 dpf for the LT group), the numbers of males and females in each tank were counted. Five males and five females from each tank were measured. Fertility tests were then performed to determine if the low temperature during early gonadal development had any effect on reproductive performance later in life.

A sample of males and females (three and four fish of each sex from CO and LT tanks, respectively) from each tank was tested for fertility. Each male and female was spawned separately with a test fish. Test fish were unrelated to the experimental fish and had previously been proven to have high fertility (over 90% fertilization rate when spawned amongst themselves). The same group of test fish was used for assessing both the CO and LT groups. Spawning between the experimental fish and the test fish was performed in the same manner as described above for the original broodstock fish.

Fertilization rate (2 – 4 cells), 1 dpf viability (number alive at 1 dpf/number fertilized), clutch size, and post-spawn weight and length were recorded. Pairs that did not spawn were attempted again one to two days later. All fish spawned within three attempts.

Statistical analysis

ln 푓𝑖푛푎푙 푤푒𝑖푔ℎ푡−ln 𝑖푛𝑖푡𝑖푎푙 푤푒𝑖푔ℎ푡 Specific growth rate (SGR) was calculated as 100 ∗ 푑푎푦푠 and was calculated from the preceding last weighing. Fertility parameters were compared between LT and CO groups using linear mixed models. In all models, temperature treatment was a fixed effect and tank was included as a random effect (random intercept) nested within temperature treatment. For testing each fertility parameter, a t-test was applied to determine if the coefficient for temperature treatment was significantly different from zero. Survival rate and sex ratio were compared between treatments using a Student’s t-test with Tank considered to be the experimental unit. F-tests demonstrated that data were homoscedastic between treatments. Variables that were expressed as proportions derived from counts (survival, sex ratio, fertilization rate, viability at 1 dpf) were Arcsine transformed prior to modeling and hypothesis testing. A type I error rate of

0.05 was used for all statistical tests. Statistical tests and modelling were performed using R statistical software ver. 3.3.2 (R Core Team, 2017) and the lme4 (Bates et al.,

2015) and lmerTest (Kuznetsova et al., 2017) packages. All values are given as mean ±

SD.

Results

Survival was high in both groups (CO and LT) throughout the life cycle (Table

3.1). There was no significant difference in survival between the CO and LT groups at any time point. Growth was expectedly more rapid at 28.5°C than at 23°C with SGR during the first 21 dpf of 35.0 ± 0.3% day-1 and 27.5 ± 0.5 % day-1, respectively (Table

3.1). After the LT group was transitioned to 28.5°C, growth rate increased, and was not different from that of the CO group (Table 3.1, Figure 3.1).

Sex ratio was not significantly different between the two temperature treatments.

The mean percent male in the CO and LT groups were 41 ± 5% and 55 ± 12%, respectively. There were no significant difference in fertility parameters (fertilization rate, viability at 1 dpf, clutch size, or relative fecundity) between the temperature groups

(Table 3.2).

Discussion

In comparison to non-vertebrate model organisms, there has been a lack of temperature-sensitive mutations identified in zebrafish. We describe a novel rearing protocol that results in high survival and acceptable growth at low temperatures (23°C), comparable or higher than previously reported for zebrafish at 35 dpf at 27°C (Uusi-

Heikkilä et al., 2015). Despite being cultured at temperatures close to the lower thermal limit for embryonic development, there was no difference in survival between the LT and

CO groups. The survival of the LT group at 21 dpf (86.1 ± 2.7%) was higher than that reported by authors using dry feed as a sole food source (Carvalho et al., 2006; Goolish et al., 1999) and comparable to results of other authors utilizing live food during larval and early juvenile stages (Aoyama et al., 2015; Best et al., 2010; Dabrowski and Miller,

2018).

Growth rate at 23°C was expectedly slower than at 28.5°C. However, growth rate at 23°C using this method was similar to that reported with other commonly utilized

40 protocols at 28.5°C. Best et al. (2010) described a protocol for first-feeding larvae with marine rotifers and maintaining low levels of salinity (5 ppt), then transitioning to a freshwater environment and feeding Artemia nauplii at 10 dpf. Utilizing this method at

28.5°C, the authors reported a mean length of 13 mm at 23 dpf. Carvalho et al. (2006) fed larvae a diet of Artemia nauplii at 28°C and at 26 dpf the fish had obtained a mean length of 14.3 mm. The current protocol yielded a mean length at 21 dpf of 12.7 ± 0.3 mm for the LT group and 18.5 ± 0.4 mm for the CO group (Table 3.1). This demonstrates that application of this protocol at low temperatures will allow temperature-sensitive mutations to be identified without reducing growth rate below currently accepted levels.

Culturing zebrafish at low temperatures also raises the question of whether the low temperature has an effect on sex determination. Temperature-dependent sex determination has been observed in several fish species (Roemer and Beisenherz, 1996;

Strüssmann et al., 1996), with higher temperatures generally causing an increased proportion of males and lower temperatures causing an increased proportion of females.

While extreme high temperatures (above 35°C) have been shown to cause an increase in the proportion of males in zebrafish (Uchida et al., 2004), this results in high mortality and is outside the range of environmental conditions generally experienced in the wild.

Therefore, it has been suggested that this is not an evolved response (Ospina-Álvarez and

Piferrer, 2008). In this study, low temperature was applied from the beginning of exogenous feeding until fish reached a mean length of 18.9 ± 0.7 mm. Based on the observation of gonad morphology in juvenile zebrafish, sex determination is thought to occur between 12 and 23 mm (Maack et al., 2003), meaning that fish in the low

41 temperature group were exposed to low temperatures through the majority of the sex determination period. We observed no significant difference in sex ratio between the LT and CO groups, suggesting that low temperature may not affect sex determination in zebrafish, but our analysis has limited statistical power due to the large variation in sex ratio among tanks.

There were no significant differences in fertility parameters (Table 3.2) between the LT and CO groups, when female size did not differ, suggesting that low temperatures during early development do not affect later reproductive function in zebrafish. Use of this rearing method will make it possible for zebrafish researchers to screen for temperature-sensitive mutations in essential genes and in genes active during later stages of metamorphosis.

Age (dpf) 21 35 62 75

Temperature Low Control Low Control Low group

Sex Juvenile Male Female Male Female

Survival (%) 86.1 ± 2.7 90.0 ± 3.0 85.8 ± 2.9 86.9 ± 2.8 82.1 ± 6.2

43 Weight (mg) 20.5 ± 1.5 67.3 ± 3.4 76.4 ± 6.7 428 ± 10 589 ± 74 434 ± 18 572 ± 46

Length (mm) 12.7 ± 0.3 18.5 ± 0.4 18.9 ± 0.7 36.4 ± 0.4 38.4 ± 1.7 36.7 ± 0.4 37.8 ± 0.8

SGR (% day-1) 27.5 ± 0.5* 35.0 ± 0.3* 9.4 ± 0.5 4.5 ± 0.1 5.3 ± 0.3 4.4 ± 0.1 5.0 ± 0.3

Table 3.1. Survival and growth of fish raised at standard and low temperatures during larval and early juvenile stages. All values are

given as tank mean ± SD. *calculated with an initial weight of 0.25 mg based on bulk weighing of 25 larvae at 5 days post-fertilization

(dpf)

Females Males Temperature Group Low (n = Control (n = Low (n = Control (n = 12) 9) 12) 9) Weight (mg) 647 ± 95 592 ± 51 447 ± 60 437 ± 64 Length (mm) 39.5 ± 1.6 39.3 ± 1.4 37.2 ± 1.6 37.3 ± 2.6 Age (dpf) 80 ± 3 67 ± 4 83 ± 2 68 ± 4 Fertilization rate (%) 82 ± 13 81 ± 15 83 ± 14 91 ± 9 Viability at 1 dpf (%) 88 ± 10 92 ± 6 95 ± 7 92 ± 6 Clutch size (oocytes/female) 327 ± 140 275 ± 85 - - Relative fecundity 0.51 ± 0.2 0.47 ± 0.1 - - (oocytes/mg) Table 3.2 Fertility of fish raised at standard and low temperatures during larval and early juvenile stages. All values are given as mean ± SD. dpf, days post-fertilization

Figure 3.1 Water temperature and mean weight of low temperature (LT) and control (CO) fish. Error bars represent standard deviation of mean weight between tanks.

CHAPTER 4. WHY ARE TRIPLOID ZEBRAFISH ALL MALE?

Abstract

Adult triploid zebrafish Danio rerio have previously been reported to be all male.

This phenomenon has only been reported in one other gonochoristic fish species, the rosy bitterling Rhodeus ocellatus, despite triploidy being induced in numerous species. To investigate the mechanism responsible, we first produced triploid zebrafish and observed gonad development. Histological sections of juvenile triploid gonads showed that primary growth oocytes were able to develop in the juvenile ovary, but no cortical alveolus or more advanced oocytes were found. All adult triploids were male. Male triploids were able to induce oviposition by diploid females during natural spawning trials, but fertilization rates were low (1.0 ± 3.1%) compared to diploid male siblings

(67.4 ± 16.6%). Embryos produced by triploid sires were aneuploid with a mean ploidy of 2.44 ± 0.10n, demonstrating that triploid males produce aneuploid spermatozoa. After confirming that adult triploids are all male, we produced an additional batch of triploid zebrafish and exposed them (and a group of diploid siblings) to 100 ng/L estradiol (E2) from 5 – 28 dpf. E2 treated triploids and non-treated triploids were all male. Non-treated diploids were also all male, but E2 treated diploids were 89% female. This demonstrates that triploidy is acting downstream of estrogen synthesis in the sex differentiation pathway to induce male development. Based on this and the observations of juvenile

46 gonad development in triploids, we suggest that triploidy inhibits development of oocytes past the primary growth stage, and this causes female to male sex reversal.

47 Introduction

Sex determination is the process by which a gonochoristic organism makes the binary choice of developing into either a male or a female. In many fish, this process is controlled by a single gene. If the female allele is dominant, the species is referred to as having female heterogamety (WZ females, ZZ males) (Dabrowski et al., 2000; Glennon et al., 2012), while if the male allele is dominant, the species is referred to as having male heterogamety (XX females, XY males) (Bongers et al., 1999; Chourrout and Quillet,

1982). In some fish, the process is controlled by multiple genes, each contributing to an underlying genetic liability, and is described as polygenic (Vandeputte et al., 2007).

Classical genetic techniques indicate that zebrafish have a polygenic sex determination system, as the sex ratio of individual families can range from 5 – 97 % male, but individual breeding pairs give highly correlated sex ratios (Liew et al., 2012).

Additionally, sex ratio has been shown to respond to selection (Liew et al., 2012), and mapping studies have found multiple quantitative trait loci (QTL) associated with sex

(Anderson et al., 2012; Bradley et al., 2011; Howe et al., 2013). In one mapping study, the authors reported finding only one QTL for sex in a selection of wild and “less- domesticated” strains, while no QTL were found in the two most common laboratory strains, AB and TU, that had been through a process of removing recessive lethal alleles.

This led the authors to propose that wild zebrafish have a system of female heterogamety, but the process of removing recessive lethal alleles had eliminated the sex determining locus (Wilson et al., 2014). However, it has been shown that one of these “less- domesticated” strains and hybrids between this strain and wild zebrafish give variable sex ratios (Brown et al., 2012), demonstrating that even though a major QTL for sex is 48 present, the sex determination system is likely polygenic. Additionally, the closely related pearl danio D. albolineatus was shown to have a polygenic sex determination system, suggesting that this system may be evolutionarily conserved among Danio (chapter 7).

Triploidy, the presence of three complete haploid sets of chromosomes, can be induced in fish and typically results in sterility due to aberrations in meiosis. Triploid males frequently produce aneuploid spermatozoa, while triploid females in most species are sterile (Benfey, 2011). However, triploid common carp Cyprinus carpio females have been found to produce aneuploid oocytes (Gomelsky et al., 2016, 2015). Triploidy is induced by applying a physical shock (temperature or pressure) to fertilized oocytes, that causes retention of the second polar body. Kavumpurath and Pandian (1990) first reported the production of triploid zebrafish, and surprisingly observed that triploid adults were all male. Mizgireuv et al. (2004) also obtained triploid zebrafish and reported that out of 120 adult triploid zebrafish, only one was female. Feitsma et al. (2007) obtained sixteen adult triploids that were produced without applying a shock (these fish were the offspring of mlh1-/- females, which exhibit chromosome missegregation during oogenesis), and all were male. While triploidy has been induced in numerous fish species, we are aware of only one other gonochoristic species, the rosy bitterling Rhodeus ocellatus, (Kawamura, 1998; Ueno and Arimoto, 1982) and a few protandrous hermaphrodites (Haffray et al., 2005; H. Kitamura et al., 1991) in which it was reported that triploids were all male. The purpose of the current study was to investigate this unusual phenomenon and determine why triploid zebrafish are all male.

49 Materials and methods

Experimental overview

Our investigation was broken into two experiments. The goals of experiment one were to observe early gonad development in triploids, establish if adult triploids are all male, and investigate the fertility of triploid males. We performed three crosses of zebrafish females with zebrafish males (one male and one female per cross). A sample of fertilized oocytes (100 – 150 oocytes) from each cross was left unshocked as a diploid control and the remaining (350 – 800 oocytes) were subjected to a heat shock to induce triploidy. At five days post-fertilization (dpf), 9 – 10 larvae from each triploid group and

3 larvae from each diploid group were tested for ploidy. Fish were then raised until 21 dpf, when they were counted and ten fish were randomly sampled and measured. At 21 and 34 days post-fertilization (dpf) five to seven fish from two of the triploid and two of the diploid groups were randomly sampled, ploidy confirmed, and preserved in formalin for histological analysis of gonad development. Juveniles from one breeding pair were not sampled due to lower numbers of surviving larvae. At 70 dpf, surviving adults were counted and ten fish of each sex were randomly sampled and measured. Sex was identified in all surviving fish. Twenty-five fish from each triploid group and six fish from each diploid group were analyzed for ploidy. Twenty-four triploid males (6 – 12 per group) and twelve diploid males (3 – 6 per group, proportional among groups to the number of triploid siblings used) were used for fertility testing by natural spawning. Six triploid males (two per group) and three diploid males (one per group) were used for fertility testing by in vitro fertilization.

50 The goal of experiment two was to determine if triploidy induces male development by acting upstream or downstream of increased body estrogen concentration in the sex determination and differentiation pathway. Increased estrogen concentration is thought to be a key event in triggering the female differentiation process in zebrafish.

This is based on observations that exposure to exogenous estrogens causes male-to- female sex reversal (Örn et al., 2003) and that inhibition of estrogen synthesis

(pharmacologically and genetically) causes female-to-male sex reversal (Dranow et al.,

2016; Takatsu et al., 2013). We collected oocytes from three females and fertilized them

(individually) with sperm collected and pooled from two males. A sample of fertilized oocytes (150 – 300 oocytes) from each cross was left untreated as a diploid control and the remaining (500 – 700 oocytes) were subjected to a heat shock to induce triploidy. The diploid and triploid larvae were each split equally into two groups: one treated with estradiol (E2) and one untreated. Exogenous E2 was previously used to sex reverse triploid rainbow trout males (Krisfalusi and Cloud, 1999). Larvae from the three crosses were combined to be proportional between the diploid and triploid groups (i.e. for every triploid larva from a given female that was stocked, there was a diploid larva from that same female stocked) with a total number of 80 larvae per group. Each female contributed 24 – 27 larvae to each group. The E2 treatment was applied from 5 – 28 dpf.

At 28 dpf, fish were counted and 10 fish from each group were randomly sampled and measured. A random sample of surviving adults (20 – 30 fish per group depending on initial survival) were raised to 90 dpf, and the sex of all surviving adults was identified.

Ten adults from each shocked group and three adults from each unshocked group were tested for ploidy. 51 Fish and fish husbandry

The broodstock used were from an AB/TL hybrid line. Broodstock husbandry and embryo incubation were performed as described in chapter 2. At 5 dpf larvae were stocked into static water tanks and raised according to the protocol described in chapter 2, with the exception that the practices for the larval/juvenile stage (continuous light, algal turbidity, live food, low salinity) were maintained until 28 dpf in experiment 2. In experiment one, stocking density during the larval/early juvenile stage ranged from 12 to

17 fish/L, and was constant within each breeding pair (triploid and diploid siblings were stocked at the same density). In experiment two, stocking density during this period was

15 fish/L. Stocking density after the larval/early juvenile stage for both experiments was

1.1 – 1.6 fish/L.

Breeding and induction of triploidy

Ovulation was induced in female zebrafish by housing females overnight in the same tank as males but separated by a mesh barrier. The following morning, approximately 15 minutes after the lights were turned on, females and males were allowed to interact. When spawning behavior was observed, females were removed, anesthetized with MS-222 (100mg/L), and oocytes collected by stripping. Sperm was collected from zebrafish males by dissection and maceration of the testes in E400 extender (Matthews and Carmichael, 2015). Oocytes were fertilized in vitro and incubated at 28.5°C. Triploidy was induced by applying a heat shock of 41.4°C

(Streisinger et al., 1981) from two to four minutes post-fertilization to the fertilized oocytes. Following the heat shock, oocytes were immediately returned to 28.5°C.

52 Confirmation of triploidy

Triploidy was assessed by flow cytometry. Larvae were assessed following the protocol of Delomas and Dabrowski (2016). Staining was performed similarly to the process described by Garcia-Abiado et al. (2001). Individual embryos were placed in microcentrifuge tubes containing 500μL staining solution (50 mg/L propidium iodide, 10 mg/L RNase A in Isoton II) with 1μL of a 0.05x dilution of rainbow trout Oncorhynchus mykiss red blood cells (RBC) in Isoton II as an internal standard. Samples were incubated overnight in the dark at 4°C, then vortexed and aspirated through a 26G needle into a

1mL syringe. Samples were then filtered through 60μm mesh and a 26G needle. Ten thousand events were recorded per sample using an Accuri C6 flow cytometer (BD

Biosciences, San Jose, CA). Relative nuclear DNA content was calculated as the ratio of fluorescence intensity between the sample peak and the RBC peak. This ratio was multiplied by the c-value (DNA content of a haploid chromosome set) for rainbow trout

(Ohno et al., 1969) to obtain the sample c-value. Sample c-value was converted to ploidy by comparing to the known c-values for zebrafish (Ciudad et al., 2002).

Juveniles and adults were assessed following the same procedure, except a fin clip was used as the tissue source. The fin clip was vortexed periodically during the incubation. Additionally, only the solution (not the entire fin clip) was drawn into the syringe and filtered for analysis.

Estradiol (E2) treatment

The E2 treatment in experiment two was applied from 5 to 28 dpf, while the fish were reared in static containers. This period was chosen so that fish reared under our protocol would reach the size (23 mm) at which indifferent gonads are no longer 53 observed (Maack et al., 2003; Maack and Segner, 2003) during the E2 treatment. This criteria for ending exogenous hormone treatment has been used successfully in other studies of the effects of exogenous hormones on zebrafish sex differentiation (Fenske and

Segner, 2004). A stock solution of E2 was dissolved in ethanol at 100 mg/mL. This was diluted in water to 1 mg/mL immediately prior to addition to the tank for a final concentration of 100 ng/L in the tank water. The final concentration of ethanol in the tank water was 2.9 μM (0.000017% v/v). Water was changed (100% of volume) every other day to replenish the concentration of E2. The water of untreated controls was also changed (100% of volume) every other day to control for handling stress. The four treatments were: diploid with 100 ng/L E2 (DE2), diploid without E2 (D), triploid with

100 ng/L E2 (TE2) and triploid without E2 (T).

Sex identification and histology

In experiment one, sex was first identified in adults by dissecting and observing gross gonad morphology. Sex was then confirmed in 25 – 30 triploid individuals and five diploid individuals from each breeding pair by histological examination of the gonads.

Gonad development in juveniles was investigated by histological examination of the gonads. Stages of early oocyte development were defined as primary growth (stage I) and cortical alveolus (II) (Selman et al., 1993). Sex was identified in all adult fish in experiment two by dissection and observation of gonad morphology. Additionally, gonads from ten triploid individuals from each treatment were examined histologically to confirm the observed sex. Histology was performed by tissue dehydration, embedding in paraffin, sectioning, and staining with hematoxylin and eosin according to standard techniques (Hewitson et al., 2010). 54 Fertility testing

Fertility of triploids was examined both in natural spawning and through in vitro fertilization. For the natural spawning trials, a group of six zebrafish females was used.

Females were assigned to males randomly, with the restriction that they be used evenly

(same number of matings for each female) and proportionately between triploid and diploid groups. Breeding and oocyte collection were performed as described in chapter 2.

Fertilization rate (2 – 4 cells) was counted in a sample of oocytes and survival of embryos in this same sample was counted at 1 dpf.

Fertility was also examined by in vitro fertilization. Sperm was collected from all males by dissection and maceration of one testis in 100μL of E400 extender (Matthews and Carmichael, 2015). Oocytes from three zebrafish females were collected as described in the “Breeding” section above, pooled, and evenly divided into nine groups. Each group was inseminated with 10μL of the prepared sperm solution from one male and activated with 500μL of aquarium water following “dry” insemination. After one minute, an additional 2 mL of water was added to each group and oocytes/embryos were incubated at 28.5°C. At one-hour post-fertilization, fertilization rate (2 – 4 cells) was counted, and survival of embryos was counted at 1 dpf.

Statistical analysis

In experiment one, weight and length were compared between triploids and diploids at 21 dpf and between triploid and diploid males at 70 dpf. Linear models were fit with weight or length as the response variable and both ploidy and breeding pair as the explanatory variables. The difference between triploid and diploid was assessed by using a t-test to determine if the coefficient for ploidy was significantly different from zero. In 55 experiment two, weight and length of the four groups at 28 dpf was compared using a one-way ANOVA. Pairwise comparisons were made using Tukey’s HSD. An alpha level of 0.05 was used for all hypothesis tests. All statistical tests were performed using R statistical software v. 3.4.1 (R Core Team, 2017). All quantities are expressed as mean ±

SD.

Results

Experiment one

Larvae in the shocked groups had a rate of triploidy of 90 ± 10% triploid (n = 9 –

10 per group, 29 total). All larvae in the unshocked groups were diploid (n = 3 per group,

9 total). Adults in the shocked groups were 100% triploid (n = 25 per group, 75 total) and all adults in the unshocked groups were diploid (n = 6 per group, 18 total). Survival of the triploid (shocked) groups was lower than that of the diploid (unshocked) groups, but the majority of mortality occurred prior to the onset of exogenous feeding (Figure 4.1). Mean weight and length of juvenile (21 dpf) triploids, 63 ± 15 mg and 20.5 ± 1.5 mm, was not significantly different from those of juvenile diploids, 67 ± 13 mg and 20.6 ± 1.4 mm

(Figure 4.2, Student’s t, p > .05).

All triploid adults were found to be male, while diploids were mixed sex with a mean sex ratio of 41 ± 15% male (Table 4.1). At 70 dpf, mean weight of triploid males,

442 ± 86 mg, was significantly larger than that of diploid males, 403 ± 74 mg, (Figure 2,

Student’s t, p < .01), but no significant difference was found in length (36.7 ± 2.2 mm and 36.0 ± 2.2 mm for triploid and diploid males, respectively). Diploid females were

56 larger than both groups of males with mean weight and length of 574 ± 168 mg and 37.2

± 2.9 mm (Figure 4.2).

Of the 24 triploid males we attempted to spawn naturally, 19 males induced oviposition by a female zebrafish. Twelve out of 13 diploid males tested induced oviposition by a female zebrafish (Table 4.2). Fertilization rate achieved during natural spawning by the triploid males (1.0 ± 3.1%) was drastically lower than that of the diploid males (67.4 ± 16.6%). Out of the 19 triploid males that induced oviposition, only eight fertilized at least one oocyte; all 12 diploid males that induced oviposition successfully fertilized oocytes. Fertilization rates achieved through in vitro fertilization showed the same pattern, with diploid males producing a higher mean rate (66.5 ± 5.7%) than triploid males (0.4 ± 0.7%) and while sperm from only two of the six triploid males fertilized oocytes, sperm from all three diploid males was able to fertilize the oocytes. Survival to 1 dpf showed a similar pattern in the natural spawning group, with embryos from diploid males having a mean survival rate of 91.3 ± 5.8% and embryos from triploid males having a mean survival rate of 55.1 ± 48.5%. Survival to 1 dpf was approximately the same between embryos produced by in vitro fertilization with sperm from diploid and triploid males; however, only six embryos total were produced by the triploid males, which further reinforces the significance of the difference in sperm quality between diploid and triploid males. Histological sections of testes from diploid males showed developing spermatocytes and densely packed spermatozoa in the lumen of the spermatogenic tubules (Figure 4.3A). Testes of triploid males contained spermatocytes, but visible spermatozoa were not observed in the lumen (Figure 4.3B).

57 Five embryos produced by five different triploid males were analyzed for ploidy, and all were aneuploid with mean ploidy 2.44 ± 0.10n. Five embryos produced by five diploid males were also analyzed for ploidy and all were found to be diploid (2n).

To investigate the effects of triploidy on sex differentiation, we examined early gonadal development in diploid and triploid juveniles. At 21 dpf, triploids (20.4 ± 2.2 mm length) either had undifferentiated gonads (6 out of 13, Figure 4.4C) or juvenile ovaries (7/13, Figure 4.4D). Diploid siblings (21.4 ± 1.0 mm) also had either undifferentiated gonads (1/10, Figure 4.4A) or juvenile ovaries (9/10, Figure 4.4B). At 34 dpf, 13/14 of triploids (27.9 ± 2.7 mm) had testes (Figure 4.4G), but five of these still contained isolated stage I oocytes (Figure 4.4H). One abnormally small triploid (23.2mm) at 34dpf retained the juvenile ovary. Diploid siblings (27.9 ± 2.5 mm) either had testes

(6/10, Figure 4.4F) or ovaries containing stage I and II oocytes (4/10, Figure 4.4E).

Experiment two

Embryonic survival of triploids and diploids was similar in experiment two to the results obtained in experiment one. Survival of embryos (2 – 4 cells) to five dpf was 96 ±

2% for the diploid groups and 53 ± 6% for the triploid groups. Survival in each group from five dpf to 28 dpf (period of the estrogen treatment) was 49%, 49%, 69%, and 71% for TE2, T, DE2, and D, respectively. Survival from 28 dpf until the end of the experiment (90 dpf) was 96%, 85%, 93%, and 90% for TE2, T, DE2, and D, respectively.

Weight and length at 28 dpf were significantly different between T and the other three treatments (Figure 4.5), but the mean weight of all groups was greater than 100 mg, the threshold previously suggested for puberty initiation (Chen and Ge, 2013). All fish tested for ploidy from T and TE2 tested were triploid (n = 10 fish per group, 20 total) and all 58 fish tested from D and DE2 were diploid (n = 3 fish per group, 6 total). All fish from T,

TE2, and D were male, while the sex ratio in DE2 was 11% male (3 out of 28, Table 4.3).

Discussion

We first confirmed previous reports that adult triploid zebrafish are all (Feitsma et al., 2007; Kavumpurath and Pandian, 1990), or almost all (Mizgireuv et al., 2004), male.

To our knowledge, the only other gonochoristic species in which this has been reported is the rosy bitterling (Kawamura, 1998; Ueno and Arimoto, 1982). The sex determination system of rosy bitterling was shown to be a system of female heterogamety, and based on strongly male-biased sex ratios in gynogenetic progeny groups, it was suggested that

WW females have reduced viability compared to ZZ males (Kawamura, 1998). These authors also suggested that the absence of female triploids could be explained by the reduced viability of the WW genotype if WWZ triploids were invaiable, while WZZ and

ZZZ triploids were male. Kavumpurath and Pandian (1990) similarly suggested that female triploid zebrafish may be inviable, and so only the males survive to adulthood.

All zebrafish develop a “juvenile ovary” containing stage I oocytes. The juvenile ovary either continues to develop, and that individual becomes a female, or the oocytes go through apoptosis, and that individual becomes a male (Takahashi, 1977; Uchida et al., 2002). In zebrafish, it has been shown that experimental ablation of primordial germ cells or oocytes can cause female to male sex reversal (Dranow et al., 2013; Siegfried and

Nüsslein-Volhard, 2008). Additionally, mutations that interfere in early oocyte development can cause female to male sex reversal (Rodríguez-Marí et al., 2010). It is possible that triploidy interferes with oocyte development and thereby induces male 59 development. Our observations of early gonad development are consistent with this hypothesis, as we observed stage I oocytes to form in the juvenile ovary of triploids, but no gonads from triploids contained stage II (or more advanced, vitellogenic) oocytes

(Figure 4.4).

The results of experiment two also support the hypothesis that triploidy interferes with the development of oocytes. Sex ratio in the DE2 group was strongly female biased

(11% male, Table 4.3) while sex ratio in the D group was strongly male-biased (100% male). This demonstrates the efficacy of the E2 treatment in overriding the sex determination system and inducing female development. It is unusual for a progeny group of zebrafish to be all male, as we observed in the D group, but this has been shown to occur at a low frequency as a result of the polygenic sex determination system of zebrafish (Liew et al., 2012). All fish in both the TE2 and T groups were male, demonstrating that E2 treatment is not able to override the male-inducing effects of triploidy. This demonstrates that triploidy is acting downstream of estrogen in the sex differentiation pathway and is consistent with the hypothesis that triploidy inhibits later stage oocyte development, thereby inducing male development.

Our results contradict the hypothesis of Kavumpurath and Pandian (1990), that triploid females are inviable. If the triploid female genotype was inviable, but not the triploid female phenotype, we would expect triploid females to be produced by E2 treatment. If the triploid female phenotype was inviable, we would expect higher mortality in the TE2 group compared to the T group. Since we made neither observation, we can conclude that differential mortality is unlikely to explain the unisexuality of triploid zebrafish. 60 Triploid males of multiple fish species have been observed to produce small amounts of aneuploid spermatozoa (Benfey, 2011; Benfey et al., 1986; Lincoln and Scott,

1984; Peruzzi et al., 2009). Our observations of triploidy in zebrafish are consistent with this pattern, as triploid males fertilized oocytes at a dramatically lower rate than diploid males, and the few surviving embryos were shown to be aneuploid. Additionally, histological examination of testes from triploids showed a lack of spermatozoa compared to testes from diploids. This demonstrates that triploid zebrafish males produce small amounts of aneuploid spermatozoa.

We observed a large proportion (19/24, Table 4.2) of triploid male zebrafish to successfully induce oviposition by diploid females. Due to the large adult size, and therefore difficulty of natural spawning in the laboratory setting, of most aquaculture species, the ability to spawn naturally has not been widely observed in triploid fish. It has been reported in masu salmon Oncorhynchus masou that triploid males demonstrate courtship behavior (“quivering” next to females) and are able to induce nest building by diploid females (S. Kitamura et al., 1991). Additionally, triploid Atlantic cod Gadus morhua males have been shown to successfully spawn with diploid females (Feindel et al., 2010). Our results further demonstrate that triploid male fish are able to naturally spawn with diploid females.

Triploid males were observed to obtain a significantly larger weight at 28 and 70 dpf in comparison to diploid males (Figures 4.5 and 4.2, respectively). Previous studies in a variety of fish species has demonstrated that the relationship between triploidy and growth varies depending upon the species and time period examined (Benfey, 2011).

Diploids are usually observed to grow faster than triploids early in the life cycle, but 61 triploids often grow faster later in the life cycle, possibly due to a lower level of reproductive investment, which leads to approximately equal growth rates in many studies (Carter et al., 1994; Felip et al., 2001). However, in all-female populations, it has been observed that triploids reach larger sizes than diploids, presumably due to the large metabolic cost of oocyte development in diploids (Sheehan et al., 1999). The significantly larger weight of juvenile and adult triploid male zebrafish observed in the current study could be a result of reduced metabolic investment in spermatogenesis compared to diploid males.

Sex ratio Breeding pair Ploidy Males Females (% male)

1 23 53 30

2 2n 14 25 36

3 32 23 58

1 72 0 100

2 3n 35 0 100

3 53 0 100

Table 4.1 Sex of diploid and triploid zebrafish offspring of three different breeding pairs.

Fertilization Ploidy Number Number Fertilization One dpf method tested induced rate (%) viability

oviposition (%)

Natural 2n 13 12 67.4 ± 16.6 91.3 ± 5.8

3n 24 19 1.0 ± 3.1 55.1 ± 48.5

In vitro 2n 3 - 66.5 ± 5.7 97.8 ± 2.4

3n 6 - 0.4 ± 0.7 100 ± 0*

Table 4.2 Fertility of diploid and triploid zebrafish males tested through natural spawning or by in vitro fertilization with sperm collected by dissection and maceration of the testes.

Dpf is days post-fertilization. *Two males yielded three embryos each (total of six), and all six embryos survived to one dpf

Ploidy Estradiol (ng/L) Males Females Sex ratio (% male)

0 27 0 100 2n 100 3 25 11

0 17 0 100 3n 100 26 0 100

Table 4.3 Sex of adult diploid and triploid zebrafish exposed to either 0 ng/L or 100 ng/L estradiol from 5 to 28 days post-fertilization.

Figure 4.2 Mean individual weight and length of diploid and triploid zebrafish at 21

(juveniles) and 70 (males and females) days post-fertilization (dpf) (n = 3 groups per treatment, 10 fish per group). Error bars represent standard deviation. *statistically significant difference (Student’s t, p < 0.01). Two comparisons were statistically tested: triploid juveniles vs diploid juveniles at 21 dpf and triploid males vs diploid males at 70 dpf.

Figure 4.3 Histological sections of testes from adult (A) diploid and (B) triploid males.

The lumen in the testis from the diploid fish contains spermatozoa while none are visible in the lumen of the testis from the triploid fish. *Lumen of spermatogenic tubules.

Figure 4.4 Histological sections of developing gonads from diploid and triploid zebrafish.

II, and S indicate stage I oocytes, stage II oocytes, and spermatocytes, respectively. Scale bars represent 50μm.

Figure 4.5 Mean weight and length of diploid (D, DE2) and triploid (T, TE2) zebrafish exposed to either 0 ng/L (D, T) or 100 ng/L E2 (DE2, TE2) at 28 days post-fertilization.

Error bars represent standard deviation. Different letters indicate statistically significant differences between groups in both weight and length (ANOVA, p < 0.01).

70 CHAPTER 5. COMMON CARP SPERM INDUCES ZEBRAFISH EMBRYONIC DEVELOPMENT

Abstract

Haploid gynogenetic screens increase the efficiency of forward genetic screens and linkage analysis in fish. Typically, UV-irradiated zebrafish sperm is used to activate zebrafish oocytes for haploid screens. We describe the use of UV-irradiated common carp sperm to activate haploid gynogenetic zebrafish development. Carp x zebrafish hybrids are shown to have a characteristic set of features during embryonic development and exhibit functional development of several tissues (muscle, heart, nervous system).

Hybrids become inviable past the embryonic stages. This technique eliminates the possibility of incompletely irradiated zebrafish spermatozoa contaminating haploid progenies. While developing this protocol, one unique zebrafish female was identified which, upon insemination with UV-irradiated carp spermatozoa, repeatedly displayed spontaneous diploidization of the maternal chromosomes in her offspring.2

2 Chapter 5 was previously published in Biology Letters (http://rsbl.royalsocietypublishing.org/) as Delomas, T.A., Dabrowski, K., 2016. Zebrafish embryonic development is induced by carp sperm. Biol. Lett. 12, 20160628. TAD designed, performed, and interpreted the results of the experiment under supervision of KD. TAD was the primary author and drafted the manuscript. Both authors contributed to the editing process. 71 Introduction

Zebrafish Danio rerio have rapidly become a key model vertebrate species.

Forward genetic screens, which entail random mutation of organisms’ genomes, screening for phenotypes, and then identifying mutated genes, have uncovered numerous developmental genes in zebrafish. This process is accelerated by screening for phenotypes in haploid generations produced by single mutagenized parents, known as a haploid screen, as parental and F1 generations can be directly assessed (Corley-Smith et al., 1996; Walker, 1999). Haploid screens have been instrumental in developing genetic linkage maps, that detail the recombination frequency between loci in a species’ genome

(Postlethwait et al., 1994). These maps are used to locate alleles responsible for a given phenotype by identifying molecular markers linked to the trait of interest. Once the location of an allele is known, it can be cloned and sequenced. This process is known as positional cloning, and has been used in zebrafish to sequence alleles responsible for phenotypes identified in forward genetic screens (Donovan et al., 2000).

Haploid screens in zebrafish are typically performed by activating zebrafish oocytes with UV-irradiated zebrafish sperm (Walker, 1999). Irradiation damages the paternal chromosomes and results in inheritance of only maternal chromosomes.

However, this method suffers first due to the fact that rare, incompletely irradiated spermatozoa may lead to fertilization and can “contaminate” the results of the experiment. Second, it is labor-intensive to obtain a sufficient volume of zebrafish sperm to successfully activate oocyte development after sperm irradiation.

Embryonic development can be induced in many fish species by activation of oocytes with heterologous sperm. The resulting zygotes can become hybrid, androgenetic 72 (inheriting only paternal chromosomes), or gynogenetic (inheriting only maternal chromosomes) embryos (Chevassus, 1983). Activation of oocytes by heterologous sperm has been demonstrated in multiple species, including the activation of Mozambique tilapia Oreochromis mossambicus oocytes with common carp Cyprinus carpio sperm

(Varadaraj, 1990), both reciprocal crosses of grass carp Ctenopharyngodon idella and common carp (Stanley, 1976), and the activation of zebrafish oocytes with cryopreserved common carp sperm (Kovács et al., 2014). The purpose of using heterologous sperm in these cases was to generate “uncontaminated” gynogenetic progeny. We hypothesized that by using irradiated heterologous sperm we would improve the efficacy of haploid screens in zebrafish, particularly if a species whose hybrids with zebrafish are distinguishable during embryonic development were used.

Methods

Fish

Broodstock zebrafish (2 – 10 month old, 500 – 1400mg weight) were an AB/TL hybrid line, Casper (Carolina Biological, Burlington, NC), and fish from a local pet shop

(GloFish brand). The carp male used in this study was a common carp x koi (ornamental common carp) hybrid. Zebrafish broodstock husbandry was performed as described in chapter 2. The carp male used was kept in a greenhouse enclosed recirculating system with temperature maintained at 26 – 29°C throughout the year and with a natural photoperiod, 14L:10D at the time of the study. Embryos were incubated as described in chapter 2.

73 Collection of Gametes and Generation of Haploids

Collection of oocytes and zebrafish sperm was performed as described in chapter

4. Zebrafish sperm was taken from males homozygous for a transgenic coloration gene

(GloFish) so that any contamination would be evident. Spermiation was induced in a carp male by injection of 3 mg/kg dried carp pituitary (Stoller Fisheries, Spirit Lake, IA) given

12 – 14 hours prior to stripping (Alsaqufi et al., 2014). For production of haploid progenies, carp sperm was irradiated similarly to the method described by Alsaqufi et al.

(2014). Sperm was first diluted in 350 mmol glucose, 30 mmol Tris extender (pH 8.0)

(Horvath et al., 2003) at a ratio of 1:9 (sperm : extender), and then irradiated at 3,000

J/m2 in a UV crosslinker (Stratagene, La Jolla, CA) while being stirred with a magnetic stir bar on ice. All oocytes were fertilized in vitro.

Oocytes from individual females were divided into three groups for separate inseminations with intact carp sperm, UV-irradiated carp sperm and intact zebrafish sperm. After eight inseminations with intact carp sperm in order to confirm absence of viable hybrids and characteristic phenotype, this cross was no longer repeated.

Flow Cytometry

A sample of 3 – 14 embryos taken from each progeny 3-5 days post-fertilization

(dpf) was analyzed by flow cytometry for nuclear DNA content according to the method described in chapter 4. Sample c-value was converted to ploidy by comparing to the known c-values for zebrafish (Ciudad et al., 2002), common carp (Tiersch et al., 1989), and the theoretical value for a zebrafish x common carp hybrid, calculated as the average of zebrafish and common carp c-values.

74 Results

Zebrafish embryonic development was successfully induced by both intact and

UV-irradiated carp sperm. Average ± SD rate of first cleavages (2 – 4 cell stage) for intact carp sperm (n = 8 propagations), UV-irradiated carp sperm (n = 11), and intact zebrafish sperm (n = 13) were 71 ± 9%, 47 ± 25%, and 76 ± 12%, respectively. Survival rates of embryos produced by insemination with UV-irradiated or intact common carp sperm were reduced at each time point compared to embryos produced by insemination with intact zebrafish sperm (Figure 5.1).

Embryos resulting from hybridization with intact carp sperm progressed normally through epiboly. At 24 hours post-fertilization (hpf), some embryos appeared normal except for shorter, curved caudal regions while other embryos had short bodies with no distinguishable caudal peduncle. By 48 hpf, all embryos became abnormal, with severely truncated caudal regions, short bodies, and typically with pericardial edema (Figure

5.2B). The heartbeat was visible in embryos from 48 to 120 hpf. The first hatched larvae were observed at 24 hpf and at 72 hpf all surviving embryos had hatched. From 48 to 120 hpf, some hatched larvae were observed to twitch (10 ± 5%) and approximately 1% swam erratically for a short distance when agitated. No larvae displayed sustained motility, utilized the adhesive gland, nor inflated the gas bladder. Flow cytometric analysis revealed that all hybrid larvae were diploid with c-value 1.7 pg, which is equal to the average c-value of zebrafish (Ciudad et al., 2002) and carp (Tiersch et al., 1989)

(Table 5.1).

Embryos resulting from insemination with UV-irradiated carp sperm developed at approximately the same rate as normal zebrafish (control) embryos. At 24 hpf, most 75 embryos displayed no obvious signs of deformity, but several embryos with a slightly shortened body or without a distinguishable head were observed. At 48 hpf, embryos had a short body and slightly curved caudal regions (haploid syndrome). The body of haploid zebrafish was shorter than diploid zebrafish, but significantly longer than the body of carp hybrids (Figures 5.2, 5.3). From 48 to 120 hpf, the heartbeat was visible, and several larvae twitched or moved a short distance when agitated. Some embryos developed pericardial edema. No morphologically abnormal larvae adhered to the container’s wall, generated sustained movement or filled the gas bladder. Flow cytometric analysis revealed that these larvae were haploid (Figure 5.4).

A portion of larvae resulting from inseminating oocytes produced by one individual AB/TL zebrafish with UV-irradiated carp sperm were repeatedly observed in three subsequent propagations (at weekly intervals) to have regular morphology with inflated swim bladders at 120 hpf. A total of 42 morphologically normal larvae with inflated gas bladders were obtained. Four were shown by flow cytometry to be diploid.

Thirty-eight were stocked into tanks and began feeding on rotifers at 120 hpf. This confirms that these larvae were diploid. Morphologically abnormal larvae (n = 25) from the same progenies were analyzed and shown to be haploid (Table 5.2). The percentage of embryos (2 – 4 cell stage) from this female that developed into diploid larvae was variable, at 34.5% per clutch, 0.5%, and 1.0% over three consecutive spawnings. In progenies produced by crossing other females with irradiated carp sperm, only one normal larva was observed out of 1730 embryos (Table 5.3).

Control zebrafish embryos developed normally (Kimmel et al., 1995), and 98% of larvae alive at 120 hpf had an inflated gas bladder. Flow cytometric analysis of larvae 76 with inflated gas bladders revealed them to be diploid (Table 5.1). A small sample (n = 4) of deformed control larvae with uninflated gas bladders were analyzed and this group was shown to be composed of both diploids (n = 2) and haploids (n = 2).

Discussion

The abnormal morphology and low survival during embryonic development of the zebrafish x carp hybrids demonstrates their inviability past embryonic development. This is an important finding as common carp successfully hybridizes in multiple intergeneric crosses, for example with grass carp (Stanley, 1976). High mortality of interspecies fish hybrids during embryonic development is commonly observed (Chevassus, 1983), presumably due to genetic incompatibility. Analysis of nuclear DNA content revealed that hybrid nuclei contained the amount of DNA predicted in a true hybrid. This demonstrates that karyogamy occurs in zebrafish x carp hybrids, but embryonic development results in severe deformities and hybrids are not capable of developing past the embryonic stages.

Development of haploid gynogenetic zebrafish produced using irradiated zebrafish sperm has been previously described in depth (Walker, 1999), and haploids produced using irradiated carp sperm follow the same pattern. Although the survival rate of haploids was lower than diploid zebrafish, there was a significant portion alive at each examined time point. Luo and Li (2003) also demonstrated that gynogenetic haploid goldfish obtained with UV-irradiated carp sperm survived beyond hatching, but did not begin feeding. This reinforces the suggestions of Walker (1999) that haploid screens can be used to investigate development of advanced features such as muscular development 77 (Martin et al., 2015), and patterning of the nervous system (Saint-Amant and Drapeau,

1998).

This new procedure for gynogenetic haploid screens has two main advantages over current methods (Kroeger et al., 2014; Walker, 1999) that utilize UV-irradiated zebrafish sperm to activate zebrafish oocytes. First, the recognizable phenotype of zebrafish x carp hybrids eliminates interference from incompletely inactivated spermatozoa. Second, sperm is more readily available from carp/koi due to their larger size, high sperm production (5 – 10 mL sperm from a 1 – 2 kg male) and the ease of hormonally inducing spermiation all-year round.

In the normal zebrafish progenies produced, 2% of larvae did not inflate their gas bladder at 5 dpf. These embryos were morphologically deformed, and a portion were haploid. This is likely due to spontaneous inactivation of either the paternal or the maternal pronucleus. Gomelsky and Recoubratsky (1990) reported similar observations in common carp and were able to make use of pigmentation inheritance to conclude the rate of spontaneous inactivation of the paternal and maternal pronuclei were 0.104% and

0.158%, respectively.

In this study, one zebrafish female repeatedly produced diploid embryos when inseminated with UV-irradiated carp sperm. This spontaneous diploidization of maternal chromosomes (SDM) occurred at a variable rate between progenies indicating variability in the underlying physiological mechanism, likely retention of the second polar body, as environmental conditions were constant. Females with oocytes predisposed to SDM have been reported in other species, such as common carp (Cherfas et al., 1995) and Nile tilapia Oreochromis niloticus (Ezaz et al., 2004). Using heterologous sperm was key in 78 the identification of this female. If UV-irradiated zebrafish sperm was used, the small number of surviving diploids would have been assumed to be the result of incompletely irradiated spermatozoa. The use of UV-irradiated carp sperm will increase the efficiency and effectiveness of haploid gynogenetic screens in zebrafish.

Sperm Number of Number of Ploidy of larvae

propagations larvae

analyzed

Carp (intact) 6 35 2n

Zebrafish (intact) 10 29 n, 2n

Carp (irradiated) 8 40 n

Table 5.1 Ploidy of fish produced by separate crosses with wild-type females (n = 6).

80 Sperm Number of Number of Ploidy of larvae

propagations larvae

analyzed

Carp (intact) 2 7 2n

Zebrafish (intact) 3 16 2n

Carp (irradiated) 3 29 n, 2n*

81 Dam Number of Total SDM (%)

propagations number of Average (SD) Range

2-cell

embryos

High frequency SDM 3 461 12.0 (19.5) 0.5-34.5

Other females 8 1730 0.02 (0.06) 0-0.2

Figure 5.1 Average percent of embryos surviving in propagations of wild-type females

(not high frequency SDM) from the first mitotic division (2-cells). Error bars represent standard deviation and only one direction is shown to avoid displaying overlapping bars.

Figure 5.2 Development of A. haploid zebrafish, B. zebrafish x carp hybrids, C. diploid zebrafish.

Figure 5.3 Variety of embryo phenotypes seen in zebrafish x carp hybrids and haploid zebrafish.

86 CHAPTER 6. EFFECTS OF HOMOZYGOSITY ON SEX DETERMINATION IN ZEBRAFISH

Abstract

Polygenic sex determination has been observed in several fish species, including zebrafish Danio rerio. To investigate this system, we utilized a broadly accepted model for sex determination studies in fish, induced gynogenesis (inheritance of only maternal chromosomes). Gynogenetic zebrafish were obtained by activating zebrafish oocytes with

UV irradiated common carp Cyprinus carpio sperm and then applying one of four different shocks (two meiotic and two mitotic shocks). Gynogens produced by three of the shocks survived to maturity. All adult gynogens (n = 52) except one were found to be male. There was no difference in growth rate between the biparental controls and gynogens produced through the most effective shock, thereby eliminating growth rate as the cause of the skewed sex ratio. Gynogen males had reduced fertility compared to biparental controls, with about half of gynogens being unable to reproduce through natural spawning (all controls reproduced successfully), and gynogen males that did reproduce gave lower fertilization rates compared to controls. This demonstrates the negative effects of increased homozygosity on male reproductive function. Families sired by meiotic gynogen males were more likely to be female-biased (33% of families) compared to families sired by biparental control males (11%). In addition to confirming the polygenic nature of sex determination in zebrafish, these observations suggest that

87 recessive and/or overdominant male-determining alleles are present in zebrafish populations.

88 Introduction

In gonochoristic species, development of the reproductive system begins with formation of a bipotential gonad. At a species-specific point in development, genetic and/or environmental factors induce the bipotential gonad to commit to either a female or a male fate. This process of commitment is referred to as sex determination. Further development of the organism into either a male or female is known as sex differentiation.

The factors that influence sex determination vary between taxonomic groups.

Most gonochoristic fish species studied have a genetic sex determination system.

However, there are notable examples of fish species where sex determination is controlled by temperature (Strüssmann et al., 1996) or a combination of temperature and pH (Roemer and Beisenherz, 1996). In species with genetic sex determination, there are many examples of male heterogamety (Chourrout and Quillet, 1982; Komen et al., 1991), and female heterogamety (Dabrowski et al., 2000; Glennon et al., 2012). There are also isolated observations of polygenic sex determining systems (Vandeputte et al., 2007), where it is inferred that multiple genes influence sex. In these species, sex may be thought of as a trait with continuous underlying genetic liability and a threshold liability the defines the border between male and female genotypes (Bulmer and Bull, 1982).

Liew et al. (2012) demonstrated that 1) sex ratios from natural crossings of zebrafish Danio rerio range from 5 – 97% male, 2) repeated crossings of the same sire and dam give families with highly correlated sex ratios, and 3) sex ratio responds to selection. Additionally, mapping studies have found several different loci associated with sex in laboratory strains (Anderson et al., 2012; Bradley et al., 2011). One study observed a major sex determining locus across several “less-domesticated” strains (Wilson et al., 89 2014), and this group suggested that wild zebrafish may have a system of female heterogamety. However, it is worth noting that variable sex ratios were still found in one of these less-domesticated strains and hybrids between this strain and wild zebrafish

(Brown et al., 2012). Together, this evidence suggests that sex in zebrafish is controlled by multiple genes, although environmental influences have been found, such as high water temperatures (35 – 37°C) during early development increasing the proportion of males in a population (Abozaid et al., 2012, 2011; Uchida et al., 2004).

Induced gynogenesis (inheritance of only maternal chromosomes) is frequently used to investigate the genetic basis of sex determination in fishes. This is accomplished by inseminating oocytes with genetically inactivated sperm and either preventing extrusion of the second polar body (meiotic gynogenesis) or the first cytokinesis (mitotic gynogenesis) by application of a physical shock. The sex ratio of the resulting gynogens can give insight into the sex determining mechanism. Gynogenesis also results in a high level of inbreeding, with meiotic gynogenesis approximately equivalent to a self-cross (F

= 0.5) and mitotic gynogenesis resulting in the production of fully homozygous individuals (F = 1) (Gomelsky, 2011).

Streisinger et al. (1981) reported obtaining both meiotic and mitotic zebrafish gynogens by applying a physical shock at different times (1.5 or 13 – 22 min, respectively) post-insemination. Mitotic gynogens of both sexes were obtained, and clonal lines were produced by gynogenetically reproducing female mitotic gynogens.

Some clonal lines contained fish of both sexes, including multiple lines that were predominately male. Bull (1983) interpreted this as being caused by environmental variation he referred to as “environmental noise”. This situation is similar to that 90 observed in a clonal Nile tilapia Oreochromis niloticus line (Sarder et al., 1999). In this completely homozygous XX line, most fish were female, but some males were also found. This was interpreted as being either the result of a recessive male-determining allele(s) with limited penetrance or developmental instability of the line’s genotype

(Sarder et al., 1999). All three of these explanations (environmental noise, limited penetrance, and developmental instability) are based on the idea that while the major determinants of sex may be genetic, there is always some level of environmental variation between individuals that could affect the phenotype, and susceptibility to this variation may be genetically determined.

Hörstgen-Schwark (1993) reported obtaining zebrafish mitotic gynogens (using the criteria of application of a late shock), and all adult gynogens were male. The author additionally performed gynogenesis using oocytes from females in the clonal line “C32”, originally produced by Streisinger et al. (1981), and only obtained male progeny even when rearing the fish under conditions matched to those described by Streisinger et al.

(1981). Pelegri and Schulte-Merker (1998) obtained meiotic gynogens, and reported they were predominantly male. However, there was variation between lines with some lines giving predominately female meiotic gynogens.

To summarize, several research groups have performed gynogenesis in zebrafish, but observations of the sex ratio in obtained gynogenetic progeny groups are inconsistent.

Additionally, all reports thus far utilize irradiated zebrafish sperm to activate oocytes for gynogenesis. This risks contamination of gynogenetic progeny groups with biparental offspring as a result of rare, incompletely inactivated spermatozoa. To eliminate this methodological issue and further investigate the sex determination system of zebrafish, 91 we produced gynogenetic progenies using irradiated common carp Cyprinus carpio sperm to activate embryonic development (Delomas and Dabrowski, 2016) and investigated the sex and fertility of obtained gynogens as well as the sex ratio in progenies produced by gynogen sires.

Materials and Methods

Broodstock care and gamete collection

Broodstock zebrafish were from an AB/TL hybrid line and were kept in a freshwater recirculating system maintained at 28 ± 1°C with a natural photoperiod,

11L:13D (light:dark) at the time of the study. Fish were fed dry feed (Otohime B2, Reed

Mariculture, Campbell, CA) supplemented with Artemia nauplii. A common carp x koi carp (ornamental form of common carp) hybrid male was kept under the same conditions and fed a dry diet (Aquamax 400, Purina Animal Nutrition, Gray Summit, MO).

Gamete collection and sperm irradiation was performed according to Delomas and Dabrowski (2016) and as described in chapter 5.

Gynogenesis

Oocytes from 3 – 6 females were pooled for each gynogen progeny group. Only oocytes from females that did not show signs of being “overripe” (highly fragmented oocytes, opaque ovarian fluid) were used. A small portion (50 – 200 oocytes) was fertilized with 10 μl zebrafish sperm solution as a biparental control. The remaining oocytes (1000 – 2600) were inseminated with 100 – 150 μl UV-irradiated carp sperm solution. After insemination and activation with water, oocytes were maintained at

28.5°C until application of the shock. A small sample (100 – 200 oocytes) was left 92 unshocked as a haploid control. All oocytes inseminated with UV-irradiated carp sperm

(shocked and unshocked groups) were kept in the dark until the 2-cell stage in order to prevent photoreactivation of the sperm pronucleus (Ijiri and Egami, 1980; Lebeda and

Flajshans, 2016; Recoubratsky et al., 2003). Both zebrafish x carp hybrids and zebrafish haploids do not survive past embryonic development, and so any viable larvae obtained are expected to be diploid zebrafish (Delomas and Dabrowski, 2016).

Two types of shock intended to produce meiotic gynogens were used. The first was an early pressure shock (EP, n = 3 progeny groups) consisting of applying a pressure shock of 6200 psi (42.7 mPa) from 1.5 until 6 minutes post-activation (mpa) (Gestl et al.,

1997). The second was an early heat shock (EH, n = 2) that consisted of incubating the oocytes at 41.4°C from 1.5 to 3.5 mpa, and then returning the oocytes back to 28.5°C.

Two types of shock intended to produce mitotic gynogens were tested. The first was a combined ether pressure shock (CEP, n = 3) that consisted of immersing the oocytes in 2% (v/v) ethyl ether solution at 17 mpa, applying a pressure shock of 8000 psi

(55.2 mPa) from 22.5 until 28 mpa, and transferring the oocytes back to freshwater (no ether) at 36 mpa (Streisinger et al., 1981). The second was a late heat shock (LH, n = 3) that consisted of incubating the oocytes at 41.4°C from 13 to 15 mpa, and then returning the oocytes back to 28.5°C (Heier et al., 2015; Streisinger et al., 1981).

Fish rearing

After application of the shock, rate of first cleavages (4 – 8 cells), and total number of oocytes were counted in all progeny groups. The resulting embryos were incubated as described in chapter 2. Total number of surviving embryos was counted at 1,

2, 3, and 5 days post-fertilization/activation (both are abbreviated as dpf for simplicity). 93 At 5 dpf, larvae with inflated gas bladders were counted and transferred to static water containers.

Fish were raised according to the protocol described in chapter 2. Fish stocking density varied between tanks depending on the number of larvae surviving in the progeny group, but was always below 17 fish L-1.

At 21 dpf, the lengths of 10 fish from each progeny group were measured (in gynogenetic progeny groups with less than 10 fish, all fish were measured), and survivors were transferred to a freshwater recirculating system with a 13L : 11D photoperiod. Fish were fed Artemia nauplii supplemented with dry feed (Otohime B2) for the remainder of the experiment. At 42 dpf, the lengths of 10 fish from each tank were measured (in gynogenetic progeny groups with less than 10 fish, all fish were measured).

Sex identification

Sex of fish was determined by observation of external morphology. Females have a rounded body shape, and males have stronger gold pigmentation of the anal fin (Parichy et al., 2009). Mating behavior of gynogens who were tested for fertility was observed to confirm their sex. Gynogens that were not tested for fertility were dissected and sex was confirmed by observation of gross gonad morphology.

Effect of shock on sex ratio

To examine the possibility that the shock applied to induce diploidization has an effect on sex ratio, we fertilized oocytes from two females separately with intact zebrafish sperm and applied an LH shock to a portion of the oocytes. Resulting embryos were anticipated to be a mixture of inviable tetraploids and viable diploids. Embryos and fish were raised according to the same method described in chapter 2 and sex ratio was 94 determined in the shocked progeny groups and non-shocked control groups. Ploidy was measured in a sample of adults by flow cytometry according to the protocol used by

Delomas and Dabrowski (2016), except a fin clip was used as the tissue sample.

Fertility testing and outcrossing of meiotic gynogens

Fertility tests were performed by natural spawning between an experimental fish and a test fish. A group of six full-sib females from an AB/TL hybrid line were used for fertility testing of males. Females were randomly paired with males. Five males from an

AB/TL hybrid line were used for fertility testing of females. Males from control, EP, and

LH treatments were tested. Males from the EH treatment were not tested due to the small number surviving to maturity. Mating and embryo incubation was performed as described in chapter 2. Fertilization rate (2 – 4 cells) and 1 dpf embryo viability (number alive at 1 dpf / number of fertilized oocytes) were recorded. Males that failed to fertilize any oocytes or failed to induce oviposition during four attempts, each separated by two days, were considered infertile by natural spawning. Offspring of EP gynogens and corresponding controls produced during the fertility testing were raised as described above until sexual maturity (60 dpf).

In vitro fertilization was attempted with sperm collected from two EP gynogen males and five LH gynogen males that failed to induce oviposition in order to determine if these males produced functional spermatozoa. Testes were dissected from these and seven control males (produced from the same oocyte groups as the seven gynogen males). One testis was macerated in extender and used to inseminate oocytes according to the protocol described above for the original broodstock. The remaining testis from each fish was preserved in 10% neutral buffered formalin for histological analysis. Tissue 95 processing and embedding in paraffin were performed according to standard methods

(Hewitson et al., 2010), and slides were stained with hematoxylin and eosin.

Statistical analysis

Survival, body length, fertilization rate, and 1 dpf embryo viability were analyzed using linear mixed models with shock as a fixed effect and oocyte group (the group of females used) as a random effect (random intercept). Significant differences among levels of the fixed effect were first assessed by ANOVA using the Kenward-Roger approximation for denominator degrees of freedom and then, if the effect was significant, multiple comparisons were performed on the least square means using Tukey’s method for p-value adjustment and the Kenward-Roger approximation for degrees of freedom.

For the analysis of percentage of survival, fertilization rate, and 1 dpf embryo viability, the respective proportions were Arcsin transformed prior to analysis. To determine if the sex ratio in families sired by gynogen/control fish was biased, Chi-square goodness-of-fit tests were used to compare the sex ratio to 1 : 1. To determine if the sex ratio of biparental families exposed to an LH shock was different from unshocked siblings, a Chi- square test of independence was used. To test the distributions of male-biased, unbiased, and female-biased families between control and gynogen sires, a Fisher’s exact test was used. For all hypothesis tests, differences were considered significant at p < 0.05. All analyses were performed using R statistical software ver. 3.3.2 (R Core Team, 2017) and the lme4 (Bates et al., 2015), lmerTest (Kuznetsova et al., 2017), pbkrtest (Halekoh and

Højsgaard, 2014), lsmeans (Lenth, 2016), and multComp packages. All values are given as mean ± SD.

Results

Survival rates of the shocked progeny groups were lower than the control at all time points (Table 6.1). The differences between the control and all four gynogenetic treatments were statistically significant (ANOVA, p < .05) at most time points during embryonic and early larval development (1, 2, and 5 dpf) and at 21 dpf. The majority of mortality occurred during the first 24 hours post-insemination. We also observed that the perivitelline space was smaller in activated oocytes subjected to the EH shock compared to oocytes subjected to the other shocks. Survival rates of gynogens produced through the different shocks was not significantly different during any one period of embryonic and early larval development (ANOVA, p > .05), but when cumulative survival from 4 – 8 cells to 5 dpf was analyzed, the survival rate of CEP groups was significantly lower than the other three shocks, and the survival rates of all shocked groups were significantly lower than the control (ANOVA, p < .05, Table 6.1). No fish in the CEP groups survived to 21 dpf. The LH group had significantly lower survival from 21 dpf to 42 dpf compared to the control and EH groups. Survival of the EH group was 100% from 21 to 42 dpf, but it must be noted that there were only six fish alive during this period in the two progeny groups.

At 21 dpf, mean length of fish in the LH groups (17.2 ± 2.2 mm) was significantly smaller than fish in the EP (20.8 ± 2.0 mm) and control (20.2 ± 2.6 mm) groups

(ANOVA, p < .05), but not fish in the EH group (p > .05, 19.2 ± 2.3 mm) (Figure 6.1).

By 42 dpf, fish in both the EH (25.7 ± 3.8 mm) and LH (25.0 ± 4.2 mm) groups were

97 significantly shorter than fish in the control (31.8 ± 3.9 mm) and EP (30.8 ± 3.0 mm) groups (ANOVA, p < .05).

Control groups were mixed sex, with mean ± SD of 57.7 ± 4.7% male. A total of

52 gynogens (29, 17, and 6 from the EP, LH, and EH groups, respectively) grew to a sexually mature size (greater than 25 mm) (Eaton and Farley, 1974). Only one (from an

EH group) was female.

The one female gynogen was naturally spawned with unrelated males a total of five times. The first three crosses yielded only unfertilized oocytes, but the fourth and fifth crosses had fertilization rates of 20% and 77%, and survival to 1 dpf of 54% and

83%, respectively.

All control males tested were able to reproduce by natural spawning, but only about half of the LH and EP males were capable (Table 6.2). Fertile EP males gave significantly lower fertilization rates compared to control males (ANOVA, p < .05).

While the fertilization rates given by fertile LH males was not significantly different from either the control or the EP males (ANOVA, p > .05) due to large individual variation and the small number of fertile LH males, there was a sizeable numeric difference in the mean fertilization rate between the LH and control males. There were no significant differences in survival of fertilized embryos to 1 dpf between treatments (ANOVA, p >

.05).

Seven and five males from the EP and LH groups, respectively, failed to induce oviposition after four attempts, each separated by two days. To determine if these males were capable of producing functional spermatozoa, in vitro fertilization was attempted using sperm collected from two EP gynogen, five LH gynogen, and seven control males. 98 Spermatozoa from both EP males, three LH males, and all seven control males were able to fertilize oocytes, with mean fertilization rates of 70 ± 4%, 50 ± 33%, and 77 ± 11%, respectively. Histological sections of the testes from the two sterile LH males showed few spermatozoa in the lumen of the spermatogenic tubules (Figure 6.2A). Sections from fertile gynogen and control males showed lumen containing densely packed spermatozoa

(Figure 6.2B, C).

In order to distinguish the effects on phenotypic sex of gynogenesis versus the shock used to induce diploidization, we fertilized oocytes with intact zebrafish sperm and applied an LH shock. Survival from fertilization (2 – 4 cells) to 5 dpf in the biparental LH groups was 3.9% and 8.2%, and survival of their unshocked siblings was 25.6% and

48.9%. Between the two biparental LH groups, a total of 14 fish reached maturity, 6 males and 8 females (2:3 and 4:5 in each group). Thirty-three control fish survived to maturity, and they exhibited a sex ratio of 19 males and 14 females (9:5 and 10:9 in each group). There was no significant difference in sex ratio between the treatments (Chi- square, p > 0.05). All fish from the biparental LH group and three fish from the control group were analyzed for ploidy and all were found to be diploid.

Sex ratio of families sired by EP males was variable, with an equal number of male-biased, female-biased, and unbiased families (Figure 6.3). Sex ratio of families sired by control males (produced from the same oocyte pools as the EP males) were mostly unbiased, with two male-biased families and one female-biased family. The notable difference between sire groups was the proportion of female-biased families:

33% (4 of 12) of families with EP sires were female-biased compared to 11% (1 of 9) of families with control sires. The difference in distribution between EP and control sires 99 was not statistically significant (Fisher’s exact test, p > 0.05). Gynogen males were observed to produce more strongly biased families, with a range of 11 – 100% male compared to a range of 34 – 68% male for the controls (Table 6.3).

Discussion

Consistent with previous reports of gynogenesis in zebrafish (Hörstgen-Schwark,

1993; Streisinger et al., 1981), we observed strongly male-biased progeny groups. The use of irradiated carp sperm to activate embryonic development allows us to exclude the possibility that the male gynogens were the result of contamination by incompletely irradiated zebrafish spermatozoa. The strong male bias is consistent with polygenic sex determination. A purely additive polygenic system (h2 = 1) predicts that a majority of gynogenetic fish would be female. This is because gynogenesis results in all-female inheritance, and the genetic liability of the parent female genotype must be above the threshold required to express the female phenotype. The average liability of the gynogenetic offspring would be equal to the parent, and therefore express the female phenotype. If environmental variance is present, but all genetic variance is additive, the proportion of the population that is female should increase after gynogenesis compared to the population that the parent females were selected from, although the majority of offspring may not be female if the parent population was male-skewed (Falconer and

Mackay, 1996). We observed a large decrease in the proportion of females after gynogenesis, which suggests that non-additive gene action is impacting the sex of gynogenetic offspring. A consequence of gynogenesis is inbreeding, which results in the

100 expression of recessive and overdominant traits in gynogenetic offspring homozygous for the corresponding alleles.

As such, we hypothesize the presence of recessive and/or overdominant male determining alleles in zebrafish populations. Because this phenomenon was observed in meiotic gynogens (EP and EH groups), a proportion of these alleles must be located close to centromeres, where chiasma interference will cause homozygosity (Streisinger et al.,

1986). Recessive male-determining mutations have been previously found in three fish species with female homogamety: common carp (Komen et al., 1992), Nile tilapia

(Karayucel et al., 2004; Sarder et al., 1999), and rainbow trout (Quillet et al., 2002).

Inbreeding depression is caused by recessive and overdominant deleterious alleles

(Falconer and Mackay, 1996). While the EP gynogen group did not exhibit growth depression, it did have lower mean fertility than control males. Artificial depletion of primordial germ cells (PGCs) and mutations that prevent early oocyte development have been observed to cause female-to-male sex reversal in zebrafish (Dranow et al., 2013;

Rodríguez-Marí et al., 2010; Siegfried and Nüsslein-Volhard, 2008). It is possible that the recessive and overdominant deleterious alleles that cause reduced fertility among gynogens did so partly be affecting PGC or early oocyte development and thereby induced male-development.

Previous reports of androgenesis (Corley-Smith et al., 1996; Hou et al., 2015) and gynogenesis (Hörstgen-Schwark, 1993; Streisinger et al., 1981) in zebrafish have also yielded all-male, or male-biased progeny groups. This is consistent with our hypothesis of recessive/overdominant male determining alleles that potentially act by interfering with early germ cell development. Pelegri and Schulte-Merker (1998) describe the 101 distribution of sex ratio in meiotic gynogenetic progeny groups produced by females of different genetic background. They demonstrated that of two strains they used for gynogenetic screens, one typically gave male-biased progeny groups while the other typically gave female-biased progeny groups. These results further demonstrate the genetic nature of the mechanism (as different genetic strains had different responses) that causes male development in gynogens.

Inbreeding in zebrafish independent of gynogenesis has also been shown to yield male-biased families (Lawrence et al., 2008; Shinya and Sakai, 2011), including in hybrids between WIK and a wild strain (Brown et al., 2012). It has also been shown that selecting for female-biased sex ratios combined with full-sib mating leads to the creation of a severely female-biased line, suggesting that these “male” alleles can be removed (or compensated for) by selection (Liew et al., 2012).

One prediction of our hypothesis is that outcrossing of gynogen males will result in a higher number of female-biased families than outcrossing of sibling biparental males.

This is because the recessive/overdominant alleles will no longer control sex, and effects of additive “female” alleles will be revealed. We tested this by outcrossing EP gynogen and control males to the same group of females and found that a higher percentage (33%) of gynogen sires produced female-biased families compared to control sires (11%)

(Figure 6.3). While our results in isolation are not statistically significant, they are consistent with other reports in the literature. Uchida et al. (2002) obtained strongly female-biased families (1.7% male) when crossing an unspecified number of gynogenetic males to biparental females. Abozaid et al. (2012, 2011) also reported crossing one mitotic gynogenetic male zebrafish to several females and repeatedly obtained female- 102 biased families (10 – 25% male). Together, these studies provide evidence for our hypothesis.

It has been shown that inbreeding in the absence of selection increases genetic variance on the population level (Wright, 1921). This is an alternative explanation for our observations of an increased number of female biased progenies with gynogen sires. This is supported by the larger range of sex ratio in families produced by gynogen sires (11 –

100% male) compared to families produced by control sires (34 – 68% male), but does not explain why the gynogens themselves were almost all male.

Applying a physical shock to oocytes soon after insemination (i.e. for induction of polyploidy or gynogenesis) is a common technique in fish genetics and has not been shown in any other fish species to cause sex reversal. Nonetheless, we examined the sex of biparental diploids that had been subjected to the LH shock to confirm that shocking the inseminated oocytes was not causing male development independent of gynogenesis.

We obtained multiple females from this biparental shocked group (8 of 14 fish) and there was no significant difference in sex ratio between the shocked and unshocked groups.

This demonstrates that the effects of gynogenesis on sex are independent of the shock, and therefore must be genetic in nature.

It has been suggested that increased growth rate can cause female development in zebrafish, and, likewise, that decreased growth rate can cause male development

(Lawrence et al., 2008). The sex ratio difference between gynogens and biparental controls in the current study cannot be due to differences in growth, as there was no significant difference in size between the control and the EP gynogens at 21 or 42 dpf.

While there were significant differences between the sizes of the control and LH and EH 103 gynogens (Figure 6.1), the LH and EH gynogens grew several times faster than biparental families described in other studies that developed into mixed sex groups. For example,

Brown et al. (2012) reported that outbred and inbred lines of zebrafish which reached 17

– 21 mm at 63 dpf were mixed sex.

Inbreeding induced male development in zebrafish has been previously suggested to be evolutionarily adaptive because inbreeding depression causes slower growth rates and small males may have increased reproductive potential compared to small females due to the positive relationship between clutch size, egg size, and body size (Brown et al.,

2012; Lawrence et al., 2008). However, body size has also been shown to be positively related to male reproductive success in zebrafish (Paull et al., 2010; Pyron, 2003; Skinner and Watt, 2007). An alternative hypothesis is that zebrafish may exhibit male-biased dispersal, and so producing more males under conditions of inbreeding would decrease the likelihood of further inbreeding in the following generation (chapter 8).

Whether presumptive mitotic gynogens in a given progeny group are completely homozygous or are spontaneous gynogens caused by non-disjunction of chromosomes during meiosis I or II has long been discussed (Komen et al., 1991; Purdom et al., 1985;

Quillet, 1994; Streisinger et al., 1981). Streisinger et al. (1981) observed inheritance of isozymes in presumptive mitotic zebrafish and found that of 476 fish, all were homozygous at the locus examined while only 14% of meiotic gynogens were homozygous. In koi (ornamental common carp), 20 presumptive mitotic gynogens were analyzed for six microsatellite loci with high recombination rates, and all were completely homozygous (Alsaqufi et al., 2014). In contrast, isozyme analysis in presumptive mitotic rainbow trout demonstrated a high frequency of gynogens 104 heterozygous at one or more loci, presumably a result of spontaneous gynogenesis

(Purdom et al., 1985) but possibly a result of incorporated DNA fragments from the irradiated sperm (Disney et al., 1988). Similarly, microsatellite analysis demonstrated that out of ten presumptive mitotic gynogen loach Misgurnus anguillicaudatus, one was heterozygous at two out of three diagnostic loci (Morishima et al., 2001). The frequency of spontaneous gynogenesis appears to be species specific, with low frequency in zebrafish (Streisinger et al., 1981), but this frequency has been shown to be genetically influenced (Delomas and Dabrowski, 2017b). While complete homozygosity was not confirmed in presumptive mitotic gynogens in the current study, the conclusions drawn would not appreciably change if these progeny groups were revealed to be a mixture of both spontaneous meiotic and mitotic gynogens.

105

Number Age (dpf)

of Shock Type Progeny 1 – 5 1 2 3 5 21 42

Groups

89.7 ± EP Meiotic 3 5.9 ± 1.5b 24.0 ± 17.5a 81.0 ± 6.7a 45.3 ± 24.4a 35.3 ± 15.8b 93.0 ± 6.1ab 12.5ab

106 90.0 ± b a a a c b

EH Meiotic 2 2.6 ± 2.1 8.3 ± 5.8 82.9 ± 8.5 40.6 ± 4.4 61.9 ± 6.7 100.0 ± 0.0 14.1ab

CEP Mitotic 3 0.2 ± 0.1a 1.6 ± 1.4a 71.8 ± 25.8a 78.9 ± 1.6a 19.6 ± 7.6a 0.0 ± 0.0a -

LH Mitotic 3 9.0 ± 4.3b 24.0 ± 5.0a 82.2 ± 5.1a 89.5 ± 7.8ab 50.4 ± 17.8a 34.9 ± 7.3b 66.4 ± 15.9a

None Control 11 68.7 ± 13.9c 74.4 ± 14.3b 98.4 ± 1.8b 98.3 ± 2.9b 95.5 ± 5.2b 86.5 ± 5.9d 96.4 ± 7.5b

Table 6.1 Arithmetic mean ± SD percent survival of gynogenetic and control progeny groups. Age ‘1 – 5’ is the percentage of

embryos (4 – 8 cells) surviving to 5 days post-fertilization (dpf). Survival at 1dpf is the percentage of embryos (4 – 8 cells) that

survived to 1 dpf. Survival at later ages is given as the percent surviving from the previous age listed. Different letters within the same

column indicate significant differences between groups.

Number Number Fertilization 1 dpf Weight Length Shock tested fertile rate (%) survival (%) (mg) (mm)

472.2 ± None 19 19 78.3 ± 24.0a 83.8 ± 13.2 38.1 ± 2.5 87.0

463.1 ± EP 20 12 49.5 ± 28.0b 82.2 ± 19.4 36.8 ± 1.3 56.0

62.1 ± 392.6 ± LH 17 7 94.7 ± 4.4 35.1 ± 1.6 33.4ab 61.3

Table 6.2 Fertility of EP and LH gynogen males during natural spawning trials in comparison to control biparental siblings. Values are given as arithmetic mean ± SD.

Different letters within the same column indicate significant differences between groups.

107

Sire Dam Number Number Male Sire Bias Treatment group male female (%) C1 2 20 38 34 Female C2 1 10 15 40 Unbiased C3 2 25 32 44 Unbiased C4 3 27 32 46 Unbiased Control C5 3 10 10 50 Unbiased C6 1 36 24 60 Unbiased C7 1 26 14 65 Unbiased C8 3 36 19 65 Male C9 2 41 19 68 Male EP1 2 7 54 11 Female EP2 1 12 48 20 Female EP3 3 9 32 22 Female EP4 2 14 44 24 Female EP5 1 7 6 54 Unbiased EP6 2 33 26 56 Unbiased EP EP7 3 36 22 62 Unbiased EP8 2 39 20 66 Unbiased EP9 1 9 3 75 Male EP10 3 42 14 75 Male EP11 1 26 0 100 Male EP12 2 25 0 100 Male Table 6.3 Sex ratio of families sired by EP gynogen and control males. Dam group refers to the groups of pooled oocytes used to produce the sires. Bias was assessed by performing a Chi-square goodness of fit test against a 1 : 1 ratio.

108

Figure 6.1 Arithmetic mean ± SD length of gynogenetic and control fish. Different letters at the same time point indicate significant differences between groups. Dpf, days post- fertilization

109

110

Figure 6.3 Frequency of male-biased, unbiased, and female-biased families sired by EP gynogens and biparental control males. Family size varied due to fertility of individual males with mean of 46 and range of 12 – 61 fish/family. Families were categorized as biased or unbiased by testing the sex ratio against the theoretical 1 : 1 ratio with a Chi- square goodness-of-fit test (p<0.05).

111

CHAPTER 7. POLYGENIC SEX DETERMINATION IN PEARL DANIO

Abstract

In the majority of gonochoristic vertebrates studied, sex is determined by inheritance of a single gene, with either the dominant allele being male-determining

(male heterogamety) or female determining (female heterogamety). The sex determination system in zebrafish has been proposed to be polygenic, although some authors suggest that wild zebrafish could have female heterogamety. Polygenic sex determination has been proposed to be evolutionarily unstable in the presence of sexually antagonistic alleles, but there is no information about the sex determining systems of other Danio species. We begin to fill this gap by investigating the sex determining system in pearl danio Danio albolineatus. We performed a full-factorial mating and induced meiotic gynogenesis. Sex ratio varied among 24 biparental families from 5 to 100% male with a mean of 52 ± 25% male. Six breeding pairs were crossed twice and there was no significant difference between sex ratio between the first and second repetitions.

Heritability of sex was estimated by Bayesian techniques and the posterior distribution had mean, median, and 95% credibility interval of 0.89, 0.87, and 0.44 – 1.40, respectively. Meiotic gynogenesis produced families (n = 6) with highly male-biased sex ratios, 91 ± 8% male. Together, these results demonstrate that the sex determination system in pearl danio is polygenic. Additionally, the results of meiotic gynogenesis

112 suggest that increased homozygosity (inbreeding) induces male development. We compare these results with investigations of zebrafish sex determination and suggest that these qualities may be evolutionarily conserved.

113

Introduction

The process by which a gonochoristic organism makes the binary choice to differentiate into a male or female is known as sex determination. Sex determination mechanisms vary between species, and include examples of both environmentally and genetically controlled processes (Bachtrog et al., 2014; Beukeboom and Perrin, 2014;

Bull, 1983). In most species studied with genetic sex determination, inheritance of a single gene determines the sex of an individual. An alternative system that has been observed in several taxonomic groups is polygenic sex determination (see Beukeboom and Perrin, 2014; Moore and Roberts, 2013). In some species, such as the platyfish

(Kallman, 1973) and African pygmy mouse (Veyrunes et al., 2010) sex is controlled by a few alleles with clear dominance relationships. In other species, such as the apple snail

(Yusa, 2007), a larger number of loci contribute to sex determination, and in these species sex may be considered as a threshold trait with a continuous underlying genetic liability (Bulmer and Bull, 1982).

Theoretical consideration has demonstrated that in the absence of other influences, a continuous path of evolutionary stable solutions connects sex determination being controlled by a single gene, one major gene and multiple minor genes, and numerous genes each having a small effect (Bull, 1983). On the other hand, if sexually antagonistic alleles are present in a population, theory predicts that polygenic sex determination is unstable (in the absence of a strong selective pressure favoring it) and a single sex determining locus will evolve with sexually antagonistic alleles tightly linked to it (Rice, 1986). Sexually antagonistic alleles have been demonstrated in multiple

114 species (Cox and Calsbeek, 2009), most notably in Drosophila (Gibson et al., 2002;

Innocenti and Morrow, 2010; Rice, 1992), and are thought to be widespread if not universal. Given the observation of polygenic sex determination in multiple species, this implies that either there is a strong selective pressure for polygenic sex determination, or it is an intermediate state that has by chance been observed in multiple species. Further investigation of taxonomic groups containing polygenic sex determination may allow us to discriminate between these two hypotheses.

Sex determination in zebrafish Danio rerio has been investigated through multiple approaches. Classical genetics techniques showed that, using both laboratory strains (AB and TU strains, that were generated by attempting to remove recessive lethal alleles from the population) and a strain obtained from a local pet shop, 1) sex ratios in families ranged from 5 to 97% male with a median of 51% 2) individual breeding pairs repeatedly produced families with highly correlated sex ratios, and 3) sex ratio responded to selection, resulting in the creation of both female-biased and male-biased strains (Liew et al., 2012). Additionally, the existence of recessive or overdominant male-determining alleles in some populations has been shown, that can cause inbred families to be male- biased (chapter 6). Mapping experiments using hybrids between the AB strain and other strains (TU, WIK, and Nadia) have identified four quantitative trait loci (QTL) associated with sex (Anderson et al., 2012; Bradley et al., 2011; Howe et al., 2013). One mapping study comparing several different strains of zebrafish found one major QTL on chromosome four in three strains maintained in captivity for various lengths of time: eight generations (Nadia); since 1997, at the latest, (WIK); and for “many years” prior to

115

2004 (Ekkwil) (Wilson et al., 2014). This QTL was also present in one population obtained directly from the wild (Cooch Behar), but was not present in the two laboratory strains (AB and TU) previously bred to eliminate recessive lethal alleles. Chromosome four has many hallmarks of a sex chromosome, including late replication, low concentration of protein coding genes, and high GC content (Wilson et al., 2014). This evidence led Wilson et al. (2014) to suggest that zebrafish have a WZ/ZZ sex determining system, but the process of eliminating recessive alleles in the AB and TU strains also eliminated the sex determining locus on chromosome 4. However, sex ratios produced by WIK and hybrids between WIK and fish recently derived from the wild have been shown to be variable under identical rearing conditions (Brown et al., 2012), implying that even if a major QTL for sex exists in zebrafish (except AB and TU strains), the sex determination mechanism is likely still polygenic.

The existence of a major QTL raises three other possible hypotheses: that zebrafish are 1) evolving female heterogamety, 2) evolved a polygenic system from a previous system of female heterogamety, or 3) are evolutionarily stable with a polygenic system containing a major QTL for sex. Testing these hypotheses will require first examining the sex determination systems of species closely related to zebrafish.

Studies of phylogenetic relationships within the Danio genus suggest that the genus can be broadly separated into three groups: the large danios, the D. choprae species group, and the D. rerio species group (Fang et al., 2009; Mayden et al., 2007; McCluskey and Postlethwait, 2015; Tang et al., 2010). A phylogenetic study utilizing RAD-Seq to obtain large numbers of loci for comparison demonstrated that the pearl danio D.

116 albolineatus is located within the D. rerio species group, but it can be considered as a separate species subgroup – the D. albolineatus species subgroup (McCluskey and

Postlethwait, 2015). Within the D. rerio species group, the D. albolineatus species subgroup was found to be most distantly related to D. rerio. This inferred phylogenetic relationship matches well with the geographical distributions of D. rerio and D. albolineatus, which are separated by the Arakan mountain range (McCluskey and

Postlethwait, 2015). To investigate the evolutionary history of sex determination in the D. rerio species subgroup, we examined the sex determination system of pearl danio D. albolineatus.

Methods

Fish and fish husbandry

Ten adult pearl danios (four males and six females) were purchased from a local pet shop in Columbus, OH. Broodstock husbandry was performed as described for zebrafish broodstock in chapter 2. At the time of these experiments, weight and length of broodstock females were 1463 ± 164 mg and 54.3 ±3.1 mm and males were 968 ± 171 mg and 49.5 ± 1.3 mm. All biparental crosses were performed through natural spawning following the same procedure as described for zebrafish in chapter 2.

When spawning behavior ceased (2 – 4 hours), embryos were collected from the breeding tank, fertilization rate was counted in a sample of 150 – 300 oocytes, and these samples were incubated as described for zebrafish embryos in chapter 2 for five days. At

5 days post-fertilization (dpf), up to 100 larvae with inflated gas bladders were stocked (if

117 less than 100 larvae were alive, all larvae were stocked) and raised according to the method described for zebrafish in chapter 2. Larvae were initially stocked at a density of

16.7 fish/L. At 21 dpf, the number of surviving juveniles were counted and up to 60 fish were randomly sampled and stocked at a density of 1.6 fish/L for continued rearing as described in chapter 2.

Mating design

A full factorial mating of six females and four males was performed. Six crosses were repeated (3 – 4 weeks apart), with each female being repeated once, two males being repeated once, and two males being repeated twice. All six females were reproduced gynogenetically.

Sex identification

At 63 dpf (68 dpf for gynogens), fish were euthanized, dissected, and sex was identified by observation of gonad morphology. Ten fish from each family were weighed and measured. Ten to fifteen fish from each of six families were kept alive after 63 dpf and sex was identified at 90 dpf based on external morphology (females having rounded abdomens and males having a streamlined shape). These identifications were confirmed by spawning five pairs from each family for unrelated experiments.

Gynogenesis

Oocytes were collected for gynogenesis by allowing a female and male pearl danio to begin spawning naturally and interrupting spawning after approximately 150 –

300 oocytes had been deposited. These naturally fertilized oocytes were used as a biparental control for oocyte quality. The females were then anesthetized in MS-222 (100

118 mg/L), and oocytes collected by manual stripping. Sperm from a common carp Cyprinus carpio male was collected, diluted 1 : 9 in a glucose – Tris extender, irradiated at 3000

J/m2 (Alsaqufi et al., 2014) in a UV crosslinker (Stratagene, La Jolla, CA) while being stirred on ice, and used to activate the collected oocytes according to the protocol described by Delomas and Dabrowski (2016) and in chapter 5 with the exception that the carp male was not injected with carp pituitary gland (the male was spermiating sufficiently without injection). Prior to the main experiment, pearl danio oocytes were fertilized with intact carp sperm to yield hybrids between pearl danio females and carp males. As was found for zebrafish x common carp hybrids (Delomas and Dabrowski,

2016), the resulting hybrid embryos were diploid (confirmed by flow cytometry), deformed, and did not survive past embryonic development (data not shown).

To induce retention of the second polar body, and thereby obtain diploid meiotic gynogens, an early heat shock was applied after activation and brief incubation of the oocytes at 28.5°C. Multiple different shocking protocols were attempted: 41.0°C from 2 to 4 minutes post-activation (mpa), and 41.0, 38.0, 38.5, and 38.8°C from 4 to 6 mpa.

After shocking, the oocytes were transferred to a water bath at 28.5°C and incubated in the dark until past the 2-cell stage (1 hour) to prevent photoreactivation of the sperm pronucleus (Ijiri and Egami, 1980; Lebeda and Flajshans, 2016; Recoubratsky et al.,

2003). A small sample of oocytes (100 – 200) was left unshocked as a haploid control.

Gynogens were then raised to maturity according to the protocol described above for biparental populations.

119

Statistical analysis

All biometric values are given as mean ± SD. In order to investigate the relationship between genetic variation and sex, sex was modeled as a threshold trait with a normally distributed underlying liability using a single-trait model with random dam and sire effects. This model was first fit with a frequentist approach in ASReml 4.1 which utilizes penalized quasi-likelihood (PQL) to approximate likelihood (Gilmour et al.,

2015). The same model was then fit using a Bayesian approach in R statistical software

(R Core Team, 2017) using the MCMCglmm package (Hadfield, 2010). Chi-square distributions with one degree of freedom were used as the priors for the variance components. The Markov Chain Monte Carlo was run for 5,000,000 iterations with a burn-in of 10,000 and a thinning interval of 500. Heritability was calculated as combined dam and sire heritability (Becker, 1984).

Results

Full-factorial mating

All pairs spawned successfully through natural spawning. Mean survival rates from fertilization (2 – 4 cells) to 5 dpf and from 5 dpf to 63 dpf were 86 ± 17% and 86 ±

10%, respectively. Fish at 63 dpf were sexual dimorphic in size, with mean male length and weight of 42.4 ± 3.4 mm, 702 ± 159 mg and female length and weight of 47.4 ± 4.0 mm, 1206 ± 319 mg. At this size and age, dissected gonads were clearly distinguishable as either testes or ovaries containing vitellogenic oocytes. Sex ratio in full-sib families ranged from 5 to 100% male with a mean of 52 ± 25% male (Figure 7.1). In the six

120 crosses that were performed twice, there were no significant differences between sex ratios in replicates (Table 7.1).

Both frequentist and Bayesian techniques showed that sex had high heritability.

Estimation of the model by PQL gave a heritability of 0.79 ± 0.33. Estimation of the model by MCMC gave a posterior distribution for heritability with mean, median, and

95% credibility interval of 0.89, 0.87, and 0.44 – 1.40, respectively.

Gynogenesis

Shocks for inducing retention of the second polar body had variable efficacy.

Oocytes shocked at 41°C (n = 4 progeny groups) did not yield surviving embryos at 1 dpf, while the biparental controls from the same batch of oocytes did (Table 7.2).

Oocytes shocked at 38.0°C (n = 3 progeny groups) developed into viable diploid larvae

(inflated gas bladder at 5 dpf) at a rate of 1.0 ± 0.1% of embryos (4 – 8 cells). Oocytes shocked at 38.5 – 38.8°C (n = 5 progeny groups) developed into viable diploid larvae at a rate of 20.2 ± 10.2% of embryos (4 – 8 cells). Survival of gynogenetic fish from 5 – 68 dpf was 65.3 ± 4.8% and 56.8 ± 27.1% for the 38.0°C and 38.5 – 38.8°C shocked groups, respectively.

Gynogens were observed to grow slower than biparental progeny, and so they were dissected at a later age, 68 dpf, to identify their sex. Mean gynogen male and female length and weight at 68 dpf were 39.1 ± 3.4 mm, 560 ± 115 mg and 42.1 ± 2.9 mm, 738 ±

162 mg, respectively. The sex ratio of gynogenetic families was 91 ± 8% male, demonstrating strong male bias. There was no apparent relationship between the

121 estimated breeding values (EBVs) for the dams and the sex ratios of their gynogenetic offspring (Table 7.3).

Discussion

Sex ratio varied from 5 to 100% male (Figure 7.1), with no significant differences between repeated spawning of the same breeding pairs (Table 7.1). Additionally, heritability of sex was high, with the mean of the posterior distribution being 0.89 and a

95% credibility interval of 0.44 – 1.40. Heritability estimated by frequentist techniques was slightly lower (0.79 ± 0.33). This disparity is likely due to the approximation of likelihood in the frequentist approach by PQL, that has been shown to underestimate heritability (de Villemereuil et al., 2013). Together, these observations demonstrate polygenic sex determination in pearl danio.

The heritability of sex (modelled as a threshold trait) was previously estimated in

European sex bass Dicentrarchus labrax, that has a mixed system of both environmental and genetic sex determination, to be 0.62 (Vandeputte et al., 2007). Heritability of sex was also estimated in Ouachita map turtles Graptemys ouachitensis and common snapping turtles Chelydra serpentina to be 0.82 (Bull et al., 1982) and 0.56 (Janzen,

1992), respectively, under environmental temperatures that give mixed sex ratio (both species exhibit temperature-sensitive sex determination). This shows that, consistent with our heritability estimate of 0.89, much of the genetic variation in sex found in species lacking a single sex determining gene can be due to additive genetic effects.

122

The sex ratios of gynogenetic pearl danio families were strongly male biased in the current study (Table 7.3). This parallels observations made in zebrafish, in which multiple groups have shown that gynogenetic families are frequently male-biased

(chapter 6; Hörstgen-Schwark, 1993; Pelegri and Schulte-Merker, 1998; Streisinger et al.,

1981). In zebrafish, inbreeding without gynogenesis has been shown to produce male biased progenies (Brown et al., 2012; Lawrence et al., 2008; Shinya and Sakai, 2011), and it was previously proposed that the slower growth of both inbred fish and gynogens caused male development. In chapter 6, we demonstrated that strong male-bias among zebrafish gynogens was independent of growth rate, and proposed that this phenomenon was instead the result of recessive or overdominant male-determining alleles. Observing that gynogenetic families are male-biased in pearl danio suggests that recessive/overdominant male determining alleles may be conserved between these species.

It was recently suggested that wild and domesticated zebrafish strains that have not been purged of recessive lethal alleles have a system of female heterogamety, while the AB and TU strains had developed a polygenic sex determination system because the natural sex determining allele was linked to a recessive lethal allele (Wilson et al., 2014).

Given the rarity of polygenic sex determination, the most parsimonious explanation for our observation of polygenic sex determination in pearl danio and other reports of the same in zebrafish (Anderson et al., 2012; Bradley et al., 2011; Liew et al., 2012) is that polygenic sex determination is evolutionarily conserved between these species.

Additionally, it has been observed that zebrafish families from the WIK strain and

123 hybrids between the WIK strain and a strain recently derived from the wild have variable sex ratios (Brown et al., 2012). Both of these observations imply that the QTL identified on chromosome 4 in a selection of zebrafish strains by Wilson et al. (2014) is a major

QTL for sex that functions in a polygenic sex determination system. Indeed, the observation of two unusual qualities of sex determination systems, 1) polygenic sex determination, and 2) the relationship between homozygosity/inbreeding and male- development, in both zebrafish and pearl danio suggests that major parts of the sex determination system may be conserved between these two species. Given the rapid rate of evolution sometimes observed in sex determination systems (Bachtrog et al., 2014;

Bull, 1983), one may suggest that this implies existence of a selective pressure for both these qualities in zebrafish and pearl danio. A selective pressure would also reconcile the contradictions between polygenic sex determination and sexually antagonistic alleles

(Rice, 1986), but it is unclear what this pressure could be.

Alternatively, the similarities between chromosome 4 and sex chromosomes in other species (Wilson et al., 2014) suggests that zebrafish could be evolving toward (or away from) a system of female heterogamety. If this is also true of pearl danio, it would be consistent with polygenic sex determination being a transitory state. Future study of sex determination in Danio presents a unique opportunity to test evolutionary theories about sex determination and the development of sex chromosomes.

124

First spawn Second spawn

Survival (%) Sex Survival (%) Sex Dam x segregation segregation sire 5 dpf 5 – 63 dpf (male : 5 dpf 5 – 63 dpf (male :

female) female)

4 x 3 92.0 71.1 41 : 3 96.7 74.7 49 : 5

3 x 3 86.5 97.0 54 : 5 88.9 91.8 54 : 4

1 x 2 93.0 96.0 15 : 45 94.8 97.4 18 : 41

6 x 4 96.0 100.0 38 : 21 95.4 88.0 36 : 24

5 x 1 53.6 82.7 30 : 27 48.9 87.2 18 : 23

2 x 1 97.2 77.1 7 : 45 96.1 96.4 9 : 43

125

Biparental control survival Gynogen survival (%) Shock (°C, (%) Dam mpa) 1 – 5 5 – 21 1 – 5 5 – 21 1 dpf 1 dpf dpf dpf dpf dpf

1 41.0, 4 – 6 0 - - 91.9 94 98

6 41.0, 2 – 4 0 - - 96 100 100

2 41.0, 2 – 4 0 - - 100 96.6 88

4 41.0, 2 – 4 0 - - 91.6* 99* 93.8*

3 38.0, 4 – 6 52.5 3.3 69.2 99.1 90.4 90

6 38.0, 4 – 6 43.1 0.9 66.7 96.8 100 86.7

5 38.0, 4 – 6 16.4 5.7 80 67.5 83.2 88

4 38.5, 4 – 6 52.2 50.2 48.5 91.6* 99* 93.8*

1 38.8, 4 – 6 52.2 59.6 89.5 92.2 96.6 97

6 38.8, 4 – 6 46.3 19.8 82.1 96 88 100

2 38.8, 4 – 6 30.2 47.6 77.5 91.3 92.1 99

126

Dam Dam EBV Males Females Males (%)

1 -0.606 82 7 92.1

2 -0.514 21 4 84.0

3 0.001 8 1 88.9

4 0.298 16 1 94.1

5 0.315 3 0 100.0*

6 0.504 17 4 81.0

127

Figure 7.1 Distribution of sex ratio in 24 pearl danio families.

128

CHAPTER 8. WHY DOES INBREEDING CAUSE MALE BIASED SEX RATIOS?

Abstract

Inbreeding has been shown to result in male-biased sex ratios in zebrafish Danio rerio, an increasingly important model species in the biological sciences, and the closely related pearl danio D. albolineatus. It was previously suggested that this strategy may be evolutionarily advantageous because of the positive relationship between body size and female reproductive performance. This argument assumes that body size is more important for female fitness than male fitness. We discuss studies that show that body size is a key determinant of male reproductive success, calling into question this key assumption. We then present an alternative hypothesis based upon previous theoretical studies: male-biased sex ratios improve parental fitness under conditions of inbreeding through male-specific dispersal and investment in mate searching.

129

Introduction

The sex determination systems of zebrafish Danio rerio and pearl danio D. albolineatus are thought to be polygenic (chapter 7; Liew et al., 2012). There is evidence that some “less-domesticated” strains of zebrafish have a major QTL for sex, which has led to the proposal that wild zebrafish have a system of female heterogamety, but this system was lost in the two most common laboratory strains (Wilson et al., 2014).

However, it has been observed that one of these “less-domesticated” strains and hybrids between this strain and wild zebrafish still produce variable sex ratios (Brown et al.,

2012), suggesting that while a major QTL exists, sex determination is still polygenic.

Inbreeding has been shown to result in male-biased sex ratios in zebrafish, independent of changes in growth rate. It was shown that outcrossing increased the proportion of female offspring (Lawrence et al., 2008) and inbreeding caused sex ratios to become male-biased (Brown et al., 2012). Induced gynogenesis (inheritance of only maternal chromosomes), which results in a level of inbreeding approximately equivalent to a self-cross (F = 0.5) or fully homozygous offspring (F = 1) depending on the method

(Gomelsky, 2011), also yields strongly male-biased progeny groups in zebrafish (chapter

6; Hörstgen-Schwark, 1993; Streisinger et al., 1981). Induced gynogenesis was recently shown to also yield strongly male-biased progeny groups in pearl danio (chapter 7), suggesting that inbreeding also induces male-development in this closely related species.

Discussion

It was previously suggested that the relationship between inbreeding and male- biased sex ratios is evolutionarily advantageous because inbreeding depression causes

130 slower growth rates and small males have higher fitness than small females due to the positive relationship between clutch size and body size (Lawrence et al., 2008) and the generally higher metabolic demand of oocyte production (Brown et al., 2012). However, this argument assumes that body size is less important for male fitness compared to that of females. Zebrafish have been shown to develop dominance hierarchies, with body size being positively related to dominance. When allowed to spawn in groups (2 males and 2 females), larger, dominant males were shown to have significantly higher reproductive success compared to smaller, subordinate males (Paull et al., 2010). This was shown not to be the case when males of different dominance ranks were removed and spawned separately (1 male and 1 female) (Spence and Smith, 2006). The difference in the effect of dominance on reproductive success in the presence and absence of direct male-male competition suggests that dominant males become more successful by physically excluding the subordinate male. This is supported by observed aggression of dominant males being primarily directed at subordinate males and having markedly increased attack frequency during the spawning period (Paull et al., 2010). Additionally, it has been demonstrated that when simultaneously presented with two males of different sizes, females prefer the larger male (Pyron, 2003). Females also allocate more oocytes to larger males, even in the absence of direct male-male competition (Skinner and Watt,

2007). Similar results can be found throughout the animal kingdom where, in the absence of monogamy, dominant males sire more offspring than subordinate males, and dominance is frequently positively related to body size (Ellis, 1995). This evidence clearly demonstrates that in zebrafish body size is of great importance to male

131 reproductive success, thereby calling into question the previous hypothesis explaining the relationship between inbreeding and male-biased sex ratios.

It has been observed that the sex with a higher potential reproductive rate

(typically the male) invests more in mate searching, and this was previously explained by the asymmetry in benefit from finding mates caused by the difference in reproductive rate

(Bateman, 1948; Glutton-Brock and Vincent, 1991; Trivers, 1971). Theoretical work has shown that it is not the higher reproductive rate itself (Hammerstein and Parker, 1987), but rather the existence of “mating-windows” during which one sex is able to mate, costs of searching extending beyond the time spent searching, or asymmetries in the cost of searching itself (Fromhage et al., 2016). While this has not been widely studied in fish, it has been observed in the guppy Poecilia reticulata that males are more likely than females to leave a given shoal (Croft et al., 2003b). Because of the larger metabolic cost of oocyte growth compared to sperm production and the periodicity of ovulation events, zebrafish females have restricted “mating windows” during which they are able to oviposit. This implies that male zebrafish are expected to invest more in mate searching

(Fromhage et al., 2016).

Differential investment in mate searching is closely related to the wider phenomenon of sex-specific dispersal, where one sex of a given species is observed to disperse further than the other. Relating sex-specific dispersal to mate-searching assumes that mate-searching is the overriding impetus for dispersal; however, there are many drivers of dispersal, and so this assumption may not hold in all cases. Theoretical work has shown that sex-specific dispersal can be selected for by a high cost of inbreeding

132

(Motro, 1991), sex-disparities in local mate competition (Perrin and Mazalov, 2000), and sex-disparities in dispersal costs (Gros et al., 2008). Sex-specific dispersal has been observed in a variety of mammals and birds, and based on these data, Greenwood (1980) suggested that mating systems involving male defense of resources (typically monogamous) lead to female-biased dispersal and systems involving male defense of mates (typically polygynous) lead to male-biased dispersal. Recent work has found that the broader social system of a given species also plays an important role (Lawson

Handley and Perrin, 2007). In fish (excluding salmonids, where the situation is more complex (Consuegra and García de Leániz, 2007; Kitanishi et al., 2017), possibly due to the existence of both resident and migratory males in anadromous species), a similar pattern may also be found. In polygamous species with local mate competition among males, male-biased dispersal is typically observed (Cano et al., 2008; Croft et al., 2003a;

Knight et al., 1999; Stiver et al., 2007), and female-biased dispersal is typically observed in species with longer inter-spawning intervals in males (Okuda, 1999) and species in which males guard valuable territories (and may be monogamous) (Schradin and

Lamprecht, 2000; Taylor et al., 2003; van Dongen et al., 2014). There are also multiple reports of species with no sex-specific dispersal. These include four nonguarder open substratum spawners with unknown levels of mate competition (Aparicio and Sostoa,

1999; Buonaccorsi et al., 2001; Gilliam and Fraser, 2001; Hoarau et al., 2004) and the white bass Morone chryops (Hayden et al., 2011) that is polygamous but shows no male- male aggression during spawning (Salek et al., 2001).

133

Mrakovcic and Haley (1979) demonstrate inbreeding depression in zebrafish, and the behavioral studies discussed above suggest that male zebrafish defend mates.

Therefore, we can reasonably predict male-specific dispersal and mate searching in zebrafish. Zebrafish males have been observed by some (Spence et al., 2007; Spence and

Smith, 2005) but not others (Pyron, 2003) to be territorial of spawning sites. If zebrafish males do continuously defend optimum spawning territories, female-specific dispersal may be expected. However, observations have shown that territoriality is limited to the spawning period, and outside of spawning, the fish shoal together (Spence and Smith,

2005). Additionally, observations of territoriality over spawning sites have been in laboratory or semi-wild conditions were spawning sites are either experimentally limited

(one per tank) (Spence and Smith, 2005), or highly varied in quality of substrate (Spence et al., 2007). It is unclear to what extent these limitations are present in the wild.

With the expectation for male-specific dispersal and mate searching in zebrafish, production of male-biased sex ratios under conditions of inbreeding may be selected for.

Under localized inbreeding, this strategy would result in increased numbers of male offspring that would be more likely to disperse, find unrelated mates, and thereby decrease the level of inbreeding depression (and increase fitness) in their offspring. This is a specific instance of the broader hypothesis that under conditions of sex-specific dispersal, optimum sex allocation in unfavorable environments will be biased towards the more-dispersing sex and vice versa. This broader hypothesis was shown to be theoretically consistent and was proposed to explain biased sex ratios in phytophagous arthropods, territorial passerines (Julliard, 2000), and as a possible explanation for

134 environmental sex determination (Reinhold, 1998). It has been suggested that unfavorable environments in general (restricted food availability, high stocking density) can cause male-biased sex ratios in zebrafish (Lawrence et al., 2008; Ribas et al., 2017a).

The broader hypothesis could explain these relationships as well, but the link between unfavorable environmental conditions and male-biased sex ratios may be specific to only some families (Delomas and Dabrowski, 2017a; Liew et al., 2012; Ribas et al., 2017b).

Empirical investigation of sex-specific mate searching and dispersal in zebrafish will allow testing of this idea. Additionally, further investigation of the Danio genus could provide insight into the evolution of complex sex allocation strategies in vertebrates.

135

CHAPTER 9. ASYMMETRIC VIABILITY IN RECIPROCAL CROSSES OF ZEBRAFISH AND PEARL DANIO

Abstract

Interspecies hybrids have long been studied to further our understanding of speciation. Frequently, reciprocal crosses will have asymmetric viability – a phenomenon termed “Darwin’s corollary to Haldane’s rule”. It has been proposed that this asymmetry is caused by Dobzhansky-Muller incompatibilities between nuclear genes and cytoplasmic factors (e.g. maternal transcripts, mitochondrial genome), but the molecular basis of this hypothesis has received little empirical investigation, presumably due to the lack of an appropriate model system. We report a case of extreme asymmetry in viability between reciprocal hybrids of zebrafish and pearl danio. Hybrids from zebrafish dams x pearl danio sires (n = 4 crosses) were viable, with 83.2 ± 9.6% surviving from fertilization to 5 days post-fertilization (dpf), 80.1 ± 14.4% surviving from 5 to 21 dpf, and were morphologically similar to zebrafish of the same age. Hybrids from pearl danio dams x zebrafish sires (n = 6) were inviable after embryonic development. These hybrids developed pericardial edema at 1 dpf, only 37.2 ± 18.0% survived from fertilization to 5 dpf, and of the 595 larvae alive at 5 dpf, only one juvenile with stunted growth survived to 21 dpf. We propose that, given the resources available for the zebrafish model system and the strong asymmetry in viability between reciprocal crosses, that these hybrids will allow investigation of the molecular basis for Darwin’s corollary to Haldane’s rule. 136

Introduction

Interspecies hybrids have long been studied as a method of investigating speciation (Darwin, 1859). It has been noted that often one reciprocal cross will have lower viability than the other cross. This phenomenon has been observed across diverse taxonomic groups, including plants (Tiffin et al., 2001), amphibians (Arntzen et al., 2009;

Brandvain et al., 2014), and fish (Bolnick and Near, 2005) and has been termed

“Darwin’s corollary to Haldane’s rule” or simply “Darwin’s corollary” (Turelli and

Moyle, 2007). Frequently, the asymmetry in viability is small, but statistically significant

(Bolnick et al., 2008), although there are cases where one cross is viable and the other is inviable past embryonic development. For example, hybrids between masu salmon

Oncorrhynchus masuo and rainbow trout O. mykiss are viable if the dam is a rainbow trout, but die during embryonic development if the dam is a masu salmon (Fujiwara et al.,

1997). Similarly, hybrids between olive flounder Paralichthys olivaceus and either summer flounder P. dentatus (Xu et al., 2009) or spotted halibut Verasper variegatus

(Kim et al., 1996) are viable when the olive flounder is the dam, but inviable past embryonic development when the olive flounder is the sire.

Inviability of hybrids can be generally explained by the existence of Dobzhansky–

Muller incompatibilities (DMIs), which are alleles from divergently evolved species that produce non-functional epistatic interactions (Coyne and Orr, 2004). Asymmetric viability is also commonly explained by this theory, as the asymmetry could be the result of DMIs between nuclear genes and cytoplasmic factors (e.g. maternal transcripts, mitochondrial genome). If one species has developed more DMIs in cytoplasmic factors

137 compared to the other species, then the reciprocal crosses will have asymmetric viability

(Tiffin et al., 2001; Turelli and Moyle, 2007).

The molecular basis of DMIs and their effects on hybrid incompatibility have been investigated empirically (Coyne et al., 1998; Moehring et al., 2006; Moyle and

Graham, 2005; Moyle and Nakazato, 2010; Presgraves, 2003). However, the molecular basis of asymmetric hybrid viability has received little attention in the scientific literature, presumably due to the absence of an appropriate model system. Most observations of asymmetric viability are in non-model organisms and are often small

(though statistically significant) differences in survival rates. We described here a case of strongly asymmetric viability in hybrids between zebrafish and pearl danio. The large difference in viability between crosses and the established resources available for zebrafish research will make this an ideal model system for investigation of asymmetric hybrid viability.

Materials and Methods

Experimental overview

Hybridization was first attempted by natural spawning of six pearl danio females with six zebrafish males and four zebrafish females with four pearl danio males. Crosses in this manuscript will be written as dam x sire. All pearl danio x zebrafish crosses and one zebrafish x pearl danio cross were unsuccessful, and so in vitro fertilization was performed with these seven crosses. Survival was monitored in all progeny groups during

138 embryonic development and through the larval and early juvenile stages. Ploidy was measured in 10 embryos of each reciprocal hybrid at 5 days post-fertilization (dpf).

Fish and broodstock husbandry

Six pearl danio females and four pearl danio males were obtained from a local pet shop in Columbus, OH. The six zebrafish males and four zebrafish females used in this experiment were of an AB/TL hybrid line maintained in our laboratory. Broodstock care was performed as described in chapter 2.

Natural crosses

Natural crosses were performed as described in chapter 2. Crosses that were not successful were attempted a total of three times with two days in between each attempt.

In vitro fertilization

Gamete collection from females of both species and zebrafish males was performed as described in chapter 4. Sperm was collected from pearl danio males by manual stripping with a capillary tube held over the gonopore. Collected pearl danio sperm was then diluted in 50 μL E-400 extender (Matthews and Carmichael, 2015).

Fertilization was performed according to the dry method, by first adding 10 μL of sperm solution to the collected oocytes and then adding 1000 μL of aquarium water. After one minute, an additional 3 mL of aquarium water was added and the embryos were incubated at 28.5°C.

Embryo incubation and larval rearing

Fertilization rate was counted (2 – 4 cells) in a sample of oocytes (200 – 300) from each cross. These samples were then incubated as described in chapter 2. The

139 number of surviving embryos was counted at 1, 2, 3, and 5 days post-fertilization (dpf).

At 5 dpf, all larvae from female pearl danio x male zebrafish crosses were stocked and

100 larvae (stocking density of 16.7 fish/L) from zebrafish x pearl danio crosses were stocked for further rearing. Larvae were stocked into static water containers and raised until 21 dpf according to the protocol described in chapter 2, except that hybrids were fed marine rotifers Brachionus plicatilis until 12 dpf, after which they were transitioned to

Artemia nauplii. At 21 dpf, surviving fish were counted and 10 fish from each cross with survivors were randomly sampled and measured.

Flow cytometry

Ploidy was measured by flow cytometry as described by Delomas and Dabrowski

(2016) and in chapter 4. Briefly, embryos were incubated overnight in a propidium iodide staining solution with rainbow trout erythrocytes as an internal standard, then filtered through a 60 μm mesh and analyzed with an Accuri C6 flow cytometer. Sample c-value was calculated using the known c-value of rainbow trout (Ohno et al., 1969) and the ratio of fluorescence intensity between the hybrid peak and the rainbow trout peak. Ploidy was calculated from c-value by comparing against the theoretical c-value of a diploid zebrafish x pearl danio hybrid (calculated as the mean c-value of these two species).

Zebrafish and pearl danio c-values used were the mean c-values measured in five embryos of each species using the same protocol.

140

Results

Of the six pearl danio x zebrafish crosses attempted, none spawned naturally despite three attempts. Of the four zebrafish x pearl danio crosses attempted, three spawned naturally on the first attempt and one failed to spawn during three attempts. All in vitro fertilizations produced embryos.

The first morphological difference between the two reciprocal hybrids became apparent at 1 dpf (Figure 9.1). Pearl danio x zebrafish hybrids developed a mild pericardial edema at 1 dpf whereas zebrafish x pearl danio hybrids appeared grossly similar to zebrafish embryos of the same age (Kimmel et al., 1995). While there was little visible variation in the severity of pericardial edema at 1 dpf, variation in severity increased at later ages. At 5 dpf, the severity of pericardial edema ranged from mild to extensive (Figure 9.1). Additionally, some pearl danio x zebrafish embryos were observed to have curved bodies.

All 10 embryos of both reciprocal crosses had nuclear DNA content corresponding to the expected value for a diploid hybrid between pearl danio and zebrafish.

Survival rates of the two reciprocal hybrids diverged after 1 dpf (Figure 9.2). At five dpf, 37.2 ± 18.0% of pearl danio x zebrafish hybrids were alive, but only 0.8 ± 1.5% of them had inflated gas bladders. The others were either swimming but uninflated (16.6

± 3.0%) or not swimming and uninflated (82.6 ± 3.7%). In contrast, 83.2 ± 9.6% of the zebrafish x pearl danio hybrids were alive at 5 dpf, and 95.3 ± 6.9%, 0%, and 4.7 ± 6.9%

141 of them had inflated gas bladders, were swimming but uninflated, and were not swimming and uninflated, respectively.

Of the 595 pearl danio x zebrafish hybrids alive at 5 dpf, only one (0.2%) appeared to consume rotifers and survived until 21 dpf. At 21 dpf this fish was 8.3 mm in length. The reciprocal hybrids had a mean survival rate from 5 to 21 dpf of 80.1 ± 14.4% and a mean length at 21 dpf of 17.6 ± 3.9 mm.

Discussion

Our observations clearly demonstrate a strong asymmetry in morphological development and viability between reciprocal crosses of zebrafish and pearl danio. The major morphological defect observed in pearl danio x zebrafish was pericardial edema, and this became evident as early as 1 dpf. Pearl danio x zebrafish hybrids were essentially inviable, with only one out of approximately 1700 fertilized oocytes surviving to 21 dpf, and this surviving juvenile was drastically smaller than juveniles from the reciprocal cross. Such extreme asymmetries in viability have been reported in a few other fish hybrids (Fujiwara et al., 1997; Kim et al., 1996; Xu et al., 2009). Fujiwara et al. (1997) found that inviable masu salmon female x rainbow trout male hybrids exhibited large- scale loss of rainbow trout chromosomes, causing aneuploidy and morphological deformity. The viable reciprocal hybrid (rainbow trout female x masu salmon male) did not show any signs of chromosome loss. The ploidy of pearl danio x zebrafish hybrids in the current experiment, as measured by flow cytometry, was always diploid,

142 demonstrating that large-scale chromosome loss cannot explain the inviability of these hybrids.

There are previous reports of hybridizing zebrafish and pearl danio (Parichy and

Johnson, 2001; Schmidt, 1930; Wong et al., 2011), although only the viable cross

(zebrafish female x pearl danio male) has been reported. Given the well-developed scientific resources available for the zebrafish model system, and the strong asymmetry in viability between reciprocal crosses, hybrids between zebrafish and pearl danio could prove to be a valuable tool for investigating the molecular basis of asymmetric hybrid viability, or “Darwin’s corollary”.

143

Figure 9.1 Embryonic development of hybrids between zebrafish and pearl danio. A.

Pearl danio female x zebrafish male B. Zebrafish female x pearl danio male. Arrow indicates pericardial edema at 1 day post-fertilization (dpf).

144

145

CHAPTER 10. MONOGENIC CONTROL OF UNIFORMITY OF GROWTH IN ZEBRAFISH X PEARL DANIO HYBRIDS

Abstract

Uniformity of growth is an economically important trait for aquaculture species as it reduces or eliminates labor intensive practices (grading and multiple harvests) and can reduce aggressive behavior between size classes of fish. We describe two phenotypes of zebrafish Danio rerio females based on the coefficients of variation (CV) in length and weight of hybrid offspring produced by crossing these females to pearl danio D. albolineatus males. “Uniform” females gave offspring with mean ± SD CV in weight and length of 18 ± 5% (range: 9 – 27%) and 6 ± 1% (range: 4 – 8%), respectively, at 21 days post-fertilization (dpf), while “variable” females gave offspring with mean CV in weight and length of 58 ± 6% (range: 38 – 79%) and 21 ± 4% (range: 13 – 30%), respectively.

We examined the inheritance of this trait and found that females homozygous for the recessive allele gave hybrid offspring with high uniformity of growth, while females with the dominant allele gave hybrid offspring with highly variable growth. CVs of oocyte diameter and larval length at 5 dpf were not significantly different between the two types of female. The CV of yolk volume was significantly different between the two types of female, but the uniform females had a higher mean CV than the variable females (13.05 ±

1.22% and 10.08 ± 0.44%), and thus does not explain the different phenotypes. These data demonstrate that the difference in uniformity of growth began after the onset of 146 exogenous feeding. Hybrid offspring of uniform females also exhibited significantly lower rates of visible morphological deformity at 5 dpf (11.7 ± 11.7% and 17.0 ± 13.8%, uniform and variable) and higher rates of survival from 5 to 21 dpf (94.6 ± 6.9% and 78.0

± 15.3%), but not significantly different survival during embryonic development (89.0 ±

13.1% and 91.6 ± 7.8%).

147

Introduction

Uniformity of growth is an economically important trait in aquaculture species, as it allows producers to reduce or eliminate labor intensive management practices such as grading and multiple harvests. Additionally, uniformity of growth can reduce aggressive behavior between different size classes of fish (Kestemont et al., 2003; Moran, 2007), thereby reducing stress and increasing growth and health. Variation in growth within a common environment can be due to both differences in genotype and differences in the microenvironment between fish. The susceptibility of fish to microenvironment variation

(i.e. the propensity to yield a different phenotype under mildly different environmental conditions) may be partly genetic in nature, and, if so, could be improved through selective breeding.

Previous studies have addressed this possibility by estimating additive genetic variation for uniformity of growth. This has been reported in terrestrial livestock (Ibáñez-

Escriche et al., 2008; Mulder et al., 2009) as well as in multiple fish species, including

Nile tilapia Oreochromis niloticus (Khaw et al., 2016), rainbow trout Oncorhynchus mykiss (Janhunen et al., 2012), and Atlantic salmon Salmo salar (Sae-Lim et al., 2017;

Sonesson et al., 2013). These studies reported low heritability for this trait, suggesting that selective breeding may be able to improve uniformity of growth.

Hybrid finfish are responsible for an important portion of current aquaculture production, as select crosses have been found that exhibit heterosis or combine beneficial traits of both parental species. For example, hybrids between channel catfish Ictalurus punctatus and blue catfish I. furcatus have superior feed conversion efficiency and

148 growth rate (Li et al., 2014), tolerance to dissolved oxygen (Dunham et al., 1983), and fillet yield (Argue et al., 2003) compared to channel catfish. While some hybrids can exhibit superior production traits compared to parental species, genetic improvement of hybrids is challenging in comparison to pure species. This is because directly evaluating breeding candidates requires the production and growout of hybrid offspring (reciprocal recurrent selection) and because heterosis is the result of complex epistatic, dominant, and overdominant interactions that are currently difficult to predict (Jiang et al., 2017; Li et al., 2008).

Hybrids between zebrafish and pearl danio have been previously described

(Parichy and Johnson, 2001; Schmidt, 1930), but little information is available on their growth. We performed several crosses of zebrafish females with pearl danio males and found a female zebrafish that produced hybrid offspring with drastically more uniform growth than other zebrafish females in our laboratory. The aim of the current study was to examine the inheritance of this trait.

Materials and methods

Experimental overview

We observed one zebrafish female in our laboratory that, when mated to three different pearl danio males, yielded offspring with more uniform growth compared to hybrids from other females in our laboratory (Figure 10.1). To investigate the inheritance of this trait, we first crossed this “uniform” female (P-1, Figure 10.2) and another

“variable” female (P-2) to a zebrafish male (P-3). Females from the resulting families

149

(Families 1 and 2, Table 10.1) were mated with the same three pearl danio males to determine if their offspring had uniform or variable growth rates. Five females from families 1 and 2 were mated twice with two different pearl danio males to ensure they gave the same phenotype when being tested with different males. We then crossed a zebrafish male from family 2 with a uniform and a variable female from family 1 (Figure

10.2). The females in the resulting families (Families 3 and 4) were mated with pearl danio males to determine if their hybrid offspring had uniform or variable growth rates.

Additionally, the oocytes and hybrid larvae of three uniform and three variable females from family 1 were measured to determine if the difference in variability was present prior to exogenous feeding.

Broodstock and breeding

Zebrafish broodstock were from an AB/TL hybrid line maintained in our laboratory. Three pearl danio males were obtained from a local pet store in Columbus,

OH and were used for all hybrid crosses in this experiment. Broodstock were maintained as described in chapter 2. Crosses between zebrafish were performed by natural spawning as described in chapter 2.

Hybridizations with females in the P generation were also performed through natural spawning as described in chapter 2. Hybridizations with females in the F1 and F2 generations were performed through in vitro fertilization in order to avoid any selection bias due to differences in females’ propensity to spawn with pearl danio males. Gamete collection, fertilization, and incubation of embryos was performed as described in chapter

150

9. Approximately one hour after fertilization, fertilization rate was measured in a sample of 100 – 300 oocytes.

Larval and juvenile husbandry

At 5 dpf, the number of surviving larvae and the number of deformed (non- inflated gas bladder, curved tail, short body, or edema) larvae were counted and a maximum of 100 non-deformed larvae (if fewer than 100 non-deformed larvae were surviving, all larvae were stocked) per cross were stocked in static water tanks at a density of 16.7 fish/L. Fish were reared based on the protocol described in chapter 2, except that hybrids were fed marine rotifers Brachionus plicatilis until 12 dpf, after which they were transitioned to Artemia nauplii. At 21 dpf, the number of surviving juveniles were counted, and a random sample was taken to measure individual weight and length. In zebrafish families and hybrid families produced by P females, 10 fish were measured. In hybrid families produced by F1 and F2 females, 15 fish were measured to increase the accuracy of CV estimates.

Oocyte and larval measurements

Three uniform and three variable females from family 1 were crossed with the three pearl danio males (one female of each phenotype with each male). Approximately

1.25 hours after fertilization, when embryos were at the eight-cell stage, photographs of embryos were taken under a dissecting microscope, and oocyte diameter, yolk length, and yolk height were measured in 30 embryos from each female using the Fiji distribution of

151

ImageJ (Schindelin et al., 2012). Yolk volume was calculated using the formula for a prolate spheroid. At five dpf, photographs of larvae were taken under a dissecting microscope, and larval length was measured in 20 larvae from each female using Fiji.

Statistical analysis

All means in this manuscript are given as mean ± SD. Coefficient of variation was

푆퐷 calculated as 퐶푉 = × 100%. To determine if weight and length of hybrids from 푀푒푎푛 uniform and variable females were unimodal or multimodal, weights and lengths of individual fish were centered based on tank means and the combined distributions of centered values were examined graphically.

Statistical comparisons for weight, length, embryonic survival, embryonic deformities, and juvenile survival were only made on offspring of females from families

1, 3, and 4 as these were the only families with females of both phenotypes present. The significance of the difference in these parameters between hybrid offspring of uniform and variable dams was assessed using mixed models with the variable of interest as the response variable, dam phenotype, dam family, and sire as fixed effects, and dam as a random effect (random intercept). Dam family, sire, and dam were included in the model to account for the non-independence of observations within each of these groups and family and sire were fitted as fixed effects due to the small number of levels in each

(three sires and three families). Linear models were fit for weight and length and the effect of dam phenotype was tested using a t-test with the Kenward-Roger method for calculating denominator degrees of freedom. It must be noted that in these two models,

152 the assumption of homogeneity of variance is violated due to the difference in variance between uniform and variable dams that is inherent in the data. Logistic models were fit for embryonic survival, embryonic deformities, and juvenile survival and the effect of dam phenotype was tested using a likelihood ratio test comparing the model with the dam phenotype effect to the model without it.

Differences in mean oocyte diameter, yolk volume, and the CV of these parameters between uniform and variable females were assessed using t-tests with

Welch’s correction for unequal variances. Differences in mean larval length and the CV of larval length were assessed using paired t-tests, with uniform and variable female mated to the same sire forming pairs. Variances of zebrafish weight and length at 21 dpf in families 1-4 were compared with Levene’s test. Ratios of uniform : variable were tested against theoretical Mendelian ratios using binomial tests. All hypothesis tests were two-sided and used a type I error rate of 0.05. All data analysis and statistical tests were performed using R statistical software ver. 3.4.3 (R Core Team, 2017). Mixed models were fit and assessed using the lme4 (Bates et al., 2015), lmerTest (Kuznetsova et al.,

2017), and pbkrtest (Halekoh and Højsgaard, 2014) packages.

Results

The original uniform female (P-1) repeatedly gave hybrid offspring with uniform growth, while the other females in the P generation gave hybrid offspring with highly variable growth (Figure 10.3).

153

Hybrid offspring from females in generations F1 and F2 fell into two distinct categories: uniform and variable (Figure 10.4). At 21 dpf, the hybrid families from uniform females had mean CV in weight and length of 18 ± 5% (range: 9 – 27%) and 6 ±

1% (range: 4 – 8%), respectively, while the hybrid families from variable females had mean CV in weight and length of 58 ± 6% (range: 38 – 79%) and 21 ± 4% (range: 13 –

30%), respectively. The CV of length in hybrids produced by uniform females in F1 and

F2 was lower than in hybrids produced by female P-1, likely due to the different sampling sizes and therefore increased precision of CV measurements for generations F1 and F2.

The five females from F1 that were crossed separately to two different pearl danio males gave hybrid families of a consistent type (one female gave two uniform hybrid families and four females each gave two variable hybrid families).

Of the 14 females tested from family 1, seven gave uniform and seven gave variable hybrid offspring, a 1:1 ratio (Table 10.1). All six of the females tested from family 2 gave variable hybrid offspring, a 0:1 ratio and significantly different from a 1:1 ratio (binomial test, p < 0.05). Of the nine females tested from family 3, six gave uniform and three gave variable hybrid offspring, which is not significantly different from a 1:1 ratio (binomial test, p > 0.10). Of the 13 females tested from family 4, two gave uniform and 11 gave variable hybrid offspring, which is not significantly different from a 1:3 ratio

(binomial test, p > 0.10) and is significantly different from a 1:1 ratio (binomial test, p <

0.05).

Hybrid survival during embryonic development, larval deformity rate, larval/juvenile survival, and size (weight and length) at 21 dpf were compared between

154 uniform and variable females from families 1, 3, and 4, as these were the families that contained females of both types. The mean survival of hybrid embryos from fertilization to 5 dpf was 89.0 ± 13.1% and 91.6 ± 7.8% for offspring of uniform and variable females, respectively. The relationship between female type (uniform vs variable) and survival to 5 dpf was not statistically significant (likelihood ratio-test, p > 0.10). The mean deformity rate (number deformed/total alive) of hybrid larvae at 5 dpf was 11.7 ±

11.7% and 17.0 ± 13.8% for offspring of uniform and variable females, respectively. The relationship between female type and deformity was statistically significant (likelihood ratio-test, p < 0.05). The mean survival of hybrids from 5 dpf to 21 dpf was 94.6 ± 6.9% and 78.0 ± 15.3% for offspring of uniform and variable females, respectively. The relationship between female type and survival from 5 dpf to 21 dpf was statistically significant (likelihood ratio-test, p < 0.001). Mean weight and length of hybrid offspring of uniform females (55.6 ± 13.4 mg and 19.4 ± 1.4 mm) were higher than those of variable females (42.1 ± 27.5 mg and 16.9 ± 3.8 mm) from families 1, 3, and 4, and this difference was statistically significant for length (t-test, p < 0.01) but not weight (t-test, p

= 0.057). Weight and length of hybrid offspring from both types of females were unimodal after centering based on tank means.

Mean oocyte diameter, yolk volume, and larval length did not significantly differ between the uniform and variable females tested (Table 10.2, t-test, p > 0.05). The CV of oocyte diameter and larval length also did not significantly differ between these females

(t-test, p > 0.10). The CV of yolk volume was significantly different (t-test, p < 0.05);

155 however, the uniform females gave oocytes with a larger mean CV (13.05 ± 1.22%) than did the variable females (10.08 ± 0.44%).

In the four zebrafish families, there was no apparent relationship between female type and the uniformity of growth of zebrafish offspring. The CV of weight in the four zebrafish families at 21 dpf were 14.7%, 21.8%, 11.5% and 8.7% for families 1, 2, 3, and

4, respectively. The CVs of length at 21 dpf were 5.3%, 8.8%, 4.3% and 3.0% for families 1, 2, 3, and 4, respectively. Variance both weight and length in all four of these families were not significantly different from each other (Levene’s test, p > 0.10).

Discussion

Inherited traits can be roughly separated into two categories: monogenic traits controlled by one gene with two alleles and complex traits that are controlled by multiple genes and their interactions (both genetic and environmental). Monogenic traits result in individuals that fall into distinct categories because they are inherited as discrete alleles with three different possible genotypes. Complex traits result in individuals with a continuum of phenotypes, or, in the case of categorical traits, individuals with a continuum of liability for a given phenotype. This is because, while the genes responsible are still inherited discreetly, there are a large enough number of genes, each with a relatively small effect, such that phenotypes appear continuous. Zebrafish females in the current study were able to be clearly categorized as uniform and variable based on the

CV of length or weight of their hybrid offspring (Figure 10.4), suggesting that this trait is monogenic. There was variation in CV within each group, demonstrating that other

156 factors (environmental and other genes) also play minor roles in determining uniformity of growth.

The inheritance of the uniform phenotype is consistent with it being caused by a recessive allele. Female P-1 would be homozygous recessive (v/v, for variable) and male

P-3 would be heterozygous (V/v), leading to ratio of 1:1 of homozygous (v/v) uniform females (including female F1-1) and heterozygous (V/v) variable females (including female F1-2) in family 1. Female P-2 had the variable phenotype, and crossing her to male P-3 yielded only variable females. The number observed (0 uniform, 6 variable) is significantly different from 1 : 1 (binomial test, p < 0.05), which suggests that female P-2 was homozygous dominant (V/V). This leads to both homozygous dominant (V/V) and heterozygous (V/v) fish in family 2. Crossing of male F1-3 from family 2 with a uniform female (F1-1) and a variable female (F1-2) from family 1 gave females with both phenotypes in families 3 and 4. This is consistent with male F1-1 being heterozygous

(V/v).

Given the presence of uniform females in family 4, the inheritance pattern could only be explained by a dominant allele that causes uniform growth if male F1-3 was heterozygous for that dominant allele. This would result in a 1:1 ratio of uniform : variable among females in family 4, but the observed count (2:11) was significantly different from a ratio of 1:1. A dominant allele causing uniformity of growth would also require that the male P-3 be heterozygous for the dominant allele (as male F1-3 would have to inherit it from P-3), and thus the females in family 2 would also display a ratio of

1:1, uniform : variable, but the observed count (0:6) was significantly different from 1:1.

157

Together, these observations allow us to eliminate the possibility of a dominant allele causing increased uniformity.

We observed no difference in CV of larval length between uniform and variable females (Table 10.2), suggesting that the difference in uniformity of growth of hybrid offspring is caused by differences in growth after exogenous feeding, and not by differences in embryonic growth. Surprisingly, the CV of yolk volume was significantly higher in oocytes from uniform females than those from variable females. However, this difference was numerically small (13.05 ± 1.22% and 10.08 ± 0.44%) which suggests that while it is statistically significant, it is not biologically significant.

The mechanism by which this gene acts to yield hybrid offspring with highly variable or highly uniform growth after exogenous feeding is unclear. It is plausible that the gene responsible acts through maternal effects, but to test this hypothesis (without knowing the mechanism) requires performing the reciprocal cross (pearl danio female x zebrafish male) and raising the reciprocal hybrid offspring. Unfortunately, the reciprocal hybrid is inviable past embryonic development (chapter 9). One possible mechanism is that the dominant allele of this gene results in a maternally deposited product that causes a developmental defect of varying severity in the resulting hybrid offspring. The varying severity of this defect could then cause varying rates of growth after the onset of exogenous feeding. Our observations that variable females gave hybrid offspring with a significantly higher probability of being visibly deformed at 5 dpf and lower probability of survival from 5 to 21 dpf are consistent with this hypothesis, but do not indicate what the nature of this defect may be.

158

Previous studies examining genetic influences on uniformity of growth have treated it as a complex trait and focused on measuring additive genetic variation for use in genetic improvement programs. It has been shown that there is additive genetic variation for this trait in multiple fish species, and selective breeding would likely lead to more uniform growing strains (Janhunen et al., 2012; Khaw et al., 2016; Sae-Lim et al.,

2017; Sonesson et al., 2013). However, no studies have addressed genetic causes of uniformity of growth in interspecies hybrids, nor have they reported large monogenic influences on uniformity of growth.

Identification of the specific gene and alleles responsible for this phenomenon and analysis of the mechanism by which they lead to increased or decreased uniformity in zebrafish x pearl danio hybrids could give insight into the genetic basis of uniformity of growth. Additionally, identification of the causal gene could allow researchers to use a candidate gene approach to identify alleles controlling uniformity of growth in production aquaculture species and hybrids.

159

Theoretical ratios Family Dam Number of Number Number Generation Dam Sire (Uniform : number phenotype females tested uniform variable Variable)

1 F1 P-1 P-3 Uniform 14 7 7 1:1

2 F1 P-2 P-3 Variable 6 0 6 0:1, 1:1*

3 F2 F1-1 F1-3 Uniform 9 6 3 1:1

4 F2 F1-2 F1-3 Variable 13 2 11 1:3, 1:1*

160

Table 10.1 Segregation of uniform and variable phenotypes in families from generations F1 and F2. See Figure 10.2 for the pedigree of

dams and sires. *significantly different with p < 0.05, binomial test

Mean CV

Oocyte Yolk Larval Oocyte Yolk Larval Female diameter volume length diameter volume length Phenotype (mm) (mm3) (mm) (%) (%) (%)

0.116 ± 3.80 ± 13.05 ± 3.53 ± Uniform 1.09 ± 0.01 2.82 ± 0.50 0.006 0.06 1.22* 0.11

0.113 ± 3.76 ± 10.08 ± 3.24 ± Variable 1.12 ± 0.02 2.99 ± 0.37 0.004 0.06 0.44* 0.57

161

Figure 10.1 Example of variation in size of hybrids at 21 days post fertilization from a variable dam (left) and a uniform dam (right)

162

Figure 10.2 Pedigree of broodstock and phenotypes of offspring. Dams and sires of zebrafish offspring are labeled with an identification code corresponding to Table 10.1.

163

Figure 10.3 Coefficient of variation (CV) of length of hybrid offspring from five females in the P generation. Ten fish from each family were measured at 21 days post-fertilization to determine CV.

164

Figure 10.4 Coefficient of variation (CV) of weight and length at 21 days post- fertilization in hybrid offspring produced by females of four different families from generations F1 (families 1 and 2) and F2 (families 3 and 4). Vertical lines mark cutoff points (8% and 13%) observed for CV of length between families from uniform females and variable females. Cutoff points for CV of weight (not shown on graph) were 27% and

37%.

165

CHAPTER 11. CONCLUSION

This series of studies was aimed at improving the zebrafish model system with a specific focus on its use in ichthyology, fisheries, and aquaculture research. We first addressed rearing methods and developed a rearing protocol that resulted in faster growth, shorter generation times, and higher fertility compared to current methods

(chapter 2). We then demonstrated the capabilities of this protocol by using it to rear zebrafish near the lower thermal limit for embryonic development with no significant decrease in survival (chapter 3).

Utilizing this improved rearing method, we were then able to investigate the sex determination system of zebrafish. We demonstrated that triploidy induces male development in zebrafish by acting downstream of estradiol to prevent oocyte development. We showed that triploid males were able to induce oviposition by diploid females and that triploid males produced small amounts of aneuploid spermatozoa that were capable of fertilizing oocytes but led to inviable aneuploid embryos (chapter 4).

Then, we improved methods for gynogenesis in zebrafish by demonstrating that UV- irradiated common carp spermatozoa induced haploid embryonic development in zebrafish oocytes. We also showed that common carp x zebrafish hybrids were inviable after embryonic development and displayed a characteristic set of morphological deformities. This will allow any contaminating hybrid embryos to be distinguished from

166 gynogenetic embryos after inseminating zebrafish oocytes with UV-irradiated common carp sperm (chapter 5). We then combined the rearing method described in chapter 2 with the improvements made in chapter 5 and obtained gynogenetic zebrafish. We found that

51 out of 52 gynogens were male, which is consistent with a polygenic sex determination system where inbreeding induces male development. We also demonstrated that crossing these gynogenetic males to unrelated females resulted in more female biased families than crossing biparental sibling males to the same group of females. This result suggests that inbreeding induces male development through recessive and/or overdominant alleles

(chapter 6).

In order to investigate the evolution of sex determining systems in the Danio genus, we assessed the sex determining system of pearl danio. We first performed a full- factorial mating and demonstrated that: 1) sex ratio varied from 5 to 100% male between families with a mean ± SD of 52 ± 25% male, 2) repeated crossings of the same breeding pairs gave sex ratios that were not significantly different between repetitions, 3) sex had a high heritability (mean of posterior distribution 0.89, 95% credibility interval of 0.44 –

1.40). Together, these observations demonstrate that pearl danio has a polygenic sex determination system, which suggests that polygenic sex determination may be conserved between zebrafish and pearl danio. We then performed gynogenesis with the same group of pearl danio females and demonstrated that gynogenetic families were strongly male- biased, suggesting that inbreeding also causes male development in pearl danio (chapter

7). After observing the relationship between inbreeding and male development in both zebrafish and pearl danio, we proposed a hypothesis for why this may be an

167 evolutionarily adaptive trait: male-biased sex ratios improve parental fitness under conditions of inbreeding through male-specific dispersal and investment in mate searching (chapter 8).

As interspecies hybridization is an important method in both fish speciation research and aquaculture genetics, we investigated the potential of zebrafish x pearl danio hybrids to serve as a model for interspecies hybridization. We show that there is severe asymmetry in viability between reciprocal crosses (Darwin’s corollary) of zebrafish and pearl danio. Zebrafish female x pearl danio male hybrids were viable, while pearl danio female x zebrafish male hybrids were inviable after embryonic development. This makes zebrafish x pearl danio hybrids an excellent model for studies into the mechanisms responsible for Darwin’s corollary (chapter 9).

Finally, we describe genetic influences on uniformity of growth in zebrafish female x pearl danio male hybrids. We discovered a recessive allele in zebrafish females that dramatically increased the uniformity of growth in hybrid offspring from females homozygous for this allele. Future identification of the gene responsible for this effect could lead to the identification of similar alleles in production aquaculture species

(chapter 10).

This series of studies has addressed obstacles to the adoption of the zebrafish model system for ichthyological, fisheries, and aquaculture research. We have also demonstrated the use of the model system to investigate key issues in evolutionary biology and aquaculture genetics.

168

REFERENCES

Abi-Ayad, A., Kestemont, P., 1994. Comparison of the nutritional status of goldfish

(Carassius auratus) larvae fed with live, mixed or dry diet. Aquaculture 128, 163–

176. doi:10.1016/0044-8486(94)90111-2

Abozaid, H., Wessels, S., Hörstgen-Schwark, G., 2012. Elevated temperature applied

during gonadal transformation leads to male bias in zebrafish (Danio rerio). Sex.

Dev. 6, 201–9. doi:10.1159/000336297

Abozaid, H., Wessels, S., Hörstgen-Schwark, G., 2011. Effect of rearing temperatures

during embryonic development on the phenotypic sex in zebrafish (Danio rerio).

Sex. Dev. 5, 259–65. doi:10.1159/000330120

Alsaqufi, A.S., Gomelsky, B., Schneider, K.J., Pomper, K.W., 2014. Verification of

mitotic gynogenesis in ornamental (koi) carp (Cyprinus carpio L.) using

microsatellite DNA markers. Aquac. Res. 45, 410–416. doi:10.1111/j.1365-

2109.2012.03242.x

Anderson, J.L., Rodríguez Marí, A., Braasch, I., Amores, A., Hohenlohe, P., Batzel, P.,

Postlethwait, J.H., 2012. Multiple sex-associated regions and a putative sex

chromosome in zebrafish revealed by RAD mapping and population genomics.

PLoS One 7, e40701. doi:10.1371/journal.pone.0040701

Aoyama, Y., Moriya, N., Tanaka, S., Taniguchi, T., Hosokawa, H., Maegawa, S., 2015.

169

A novel method for rearing zebrafish by using freshwater rotifers (Brachionus

calyciflorus). Zebrafish 12, 288–295. doi:10.1089/zeb.2014.1032

Aparicio, E., Sostoa, A., 1999. Pattern of movements of adult Barbus haasi in a small

Mediterranean stream. J. Fish Biol. 55, 1086–1095. doi:10.1111/j.1095-

8649.1999.tb00743.x

Argue, B.J., Liu, Z., Dunham, R.A., 2003. Dress-out and fillet yields of channel catfish,

Ictalurus punctatus, blue catfish, Ictalurus furcatus, and their F1, F2 and backcross

hybrids. Aquaculture 228, 81–90. doi:10.1016/S0044-8486(03)00245-X

Arntzen, J.W., Jehle, R., Bardakci, F., Burke, T., Wallis, G.P., 2009. Asymmetric

viability of reciprocal-cross hybrids between crested and marbled newts (Triturus

cristatus and T. marmoratus). Evolution (N. Y). 63, 1191–1202. doi:10.1111/j.1558-

5646.2009.00611.x

Attramadal, K.J.K., Tøndel, B., Salvesen, I., Øie, G., Vadstein, O., Olsen, Y., 2012.

Ceramic clay reduces the load of organic matter and bacteria in marine fish larval

culture tanks. Aquac. Eng. 49, 23–34. doi:10.1016/j.aquaeng.2012.02.003

Bachtrog, D., Mank, J.E., Peichel, C.L., Kirkpatrick, M., Otto, S.P., Ashman, T.-L.,

Hahn, M.W., Kitano, J., Mayrose, I., Ming, R., Perrin, N., Ross, L., Valenzuela, N.,

Vamosi, J.C., 2014. Sex determination: Why so many ways of doing it? PLoS Biol.

12, e1001899. doi:10.1371/journal.pbio.1001899

Barahona-Fernandes, M.H., 1979. Some effects of light intensity and photoperiod on the

sea bass larvae (Dicentrarchus labrax (L.)) reared at the Centre Oceanologique de

Bretagne. Aquaculture 17, 311–321. doi:10.1016/0044-8486(79)90086-3

170

Barton, N.H., Etheridge, A.M., Véber, A., 2017. The infinitesimal model: Definition,

derivation, and implications. Theor. Popul. Biol. 118, 50–73.

doi:10.1016/J.TPB.2017.06.001

Bateman, A.J., 1948. Intra-sexual selection in Drosophila. Heredity. 2, 349–368.

doi:10.1038/hdy.1948.21

Bates, D., Mächler, M., Bolker, B., Walker, S., 2015. Fitting linear mixed-effects models

using lme4. J. Stat. Softw. 67, 1–48. doi:10.18637/jss.v067.i01

Bayless, J.D., 1968. Striped bass hatching and hybridization experiments. Proc. Annu.

Conf. Southeast. Assoc. Game Fish Comm. 21, 233–244.

Becker, W.A., 1984. Manual of quantitative genetics. Academic Enterprises, Pullman,

WA, USA.

Benfey, T.J., 2011. Physiology of triploid fish, in: Farrell, A.P. (Ed.), Encyclopedia of

Fish Physiology: From Genome to Environment. Academic Press, San Diego, CA,

pp. 2009–2015.

Benfey, T.J., Solar, I.I., De Jong, G., Donaldson, E.M., 1986. Flow-cytometric

confirmation of aneuploidy in sperm from triploid rainbow trout. Trans. Am. Fish.

Soc. 115, 838–840. doi:10.1577/1548-8659(1986)115<838:FCOAIS>2.0.CO;2

Best, J., Adatto, I., Cockington, J., James, A., Lawrence, C., 2010. A novel method for

rearing first-feeding larval zebrafish: polyculture with Type L saltwater rotifers

(Brachionus plicatilis). Zebrafish 7, 289–295. doi:10.1089/zeb.2010.0667

Beukeboom, L.W., Perrin, N., 2014. The evolution of sex determination. Oxford

University Press, Oxford, UK.

171

Bishop, R.D., 1968. Evaluation of the striped bass (Roccus saxatilis) and white bass (R.

chrysops) hybrids after two years. Proc. Annu. Conf. Southeast. Assoc. Game Fish

Comm. 21, 245–253.

Bolnick, D.I., Near, T.J., 2005. Tempo of hybrid inviability in centrarchid fishes

(teleostei: centrarchidae). Evolution (N. Y). 59, 1754. doi:10.1554/04-563.1

Bolnick, D.I., Turelli, M., López-Fernández, H., Wainwright, P.C., Near, T.J., 2008.

Accelerated mitochondrial evolution and “Darwin’s corollary”: Asymmetric

viability of reciprocal F1 hybrids in Centrarchid fishes. Genetics 178, 1037–48.

doi:10.1534/genetics.107.081364

Bongers, A., Zandieh-Doulabi, B., Richter, C., Komen, J., 1999. Viable androgenetic YY

genotypes of common carp (Cyprinus carpio L.). J. Hered. 90, 195–198.

doi:10.1093/jhered/90.1.195

Bosworth, B.G., Libey, G.S., Notter, D.R., 1997. Egg, larval, and fingerling traits of

crosses among striped bass (Morone saxatilis), white bass (M. chrysops), and their

F1 hybrids. Aquaculture 154, 201–217. doi:10.1016/S0044-8486(97)00031-8

Bradley, K.M., Breyer, J.P., Melville, D.B., Broman, K.W., Knapik, E.W., Smith, J.R.,

2011. An SNP-based linkage map for zebrafish reveals sex determination loci. G3

Genes, Genomes, Genet. 1, 3–9. doi:10.1534/g3.111.000190

Brandvain, Y., Pauly, G.B., May, M.R., Turelli, M., 2014. Explaining Darwin’s corollary

to Haldane’s rule: the role of mitonuclear interactions in asymmetric postzygotic

isolation among toads. Genetics 197, 743–7. doi:10.1534/genetics.113.161133

Brown, A.R., Bickley, L.K., Ryan, T.A., Paull, G.C., Hamilton, P.B., Owen, S.F., Sharpe,

172

A.D., Tyler, C.R., 2012. Differences in sexual development in inbred and outbred

zebrafish (Danio rerio) and implications for chemical testing. Aquat. Toxicol. 112–

113, 27–38. doi:10.1016/j.aquatox.2012.01.017

Bull, J.J., 1983. Evolution of sex determining mechanisms. Benjamin/Cummings Pub.

Co., Menlo Park, USA.

Bull, J.J., 1980. Sex Determination in Reptiles. Q. Rev. Biol. 55, 3–21.

Bull, J.J., Vogt, R.C., Bulmer, M.G., 1982. Heritability of sex ratio in turtles with

environmental sex determination. Evolution (N. Y). 36, 333–341.

doi:10.1111/j.1558-5646.1982.tb05049.x

Bulmer, M.G., Bull, J.J., 1982. Models of polygenic sex determination and sex ratio

control. Evolution (N. Y). 36, 13. doi:10.2307/2407962

Buonaccorsi, V.P., Mcdowell, J.R., Graves, J.E., 2001. Reconciling patterns of inter-

ocean molecular variance from four classes of molecular markers in blue marlin

(Makaira nigricans). Mol. Ecol. 10, 1179–1196. doi:10.1046/j.1365-

294X.2001.01270.x

Cano, J.M., Makinen, H.S., Merila, J., 2008. Genetic evidence for male-biased dispersal

in the three-spined stickleback (Gasterosteus aculeatus). Mol. Ecol. 17, 3234–3242.

doi:10.1111/j.1365-294X.2008.03837.x

Carrillo, A., McHenry, M.J., 2016. Zebrafish learn to forage in the dark. J. Exp. Biol.

219, 582–589. doi:10.1242/jeb.128918

Carter, C.G., McCarthy, I.D., Houlihan, D.F., Johnstone, R., Walsingham, M. V.,

Mitchell, A.I., 1994. Food consumption, feeding behaviour, and growth of triploid

173

and diploid Atlantic salmon, Salmo salar L., parr. Can. J. Zool. 72, 609–617.

doi:10.1139/z94-083

Carvalho, A.P., Araujo, L., Santos, M.M., 2006. Rearing zebrafish (Danio rerio) larvae

without live food: evaluation of a commercial, a practical and a purified starter diet

on larval performance. Aquac. Res. 37, 1107–1111. doi:10.1111/j.1365-

2109.2006.01534.x

Castranova, D., Lawton, A., Lawrence, C., Baumann, D.P., Best, J., Coscolla, J.,

Doherty, A., Ramos, J., Hakkesteeg, J., Wang, C., Wilson, C., Malley, J., Weinstein,

B.M., 2011. The effect of stocking densities on reproductive performance in

laboratory zebrafish (Danio rerio). Zebrafish 8, 141–146.

doi:10.1089/zeb.2011.0688

Charlon, N., Bergot, P., 1984. Rearing system for feeding fish larvae on dry diets. Trial

with carp (Cyprinus carpio L.) larvae. Aquaculture 41, 1–9. doi:10.1016/0044-

8486(84)90384-3

Chen, W.T., Ge, W., 2013. Gonad differentiation and puberty onset in the zebrafish:

Evidence for the dependence of puberty onset on body growth but not age in

females. Mol. Reprod. Dev. 80, 384–392. doi:10.1002/mrd.22172

Chengfei, S., Lan, Q., Xing, Y., Junjian, D., Yuanyuan, T., Maixin, L., 2017. Establishing

a zebrafish transgenic line expressing tilapia lysozyme with enhanced antibacterial

activity. Aquac. Res. 48, 760–766. doi:10.1111/are.12920

Cherfas, N., Gomelsky, B., Ben-Dom, N., Hulata, G., 1995. Evidence for the heritable

nature of spontaneous diploidization in common carp, Cyprinus carpio L., eggs.

174

Aquac. Res. 26, 289–292. doi:10.1111/j.1365-2109.1995.tb00914.x

Cherfas, N.B., Gomelsky, B.I., Emelyanova, O.V., Recoubratsky, A.V., 1994. Induced

diploid gynogenesis and polyploidy in crucian carp, Carassius auratus gibelio

(Bloch), x common carp, Cyprinus carpio L., hybrids. Aquac. Res. 25, 943–954.

doi:10.1111/j.1365-2109.1994.tb01356.x

Chevassus, B., 1983. Hybridization in fish. Aquaculture 33, 245–262. doi:10.1016/0044-

8486(83)90405-2

Chiu, C.N., Rihel, J., Lee, D.A., Singh, C., Mosser, E.A., Chen, S., Sapin, V., Pham, U.,

Engle, J., Niles, B.J., Montz, C.J., Chakravarthy, S., Zimmerman, S., Salehi-

Ashtiani, K., Vidal, M., Schier, A.F., Prober, D.A., 2016. A zebrafish genetic screen

identifies neuromedin U as a regulator of sleep/wake states. Neuron 89, 842–856.

doi:10.1016/j.neuron.2016.01.007

Chourrout, D., Quillet, E., 1982. Induced gynogenesis in the rainbow trout: Sex and

survival of progenies production of all-triploid populations. Theor. Appl. Genet. 63,

201–205. doi:10.1007/BF00303993

Ciudad, J., Cid, E., Velasco, A., Lara, J.M., Aijon, J., Orfao, A., 2002. Flow cytometry

measurement of the DNA contents of G0/G1 diploid cells from three different

teleost fish species. Cytometry 48, 20–25. doi:10.1002/cyto.10100

Clinton, M., Haines, L.C., 1999. An overview of factors influencing sex determination

and gonadal development in birds. Cell. Mol. Life Sci. 55, 876.

doi:10.1007/s000180050341

Consuegra, S., García de Leániz, C., 2007. Fluctuating sex ratios, but no sex-biased

175

dispersal, in a promiscuous fish. Evol. Ecol. 21, 229–245. doi:10.1007/s10682-006-

9001-7

Conte, F.P., Hootman, S.R., Harris, P.J., 1972. Neck organ of Artemia salina nauplii. J.

Comp. Physiol. 80, 239–246. doi:10.1007/BF00694838

Corley-Smith, G.E., Lim, C.T.J., Brandhorst, B.P., 1996. Production of androgenetic

zebrafish (Danio rerio). Genetics 142, 1265–1276.

Cortemeglia, C., Beitinger, T.L., 2005. Temperature tolerances of wild-type and red

transgenic zebra danios. Trans. Am. Fish. Soc. 134, 1431–1437. doi:10.1577/T04-

197.1

Cox, R.M., Calsbeek, R., 2009. Sexually antagonistic selection, sexual dimorphism, and

the resolution of intralocus sexual conflict. Am. Nat. 173, 176–87.

doi:10.1086/595841

Coyne, J.A., Orr, H.A., 2004. Speciation. Sinauer Associates, Sunderland, MA.

Coyne, J.A., Simeonidis, S., Rooney, P., 1998. Relative paucity of genes causing

inviability in hybrids between Drosophila melanogaster and D. simulans. Genetics

150, 1091–1103.

Creaser, C.W., 1934. The technic of handling the zebra fish (Brachydanio rerio) for the

production of eggs which are favorable for embryological research and are available

at any specified time throughout the year. Copeia 4, 159–161.

Croft, D.P., Albanese, B., Arrowsmith, B.J., Botham, M., Webster, M., Krause, J., 2003a.

Sex-biased movement in the guppy (Poecilia reticulata). Oecologia 137, 62–68.

doi:10.1007/s00442-003-1268-6

176

Croft, D.P., Arrowsmith, B.J., Bielby, J., Skinner, K., White, E., Couzin, I.D., Magurran,

A.E., Ramnarine, I., Krause, J., 2003b. Mechanisms underlying shoal composition in

the Trinidadian guppy, Poecilia reticulata. Oikos 100, 429–438. doi:10.1034/j.1600-

0706.2003.12023.x

Dabrowski, K., 1984. Influence of initial weight during the change from live to

compound feed on the survival and growth of four cyprinids. Aquaculture 40, 27–

40. doi:10.1016/0044-8486(84)90213-8

Dabrowski, K., Miller, M.E., 2018. Contested paradigm in raising zebrafish (Danio

rerio). Zebrafish 15, (available online). doi:10.1089/zeb.2017.1515

Dabrowski, K., Rinchard, J., Lin, F., Garcia-Abiado, M.-A., Schmidt, D., 2000. Induction

of gynogenesis in muskellunge with irradiated sperm of yellow perch proves diploid

muskellunge male homogamety. J. Exp. Zool. 287, 96–105. doi:10.1002/1097-

010X(20000615)287:1<96::AID-JEZ13>3.0.CO;2-N

Dabrowski, K., Zhang, Y., Kwasek, K., Hliwa, P., Ostaszewska, T., 2010. Effects of

protein-, peptide- and free amino acid-based diets in fish nutrition. Aquac. Res. 41,

668–683. doi:10.1111/j.1365-2109.2010.02490.x

Dahm, R., Geisler, R., 2006. Learning from small fry: The zebrafish as a genetic model

organism for aquaculture fish species. Mar. Biotechnol. doi:10.1007/s10126-006-

5139-0

Darwin, C., 1859. On the origin of species. John Murray, London, England. de Villemereuil, P., Gimenez, O., Doligez, B., 2013. Comparing parent-offspring

regression with frequentist and Bayesian animal models to estimate heritability in

177

wild populations: a simulation study for Gaussian and binary traits. Methods Ecol.

Evol. 4, 260–275. doi:10.1111/2041-210X.12011

Delomas, T.A., Dabrowski, K., 2017a. The importance of controlling genetic variation -

Remarks on ’Appropriate rearing density in domesticated zebrafish to avoid

masculinization: Links with the stress response. J. Exp. Biol. 220, 4078.

doi:10.1242/jeb.164293

Delomas, T.A., Dabrowski, K., 2017b. Heritability of spontaneous diploidization of

maternal chromosomes in zebrafish Danio rerio, in: Hendry, C.I. (Ed.), LARVI ’17

- Fish and Shellfish Larviculture Symposium. Ghent, Belgium, pp. 123–126.

Delomas, T.A., Dabrowski, K., 2016. Zebrafish embryonic development is induced by

carp sperm. Biol. Lett. 12, 20160628.

Delomas, T.A., Gomelsky, B., Anil, A., Schneider, K.J., Warner, J.L., 2017. Spontaneous

polyploidy, gynogenesis and androgenesis in second generation (F2) koi Cyprinus

carpio × goldfish Carassius auratus hybrids. J. Fish Biol. 90, 80–92.

doi:10.1111/jfb.13157

Devlin, R.H., Nagahama, Y., 2002. Sex determination and sex differentiation in fish: an

overview of genetic, physiological, and environmental influences. Aquaculture 208,

191–364. doi:10.1016/S0044-8486(02)00057-1

Disney, J.E., Johnson, K.R., Banks, D.K., Thorgaard, G.H., 1988. Maintenance of foreign

gene expression and independent chromosome fragments in adult transgenic

rainbow trout and their offspring. J. Exp. Zool. 248, 335–344.

doi:10.1002/jez.1402480312

178

Donovan, A., Brownlie, A., Zhou, Y., Shepard, J., Pratt, S.J., Moynihan, J., Paw, B.H.,

Drejer, A., Barut, B., Zapata, A., Law, T.C., Brugnara, C., Lux, S.E., Pinkus, G.S.,

Pinkus, J.L., Kingsley, P.D., Palis, J., Fleming, M.D., Andrews, N.C., Zon, L.I.,

2000. Positional cloning of zebrafish ferroportin1 identifies a conserved vertebrate

iron exporter. Nature 403, 776–81. doi:10.1038/35001596

Dranow, D.B., Hu, K., Bird, A.M., Lawry, S.T., Adams, M.T., Sanchez, A., Amatruda,

J.F., Draper, B.W., 2016. Bmp15 Is an oocyte-produced signal required for

maintenance of the adult female sexual phenotype in zebrafish. PLOS Genet. 12,

e1006323. doi:10.1371/journal.pgen.1006323

Dranow, D.B., Tucker, R.P., Draper, B.W., 2013. Germ cells are required to maintain a

stable sexual phenotype in adult zebrafish. Dev. Biol. 376, 43–50.

doi:10.1016/j.ydbio.2013.01.016

Driever, W., Solnica-Krezel, L., Schier, A.F., Neuhauss, S.C., Malicki, J., Stemple, D.L.,

Stainier, D.Y., Zwartkruis, F., Abdelilah, S., Rangini, Z., Belak, J., Boggs, C., 1996.

A genetic screen for mutations affecting embryogenesis in zebrafish. Development

123, 37–46. doi:9007227

Dunham, R.A., 2011. Aquaculture and fisheries biotechnology: genetic approaches, 2nd

ed. CABI, Cambridge, MA, USA. doi:10.1079/9781845936518.0000

Dunham, R.A., Smitherman, R.O., Webber, C., 1983. Relative tolerance of channel x

blue hybrid and channel catfish to low oxygen concentrations. Progress. Fish Cult.

45, 55–57.

Duray, M., Kohno, H., 1988. Effects of continuous lighting on growth and survival of

179

first-feeding larval rabbitfish, Siganus guttatus. Aquaculture 72, 73–79.

doi:10.1016/0044-8486(88)90147-0

Eaton, R.C., Farley, R.D., 1974. Growth and the reduction of depensation of zebrafish,

Brachydanio rerio, reared in the laboratory. Copeia 1974, 204. doi:10.2307/1443024

Ellis, L., 1995. Dominance and reproductive success among nonhuman animals: A cross-

species comparison. Ethol. Sociobiol. 16, 257–333. doi:10.1016/0162-

3095(95)00050-U

Ezaz, M.T., McAndrew, B.J., Penman, D.J., 2004. Spontaneous diploidization of the

maternal chromosome set in Nile tilapia (Oreochromis niloticus L.) eggs. Aquac.

Res. 35, 271–277. doi:10.1111/j.1365-2109.2004.01010.x

Falconer, D.S., Mackay, T.F.C., 1996. Introduction to quantitative genetics, 4th ed.

Pearson, Harlow, UK.

Fang, F., Noren, M., Liao, T.Y., Källersjö, M., Kullander, S.O., 2009. Molecular

phylogenetic interrelationships of the south Asian cyprinid genera Danio, Devario

and Microrasbora (Teleostei, Cyprinidae, Danioninae). Zool. Scr. 38, 237–256.

doi:10.1111/j.1463-6409.2008.00373.x

Feindel, N.J., Benfey, T.J., Trippel, E.A., 2010. Competitive spawning success and

fertility of triploid male Atlantic cod Gadus morhua. Aquac. Environ. Interact. 1,

47–55. doi:10.2307/24864017

Feitsma, H., Leal, M.C., Moens, P.B., Cuppen, E., Schulz, R.W., 2007. Mlh1 deficiency

in zebrafish results in male sterility and aneuploid as well as triploid progeny in

females. Genetics 175, 1561–9. doi:10.1534/genetics.106.068171

180

Felip, A., Piferrer, F., Zanuy, S., Carrillo, M., 2001. Comparative growth performance of

diploid and triploid European sea bass over the first four spawning seasons. J. Fish

Biol. 58, 76–88. doi:10.1111/j.1095-8649.2001.tb00500.x

Fenske, M., Segner, H., 2004. Aromatase modulation alters gonadal differentiation in

developing zebrafish (Danio rerio). Aquat. Toxicol. 67, 105–126.

doi:10.1016/J.AQUATOX.2003.10.008

Fisher, R.A., 1930. Genetical theory of natural selection. Clarendon Press, Oxford.

Fisher, R.A., 1918. The correlation between relatives on the supposition of Mendelian

inheritance. Trans. R. Soc. Edinburgh 52, 399–433.

Fredga, K., Gropp, A., Winking, H., Frank, F., 1976. Fertile XX- and XY-type females in

the wood lemming Myopus schisticolor. Nature 261, 225–227.

doi:10.1038/261225a0

Froese, R., Pauly, D., 2017. FishBase [WWW Document]. World Wide Web Electron.

Publ. URL www.fishbase.org (accessed 2.1.18).

Fromhage, L., Jennions, M., Kokko, H., 2016. The evolution of sex roles in mate

searching. Evolution. 70, 617–624. doi:10.1111/evo.12874

Fujiwara, A., Abe, S., Yamaha, E., Yamazaki, F., Yoshida, M.C., 1997. Uniparental

chromosome elimination in the early embryogenesis of the inviable salmonid

hybrids between masu salmon female and rainbow trout male. Chromosoma 106,

44–52. doi:10.1007/s004120050223

Garber, A.F., Sullivan, C. V., 2006. Selective breeding for the hybrid striped bass

(Morone chrysops, Rafinesque x M. saxatilis, Walbaum) industry: status and

181

perspectives. Aquac. Res. 37, 319–338. doi:10.1111/j.1365-2109.2005.01439.x

Garcia-Abiado, M.A.R., Lynch, W.E., Dabrowski, K., Hartman, T., 2001. Use of thermal

and pressure shocks to induce triploid hybrid saugeyes. N. Am. J. Aquac. 63, 83–91.

doi:10.1577/1548-8454(2001)063<0083:UOTAPS>2.0.CO;2

Gestl, E.E., Kauffman, E.J., Moore, J.L., Cheng, K.C., 1997. New conditions for

generation of gynogenetic half-tetrad embryos in the zebrafish (Danio rerio). J.

Hered. 88, 76–79.

Gibson, J.R., Chippindale, A.K., Rice, W.R., 2002. The X chromosome is a hot spot for

sexually antagonistic fitness variation. Proc. R. Soc. B 269, 499–505.

doi:10.1098/rspb.2001.1863

Gilliam, J.F., Fraser, D.F., 2001. Movement in corridors: enhancement by predation

threat, disturbance, and habitat structure. Ecology 82, 258–273.

Gilmour, A.R., Gogel, B.J., Cullis, B.R., Welham, S.J., Thompson, R., 2015. ASReml

User Guide, Release 4.1. VSN International Ltd, Hemel Hempstead, UK.

Glennon, R.P., Gomelsky, B., Schneider, K.J., Kelly, A.M., Haukenes, A., 2012.

Evidence of female heterogamety in largemouth bass, based on sex ratio of

gynogenetic progeny. N. Am. J. Aquac. 74, 537–540.

doi:10.1080/15222055.2012.700906

Glutton-Brock, T.H., Vincent, A.C.J., 1991. Sexual selection and the potential

reproductive rates of males and females. Nature 351, 58–60. doi:10.1038/351058a0

Gomelsky, B., 2011. Fish genetics: theory and practice. VDM Verlag, Saarbrucken,

Germany.

182

Gomelsky, B., Delomas, T.A., Warner, J.L., 2016. Ploidy variation and viability of

aneuploid ornamental koi carp obtained by crossing triploid females with diploid

males. N. Am. J. Aquac. 78, 218–223. doi:10.1080/15222055.2016.1150923

Gomelsky, B., Schneider, K.J., Anil, A., Delomas, T.A., 2015. Gonad development in

triploid ornamental koi carp and results of crossing triploid females with diploid

males. N. Am. J. Aquac. 77. doi:10.1080/15222055.2014.963766

Gomelsky, B.I., Recoubratsky, A. V, 1990. On the possibility of spontaneous andro- and

gynogenesis in fishes. J. Ichthyol. 30, 152–155.

Goolish, E.M., Okutake, K., Lesure, S., 1999. Growth and survivorship of larval

zebrafish Danio rerio on processed diets. N. Am. J. Aquac. 61, 189–198.

doi:10.1577/1548-8454(1999)061<0189:GASOLZ>2.0.CO;2

Greenwood, P.J., 1980. Mating systems, philopatry and dispersal in birds and mammals.

Anim. Behav. 28, 1140–1162. doi:10.1016/S0003-3472(80)80103-5

Gros, A., Hovestadt, T., Poethke, H.J., 2008. Evolution of sex-biased dispersal: The role

of sex-specific dispersal costs, demographic stochasticity, and inbreeding. Ecol.

Modell. 219, 226–233. doi:10.1016/j.ecolmodel.2008.08.014

Grützner, F., Rens, W., Tsend-Ayush, E., El-Mogharbel, N., O’Brien, P.C.M., Jones,

R.C., Ferguson-Smith, M.A., Marshall Graves, J.A., 2004. In the platypus a meiotic

chain of ten sex chromosomes shares genes with the bird Z and mammal X

chromosomes. Nature 432, 913–917. doi:10.1038/nature03021

Hadfield, J.D., 2010. MCMC methods for multi-response generalized linear mixed

models: The MCMCglmm R package. J. Stat. Softw. 33, 1–22.

183

Haffray, P., Bruant, J.S., Facqueur, J.M., Fostier, A., 2005. Gonad development, growth,

survival and quality traits in triploids of the protandrous hermaphrodyte gilthead

seabream Sparus aurata (L.). Aquaculture 247, 107–117.

doi:10.1016/J.AQUACULTURE.2005.02.037

Haffter, P., Granato, M., Brand, M., Mullins, M.C., Hammerschmidt, M., Kane, D. a,

Odenthal, J., van Eeden, F.J., Jiang, Y.J., Heisenberg, C.P., Kelsh, R.N., Furutani-

Seiki, M., Vogelsang, E., Beuchle, D., Schach, U., Fabian, C., Nüsslein-Volhard, C.,

1996. The identification of genes with unique and essential functions in the

development of the zebrafish, Danio rerio. Development 123, 1–36. doi:9007226

Halekoh, U., Højsgaard, S., 2014. A Kenward-Roger approximation and parametric

bootstrap methods for tests in linear mixed models - The R package pbkrtest. J. Stat.

Softw. 59, 1–30.

Hammerstein, P., Parker, G.A., 1987. Sexual selection: games between the sexes, in:

Bradbury, J.W., Anderson, B. (Eds.), Sexual Selection: Testing the Alternatives.

Wiley, Chichester, UK, pp. 119–142.

Hartwell, L.H., Culotti, J., Reid, B., 1970. Genetic control of the cell-division cycle in

yeast. I. Detection of mutants. Proc. Natl. Acad. Sci. U. S. A. 66, 352–9.

Hayden, T.A., Miner, J.G., Farver, J.R., Fryer, B.J., 2011. Philopatry and vagrancy of

white bass (Morone chrysops) spawning in the Sandusky River: Evidence of

metapopulation structure in western Lake Erie using otolith chemistry. J. Great

Lakes Res. 37, 691–697. doi:10.1016/j.jglr.2011.08.012

Heier, J., Takle, K.A., Hasley, A.O., Pelegri, F., 2015. Ploidy manipulation and induction

184

of alternate cleavage patterns through inhibition of centrosome duplication in the

early zebrafish embryo. Dev. Dyn. 244, 1300–1312. doi:10.1002/dvdy.24315

Herbst, E.W., Fredga, K., Frank, F., Winking, H., Gropp, A., 1978. Cytological

identification of two X-chromosome types in the wood lemming (Myopus

schisticolor). Chromosoma 69, 185–191. doi:10.1007/BF00329917

Hewitson, T.D., Wigg, B., Becker, G.J., 2010. Tissue preparation for histochemistry:

Fixation, embedding, and antigen retrieval for light microscopy, in: Hewitson, T.D.,

Darby, I.A. (Eds.), Histology Protocols. Humana Press, New York, NY, pp. 3–18.

doi:10.1007/978-1-60327-345-9_1

Hirsh, D., Vanderslice, R., 1976. Temperature-sensitive developmental mutants of

Caenorhabditis elegans. Dev. Biol. 49, 220–235. doi:10.1016/0012-1606(76)90268-

Hoarau, G., Piquet, A.M.-T., van der Veer, H.., Rijnsdorp, A.., Stam, W.., Olsen, J..,

2004. Population structure of plaice (Pleuronectes platessa L.) in northern Europe: a

comparison of resolving power between microsatellites and mitochondrial DNA

data. J. Sea Res. 51, 183–190. doi:10.1016/j.seares.2003.12.002

Holtzman, N.G., Iovine, M.K., Liang, J.O., Morris, J., 2016. Learning to fish with

genetics: A primer on the vertebrate model Danio rerio. Genetics 203.

Hörstgen-Schwark, G., 1993. Production of homozygous diploid zebra fish (Brachydanio

rerio). Aquaculture 112, 25–37. doi:10.1016/0044-8486(93)90155-R

Horvath, A., Miskolczi, E., Urbanyi, B., Horvath, L., Urbányi, B., 2003.

Cryopreservation of common carp sperm. Aquat. Living Resour. 16, 457–460.

185

doi:10.1016/s0990-7440(03)00084-6

Hou, J., Fujimoto, T., Saito, T., Yamaha, E., Arai, K., 2015. Generation of clonal

zebrafish line by androgenesis without egg irradiation. Sci. Rep. 5, 13346.

doi:10.1038/srep13346

Howe, K., Clark, M.D., Torroja, C.F., Torrance, J., Berthelot, C., Muffato, M., Collins,

J.E., Humphray, S., McLaren, K., Matthews, L., McLaren, S., Sealy, I., Caccamo,

M., Churcher, C., Scott, C., Barrett, J.C., Koch, R., Rauch, G.-J., White, S., Chow,

W., Kilian, B., Quintais, L.T., Guerra-Assunção, J.A., Zhou, Y., Gu, Y., Yen, J.,

Vogel, J.-H., Eyre, T., Redmond, S., Banerjee, R., Chi, J., Fu, B., Langley, E.,

Maguire, S.F., Laird, G.K., Lloyd, D., Kenyon, E., Donaldson, S., Sehra, H.,

Almeida-King, J., Loveland, J., Trevanion, S., Jones, M., Quail, M., Willey, D.,

Hunt, A., Burton, J., Sims, S., McLay, K., Plumb, B., Davis, J., Clee, C., Oliver, K.,

Clark, R., Riddle, C., Elliot, D., Eliott, D., Threadgold, G., Harden, G., Ware, D.,

Begum, S., Mortimore, B., Mortimer, B., Kerry, G., Heath, P., Phillimore, B.,

Tracey, A., Corby, N., Dunn, M., Johnson, C., Wood, J., Clark, S., Pelan, S.,

Griffiths, G., Smith, M., Glithero, R., Howden, P., Barker, N., Lloyd, C., Stevens,

C., Harley, J., Holt, K., Panagiotidis, G., Lovell, J., Beasley, H., Henderson, C.,

Gordon, D., Auger, K., Wright, D., Collins, J., Raisen, C., Dyer, L., Leung, K.,

Robertson, L., Ambridge, K., Leongamornlert, D., McGuire, S., Gilderthorp, R.,

Griffiths, C., Manthravadi, D., Nichol, S., Barker, G., Whitehead, S., Kay, M.,

Brown, J., Murnane, C., Gray, E., Humphries, M., Sycamore, N., Barker, D.,

Saunders, D., Wallis, J., Babbage, A., Hammond, S., Mashreghi-Mohammadi, M.,

186

Barr, L., Martin, S., Wray, P., Ellington, A., Matthews, N., Ellwood, M.,

Woodmansey, R., Clark, G., Cooper, J.D., Cooper, J., Tromans, A., Grafham, D.,

Skuce, C., Pandian, R., Andrews, R., Harrison, E., Kimberley, A., Garnett, J.,

Fosker, N., Hall, R., Garner, P., Kelly, D., Bird, C., Palmer, S., Gehring, I., Berger,

A., Dooley, C.M., Ersan-Ürün, Z., Eser, C., Geiger, H., Geisler, M., Karotki, L.,

Kirn, A., Konantz, J., Konantz, M., Oberländer, M., Rudolph-Geiger, S., Teucke,

M., Lanz, C., Raddatz, G., Osoegawa, K., Zhu, B., Rapp, A., Widaa, S., Langford,

C., Yang, F., Schuster, S.C., Carter, N.P., Harrow, J., Ning, Z., Herrero, J., Searle,

S.M.J., Enright, A., Geisler, R., Plasterk, R.H.A., Lee, C., Westerfield, M., de Jong,

P.J., Zon, L.I., Postlethwait, J.H., Nüsslein-Volhard, C., Hubbard, T.J.P., Roest

Crollius, H., Rogers, J., Stemple, D.L., 2013. The zebrafish reference genome

sequence and its relationship to the human genome. Nature 496, 498–503.

doi:10.1038/nature12111

Ibáñez-Escriche, N., Varona, L., Sorensen, D., Noguera, J.L., 2008. A study of

heterogeneity of environmental variance for slaughter weight in pigs. Animal 2, 19–

26. doi:10.1017/S1751731107001000

Ijiri, K.-I., Egami, N., 1980. Hertwig effect caused by UV-irradiation of sperm of Oryzias

latipes (teleost) and its photoreactivation. Mutat. Res. 69, 241–248.

doi:10.1016/0027-5107(80)90089-5

Innocenti, P., Morrow, E.H., 2010. The sexually antagonistic genes of Drosophila

melanogaster. PLoS Biol. 8, e1000335. doi:10.1371/journal.pbio.1000335

Jaenike, J., 1978. A hypothesis to account for the maintenance of sex within populations.

187

Evol. Theory 3, 191–194.

Janhunen, M., Kause, A., Vehviläinen, H., Järvisalo, O., 2012. Genetics of

microenvironmental sensitivity of body weight in rainbow trout (Oncorhynchus

mykiss) selected for improved growth. PLoS One 7, e38766.

doi:10.1371/journal.pone.0038766

Janzen, F.J., 1992. Heritable variation for sex ratio under environmental sex

determination in the common snapping turtle (Chelydra serpentina). Genetics 131,

155–161.

Jiang, Y., Schmidt, R.H., Zhao, Y., Reif, J.C., 2017. A quantitative genetic framework

highlights the role of epistatic effects for grain-yield heterosis in bread wheat. Nat.

Genet. 49, 1741–1746. doi:10.1038/ng.3974

Johnson, S.L., Weston, J.A., 1995. Temperature-sensitive mutations that cause stage-

specific defects in zebrafish fin regeneration. Genetics 141, 1583–1595.

Julliard, R., 2000. Sex-specific dispersal in spatially varying environments leads to

habitat-dependent evolutionary stable offspring sex ratios. Behav. Ecol. 11, 421–

428. doi:10.1093/beheco/11.4.421

Kallman, K.D., 1973. The sex-determining mechanism of the platyfish, Xiphophorus

maculatus, in: Schröder, J.H. (Ed.), Genetics and Mutagenesis of Fish. Springer,

Berlin, Germany, pp. 19–28.

Kamiya, T., Kai, W., Tasumi, S., Oka, A., Matsunaga, T., Mizuno, N., Fujita, M.,

Suetake, H., Suzuki, S., Hosoya, S., Tohari, S., Brenner, S., Miyadai, T., Venkatesh,

B., Suzuki, Y., Kikuchi, K., 2012. A trans-species missense SNP in Amhr2 is

188

associated with sex determination in the tiger pufferfish, Takifugu rubripes (Fugu).

PLoS Genet. 8, e1002798. doi:10.1371/journal.pgen.1002798

Karayucel, I., Ezaz, T., Karayucel, S., McAndrew, B.J., Penman, D.J., 2004. Evidence for

two unlinked sex reversal loci in the Nile tilapia, Oreochromis niloticus, and for

linkage of one of these to the red body colour gene. Aquaculture 234, 51–63.

doi:10.1016/j.aquaculture.2003.12.016

Kavumpurath, S., Pandian, T.J., 1990. Induction of triploidy in the zebrafish,

Brachydanio rerio (Hamilton). Aquac. Res. 21, 299–306. doi:10.1111/j.1365-

2109.1990.tb00468.x

Kawamura, K., 1998. Sex determination system of the rosy bitterling, Rhodeus ocellatus

ocellatus, in: Yuma, M., Nakamura, I., Fausch, K.D. (Eds.), Fish Biology in Japan:

An Anthology in Honour of Hiroya Kawanabe. Springer, Dordrecht, Dordrecht,

Netherlands, pp. 251–260. doi:10.1007/978-94-015-9016-7_19

Kestemont, P., Jourdan, S., Houbart, M., Mélard, C., Paspatis, M., Fontaine, P., Cuvier,

A., Kentouri, M., Baras, E., 2003. Size heterogeneity, cannibalism and competition

in cultured predatory fish larvae: biotic and abiotic influences. Aquaculture 227,

333–356. doi:10.1016/S0044-8486(03)00513-1

Khaw, H.L., Ponzoni, R.W., Yee, H.Y., Aziz, M.A. bin, Mulder, H.A., Marjanovic, J.,

Bijma, P., 2016. Genetic variance for uniformity of harvest weight in Nile tilapia

(Oreochromis niloticus). Aquaculture 451, 113–120.

doi:10.1016/J.AQUACULTURE.2015.09.003

Kim, K.-K., Bang, I.C., Kim, Y., Nam, Y.K., Kim, D.S., 1996. Early survival and

189

chromosomes of intergeneric hybrids between Japanes flounder Paralichthys

olivaceus and spotted halibut Verasper variegatus. Fish. Sci. 62, 490–491.

Kimmel, C.B., Ballard, W.W., Kimmel, S.R., Ullmann, B., Schilling, T.F., 1995. Stages

of embryonic development of the zebrafish. Dev. Dyn. 203, 253–310.

Kitamura, H., Yong Teong, O., Arakawa, T., 1991. Gonadal development of artificially

induced triploid red sea bream Pagrus major. Nippon Suisan Gakkaishi 57, 1657–

1660.

Kitamura, S., Ogata, H., Onozato, H., 1991. Triploid male masu salmon Oncorhynchus

masou shows normal courtship behavior. Nippon Suisan Gakkaishi 57, 2157.

Kitanishi, S., Yamamoto, T., Ishii, H., Yamaguchi, Y., Kobayashi, T., 2017. Dispersal

patterns of anadromous and freshwater resident masu salmon at different spatial

scales in mid-western Hokkaido, Japan. Ichthyol. Res. 64, 111–115.

doi:10.1007/s10228-016-0525-8

Knight, M.E., Oppen, M.J.H. van, Smith, H.L., Rico, C., Hewitt, G.M., Turner, G.F.,

1999. Evidence for male-biased dispersal in Lake Malawi cichlids from

microsatellites. Mol. Ecol. 8, 1521–1527. doi:10.1046/j.1365-294x.1999.00740.x

Komen, J., Bongers, A.B.J., Richter, C.J.J., van Muiswinkel, W.B., Huisman, E.A., 1991.

Gynogenesis in common carp (Cyprinus carpio L.): II. the production of

homozygous gynogenetic clones and F1 hybrids. Aquaculture 92, 127–142.

doi:10.1016/0044-8486(91)90015-Y

Komen, J., de Boer, P., Richter, C.J.J., 1992. Male sex reversal in gynogenetic XX

females of common carp (Cyprinus carpio L.) by a recessive mutation in a sex-

190

determining gene. J. Hered. 83, 431–434.

doi:10.1093/oxfordjournals.jhered.a111246

Kondrashov, A.S., 1988. Deleterious mutations and the evolution of sexual reproduction.

Nature 336, 435–440. doi:10.1038/336435a0

Kondrashov, A.S., 1982. Selection against harmful mutations in large sexual and asexual

populations. Genet. Res. 40, 325. doi:10.1017/S0016672300019194

Koopman, P., Gubbay, J., Vivian, N., Goodfellow, P., Lovell-Badge, R., 1991. Male

development of chromosomally female mice transgenic for Sry. Nature 351, 117–

121. doi:10.1038/351117a0

Kovács, B., Bercsényi, M., Csenki, Z., Bakos, K., Kovács, R., Csepeli, A., Németh, S.,

Uri, C., Bernáth, G., Horváth, Á., Bokor, Z., Urbányi, B., Orbán, L., 2014.

Interspecific gynogenesis with cryopreserved sperm in zebrafish (Danio rerio)

[WWW Document]. Aquac. Eur. 2014. URL

https://www.was.org/easonline/AbstractDetail.aspx?i=3714 (accessed 10.3.16).

Krisfalusi, M., Cloud, J.G., 1999. Gonadal sex reversal in triploid rainbow trout

(Oncorhynchus mykiss). J. Exp. Zool. 284, 466–472. doi:10.1002/(SICI)1097-

010X(19990901)284:4<466::AID-JEZ13>3.0.CO;2-G

Kroeger, P.T., Poureetezadi, S.J., McKee, R., Jou, J., Miceli, R., Wingert, R.A., 2014.

Production of haploid zebrafish embryos by in vitro fertilization. Jove-Journal Vis.

Exp. 9. doi:10.3791/51708

Kuznetsova, A., Brockhoff, P.B., Christensen, R.H.B., 2017. lmerTest package: Tests in

linear mixed effects models. J. Stat. Softw. 82, 1–26. doi:10.18637/jss.v082.i13

191

Lawrence, C., 2011. Advances in zebrafish husbandry and management, in: Detrich,

H.W., Westerfield, M., Zon, L.I. (Eds.), The Zebrafish : Genetics, Genomics and

Informatics. Elsevier, Boston, MA, USA, pp. 431–451.

Lawrence, C., Adatto, I., Best, J., James, A., Maloney, K., 2012. Generation time of

zebrafish (Danio rerio) and medakas (Oryzias latipes) housed in the same

aquaculture facility. Lab Anim. (NY). 41, 158–165.

Lawrence, C., Ebersole, J.P., Kesseli, R. V., 2008. Rapid growth and out-crossing

promote female development in zebrafish (Danio rerio). Environ. Biol. Fishes 81,

239–246. doi:10.1007/s10641-007-9195-8

Lawson Handley, L.J., Perrin, N., 2007. Advances in our understanding of mammalian

sex-biased dispersal. Mol. Ecol. 16, 1559–1578. doi:10.1111/j.1365-

294X.2006.03152.x

Lebeda, I., Flajshans, M., 2016. Influence of photoreactivation on induction of

gynogenesis in sterlet, Acipenser ruthenus. Aquac. Res. 47, 1364–1368.

doi:10.1111/are.12596

Lenth, R. V, 2016. Least-squares means: The R package lsmeans. J. Stat. Softw. 69, 1–

33. doi:10.18637/jss.v069.i01

Li, L., Lu, K., Chen, Z., Mu, T., Hu, Z., Li, X., 2008. Dominance, overdominance and

epistasis condition the heterosis in two heterotic rice hybrids. Genetics 180, 1725–

42. doi:10.1534/genetics.108.091942

Li, M.H., Robinson, E.H., Bosworth, B.G., Torrans, E.L., 2014. Growth and feed

conversion ratio of pond-raised hybrid catfish harvested at different sizes. N. Am. J.

192

Aquac. 76, 261–264. doi:10.1080/15222055.2014.911225

Li, Z., Vizeacoumar, F.J., Bahr, S., Li, J., Warringer, J., Vizeacoumar, F.S., Min, R.,

VanderSluis, B., Bellay, J., DeVit, M., Fleming, J.A., Stephens, A., Haase, J., Lin,

Z.-Y., Baryshnikova, A., Lu, H., Yan, Z., Jin, K., Barker, S., Datti, A., Giaever, G.,

Nislow, C., Bulawa, C., Myers, C.L., Costanzo, M., Gingras, A.-C., Zhang, Z.,

Blomberg, A., Bloom, K., Andrews, B., Boone, C., 2011. Systematic exploration of

essential yeast gene function with temperature-sensitive mutants. Nat. Biotechnol.

29, 361–367. doi:10.1038/nbt.1832

Lieschke, G.J., Currie, P.D., 2007. Animal models of human disease: zebrafish swim into

view. Nat. Rev. Genet. 8, 353–367. doi:10.1038/nrg2091

Liew, W.C., Bartfai, R., Lim, Z.J., Sreenivasan, R., Siegfried, K.R., Orban, L., 2012.

Polygenic sex determination system in zebrafish. PLoS One 7, e34397.

doi:10.1371/journal.pone.0034397

Liew, W.C., Orban, L., 2014. Zebrafish sex: a complicated affair. Brief. Funct. Genomics

13, 172–187. doi:10.1093/bfgp/elt041

Lincoln, R.F., Scott, A.P., 1984. Sexual maturation in triploid rainbow trout, Salmo

gairdneri Richardson. J. Fish Biol. 25, 385–392. doi:10.1111/j.1095-

8649.1984.tb04886.x

Liu, S., Liu, Y., Zhou, G., Zhang, X., Luo, C., Feng, H., He, X., Zhu, G., Yang, H., 2001.

The formation of tetraploid stocks of red crucian carp × common carp hybrids as an

effect of interspecific hybridization. Aquaculture 192, 171–186. doi:10.1016/S0044-

8486(00)00451-8

193

Liu, S., Sun, Y., Zhang, C., Luo, K., Liu, Y., 2004. Production of gynogenetic progeny

from allotetraploid hybrids red crucian carp×common carp. Aquaculture 236, 193–

200. doi:10.1016/J.AQUACULTURE.2003.10.001

López-Olmeda, J.F., Sánchez-Vázquez, F.J., 2011. Thermal biology of zebrafish (Danio

rerio). J. Therm. Biol. 36, 91–104. doi:10.1016/J.JTHERBIO.2010.12.005

Luo, C., Li, B., 2003. Diploid-dependent regulation of gene expression: a genetic cause

of abnormal development in fish haploid embryos. Heredity (Edinb). 90, 405–409.

doi:10.1038/sj.hdy.6800263

Maack, G., Segner, H., 2003. Morphological development of the gonads in zebrafish. J.

Fish Biol. 62, 895–906. doi:10.1046/j.1095-8649.2003.00074.x

Maack, G., Segner, H., Tyler, C.R., 2003. Ontogeny of sexual differentiation in different

strains of zebrafish (Danio rerio). Fish Physiol. Biochem. 28, 125–128.

doi:10.1023/B:FISH.0000030497.59378.88

Martin, B.L., Gallagher, T.L., Rastogi, N., Davis, J.P., Beattie, C.E., Amacher, S.L.,

Janssen, P.M.L., 2015. In vivo assessment of contractile strength distinguishes

differential gene function in skeletal muscle of zebrafish larvae. J. Appl. Physiol.

119, 799–806. doi:10.1152/japplphysiol.00447.2015

Matsuda, M., Nagahama, Y., Shinomiya, A., Sato, T., Matsuda, C., Kobayashi, T.,

Morrey, C.E., Shibata, N., Asakawa, S., Shimizu, N., Hori, H., Hamaguchi, S.,

Sakaizumi, M., 2002. DMY is a Y-specific DM-domain gene required for male

development in the medaka fish. Nature 417, 559–563. doi:10.1038/nature751

Matthews, J., Carmichael, C., 2015. ZIRC E400/RMMB sperm cryopreservation protocol

194

[WWW Document]. Zebrafish Int. Resour. Cent. URL

https://zebrafish.org/documents/protocols/pdf/Cryopreservation_IVF/zirc_rmmb_fre

ezing_protocol-2015-07-30.pdf (accessed 8.30.15).

Mayden, R.L., Tang, K.L., Conway, K.W., Freyhof, J., Chamberlain, S., Haskins, M.,

Schneider, L., Sudkamp, M., Wood, R.M., Agnew, M., Bufalino, A., Sulaiman, Z.,

Miya, M., Saitoh, K., He, S., 2007. Phylogenetic relationships of Danio within the

order Cypriniformes: a framework for comparative and evolutionary studies of a

model species. J. Exp. Zool. Part B Mol. Dev. Evol. 308B, 642–654.

doi:10.1002/jez.b.21175

McCluskey, B.M., Postlethwait, J.H., 2015. Phylogeny of zebrafish, a model species,

within Danio, a model genus. Mol. Biol. Evol. 32, 635–652.

doi:10.1093/molbev/msu325

McElligott, M.B., O’malley, D.M., 2005. Prey tracking by larval zebrafish: axial

kinematics and visual control. Brain. Behav. Evol. 66, 177–196.

doi:10.1159/000087158

McEntire, M., Riche, M., Beck, B.H., Carter, D., 2015. Effect of contrasting agents on

survival, performance, and condition of larval hybrid striped bass Morone chrysops

x M. saxatilis in tanks. J. Appl. Aquac. 27, 1–28.

doi:10.1080/10454438.2014.959814

McMenamin, S.K., Chandless, M.N., Parichy, D.M., 2016. Working with zebrafish at

postembryonic stages, in: Detrich, H.W., Westerfield, M., Zon, L.I. (Eds.), Methods

in Cell Biology. Academic Press, New York, pp. 587–607.

195

doi:10.1016/BS.MCB.2015.12.001

Mendel, G., 1865. Versuche über Plflanzenhybriden, in: Verhandlungen Des

Naturforschenden Vereines in Brünn, Bd. IV Für Das Jahr 1865. pp. 3–47.

Mizgireuv, I. V, Majorova, I.G., Gorodinskaya, V.M., Khudoley, V. V, Revskoy, S.Y.,

2004. Carcinogenic effect of N-nitrosodimethylamine on diploid and triploid

zebrafish (Danio rerio). Toxicol. Pathol. 32, 514–8.

Moehring, A.J., Llopart, A., Elwyn, S., Coyne, J.A., Mackay, T.F.C., 2006. The genetic

basis of postzygotic reproductive isolation between Drosophila santomea and D.

yakuba due to hybrid male sterility. Genetics 173, 225–33.

doi:10.1534/genetics.105.052985

Moore, E.C., Roberts, R.B., 2013. Polygenic sex determination. Curr. Biol. 23, R510–

R512. doi:10.1016/j.cub.2013.04.004

Moran, D., 2007. Size heterogeneity, growth potential and aggression in juvenile

yellowtail kingfish (Seriola lalandi Valenciennes). Aquac. Res. 38, 1254–1264.

doi:10.1111/j.1365-2109.2007.01769.x

Morgan, T.H., 1913. Heredity and sex. Columbia University Press, New York.

Morishima, K., Nakayama, I., Arai, K., 2001. Microsatellite-centromere mapping in the

loach, Misgurnus anguillicaudatus. Genetica 111, 59–69.

doi:10.1023/A:1013701128529

Motro, U., 1991. Avoiding inbreeding and sibling competition: The evolution of sexual

dimorphism for dispersal. Am. Nat. 137, 108–115. doi:10.1086/285148

Moyle, L.C., Graham, E.B., 2005. Genetics of hybrid incompatibility between

196

Lycopersicon esculentum and L. hirsutum. Genetics 169, 355–73.

doi:10.1534/genetics.104.029546

Moyle, L.C., Nakazato, T., 2010. Hybrid incompatibility “snowballs” between Solanum

species. Science (80-. ). 329, 1521–3. doi:10.1126/science.1193063

Mrakovcic, M., Haley, L.E., 1979. Inbreeding depression in the zebra fish Brachydanio

rerio (Hamilton Buchanan). J. Fish Biol. 15, 323–327. doi:10.1111/j.1095-

8649.1979.tb03612.x

Mulder, H.A., Hill, W.G., Vereijken, A., Veerkamp, R.F., 2009. Estimation of genetic

variation in residual variance in female and male broiler chickens. Animal 3, 1673–

1680. doi:10.1017/S1751731109990668

Muller, H.J., 1964. The relation of recombination to mutational advance. Mutat. Res. 1,

2–9. doi:10.1016/0027-5107(64)90047-8

Muller, H.J., 1932. Some genetic aspects of sex. Am. Nat. 66, 118–138.

doi:10.1086/280418

Myosho, T., Otake, H., Masuyama, H., Matsuda, M., Kuroki, Y., Fujiyama, A., Naruse,

K., Hamaguchi, S., Sakaizumi, M., 2012. Tracing the emergence of a novel sex-

determining gene in medaka, Oryzias luzonensis. Genetics 191, 163–170.

doi:10.1534/genetics.111.137497

Naas, K.E., N˦ss, T., Harboe, T., 1992. Enhanced first feeding of halibut larvae

(Hippoglossus hippoglossus L.) in green water. Aquaculture 105, 143–156.

doi:10.1016/0044-8486(92)90126-6

Nechiporuk, A., Poss, K.D., Johnson, S.L., Keating, M.T., 2003. Positional cloning of a

197

temperature-sensitive mutant emmental reveals a role for sly1 during cell

proliferation in zebrafish fin regeneration. Dev. Biol. 258, 291–306.

doi:10.1016/S0012-1606(03)00129-5

Nurse, P., Thuriaux, P., Nasmyth, K., 1976. Genetic control of the cell division cycle in

the fission yeast Schizosaccharomyces pombe. MGG Mol. Gen. Genet. 146, 167–

178. doi:10.1007/BF00268085

O’Connell, K.F., Leys, C.M., White, J.G., 1998. A genetic screen for temperature-

sensitive cell-division mutants of Caenorhabditis elegans. Genetics 149, 1303–

1321.

Ohno, S., 1967. Sex Chromosomes and Sex-Linked Genes, Monographs on

Endocrinology. Springer Berlin Heidelberg, Berlin, Heidelberg. doi:10.1007/978-3-

642-88178-7

Ohno, S., Muramoto, J., Klein, J., Atkin, N.B., 1969. Diploid-tetraploid relationship in

clupeoid and salmonoid fish, in: Darlington, C.D., Lewis, K.R. (Eds.),

Chromosomes Today. Oliver and Boyd, Edinburgh, pp. 139–147.

Okuda, N., 1999. Female mating strategy and male brood cannibalism in a sand-dwelling

cardinalfish. Anim. Behav. 58, 273–279. doi:10.1006/anbe.1999.1148

Örn, S., Holbech, H., Madsen, T.H., Norrgren, L., Petersen, G.I., 2003. Gonad

development and vitellogenin production in zebrafish (Danio rerio) exposed to

ethinylestradiol and methyltestosterone. Aquat. Toxicol. 65, 397–411.

doi:10.1016/S0166-445X(03)00177-2

Ospina-Álvarez, N., Piferrer, F., 2008. Temperature-dependent sex determination in fish

198

revisited: Prevalence, a single sex ratio response pattern, and possible effects of

climate change. PLoS One 3, e2837. doi:10.1371/journal.pone.0002837

Parichy, D., Johnson, S., 2001. Zebrafish hybrids suggest genetic mechanisms for

pigment pattern diversification in Danio. Dev. Genes Evol. 211, 319–328.

doi:10.1007/s004270100155

Parichy, D.M., Elizondo, M.R., Mills, M.G., Gordon, T.N., Engeszer, R.E., 2009. Normal

table of postembryonic zebrafish development: Staging by externally visible

anatomy of the living fish. Dev. Dyn. 238, 2975–3015. doi:10.1002/dvdy.22113

Paull, G.C., Filby, A.L., Giddins, H.G., Coe, T.S., Hamilton, P.B., Tyler, C.R., 2010.

Dominance hierarchies in zebrafish (Danio rerio) and their relationship with

reproductive success. Zebrafish 7, 109–117. doi:10.1089/zeb.2009.0618

Pelegri, F., Schulte-Merker, S., 1998. A gynogenesis-based screen for maternal-effect

genes in the zebrafish Danio rerio, in: Detrich W, Zon LI, Westerfield M (Eds.),

The Zebrafish: Genetics and Genomics. Academic Press, San Diego, pp. 1–20.

doi:10.1016/S0091-679X(08)61891-9

Pereira, A., Carvalho, A.P., Cruz, C., Saraiva, A., 2017. Histopathological changes and

zootechnical performance in juvenile zebrafish (Danio rerio) under chronic

exposure to nitrate. Aquaculture 473, 197–205.

doi:10.1016/j.aquaculture.2017.02.018

Perrin, N., Mazalov, V., 2000. Local competition, inbreeding, and the evolution of sex-

biased dispersal. Am. Nat. 155, 116–127. doi:10.1086/303296

Peruzzi, S., Rudolfsen, G., Primicerio, R., Frantzen, M., Kaurić, G., 2009. Milt

199

characteristics of diploid and triploid Atlantic cod (Gadus morhua L.). Aquac. Res.

40, 1160–1169. doi:10.1111/j.1365-2109.2009.02212.x

Postlethwait, J.H., Johnson, S.L., Midson, C.N., Talbot, W.S., Gates, M., Ballinger, E.W.,

Africa, D., Andrews, R., Carl, T., Eisen, J.S., 1994. A genetic linkage map for the

zebrafish. Science 264, 699–703.

Presgraves, D.C., 2003. A fine-scale genetic analysis of hybrid incompatibilities in

Drosophila. Genetics 163, 955–972.

Purdom, C.E., Thompson, D., Lou, Y.D., 1985. Genetic engineering in rainbow trout,

Salmo gairdnerii Richardson, by the suppression of meiotic and mitotic metaphase.

J. Fish Biol. 27, 73–79. doi:10.1111/j.1095-8649.1985.tb04010.x

Pyron, M., 2003. Female preferences and male–male interactions in zebrafish (Danio

rerio). Can. J. Zool. 81, 122–125. doi:10.1139/z02-229

Quillet, E., 1994. Survival, growth and reproductive traits of mitotic gynogenetic rainbow

trout females. Aquaculture 123, 223–236. doi:10.1016/0044-8486(94)90061-2

Quillet, E., Aubard, G., Quéau, I., 2002. Mutation in a sex-determining gene in rainbow

trout: Detection and genetic analysis. J. Hered. 93, 91–99.

doi:10.1093/jhered/93.2.91

R Core Team, 2017. R: A language and environment for statistical computing.

Recoubratsky, A. V., Grunina, A.S., Barmintsev, V.A., Golovanova, T.S., Chudinov,

O.S., Abramova, A.B., Panchenko, N.S., Kupchenko, S.A., 2003. Meiotic

gynogenesis in the stellate and russian sturgeons and sterlet. Russ. J. Dev. Biol. 34,

92–101. doi:10.1023/A:1023396213126

200

Reinhold, K., 1998. Nest-site philopatry and selection for environmental sex

determination. Evol. Ecol. 12, 245–250. doi:10.1023/A:1006591914859

Reitan, K.I., Rainuzzo, J.R., Øie, G., Olsen, Y., 1997. A review of the nutritional effects

of algae in marine fish larvae. Aquaculture 155, 207–221. doi:10.1016/S0044-

8486(97)00118-X

Ribas, L., Piferrer, F., 2014. The zebrafish (Danio rerio) as a model organism, with

emphasis on applications for finfish aquaculture research. Rev. Aquac. 6, 209–240.

doi:10.1111/raq.12041

Ribas, L., Valdivieso, A., Díaz, N., Piferrer, F., 2017a. On the proper rearing density in

domesticated zebrafish to avoid unwanted masculinization. Links with the stress

response. J. Exp. Biol. 220, 1056–1064. doi:10.1242/jeb.144980

Ribas, L., Valdivieso, A., Dı́az, N., Piferrer, F., 2017b. Response to “The importance of

controlling genetic variation - remarks on ‘Appropriate rearing density in

domesticated zebrafish to avoid masculinization: links with the stress response.’” J.

Exp. Biol. 220, 4079–4080. doi:10.1242/jeb.167437

Rice, W.R., 1992. Sexually antagonistic genes: Experimental evidence. Science (80-. ).

256, 1436–1439. doi:10.2307/2876947

Rice, W.R., 1986. On the instability of polygenic sex determination: the effect of sex-

specific selection. Evolution (N. Y). 40, 633–639. doi:10.1111/j.1558-

5646.1986.tb00514.x

Rieger, P.W., Summerfelt, R.C., 1997. The influence of turbidity on larval walleye,

Stizostedion vitreum, behavior and development in tank culture. Aquaculture 159,

201

19–32. doi:10.1016/S0044-8486(97)00187-7

Rodríguez-Marí, A., Cañestro, C., BreMiller, R.A., Nguyen-Johnson, A., Asakawa, K.,

Kawakami, K., Postlethwait, J.H., 2010. Sex reversal in zebrafish fancl mutants is

caused by Tp53-mediated germ cell apoptosis. PLoS Genet. 6, e1001034.

doi:10.1371/journal.pgen.1001034

Roemer, U., Beisenherz, W., 1996. Environmental determination of sex in Apistogramma

(Cichlidae) and two other freshwater fishes (Teleostei). J. Fish Biol. 48, 714–725.

doi:10.1111/j.1095-8649.1996.tb01467.x

Rottmann, R.W., Shireman, J.V., Lincoln, E.P., 1991. Comparison of three live foods and

two dry diets for intensive culture of grass carp and bighead carp larvae.

Aquaculture 96, 269–280. doi:10.1016/0044-8486(91)90157-3

Sae-Lim, P., Kause, A., Lillehammer, M., Mulder, H.A., 2017. Estimation of breeding

values for uniformity of growth in Atlantic salmon (Salmo salar) using pedigree

relationships or single-step genomic evaluation. Genet. Sel. Evol. 49, 33.

doi:10.1186/s12711-017-0308-3

Saint-Amant, L., Drapeau, P., 1998. Time course of the development of motor behaviors

in the zebrafish embryo. J. Neurobiol. 37, 622–632. doi:10.1002/(sici)1097-

4695(199812)37:4<622::aid-neu10>3.0.co;2-s

Salek, S.J., Godwin, J., Sullivan, C. V., Stacey, N.E., 2001. Courtship and tank spawning

behavior of temperate basses (genus Morone ). Trans. Am. Fish. Soc. 130, 833–847.

doi:10.1577/1548-8659(2001)130<0833:CATSBO>2.0.CO;2

Sarder, M.R.I., Penman, D.J., Myers, J.M., McAndrew, B.J., 1999. Production and

202

propagation of fully inbred clonal lines in the Nile tilapia (Oreochromis niloticus

L.). J. Exp. Zool. 284, 675–685. doi:10.1002/(SICI)1097-

010X(19991101)284:6<675::AID-JEZ9>3.0.CO;2-D

Schindelin, J., Arganda-Carreras, I., Frise, E., Kaynig, V., Longair, M., Pietzsch, T.,

Preibisch, S., Rueden, C., Saalfeld, S., Schmid, B., Tinevez, J.-Y., White, D.J.,

Hartenstein, V., Eliceiri, K., Tomancak, P., Cardona, A., 2012. Fiji: an open-source

platform for biological-image analysis. Nat. Methods 9, 676–682.

doi:10.1038/nmeth.2019

Schirone, R.C., Gross, L., 1968. Effect of temperature on early embryological

development of the zebra fish, Brachydanio rerio. J. Exp. Zool. 169, 43–52.

doi:10.1002/jez.1401690106

Schmidt, V.H., 1930. Speciesbastard zwischen Brachydanio rerio und Brachydanio

albolineatus. Das Aquarium 58–61.

Schradin, C., Lamprecht, J., 2000. Female-biased immigration and male peace-keeping in

groups of the shell-dwelling cichlid fish Neolamprologus multifasciatus. Behav.

Ecol. Sociobiol. 48, 236–242. doi:10.1007/s002650000228

Scott, G.R., Johnston, I.A., 2012. Temperature during embryonic development has

persistent effects on thermal acclimation capacity in zebrafish. Proc. Natl. Acad. Sci.

U. S. A. 109, 14247–52. doi:10.1073/pnas.1205012109

Selman, K., Wallace, R.A., Sarka, A., Qi, X., 1993. Stages of oocyte development in the

zebrafish, Brachydanio rerio. J. Morphol. 218, 203–224.

doi:10.1002/jmor.1052180209

203

Sheehan, R.J., Shasteen, S.P., Suresh, A. V., Kapuscinski, A.R., Seeb, J.E., 1999. Better

growth in all-female diploid and triploid rainbow trout. Trans. Am. Fish. Soc. 128,

491–498. doi:10.1577/1548-8659(1999)128<0491:BGIAFD>2.0.CO;2

Shelton, W.L., Rothbard, S., 2006. Exotic species in global aquaculture - a review. Isr. J.

Aquac. – Bamidgeh 58, 3–28.

Shinya, M., Sakai, N., 2011. Generation of highly homogeneous strains of zebrafish

through full sib-pair mating. G3 Genes, Genomes, Genet. 1, 377–86.

doi:10.1534/g3.111.000851

Siegfried, K.R., Nüsslein-Volhard, C., 2008. Germ line control of female sex

determination in zebrafish. Dev. Biol. 324, 277–287.

doi:10.1016/j.ydbio.2008.09.025

Singleman, C., Holtzman, N.G., 2014. Growth and maturation in the zebrafish, Danio

rerio: A staging tool for teaching and research. Zebrafish 11, 396–406.

doi:10.1089/zeb.2014.0976

Skinner, A.M.J., Watt, P.J., 2007. Strategic egg allocation in the zebra fish, Danio rerio.

Behav. Ecol. 18, 905–909. doi:10.1093/beheco/arm059

Sonesson, A.K., Ødegård, J., Rönnegård, L., 2013. Genetic heterogeneity of within-

family variance of body weight in Atlantic salmon (Salmo salar). Genet. Sel. Evol.

45, 41. doi:10.1186/1297-9686-45-41

Spence, R., Ashton, R., Smith, C., 2007. Oviposition decisions are mediated by spawning

site quality in wild and domesticated zebrafish, Danio rerio. Behaviour 144, 953–

966.

204

Spence, R., Smith, C., 2006. Mating preference of female zebrafish, Danio rerio, in

relation to male dominance. Behav. Ecol. 17, 779–783. doi:10.1093/beheco/arl016

Spence, R., Smith, C., 2005. Male territoriality mediates density and sex ratio effects on

oviposition in the zebrafish, Danio rerio. Anim. Behav. 69, 1317–1323.

doi:10.1016/j.anbehav.2004.10.010

Spitsbergen, J., Kent, M., 2003. The state of the art of the zebrafish model for toxicology

and toxicologic pathology research - advantages and current limitations. Toxicol.

Pathol. 31, 62–87. doi:10.1080/01926230390174959

Stanley, J.G., 1976. Production of hybrid, androgenetic, and gynogenetic grass carp and

carp. Trans. Am. Fish. Soc. 105, 10–16. doi:10.1577/1548-

8659(1976)105<10:pohaag>2.0.co;2

Stiver, K.A., Desjardins, J.K., Fitzpatrick, J.L., Neff, B., Quinn, J.S., Balshine, S., 2007.

Evidence for size and sex-specific dispersal in a cooperatively breeding cichlid fish.

Mol. Ecol. 16, 2974–2984. doi:10.1111/j.1365-294X.2007.03350.x

Streisinger, G., Singer, F., Walker, C., Knauber, D., Dower, N., 1986. Segregation

analyses and gene-centromere distances in zebrafish. Genetics 112, 311–319.

Streisinger, G., Walker, C., Dower, N., Knauber, D., Singer, F., 1981. Production of

clones of homozygous diploid zebra fish (Brachydanio rerio). Nature 291, 293–296.

doi:10.1038/291293a0

Strüssmann, C.A., Moriyama, S., Hanke, E., Calsina, J., Takashima, F., 1996. Evidence

of thermolabile sex determination in pejerrey. J. Fish Biol. 48, 643–651.

doi:10.1006/jfbi.1996.0064

205

Sturtevant, A.H., Mather, K., 1938. The interrelations of inversions, heterosis and

recombination. Am. Nat. 72, 447–452.

Szyper, J.P., 1989. Nutritional depletion of the aquaculture feed organisms Euterpina

acutifrons, Artemia sp. and Brachionus plicatilis during starvation. J. World Aquac.

Soc. 20, 162–169. doi:10.1111/j.1749-7345.1989.tb00570.x

Takahashi, H., 1977. Juvenile hermaphroditism in the zebrafish, Brachydanio rerio. Bull.

Fac. Fish. Hokkaido Univ. 28, 57–65.

Takatsu, K., Miyaoku, K., Roy, S.R., Murono, Y., Sago, T., Itagaki, H., Nakamura, M.,

Tokumoto, T., 2013. Induction of female-to-male sex change in adult zebrafish by

aromatase inhibitor treatment. Sci. Rep. 3, 227–263. doi:10.1038/srep03400

Tang, K.L., Agnew, M.K., Hirt, M.V., Sado, T., Schneider, L.M., Freyhof, J., Sulaiman,

Z., Swartz, E., Vidthayanon, C., Miya, M., Saitoh, K., Simons, A.M., Wood, R.M.,

Mayden, R.L., 2010. Systematics of the subfamily Danioninae (Teleostei:

Cypriniformes: Cyprinidae). Mol. Phylogenet. Evol. 57, 189–214.

doi:10.1016/j.ympev.2010.05.021

Taylor, M.I., Morley, J.I., Rico, C., Balshine, S., 2003. Evidence for genetic monogamy

and female-biased dispersal in the biparental mouthbrooding cichlid Eretmodus

cyanostictus from Lake Tanganyika. Mol. Ecol. 12, 3173–3177. doi:10.1046/j.1365-

294X.2003.01964.x

Tian, J., Yam, C., Balasundaram, G., Wang, H., Gore, A., Sampath, K., 2003. A

temperature-sensitive mutation in the nodal-related gene cyclops reveals that the

floor plate is induced during gastrulation in zebrafish. Development 130, 3331–

206

3342. doi:10.1242/dev.00544

Tiersch, T.R., Chandler, R.W., Wachtel, S.S., Elias, S., 1989. Reference-standards for

flow cytometry and application in comparative studies of nuclear-DNA content.

Cytometry 10, 706–710. doi:10.1002/cyto.990100606

Tiffin, P., Olson, M.S., Moyle, L.C., 2001. Asymmetrical crossing barriers in

angiosperms. Proc. R. Soc. B 268, 861–7. doi:10.1098/rspb.2000.1578

Triantaphyllidis, G., Abatzopoulos, T., Sorgeloos, P., 1998. Review of the biogeography

of the genus Artemia (Crustacea, Anostraca). J. Biogeogr. 25, 213–226.

doi:10.1046/j.1365-2699.1998.252190.x

Trivers, R.L., 1971. Parental investment and sexual selection, in: Campbell, B. (Ed.),

Sexual Selection and the Descent of Man. Aldine, Chicago, IL, pp. 136–179.

Turelli, M., Moyle, L.C., 2007. Asymmetric postmating isolation: Darwin’s corollary to

Haldane’s rule. Genetics 176, 1059–88. doi:10.1534/genetics.106.065979

Uchida, D., Yamashita, M., Kitano, T., Iguchi, T., 2004. An aromatase inhibitor or high

water temperature induce oocyte apoptosis and depletion of P450 aromatase activity

in the gonads of genetic female zebrafish during sex-reversal. Comp. Biochem.

Physiol. a-Molecular Integr. Physiol. 137, 11–20. doi:10.1016/s1095-

6433(03)00178-8

Uchida, D., Yamashita, M., Kitano, T., Iguchi, T., 2002. Oocyte apoptosis during the

transition from ovary-like tissue to testes during sex differentiation of juvenile

zebrafish. J. Exp. Biol. 205, 711–718.

Ueno, K., Arimoto, B., 1982. Induction of triploids in Rhodeus ocellatus ocellatus by

207

cold shock treatment of fertilized eggs. Experientia 38, 544–546.

doi:10.1007/BF02327041

Ulloa, P.E., Iturra, P., Neira, R., Araneda, C., 2011. Zebrafish as a model organism for

nutrition and growth: towards comparative studies of nutritional genomics applied to

aquacultured fishes. Rev. Fish Biol. Fish. 21, 649–666. doi:10.1007/s11160-011-

9203-0

Uusi-Heikkilä, S., Lindström, K., Parre, N., Arlinghaus, R., Alós, J., Kuparinen, A.,

2016. Altered trait variability in response to size-selective mortality. Biol. Lett. 12,

20160584. doi:10.1098/rsbl.2016.0584

Uusi-Heikkilä, S., Whiteley, A.R., Kuparinen, A., Matsumura, S., Venturelli, P.A.,

Wolter, C., Slate, J., Primmer, C.R., Meinelt, T., Killen, S.S., Bierbach, D.,

Polverino, G., Ludwig, A., Arlinghaus, R., 2015. The evolutionary legacy of size-

selective harvesting extends from genes to populations. Evol. Appl. 8, 597–620.

doi:10.1111/eva.12268 van Dongen, W.F.D., Wagner, R.H., Moodley, Y., Schaedelin, F.C., 2014. Sex biases in

kin shoaling and dispersal in a cichlid fish. Oecologia 176, 965–974.

doi:10.1007/s00442-014-3079-3

Vandeputte, M., Dupont-Nivet, M., Chavanne, H., Chatain, B., 2007. A polygenic

hypothesis for sex determination in the European sea bass Dicentrarchus labrax.

Genetics 176, 1049–1057.

Varadaraj, K., 1990. Production of diploid Oreochromis mossambicus gynogens using

heterologous sperm of Cyprinus carpio. Indian J. Exp. Biol. 28, 701–705.

208

Veyrunes, F., Chevret, P., Catalan, J., Castiglia, R., Watson, J., Dobigny, G., Robinson,

T.J., Britton-Davidian, J., 2010. A novel sex determination system in a close relative

of the house mouse. Proc. R. Soc. B 277, 1049–56. doi:10.1098/rspb.2009.1925

Walker, C., 1999. Haploid screens and gamma-ray mutagenesis. Methods Cell Biol. Vol

60 60, 43–70.

Walker, K.F., 1981. A synopsis of ecological information on the saline lake rotifer

Brachionus plicatilis Muller 1786. Hydrobiologia 81–82, 159–167.

doi:10.1007/BF00048713

Watts, S.A., Lawrence, C., Powell, M., D’Abramo, L.R., 2016. The vital relationship

between nutrition and health in zebrafish. Zebrafish 13, 72–76.

doi:10.1089/zeb.2016.1299

Watts, S.A., Powell, M., D’Abramo, L.R., 2012. Fundamental approaches to the study of

zebrafish nutrition. ILAR J. 53, 144–60. doi:10.1093/ilar.53.2.144

Weismann, A., 1889. Essays upon heredity and kindred biological problems. Clarendon

Press, Oxford.

Williams, G.C., 1975. Sex and evolution. Princeton University Press, Princeton, NJ.

Wilson, C.A., High, S.K., McCluskey, B.M., Amores, A., Yan, Y.L., Titus, T.A.,

Anderson, J.L., Batzel, P., Carvan, M.J., Schartl, M., Postlethwait, J.H., 2014. Wild

sex in zebrafish: Loss of the natural sex determinant in domesticated strains.

Genetics 198, 1291–1308. doi:10.1534/genetics.114.169284

Wong, T.-T., Saito, T., Crodian, J., Collodi, P., 2011. Zebrafish germline chimeras

produced by transplantation of ovarian germ cells into sterile host larvae. Biol.

209

Reprod. 84, 1190–1197. doi:10.1095/biolreprod.110.088427

Wright, S., 1934. An analysis of variability in number of digits in an inbred strain of

guinea pigs. Genetics 19, 506–536.

Wright, S., 1921. Systems of mating V. General considerations. Genetics 6, 167–178.

Xu, D., You, F., Wu, Z., Li, J., Ni, J., Xiao, Z., Zhang, P., Xu, Y., 2009. Genetic

characterization of asymmetric reciprocal hybridization between the flatfishes

Paralichthys olivaceus and Paralichthys dentatus. Genetica 137, 151–158.

doi:10.1007/s10709-009-9373-z

Yano, A., Guyomard, R., Nicol, B., Jouanno, E., Quillet, E., Klopp, C., Cabau, C.,

Bouchez, O., Fostier, A., Guiguen, Y., 2012. An immune-related gene evolved into

the master sex-determining gene in rainbow trout, Oncorhynchus mykiss. Curr. Biol.

22, 1423–1428. doi:10.1016/j.cub.2012.05.045

Yano, A., Nicol, B., Jouanno, E., Quillet, E., Fostier, A., Guyomard, R., Guiguen, Y.,

2013. The sexually dimorphic on the Y-chromosome gene (sdY) is a conserved

male-specific Y-chromosome sequence in many salmonids. Evol. Appl. 6, 486–496.

doi:10.1111/eva.12032

Yant, R., Smitherman, R.O., Green, O.L., 1975. Production of hybrid (blue X channel)

catfish and channel catfish in ponds. Proc. Annu. Conf. Southeast. Assoc. Game

Fish Comm. 29, 86–91.

Yossa, R., Sarker, P.K., Proulx, É., Vandenberg, G.W., 2015. The effects of the dietary

biotin on zebrafish Danio rerio reproduction. Aquac. Res. 46, 117–130.

doi:10.1111/are.12166

210

Yusa, Y., 2007. Nuclear sex-determining genes cause large sex-ratio variation in the

apple snail Pomacea canaliculata. Genetics 175, 179–84.

doi:10.1534/genetics.106.060400

211