Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Genome-wide analysis of genetic load and larval mortality in a highly fecund marine invertebrate, the Pacific Oyster Crassostrea gigas
(USC Thesis Other)
Genome-wide analysis of genetic load and larval mortality in a highly fecund marine invertebrate, the Pacific Oyster Crassostrea gigas
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
GENOME-WIDE ANALYSIS OF GENETIC LOAD AND LARVAL MORTALITY IN A HIGHLY FECUND MARINE INVERTEBRATE, THE PACIFIC OYSTER CRASSOSTREA GIGAS by Louis Valentine Plough A Dissertation Presented to the FACULTY OF THE USC GRADUATE SCHOOL UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements for the Degree DOCTOR OF PHILOSOPHY (BIOLOGY) December 2011 Copyright 2011 Louis Valentine Plough ii Acknowledgements This dissertation would not have been possible without a number of people who generously gave me their time, effort, and inspiration over the course of my time in the graduate program at USC. First I would like to thank my advisor, Dennis Hedgecock, for countless insightful discussions, the intellectual passion and energy that helped keep me motivated when I needed it, and for being a fantastic scientific mentor. I would also like to thank the other members of my dissertation committee whose guidance and critiques have significantly improved my writing and research; members of the Hedgecock lab for their assistance in the lab and for providing open minds on rough drafts, talks, and other newly hatched ideas; Jason Curole and Eli Meyer for insightful discussions and valuable training early in my graduate career; Joth Davis at Taylor Shellfish Farms (and the Taylor research lab) for providing oyster broodstock, larvae, and basically anything and everything I needed for the experiments that made up my thesis; the Wrigley Institute for long term summer funding and lab space on Catalina Island; and Linda Bazilian and Don Bingham in the MEB program for their help and guidance for everything large and small. Finally, I would like to thank my family for support and encouragement, especially Aileen, who has been with me at each step of the way, from marine labs to LA and beyond. iii Table of Contents Acknowledgements ii List of Tables iv List of Figures v Abstract vii Introduction: Genetic load in highly fecund marine organisms. 1 Introduction References 16 Chapter 1: Genome-wide analysis of stage-specific inbreeding Depression in the Pacific oyster Crassostrea gigas. 22 Chapter 1 References 66 Chapter 2: Fine-scale temporal analysis of mortality and genetic load during metamorphosis in the Pacific oyster Crassostrea gigas. 73 Chapter 2 References 100 Chapter 3: Inbreeding-by-environment interaction affects the expression of genetic load in the Pacific oyster Crassostrea gigas. 105 Chapter 3 References 153 Chapter 4: Mutations cause massive mortality in wild full-sib families of the Pacific oyster Crassostrea gigas. 161 Chapter 4 References 203 Bibliography 212 Appendix: Quantitative trait locus analysis of settlement timing in the Pacific oyster reveals significant genetic variance in early vs. late settlement. 233 Appendix References 255 iv List of Tables Table 1: Day 700 segregation results for two F 2 families 36 Table 2: Viability QTL, selection, dominance, and timing of selection at nearest markers, in families 46×10 and 51×35 43 Table 3: Selective mortality estimates for each family based on viability QTL results 53 Table 4: Marker segregation data for day 18 larvae and day 60 spat 89 Table 5: Genotype numbers for Cg205 during settlement 92 Table 6: Algal feeding treatments for the two experiments 113 Table 7: Counts of distorted markers at the α= 0.05 level arranged by cross types and diet (1-algal or 3-algal) in the two families 126 Table 8: Mortality and QTL results for 07×5 136 Table 9: Mortality and QTL results for 08×3 138 Table 10: Two-locus model results for families 07×5 and 08×3 141 Table 11: Information for 11 novel SNP markers 169 Table 12: General segregation results in the progeny of the four wild crosses 177 Table 13: Viability QTL results, genetic mortality, and genetic effects 183 Table 14: Epistasis results 188 Table 15: Segregation data for day 19 and day 24 spat samples 241 Table 16: Settlement timing QTL and ANOVA model results 247 v List of Figures Figure 1: Pedigree of inbred lines and crosses used in the dissertation 12 Figure 2: QTL Mapping results for family 46×10 and 51×35 40 Figure 3: Goodness-of-fit chi-square P-values vs. time for markers on linkage group 3 49 Figure 4: Distribution of viability loci expression across life stages of the Pacific oyster 51 Figure 5: Explanation of two linked viability loci, underlying a single QTL peak 59 Figure 6: Relative survival vs. genetic mortality during the life cycle of the Pacific oyster 64 Figure 7: Pictures of shell deployment and bags with spat for grow-out. 81 Figure 8: Settlement and Mortality data 87 Figure 9: Cg205 genotype proportions in larvae and spat during settlement 93 Figure 10: Survival data for 07×5 and 08×3 123 Figure 11: Growth data for family 07×5 and 08×3 125 Figure 12: QTL mapping results in the 1-algal and 3-algal diets for family 07×5 130 Figure 13: QTL mapping results in the 1-algal and 3-algal diets for family 08×3 133 Figure 14: Model of enzyme flux with dominance in normal (A) and stressful (B) environments 149 Figure 15: QTL mapping results for family 12 179 vi Figure 16: QTL mapping results for family 20 180 Figure 17: QTL mapping results for family 24 181 Figure 18: QTL mapping results for family 45 182 Figure 19: Heat map of epistasis results for family 45 187 Figure 20: Genotype numbers at two markers illustrating the effects of two deleterious alleles 193 Figure 21: Allelic effects on linkage group 4, family 45 195 Figure 22: Settlement timing data for the F 2 family 236 Figure 23: Plot of genotype effects on settlement timing for Cg140 249 Figure 24: Plot of genotype effects on settlement timing for Cg205 and Cg109 250 vii Abstract Most marine fish and shellfish are very fecund, with females often producing many millions of eggs per spawn (Thorson 1950, Winemiller and Rose 1992). Their larval offspring characteristically suffer high (type-III) mortality, while developing for days to weeks in the plankton, but this mortality is largely attributed by most marine scientists to environmental factors. The biological and ecological characteristics of this life history mode have profound implications for population structure, adaptation, and recruitment variation, yet relatively little is known about the role of endogenous variation during the early life stages. Experimental inbred crosses of the highly fecund Pacific oyster Crassostrea gigas have previously revealed a high genetic load, which causes substantial genotype-dependent mortality, explains widespread observations of segregation distortion in crosses of other highly fecund marine bivalves, and fits predictions (Williams 1975) that high fecundity species should suffer substantial inbreeding depression (Launey and Hedgecock 2001, Bucklin 2003). Though these previous studies provided important experimental verification of high genetic load in the oyster, they left unanswered a number of questions about the genetic basis of this load and whether it could explain the high, early life-history mortality observed in the oyster. These questions can now be addressed, thanks to the advent of genome-wide statistical approaches, such as quantitative-trait locus (QTL) mapping. In this dissertation, a series of experiments were performed to address these and other questions, by creating experimental inbred and outbred families and analyzing molecular marker segregation viii data across the genome to detect the number, location, and effect of deleterious mutations in the Pacific oyster. There were four major findings. First, inbred families exhibited 10- 15, mainly recessive or partially dominant, deleterious mutations, with little evidence for epistatic interaction between them. Genotype-dependent mortality calculated from the multiplicative fitness effects of these deleterious loci was ~96%, accounting for nearly all observed actual mortality. Second, the expression of genetic load occurred primarily during the larval stages in inbred crosses, with half of all deleterious mutations expressed during settlement and metamorphosis; further analysis revealed that mutations expressed at metamorphosis occurred over a relatively small temporal window, just prior to and during metamorphosis, but not after. Third, the environment significantly affected the fitness of deleterious alleles; dominance of and selection against some deleterious alleles increased significantly in a stressful, nutrient-deficient algal diet, while other alleles were un-affected, suggesting that inbreeding depression is caused by different loci in different environments. Fourth, seven to nine deleterious mutations were detected in wild families, causing an estimated 87-98% cumulative genetic mortality. Overall, there appears to be remarkable genetic variation in viability during the early life-stages of the Pacific oyster, causing much of the mortality that is observed in wild and inbred cultures. These results suggest that larvae of highly fecund marine animals in the natural environment may also be subject to substantial genotype-dependent mortality, which possibly contributes to the high recruitment variation and fluctuations in abundance of fished populations. The results of the dissertation also have broad implications for ix understanding the genetic basis of inbreeding depression, conservation genetics, and shellfish aquaculture. 1 Introduction Genetic load in highly fecund marine organisms The evolution of marine metazoans has produced a variety of complex life history strategies, which are critical for understanding the biology and ecology of these organisms. Life history strategies in marine animals often include a dispersive larval stage that spends days to weeks or months in the water column, but species vary substantially in the duration of the larval period, the energy investment per offspring, and the process of settlement and metamorphosis (e.g. Strathmann 1985, Stearns 1980, Giagrande et al. 1994, Llodora 2003). Two major life history modes are recognized in marine animals and the differences between them depend largely on energy allocation, which has adaptive significance. The first strategy, often considered the K-strategy in r- K selection theory (e.g. MacArthur and Wilson and 1967), is characterized by relatively high parental investment in offspring, which results in the production of fewer eggs, but with the provisioning of greater energy content and larger size at the outset of larval life. The second general strategy (r-selected) is characterized by the production of a relatively large number of eggs (higher fecundity), but with less energy per offspring, and smaller size at the outset of the larval period. These two strategies represent two ends of the life history continuum and illustrate important tradeoffs in energy allocation between growth and fecundity, which result in substantial biological and ecological differences between species. For example, in high fecundity, ―r-selected‖ species, smaller, feeding-larvae 2 may require more time in the plankton to reach maturity, which results in potentially longer larval dispersal distances and increased likelihood of larval mortality (e.g. Thorson 1950, Strathmann 1985, Caley et al. 1996, Llodora 2003). In contrast, species with direct developing larvae, or non-feeding lecithotrophic forms, spend less time in the plankton and thus exhibit smaller dispersal distances and reduced larval mortality. The length of larval period is particularly important because it affects population genetics, connectivity, and population dynamics of species (Cowen et al. 2007). Planktonic (feeding) species largely show low genetic differentiation across populations with little local adaptation, which contrasts with direct developing species that, in general, show greater genetic divergence and local adaptation (Palumbi 1994, 1995, Caley et al. 1996). The duration of larval period and the higher fecundity associated with it, may also affect short term population dynamics. Species with higher fecundity and planktonic larvae show greater year to year variation in population abundances, while species with non- feeding and direct developing larvae or larvae that spend less time in the water column have more stable population sizes over time (e.g. Thorson 1950, Cushing 1971, Rickman et al. 2000). In summary, mode of larval development and tradeoffs between fecundity and energy provision in reproduction are important for the short- and long-term evolutionary dynamics of marine animals. Of course, these tradeoffs exist for terrestrial animals as well, but what is unique about marine animals is the sheer scale of their fecundity (up to 10 8 eggs per reproductive event), which is unknown for terrestrial animals and matched only by certain long-lived plants (e.g. conifers). Furthermore, somewhere between 50 and 80% of marine invertebrates (Thorson 1950) and the majority 3 of marine teleosts (Winemiller and Rose 1992) share this high fecundity life-history strategy, making it a very important and evolutionarily significant life-history strategy in the ocean. Demographic models of life history in the ocean While the energetic trade-offs associated with fecundity are central to understanding the evolution of life history in marine animals, demographic factors are also important and provide critical insight into the population dynamics and genetics of highly fecund marine animals. Early efforts to describe life history evolution in the ocean relied on the deterministic r-K selection framework (MacArthur and Wilson 1967, Pianka 1970) derived from terrestrial animal systems. In these models, the density of a population, with respect to resources, is the main pressure that explains the selection for major life history traits. While the r-K model provides somewhat realistic predictions about energy allocation relative to environmental conditions, it is not adequate to characterize life history strategies of marine animals for a number of important reasons. First, the r-K model does not appropriately address species that are long lived, late- maturing, and have extremely high fecundity, which again is one of the most common strategies for marine invertebrates and many teleost marine fish in the ocean (Thorson 1950, Winemiller and Rose 1992, Winemiller and Rose 1993). The deterministic r-K model also fails to incorporate environmental variability and stochasticity (e.g. non- uniformity of food availability), which is common in marine environments. Finally, the model is centered on the concept of population regulation and does little to address the 4 demographic aspects of the population including age-specific mortality and age-specific fecundity, which play central roles in the population dynamics of these species and contribute to the evolution of life history traits. It wasn‘t until the emergence of demography as a central parameter in life history models that life history strategies in the ocean were appropriately considered. In their demographic-based model, Winemiller (1992) and Winemiller and Rose (1992; also see Stearns 1980, 1992) proposed that fecundity, survivorship of juveniles, and age at maturity were the three main components of fitness affecting life history, and tradeoffs among these three traits (often optimization of one component at the expense of the other two) shape the strategies that we observe in marine animals. The resulting strategies are described by a three dimensional trait continuum that splits the ―r‖ strategy into the periodic strategy (high fecundity, long living, late maturation) and the opportunistic strategy (low to medium fecundity, quick maturation, short life span) and the ―K‖ strategy is termed the equilibrium strategy (long lived, low fecundity, high parental investment). Importantly, this model distinguishes the periodic strategy from the opportunistic strategy, and highlights the importance of high larval and juvenile mortality in the ―episodic‖ life history strategy, which is widely observed in marine invertebrates and many commercially important marine fish. Implications of high fecundity life history strategies in the ocean High fecundity in marine animals has a number of important demographic and genetic implications. It has long been recognized that reproductively prolific marine fish 5 such as Atlantic cod, have highly variable population sizes over space and time and that stock abundances are largely independent of, and even negatively associated with, high fecundity (Hjort 1914, Cushing 1971, Williams 1975, Rickman et al. 2000). In his classic review, Thorson (1950) similarly showed, for marine invertebrates, the positive relationship between high fecundity and high variation in population abundance, compared with lower fecundity species with lecithotrophic or direct developing larvae that exhibited more constant and predictable recruitment and abundances. Understanding variation in recruitment, the surviving numbers of individuals from the larval stage to a particular point in juvenile development, has been a critical area of research in marine ecology and fisheries management for over a century (e.g. Hjort 1914, Cushing 1971). Certainly, high mortality during the larval stages of these species must play a role in the overall population variability; Thorson (1950) noted, for example, the high ―waste‖ of eggs and larvae that existed in species with such high fecundity. Recent reviews also highlight the substantial mortality in the early life-history stages observed in numerous field studies of recruitment and settlement (e.g. Hunt and Shiebling 1997, Gosselin and Qian 1997). Though few good estimates of the magnitude of larval mortality in the field exist, Korringa (1941) found, in field observations of wild swarms of the European flat oyster larvae Ostrea edulis off the coast of Holland, that approximately 10% of oyster embryos survive to become mature larvae, and of those, only 5-10 % successfully survive metamorphosis. High mortality during the larval stages of highly fecund marine animals has largely been attributed to predation, lack of food availability, and oceanographic factors such as currents and temperature (Thorson 1950, Peterman and Bradford 1987, 6 Roughgarden et al. 1988, Gosselin and Qian 1997, Hunt and Schiebling 1997). Though environmental sources of mortality no doubt are important in structuring populations and affect recruitment variability, selective differences in the viability of individuals may also explain high mortality at the early stages. G. C. Williams (1975), in his Elm-Oyster Model, first suggested that species with high fecundity might have large variance in adaptive performance of offspring, and he suggested that the diversity of genotypes resulting from the generation of millions of progeny should result in substantial selection during the vulnerable early life-history stages. He also predicted that species with high fecundity would carry a high genetic load and exhibit substantial inbreeding depression for fitness and physiological traits. Evidence of differential viability in seedlings of high fecundity pines existed at the time of Williams‘ initial hypothesis; for example, it was shown that pines had a substantial number of embryonic lethals expressed upon self-fertilization (e.g. Franklin 1972, Bannister 1965) and more recent findings have confirmed this high load (e.g. Savolainen et al.1992, Remington and O‘Malley 2000). Fewer data on genetic load exist for highly fecund marine animals, but evidence has been accumulating. Initially puzzling observations of high frequencies of null alleles and distortions of Mendelian segregation ratios for allozyme and microsatellite markers in families of marine bivalves (e.g. Beaumont et al. 1983, Mallet et al. 1985, Foltz 1986a,b, Beaumont et al. 1991, Gaffney 1993, McGoldrick et al. 1997, McGoldrick et al. 2000) were early indications of a high genetic load (Launey and Hedgecock 2001). A study of four microsatellite markers in inbred crosses of European flat oyster also demonstrated significant deficiencies of 7 homozygous genotypes at the larval stage, owing to selection against recessive deleterious alleles (Bierne et al. 1998). Studies of the Pacific oyster agree with these findings, showing substantial inbreeding depression (the expression of genetic load) for larval survival and growth (e.g. Hedgecock et al. 1995). Similar experimental studies of the fitness effects of genetic load in high fecundity marine fish are non-existent, but substantial variation in growth rate, behavior, survival and other performance traits has been widely observed in experimental studies of individual larvae and among batches of wild larvae in marine fish (e.g. Houde 1996,Cowan et al. 1996, Cowan et al. 1997, Hare and Cowen 1997, Heath and Gallego 1997, Houde 2002, Fuiman and Cowan 2003), which suggests great variation in viability of these early life stages, and that a high genetic load may also be present. Genetic load in the Pacific oyster: Background and rationale for dissertation research Following up on Williams‘ (1975) prediction of high genetic load in the oyster and the intriguing genetic findings of null alleles and segregation distortion in marine bivalves, Launey and Hedgecock (2001) took a molecular marker-based approach to examine the expression of genetic load in F 2 families of the Pacific oyster. They examined the segregation patterns of microsatellite markers in the progeny of inbred crosses, looking for distortion of marker segregation ratios, as an indication of selection against nearby, linked, deleterious mutations. Launey and Hedgecock (2001) found that oysters possessed a large load of deleterious recessive mutations (8-14 in the wild 8 founder individuals) causing severe distortions of segregation ratios at half of the marker loci assayed. Bucklin (2003) followed up on this work and confirmed an average of 12 highly deleterious loci, using more markers and a linkage map in a number of F 2 families. Bucklin (2003) also demonstrated that these deleterious alleles were inherited, (i.e. not synergistic), and found that genetic load expression was focused during larval development. While these studies served to characterize the high number of deleterious mutations in the Pacific oyster and provided a baseline of important experimental data for future studies of genetic load in this species, they left a number of questions unanswered and raised new questions that are now possible to address with new statistical approaches. For example, the finding of early stage effects of genetic load is especially intriguing, and begs for more detailed study of the stage-specific expression of this load. Furthermore, the statistical methods used to make genome-wide inferences about genetic load were limited to single-marker or two-locus analyses. More comprehensive, genome-wide models and statistical approaches are now available, which greatly enhance our ability to characterize genetic load in this species. Extending upon the methods used and the questions posed by previous studies, this dissertation addresses a number long-standing, novel, and important aspects of genetic load in the Pacific oyster, with particular insight into the connection between genetic load and high mortality during the early life history stages of high fecundity marine animals. Some of the questions that remain unanswered from previous work are: How many viability genes are there on average in the genome of the oyster, and where in the genome are they located? Launey and Hedgecock (2001) used a limited number of markers, with little knowledge of their linkage or genomic 9 locations, and Bucklin (2003) had only preliminary genetic maps and no comprehensive (QTL) mapping methods. When, precisely, during development, does viability selection occur and does it affect certain life stages or life history transitions more than others? Bucklin (2003) localized the temporal expression of deleterious loci to the larval stages, but sampling within this life history stage was sparse. Are the deleterious mutations causing viability selection and segregation ratio distortion unique to inbred crosses or would one see them in progeny from crosses of wild-caught parents? If so, what is their effect on viability in wild crosses? In the course of research into these standing questions, new questions surfaced, such as: What are the patterns and timing of selection against deleterious mutations during metamorphosis? Does the expression of genetic load depend on the environment? These questions and others are addressed in the dissertation, which is organized as follows: Chapter 1 examines the stage-specific expression of viability loci in two F 2 families, by genotyping viability QTL-associated microsatellite markers at multiple time- points throughout larval, juvenile, and adult development. This chapter marks the introduction of the QTL based, ‗viability selection model‘ (Luo and Xu 2003; reviewed below), which forms the foundation for similar analyses across most of the other chapters. This chapter also analyzes the connection between mortality caused by deleterious loci and mortality observed in inbred cultures. Chapter 2 builds on the temporal analysis of chapter 1, focusing in on selection and mortality during metamorphosis in an inbred, F 2 cross, through careful, daily sampling of settlers and larvae remaining in the water column. Chapter 3 examines the effect of a stressful environment on the expression of individual viability QTL in an F 2 family split and 10 reared in two environments differing in micro-algal diet and nutritional value. Chapter 4 examines, for the first time, viability selection in four wild pair-crosses, using the QTL viability model to estimate the magnitude of genotype-dependent mortality in these families. Finally, in the appendix I examine genetic variance in settlement timing through a traditional QTL analysis of early versus late time to settlement in an F 2 population. General approach and review of marker-based methods for analyzing genetic load The general experimental approach of the dissertation involves the creation of inbred and outbred experimental families and the construction of family-specific genetic linkage maps with molecular marker segregation data. Genetic linkage analysis provides information on the relative location of genetic markers throughout the genome (statistical linkage between markers indicates that they are located on the same chromosome, while un-linked markers are on different chromosomes) and with this information, the differential fitness effects of linked, deleterious loci (those loci that make up the genetic load) can be inferred within each family. Genomic resources, such as the large number of molecular markers and moderately dense genetic linkage maps, are now available for the Pacific oyster (Hubert and Hedgecock 2004, Hedgecock et al. 2005, Hubert et al. 2009), which facilitate such analyses. The availability of well documented pedigreed inbred lines is critical for these experiments, as well. Figure 1 displays the pedigrees of all inbred and outbred lines used in this dissertation through the F 2 crossing design. The 11 goal of the F 2 crossing design is to create families segregating for identical-by-descent deleterious alleles, and to maximize linkage disequilibrium (LD) between deleterious loci and genetic markers. This is accomplished first through a number of generations of sib- matings, which increases the inbreeding coefficient of lines and, thus, the probability of producing identical-by-descent homozygotes in the progeny. This is followed by intercrossing two inbred lines to produce a hybrid family, which is then sib-mated for a single generation, creating an F 2 in which deleterious alleles from both lines are segregating. Finally, the statistical analysis of marker segregation data in these families, from which one ultimately infers the fitness effects of deleterious loci, is a critical methodology of the dissertation. The novel viability QTL methods described below represent a major technical advance in these analyses. Statistical methods to analyze inbreeding depression and genetic load from molecular marker segregation data have evolved steadily over the last two decades and currently, these methods allow a much more comprehensive analysis than at the outset of their development. One of the first models developed was the linked-selection model, or the ―two- locus‖ model for detecting the fitness effects of a linked deleterious allele on marker genotypes in selfing organisms (Hedrick and Muona 1990; though Sorenson (1967) appears to be the first to consider a linked recessive lethal model to explain the observation of segregation ratios differing from 3:1 for marker-gene expectations in recessive traits). Under the ―two-locus model‖, which was subsequently extended to inbred (non-selfing) F 2 populations with varying levels of dominance (Fu and Ritland 1 Figure 1 Pedigree of inbred lines and crosses used in the dissertation Generations of inbreeding, G 0 (no inbreeding) to G 4 (four generations of inbreeding) are displayed for the inbred lines at each successive cross alongside the line name (51, 35, 46, or 10) to aid in the following of inbred lines, where necessary; F 1 crosses are marked as well. Parents connected by dashed lines are the same individual, used in two crosses. Inbreeding coefficient, f, of each F 2 family is displayed below each cross, along with the dissertation chapter in which the cross was used. 12 13 1994, Launey and Hedgecock 2001), the effects of a deleterious allele at a fitness locus, which had been made homozygous by inbreeding, is determined by its apparent fitness effect on linked markers, through observations of non-Mendelian inheritance of marker genotypes. The fitness of the marker genotypes depends on the recombination distance between the marker and the fitness gene (c), selection against the deleterious allele (s) and the dominance (h) of the deleterious allele. For example, in a cross of a marker with two alleles (AB×AB), the linked fitness gene with alleles l (lethal) and + (wild-type), the relative fitnesses [in brackets] of the multi-locus genotypes in the progeny, assuming no recombination, are: BB / ++ [1], AB / l+ [1-hs], and AA / ll [1-s]. Using a maximum- likelihood framework, Hedrick and Muona (1990) estimated, simultaneously, the selection coefficient (s) of this deleterious gene and the map distance (c) between it and the marker, which would account for the observed departures from Mendelian segregation. However, under the two locus model, power is low when markers are far from a deleterious fitness gene and the effects of multiple linked deleterious loci can bias estimates of selection and gene action. Fu and Ritland (1994) found that using sets of two flanking markers vastly improved the localization and estimation of s and thus determined that more advanced methods were needed to characterize deleterious loci at the genome-wide level. More recent molecular marker methods have sought to incorporate the genome- wide perspective of quantitative trait locus (QTL) analysis into the search for deleterious loci affecting viability. The statistical framework behind QTL analysis involves testing for significant associations between flanking DNA markers and a quantitative trait of 14 interest (e.g. height or grain yield), scanned across relatively small increments of the genome (interval mapping). Mitchell-Olds (1995) developed a statistical model that scanned the genome along each marker interval testing for the presence of a putative fitness locus that caused deviations from Mendelian segregation (viability selection) at linked marker loci. QTL methods such as those described in Mitchell-Olds (1995; also see Cheng et al. 1996) required the establishment of a linkage map with known recombination distances, which was, at the time, rather difficult to achieve for non-model organisms. Nevertheless, this methodology greatly improved previous efforts to map viability loci because all marker data could be evaluated simultaneously, reducing the bias of linked viability genes and improving the power to position viability loci. Following this work, Vogl and Xu (2000) developed a model for backcross and selfing populations that used Bayesian inference and QTL methods and finally, Luo and Xu (2003) and Luo et al. (2005) developed more general models for F 2 and out-bred crosses, which were able to estimate selection and gene-action, in addition to the location and number of viability QTL. The viability QTL (vQTL) methods described above also make possible another important technical advance of the dissertation, the calculation of genetic mortality. Used in chapters 1, 3, and 4, this calculation quantifies the average survival at each vQTL, based on the relative fitnesses of each genotype. Applied across all QTL, and corrected for sampling error, this calculation estimates how much mortality is attributable to the expression of genetic load. The genetic mortality calculation in turn, is dependent on applying tests of epistasis across the genome (chapters 1, 3, and 4), and specifically 15 determining the independence of viability QTL, so that a calculation of multiplicative survival can be justified. 17 Introduction References Bannister, M. H., 1965 Variation in the breeding system of Pinus radiata. In: The Genetics of Colonizing Species (pp.353-374), H. G. Baker and G. L. Stebbins, editors, New York, Acadmic, xv + 588 pp. Beaumont, A. R., 1991 Genetic studies of laboratory reared mussels, Mytilus edulis: heterozygote deficiencies, heterozygosity and growth. Biol. J. Linn. Soc. 44: 273-285. Beaumont, A. R., C.M. Beveridge, and M. D. Budd, 1983 Selection and heterozygosity within single families of the mussel, Mytilus edulis (L.) Mar. Biol. 4: 151-161. Bierne, N., S. Launey, Y. Naciri-Graven, and F. Bonhomme, 1998 Early effect of inbreeding as revealed by microsatellite analyses on Ostrea edulis larvae. Genetics 148: 1893-1906. Bucklin, K. A., 2003 Analysis of the Genetic Basis of Inbreeding Depression in the Pacific Oyster Crassostrea gigas. PhD Dissertation, University of California at Davis. Caley, M., M. Carr, M. Hixon, T. Hughes, G. Jones and B. Menge, 1996 Recruitment and the local dynamics of open marine populations. Annual Review of Ecology and Systematics, 27: 477 –500. Carr, D.E., and M. R. Dudash 2003 Recent approaches into the genetic basis of inbreeding depression in plants. Phil. Trans. Roy. Soc. L. B 358:1071-1084. Cheng, R., A. Saito and Y. Ukai, 1996 Estimation of the position and effect of a lethal factor locus on a molecular marker linkage map. Theor. Appl. Genet. 93: 494–502 Cowan, J. H., Jr,, E. D. Houde, and K. A. Rose, 1996 Size dependent vulnerability of marine fish larvae to predation: an individual-based numerical experiment. ICES J. Mar. Sci. 53: 23-37. Cowan, J. H. Jr., K. A. Rose, and E. D. Houde, 1997 Size based foraging success and models of larval fish populations. Pages 357-386 in, Early life history and recruitment in fish populations, R. C. Chambers and E. A. Trippel, editors Chapman and Hall, London, UK. 18 Cowen, R. K., G. Gawarkiewicz, J. Pineda, S. R. Thorrold, F.E. Werner, 2007 Population connectivity in marine systems: An Overview. Oceanography 20(3):14-21. Crowder, L. W., J. A. Rice, T. J. Miller, and E. A. Marschall, 1992 Empirical and theoretical approaches to size-based interactions and recruitment variability in fishes. Pages 237-255 in, Individual-based models and approaches in ecology, D. L. DeAngelis and L. J. Gross, editors. Chapman and Hall, New York, New York, USA. Cushing, D. H., 1971 The dependence of recruitment on parent stock in different groups of fishes. J. du Conseil Intern. Expl. Mer. 33: 340-362. Franklin, E. C., 1972 Genetic load in the loblolly pine. Amer. Naturalist 106: 262-265. Fu, Y.B. and K. Ritland, 1994 On estimating linkage of marker genes to viability genes controlling inbreeding depression. Theoretical and Applied Genetics 88: 925-93 Foltz, D. W., 1986a Null alleles as a possible cause of heterozygote deficiencies in the oyster Crassostrea virginica and other bivalves. Evolution 40: 869-870. Foltz, D. W., 1986b Segregation and linkage studies of allozyme loci in Pair crosses of the oyster Crassostrea virginica. Biochem. Genet. 24: 941-956. Fuiman, L. and J. H. Cowan, Jr., 2003 Behavior and recruitment success in fish larvae: repeatability and covariation of survival skills. Ecology 84(1): 53-67. Giagrande, A., S. Geraci, and G. Belmonte, 1994 Life-cycle and life-history diversity in marine invertebrates and the the implications in community dynaimcs. Oceanography and Marine Biology, an Annual review 32: 305-333. Gosselin, L. A., and P. Y. Qian, 1997 Juvenile mortality in benthic marine invertebrates. Mar. Ecol. Prog. Ser. 146: 265–282. Hare, J. A., and R. K. Cowen, 1997 Size, growth, development, and survival of the planktonic larvae of Pomatomus saltatrix (pisces: Pomatomidae). Ecology 78: 2415- 2431. Heath, M., and A. Gallego, 1997 From the biology of the individual to the dynamics of the population: bridging the gap in fish early life studies. J. Fish Biol. 51(Supplement A): 1-29. Hedrick, P. W., and O. Muona, 1990 Linkage of viability genes to marker loci in selfing organisms. Heredity 64: 67-72. 19 Hedgecock, D., P. M. Gaffney, P. Goulletquer, X. Guo, K. Reece, G. Warr, 2005 The case for sequencing the Pacific oyster genome. J. Shellfish Res. 24(2): 429–441. Hedgecock, D., D. J. McGoldrick, and B. L. Bayne, 1995 Hybrid vigor in Pacific oysters: an experimental approach using crosses among inbred lines. Aquaculture 137(1- 4): 285-298. Hjort, J., 1914 Fluctuations in the great fisheries of Northern Europe. Rapports et Proce`s-Verbaux des Re´unions du Conseil International pour l‘Exploration de la Mer, 20:1–228. Houde, E. D., 1996 Evaluating stage-specific survival during the early life of fish. Pages 51-66 in, Survival strategies in early life stages of marine resources, Y. Watanabe, Y. Ya- mashita, and Y. Oozeki, editors. A. A. Balkema, Rotterdam, The Netherlands. Houde, E. D., 2002 Mortality. Pages 64-87 in Fishery science: the unique contributions of early life stages, L. A. Fuiman and R. G. Werner, editors. Blackwell Science, Oxford, UK. Hubert, S., E. Cognard, and D. Hedgecock, 2009 Centromere-mapping in triploid families of the Pacific oyster Crassostrea gigas (Thunberg). Aquaculture 288: 172–183. Hubert, S., and D. Hedgecock, 2004 Linkage maps of microsatellite DNA markers for the Pacific oyster Crassostrea gigas. Genetics 168: 351-362. Hunt, H. L., and R. E. Scheibling, 1997 Role of early post-settlement mortality in recruitment of benthic marine invertebrates. Mar. Ecol. Prog. Ser. 155: 269–301. Korringa, P., 1941 Experiments and observations on swarming, pelagic life and setting in the European flat oyster, Ostrea edulis L. Arch. neerl. Zool. 5: 1-249 [Thesis, Univ. Amsterdam, 1940]. Launey, S., and D. Hedgecock, 2001 High genetic load in the Pacific oyster Crassostrea gigas. Genetics 159: 255-265. Llodra, E. R., 2002 Fecundity and life-history strategies in marine invertebrates. Adv. Mar. Bio. 43: 87-170. Luo, L. and S. Xu, 2003 Mapping viability loci using molecular markers. Heredity 90: 459-467. 20 Luo, L, Y. M. Zhang and S. Xu, 2005 A quantitative genetics model for viability selection. Heredity 94: 347-355. MacArthur, R. H., and E. O. Wilson, 1967 Theory of Island Biogeography. Princeton University Press, Princeton. Mallet, A. L., E. Zouros, K. E. Gartner-Keykay, K. R. Freeman, and L. M. Dickie, 1985 Larval viability and heterozygote deficiency in populations of marine bivalves: evidence from pair matings of mussels. Mar. Biol. 87: 165-172. McGoldrick, D. J., Hedgecock, D., 1997 Fixation, segregation and linkage of allozyme loci in inbred families of the Pacific oyster Crassostrea gigas (Thunberg): implications for the causes of inbreeding depression. Genetics 146: 321-334. McGoldrick D. J., D. Hedgecock, L. J. English, P. Baoprasertkul, and R. D. Ward, 2000 The transmission of microsatellite alleles in Australian and North American stocks of the Pacific oyster (Crassostrea gigas): selection and null alleles. J. Shellfish Res. 19: 779–788. Mitchell-Olds, T., 1995 Interval Mapping of Viability Loci Causing Heterosis in Arabidopsis. Genetics 140: 1105-1109. Palumbi, S., 1994 Genetic divergence, reproductive isolation, and marine speciation. Ann. Rev. Ecol. Syst. 25: 547 –572. Palumbi, S., 1995 Using genetics as an indirect estimator of larval dispersal. Pages 369 – 387, in Ecology of marine invertebrate larvae, L. McEdward, ed., CRC Press, Boca Raton, Florida Pianka, E. R., 1970 On ―r‖ and ―k‖ selection . Amer. Nat. 104: 592-597. Peterman, R. M, and M. J. Bradford, 1987 Wind-speed and mortality-rate of a marine fish, the northern Anchovy (Engraulis-Mordax). Science 23 (4786): 354-356. Rickman, S., Dulvy, N., Jennings, S., and Reynolds, J., 2000 Recruitment variation related to fecundity in marine fishes. Can. Jour. Fish. Aquat. Sci. 57: 116–124. Ritland, K., 1996 Inferring the genetic basis of inbreeding depression in plants. Genome 39: 1-8. Roughgarden, J., S. D. Gaines, and H. Possingham, 1988 Recruitment dynamics in complex life-cycles. Science, 241: 1460–1466. 21 Remington, D. L., and D. M. O‘Malley, 2000 Whole genome characterization of embryonic stage inbreeding depression in a selfed loblolly pine family. Genetics 155: 337-348. Stearns, S. C., 1980 A new view of life history evolution. Oikos 35: 266-281 Stearns, S. C., 1992 The Evolution of Life Histories. Oxford University Press, Oxford. Strathmann, R. R., 1985 Feeding and nonfeeding larval development and life-history evolution in marine invertebrates. Ann. Rev. Ecol. Syst. 16: 339. Savolainen, O., K. Karkkainen, and H. Kuittinen, 1992 Estimating numbers of embryonic lethals in conifers. Heredity 69: 308-314. Sørensen, F. C., 1967 Linkage between marker genes and lethal factors may cause distrubed segregation rations. Silvae. Genet. 16: 132–134 Thorson, G., 1950 Reproductive and larval ecology of marine bottom invertebrates Biological Reviews 25(1):1-45. Vogl, C. and S. Xu, 2000 Multipoint mapping of segregation distorting loci using molecular markers. Genetics 155: 1439-1447 Williams, G. C., 1975 Sex and Evolution. Princeton University Press, Princeton, NJ. Winemiller, K. O., and K. A. Rose, 1993 Why do most fish produce so many tiny offspring? Amer. Nat. 142: 585-603. Winemiller, K.O. and K.A. Rose, 1992 Patterns of life-history diversification in North American fishes: implications for population regulation. Can. J. Fish. Aquat. Sci. 49: 2196-2218. Winemiller, K.O. 1992 Life history strategies and the effectiveness of sexual selection. Oikos 62: 318-327. 22 Chapter 1 Genome-wide analysis of stage-specific inbreeding depression in the Pacific oyster Crassostrea gigas 1.1 Abstract Inbreeding depression and genetic load have been widely observed, but their genetic basis and effects on fitness during the life cycle remain poorly understood, especially for marine animals with high fecundity and high, early mortality (Type-III survivorship). A high load of recessive mutations was previously inferred for the Pacific oyster Crassostrea gigas, from massive distortions of zygotic, marker segregation-ratios in F 2 families. However, the number, genomic location, and stage-specific onset of mutations affecting viability have not been thoroughly investigated. Here, we again report massive distortions of microsatellite-marker segregation ratios in two F 2 hybrid families, but we now locate the causative deleterious mutations, using a quantitative trait locus (QTL) interval-mapping model, and we characterize their mode of gene action. We find 14-15 viability QTL (vQTL) in the two families. Genotypic frequencies at vQTL generally suggest selection against recessive or partially recessive alleles, supporting the dominance theory of inbreeding depression. No epistasis was detected among vQTL, so unlinked vQTL presumably have independent effects on survival. For the first time, we track segregation ratios of vQTL-linked markers through the life-cycle, to determine their 23 stage-specific expression. Almost all vQTL are absent in the earliest life-stages examined, confirming zygotic viability selection; vQTL are predominantly expressed before the juvenile stage (90%), mostly at metamorphosis (50%). We estimate that, altogether, selection on vQTL caused 96% mortality in these families, accounting for nearly all of the actual mortality. Thus, genetic load causes substantial mortality in inbred Pacific oysters, particularly during metamorphosis, a critical developmental transition warranting further investigation. 1.2 Introduction Inbreeding depression, the detrimental fitness consequences associated with consanguineous mating, has been observed for well over a century in domesticated, as well as natural populations of plants and animals (Darwin 1876, Charlesworth and Charlesworth 1987, Charlesworth and Willis 2009). In normally out-crossing populations, inbreeding may decrease the mean of metric characters, such as body size, decreases fitness components, such as fecundity and survival, and increases the risk of extinction for small populations (Wright 1977, Charlesworth and Charlesworth 1987, Charlesworth and Charlesworth 1999, Hedrick and Kalinowski 2000). Despite the significance of inbreeding depression, much of the foregoing research has focused on population-level differences in trait means, so that details of its genetic basis remain poorly understood. Whether the genetic load underlying inbreeding depression is maintained by dominance, overdominance, or epistasis is still debated, for example. Although many accept dominance as the major cause of inbreeding depression (Roff 24 2002, Charlesworth and Willis 2009), evidence exists for overdominance (e.g. Mitchell- Olds 1995, Williams et al. 2003) and epistasis (e.g. Li et al. 2001). Why genetic load varies among organisms with different life-histories (Lynch and Walsh 1998, Tables 10.4, 10.6) or why the number of lethal mutations might be uniform among vertebrates that apparently differ in mutation rate, genome size, and number of essential loci (McCune et al. 2002) are fundamental questions that will remain unanswered until detailed studies of the genetic basis of inbreeding depression are carried out in a variety of organisms. Another aspect of inbreeding depression that bears detailed investigation is its developmental timing. If mutations in early-acting genes tend to be highly detrimental or lethal, as often assumed, and mutations in late-acting genes, mildly detrimental, then mating system (inbred vs. outbred) should affect the life-stages, at which inbreeding depression and natural selection against harmful mutations are manifested. In support of this notion, inbreeding depression is detected in both the very early and late life-stages of outcrossing plants but is generally shifted solely to the later stages in selfing plants, because early-acting mutations have been purged (Husband and Schemske 1996, Koelewijn et al. 1999). A similar temporal pattern of inbreeding depression exists in selfing vs. outcrossing species of the ascidian Corella (Cohen 1992). Animals are generally outcrossing, so substantial inbreeding depression is expressed early in development (Drosophila, Rizski 1952, Seto 1954, 1961; the European flat oyster, Bierne et al. 1998; the bluefin killifish, McCune et al. 2002; the snail Physia acuta, Escobar et al. 2008; and the purple sea urchin Stronglyocentrotus purpuratus, Anderson and 25 Hedgecock 2010). This pattern of early inbreeding depression might be particularly exaggerated in highly fecund organisms, such trees (Koelewijn et al. 1999) and the majority of marine animals (Thorson 1950, Winemiller and Rose 1992), because a high mutational load is generated as a by-product of germ-cell division (G. C. Williams 1975). Using a molecular marker-based approach, Launey and Hedgecock (2001) showed that the Pacific oyster Crassostrea gigas carries a large number of highly deleterious, recessive mutations (~12 per genome), which not only confirmed Williams‘s (1975, p. 80) specific prediction of a high genetic load in oysters but also simultaneously explained a suite of long-standing observations about the genetics of bivalve molluscs, including segregation distortion in inbred crosses, heterosis in crosses between inbred lines, and heterozygosity-fitness correlations in the wild (reviewed by Launey and Hedgecock 2001). By comparing marker segregation ratios in the earliest swimming larval stage, which were Mendelian, to those in 2-3 month-old juveniles, which were significantly distorted, Launey and Hedgecock (2001) demonstrated conclusively that inbreeding depression was caused by zygotic, genotype-dependent selection against deleterious recessive mutations. Yet, this study left unresolved the number and genome distribution of these mutations, their contribution to mortality, and the timing of their expression during the complex life cycle of the oyster, which comprises a microscopic, free-swimming planktonic larva that metamorphoses into a sessile, benthic, filter-feeding form (Kennedy et al. 1996). Following up on the materials and methods of Launey and Hedgecock (2001), Bucklin (2003) showed that these highly deleterious mutations were heritable, not ―synthetic lethals‖ (Dobzhansky 1946), and that individuals homozygous 26 for these mutations died sometime during the larval or early juvenile stages. Unfortunately, temporal sampling in this study was sparse, so details about the expression of deleterious mutations during larval development, metamorphosis, and the juvenile and adult stages remain unknown. Understanding the developmental expression of deleterious alleles is an important step in uncovering the genetic basis of mutational load and should shed light on the causes and consequences of the substantial early mortality of highly fecund marine animals. In this study, we again examine segregation of microsatellite markers across the genome in F 2 families of the Pacific oyster to characterize deleterious loci responsible for genetic load. Improving upon the single marker or linked-marker approach used previously (Hedrick and Muona 1990, Launey and Hedgecock 2001), we use a quantitative trait locus (QTL) interval mapping model (Vogl and Xu 2000, Luo and Xu 2003) to integrate all marker segregation information, to localize viability loci of major effect in the Pacific oyster genome, and with the aid of two-locus models, to characterize the degree of dominance of alleles at these viability loci. In a second, larger part of the study, we determine, for the first time in any non-human species, the stage-specific expression of major viability loci. This is accomplished by examining marker segregation ratios in samples taken at sub-daily and daily time points during the larval period, at 30 days (juvenile stage), and at 700 days (adult stage) post-fertilization. We test whether the timing of viability selection on deleterious alleles is focused on transitions in larval development, from the trochophore to the veliger stage, the late umbo to the pediveliger stage, or at metamorphosis (Kennedy et al. 1996). Finally, after testing 27 for synergistic interactions between loci in their effect on viability (epistasis), we estimate the total selective mortality attributable to genetic load and examine the magnitude and temporal pattern of mortality associated with inbreeding depression in the Pacific oyster. 1.3 Materials and Methods Biological material and molecular methods Crosses and culturing Families 46, 10, 51 and 35 were established by pair crosses of wild Crassostrea gigas from Dabob Bay, WA, in 1996 (G 0 ; Langdon et al. 2003). In 1998, inbred lines (G 1 , f = 0.25) were created by a full-sib mating within each of the four families. In 2001, F 1 families 46×10 and 51×35 were produced by pair crosses between inbred lines (sire×dam; Hedgecock and Davis 2007). In 2004, we crossed full siblings from each of the 46×10 and 51×35 F 1 hybrid families at the University of Southern California (USC), Wrigley Marine Science Center (WMSC) on Catalina Island, CA, to produce the F 2 families used in these experiments (f = 0.25×1.25 = 0.3125, where 0.25 represents the inbreeding expected from a full-sib mating and the factor 1.25 accounts for inbreeding of the G 1 common ancestors). Gamete collection, fertilization, and standard larval rearing methods for the Pacific oyster follow Breese and Malouf (1975), as detailed in Hedgecock et al. (1996). Pedigrees of these families were verified with microsatellite DNA markers (Hedgecock and Davis 2007). 28 Sampling The study comprised two components: (1) mapping of QTL with major effects on viability followed by characterization of their mode of gene action and (2) determination of the life stages at which viability selection and mortality occurred. The first part of the study was done with samples taken in May 2006, when the F2 oysters were approximately 700 day-old adults (n = 96 for 46×10, n = 49 for 51×35); these adult oysters had been reared, since 2005, under the WSMC dock in Vexar® mesh cages (19 mm2, Aquapurse®). Since the adult samples were the basis for viability QTL mapping and identifying nearest markers for the temporal study, power analyses were conducted to determine an adequate sample size for detecting major viability genes, as described by Launey and Hedgecock (2001). This analysis revealed good power with n = 49 (0.65- 0.8) and excellent power with n = 96 (>0.9) to reject the null hypothesis of Mendelian segregation ratios in favor of the alternative hypothesis of strong viability selection. For the temporal part of the study, samples of 96 larvae were taken from each family at 6 h, 12 h, and 24 h post fertilization, and daily thereafter until larvae were ready to settle at approximately day 18. When competent for settling, larvae were screened off, treated with epinephrine to promote settlement without attachment (Coon et al. 1986), and set at low density in a downwelling/upwelling nursery system (Hedgecock and Davis 2007). A sample of 96 juvenile oysters (spat) was taken at day 30 for genotyping, and the remaining oysters were reared in the nursery system until large enough for deployment in Aquapurses®. Allocation of genotyping effort across these samples is discussed below, in ―Stage-specific expression of vQTL.‖ Sampling days are abbreviated 29 as d1, d2, etc. Oysters sampled for the temporal study were classified into five life stages: trochophore larvae (6-24 hours), veliger larvae (d1 to d13), pediveliger larvae (d14 to d18), spat or recently settled juveniles (d30), and adults (d700) (Kennedy et al. 1996). DNA extraction, PCR, and Electrophoresis Adult tissues (both from the parents and d700 progeny samples) were preserved in 70% ethanol prior to extraction. Larvae from each time point and the d30 juvenile samples were killed with 3 drops of household bleach in 15 ml of seawater, then rinsed and stored in bulk in 70% ethanol. DNA from parents of the F 2 crosses and from the F 2 adult progeny was extracted using the DNeasy animal tissue kit (Qiagen Inc., Valencia, CA). Larval F 2 progeny were extracted in 40µl volumes of 1x PCR buffer (Promega, Madison, WI), 1mM EDTA, and 1µl of 20mg/ml proteinase K (Shelton Scientific, Shelton, CT). Digestion of the larvae was carried out at 56°C for 3 hours followed by 10 minutes at 95°C to denature the proteinase K. Over 80 microsatellite markers cloned from the Pacific oyster were tested for use in this study (Magoulas et al. 1998, McGoldrick et al. 2000, Huvet 2000, Li et al. 2003, Sekino et al. 2003, Yamtich et al. 2005), most of which were on published linkage maps (Hubert and Hedgecock 2004, Hubert et al. 2009). Markers were named as in the source publication, except for those developed in this laboratory (McGoldrick et al. 2000, Li et al. 2003, Yamtich et al. 2005), which were abbreviated from their original description, e.g. ucdCgi195 to Cg195. Polymerase chain reactions were carried out in two phases; the 30 first to amplify the target microsatellite marker, and the second to incorporate a fluorescently labeled dye attached to a ―zip‖ sequence that is complementary to the 3‘ tail of the forward primer (Shuelke 2000). PCR cycle conditions consisted of an initial denaturing step at 94°C for 2 minutes, then 20 phase-one cycles (30 sec at 94°C, 45 sec at T m , 45 sec at 72°C), followed by 10 ―zip‖ cycles (30 sec at 94°C, 45 sec at 59°C and 45 sec at 72°C) and a final 10-min elongation step at 72°C. The number of zip cycles was often increased to 30 or 40 when using larval DNA as template. T m (the optimum annealing temperature) and the MgCl 2 concentration varied, depending on the locus (Magoulas et al. 1998, McGoldrick et al. 2000, Huvet 2000, Li et al. 2003, Sekino et al. 2003, Yamtich et al. 2005). PCR products were electrophoresed on a 4% denaturing PAGE gel (acrylamide: bisacrylamide 19:1, 3M Urea) using 1 TBE (Tris(base), boric acid, and EDTA di-sodium salt) on the ABI Prism 377 DNA Sequencer (Perkin Elmer, Waltham Massachusetts). Gel images were used to score the fluorescently labeled microsatellites by eye, comparing progeny alleles to adult alleles along with fluorescently labeled size standards (DeWoody et al. 2004). Data analyses Segregations and null alleles Because the grandparents of experimental crosses were not completely inbred, parental cross types have two alleles (symbolically, AA AB or AB AB), three alleles (AB AC), or four alleles (AB CD). Non-amplifying or null alleles, which are common 31 in the Pacific oyster (McGoldrick et al. 2000, Launey and Hedgecock 2001, Hedgecock et al. 2004), require modification of classical segregation ratios only for cross type AØ AB, since AA cannot be distinguished from AØ and the resultant progeny genotypes, A-, AB, and BØ have expected frequencies of 2:1:1, respectively. Under the null hypothesis of no viability selection, progeny genotypes should conform to an expected Mendelian ratio (either 1:1, 1:2:1, 1:1:1:1, or 2:1:1 for the null allele case). Deviations from expected Mendelian proportions were tested with goodness-of-fit chi-square tests, with the level of significance adjusted for multiple simultaneous tests within each type of segregation (Rice 1989, Launey and Hedgecock 2001). Adjustment for multiple tests increases type II error, so statistical significance at both nominal = 0.05 and Bonferroni adjusted levels are reported (Rice 1989). Mapping viability QTL (vQTL) Linkage maps for the two experimental families were constructed with the d700 genotype data, using the CP (cross pollinator) population type in JoinMap 3.0 (Van Ooijen and Voorrips 2001). The Kosambi mapping function with a minimum likelihood of the odds (LOD) score of 2.0 was used for linkage group assignments. Because distortions of Mendelian segregation ratios may affect linkage mapping (Lorieux et al. 1995, Zhu et al. 2007), we compared marker orders and distances with previously published linkage maps constructed from samples with little segregation distortion (Hubert and Hedgecock 2004, Hubert et al. 2009). When markers that should have mapped failed to map (only one or two per family), we used locations from published 32 maps (Hubert and Hedgecock 2004). Linkage phase was determined by JoinMap from the frequencies of parental and recombinant types. Percentages of the genome covered by these maps were estimated by ⁄ where d = mean inter-marker distance and n = total number of markers assigned to linkage groups, and L = map length estimated by adding twice the average inter-marker interval to the sum of all intervals per linkage group (LG) and then summing the lengths of all LG (Bishop et al. 1983). The linkage map, parental linkage phases, and parent and progeny genotypes were input to the viability QTL model of Luo and Xu (2003), as implemented in PROC QTL (Hu and Xu 2009), a user-defined procedure for SAS (version 9.2, SAS Institute, Inc., Cary, NC). Four markers with ambiguous AA/AØ genotypes were dropped from the linkage map used for family 46×10, and two markers, one which was not known to be linked and did not link and another with ambiguous genotypes, were dropped from the map used for family 51×35. Genomes were scanned in 1 cM increments, under the maximum-likelihood QTL framework, and a likelihood ratio test (LRT) statistic and estimates of genotype proportions were obtained for each increment (Luo and Xu 2003; all cross types are interpreted by the model as AB CD, yielding progeny genotypes AC, AD, BC, BD). Since we could not use permutation tests to set QTL significance thresholds—because there is no phenotype to permute—we used an approximate method, based on the LRT profile, to establish thresholds at the = 0.05 level (Piepho 2003). We calculated genome- and chromosome-wise significance thresholds for vQTL. Multiple QTL on a single linkage group were identified when the LRT statistic fell by at least 4.60 (~1 LOD) between two QTL peaks (Lander and Botstein 1989). 33 Mortality In the absence of selection, one expects genotypic proportions in a family to be in Mendelian ratios at all positions in the genome, whereas, at a vQTL, these ratios are distorted by genotype-dependent mortality. We can calculate the magnitude of this mortality from the relative survival of genotypes at each QTL peak. The relative survival of each genotype at a QTL is , where w max is the highest-frequency genotype (Luo and Xu 2003). Average relative survival, S, at a QTL is thus: 4 max 22 max 21 max 12 max 11 w w w w w w w w S = max 4 1 w . (1) Average relative mortality at this QTL, M, is 1- S or max 4 1 1 w . In finite samples, chance fluctuations in genotype numbers will produce a non- zero estimate of mortality by this equation, even in the absence of selection, so to correct for this sampling error, we randomly generated 1000 datasets of progeny genotypes (i.e. no selection), using the parental genotypes and sample sizes specific to each family. We then ran each of the 1000 simulated datasets through the QTL model to obtain the average maximum genotype frequency in the absence of selection, w maxØ . The correction for sampling error was then applied to the estimate of average survival at a vQTL, by adjusting the maximum frequency of the four genotypes: 25 . 0 4 1 Ø max max w w S V adj , (2) 34 where w maxV is estimated from the data by the QTL model and 0.25 is the Mendelian expectation in a sufficiently large sample without selection (code for the simulations available from the author on request). Selection, dominance, and epistasis of viability QTL We adapted the two-locus selection model of Hedrick and Muona (1990; see also Launey and Hedgecock 2001) to estimate jointly the selection coefficient (s) and the dominance deviation (h) for each viability locus, given the value of c, the recombination distance between a vQTL peak and the nearest microsatellite DNA marker. In each case, we took the estimates of s and h that maximized the ratio of the likelihood of genotype data, with selection, to the likelihood of the data, without selection. We produced 95% confidence intervals for estimates of s and h, by subtracting one from the log of the odds (LOD) score of the maximized model. (R code for this function available in the Supplementary information, File S5). Only markers of cross type AB×AB, AB×AC, and ØA×ØB that were deficient for one homozygote could be used in the two locus model; otherwise, gene effects were evaluated by inspection of genotypic proportions at the nearest marker and modified chi-square goodness-of-fit tests (e.g. for AB×AB markers that were deficient for both homozygotes, we tested the fit of AB:AA or AB:BB to a 2:1 ratio). Epistatic interactions were assessed in two ways, first using the regression method of Fu and Ritland (1996) and second, using contingency chi-square tests of genotypic associations across all pairs of markers. Under the univariate model of recessive viability 35 selection, (equation 2d, Fu and Ritland 1996) the frequency of individuals homozygous for i markers linked to viability alleles, divided by its binomial expectation, are log transformed and regressed on i. This relationship is expected to be linear if there are no interactions between loci. Because this regression model assumes that deleterious alleles are recessive, we used 10 and 11 unlinked markers, for families 46×10 and 51×35, respectively, which appeared to have completely or nearly completely recessive mutations and to be associated each with a different viability QTL. Regression analysis was carried out in the R statistical package, version 6.2.2. Contingency chi-square tests for pairwise associations between all mapped markers were also carried out in R (code available in Supplementary information, file S11), with the significance level adjusted to control for the false discovery rate (Benjamini and Hotchberg 1995). Stage-specific expression of vQTL To estimate the life stage at which selection acted on each vQTL, we determined the first culture day or interval, when the marker closest to a vQTL became distorted. As most genotype-dependent mortality occurs by the juvenile stage (Launey and Hedgecock 2001, Bucklin 2003), we examined those markers that were significantly distorted at d700, at successively earlier time points, to determine when the distortion first appeared. When multiple genotypes at a distorted marker were deficient (e.g. both AA and BB genotypes deficient in AB AB crosses), we tracked the timing of distortion for each homozygote separately (see ―Discussion‖). 36 Table 1 Day 700 segregation results for two F 2 families Family Cross type Markers scored Distorted markers Linkage group 46×10 AA AB 5 3 (3) 1,4,6,7,10 AB AB 20 12 (8) 2,4,5-8,10 AB BC 11 6 (6) 1-7, 10 AB CD 2 1 (1) 1,2 AB CØ 1 1 (1) 1 AØ CØ 6 4 (4) 1,2,4,6 AØ AB 4 2 (1) 1,6,9 Total 49 29 (24) 51×35 AA AB 5 3 (2) 1,4,7,9 AB AB 27 19 (13) 1-10 AB BC 17 11 (6) 1,3,5-7,8,10 AB CD 2 1 (0) 4 AB CØ 3 3 (1) 1,3 AB AØ 1 0 2 Total 55 37 (22) Numbers in parentheses represent significant tests (α = 0.05) after correction for multiple comparisons. 37 In order to reduce genotyping effort, we focused on samples from days 18, 10, 5 and 2, to locate when the P-value of the goodness-of-fit test went, retrospectively, from significant to non-significant. Gaps in the time series were filled by genotyping samples from intervening days, as needed, to place vQTL expression into a particular life stage. To test for the stability of segregation ratios at a marker, on days before and after the day on which genotypic ratios became significantly distorted, we performed R×C contingency tests, using the Chirxc program (Zaykin and Pudovkin 1993). 1.4 Results Analysis of segregation ratios in F 2 adults Testing over 80 microsatellite markers in the F 1 parents of families 46×10 and 51×35, revealed 49 and 56 informative markers, respectively (Table 1). Typing of informative markers in the adult (day 700 or d700) samples for each of the F 2 families yielded a total of 7024 genotypes, which constitute the ―end-point‖ data set used to assess and map viability selection in this study (supplementary Tables S1 and S2). Significantly distorted segregation ratios at the = 0.05 level were found at 29 (60%) markers in 46×10 and 37 (66%) markers in 51×35; after adjustment for multiple tests within segregation types, 24 (50%) and 22 (39%) remained significant (Table 1). These distorted ratios were distributed broadly across the genome, on nine of the 10 linkage groups (LG), with many instances of multiple markers exhibiting distorted segregation ratios on the same linkage group (Tables S1, S2). Some markers showed distorted ratios in both families (e.g. Cg140, Cg212), but segregation ratios of many markers and regions 38 of the genome were distorted in only one family, suggesting that different viability loci were segregating in each family. Inferences about the causes of these distortions were made after the vQTL analysis (see below), but we note here that the majority of segregation-ratio distortions (18 of 29 for 46×10 and 29 of 37 for 51×35) were attributable to deficiencies of homozygous genotypes (Tables S1, S2). For some AB AB cross types (four in 46×10 and 11 in 51×35), both homozygotes were deficient in the progeny. A minority of segregation-ratio distortions from three- or four-allele cross types (seven for 46×10 and five for 51×35), yielded deficiencies of heterozygous but not homozygous genotypes (e.g. ―AB‖ in Cg162, family 46×10, Table S1). Many of these cases of heterozygote deficiency were observed in crosses with null alleles, with AØ, BØ, or both being deficient (e.g. cmrCg02 in family 51×35 and Cg194 from family 46×10, Tables S1, S2). Mapping of vQTL Linkage maps were constructed from 45 and 52 markers scored in adult samples for families 46×10 ( ̅ = 89.8 per locus) and 51×35 ( ̅ = 46.9 per locus), resulting in estimated genome coverages of 80.0% and 81.3%, respectively. Four previously unmapped markers (Crgi26, Cg9, Crgi010, cmrCg02) were added to the map for family 51×35 (Tables S1, S2). For both families, JoinMap grouped markers previously assigned to LG 1A or LG 10 (Hubert and Hedgecock 2004) into one linkage group (designated as LG 1 in this study), a finding supported by gene-centromere mapping (Hubert et al. 2009). With the joining of previously assigned LG 1A and 10, numbers of the remaining 39 linkage groups were shifted by one (e.g. LG 1B became LG 2, the previously assigned LG 2 became LG 3, etc.). Linkage maps consisted of 10 LG for each family, with markers assigned to the same LG (adjusted by one) and in the same orders as reported previously (Hubert and Hedgecock 2004, Hubert et al. 2009). Only LG 10 in family 46×10 exhibited marker orders and distances that differed from previous data; however, Hubert and Hedgecock (2004) also found significant heterogeneity in marker order and distances across families for this linkage group. Overall, the large numbers of distorted markers linked to viability loci did not appear to affect linkage mapping severely. Eleven significant vQTL were identified in family 46×10 and eight vQTL in family 51×35 by peak likelihood ratios above genome-wise thresholds of 17.5 and 18.2 for families 46×10 and 51×35, respectively (Fig. 2). Additional vQTL (one in 46×10 and four in 51×35) were identified by peaks above the chromosome-wise thresholds, which were 13 or lower for all LG in both families, except for LG 10, which had a threshold of 14.2 for family 46×10 and 14.4 for family 51×35 (Fig. 2). LRT peaks below the genome-wise threshold but above the chromosome-wise threshold were associated with significant distortions of nearby markers. Thus, a total of 12 vQTL were found in each of the families (Table 2; Fig. 2), and these were distributed across seven of the nine linkage groups studied in family 46×10 (LG 9 was dropped because its only two informative markers had ambiguous AA/AØ genotypes) and eight of the 10 linkage groups in family 51×35; neither family had a significant vQTL on LG 2. Multiple vQTL were found on over half of the linkage groups in each family. 40 Figure 2 QTL Mapping results for family 46×10 and 51×35 Likelihood ratio test statistic vs. map distance (cM) for F 2 families 4610 (A) and 51×35 (B). Long vertical lines mark the ends of linkage groups which are numbered 1-10. Small triangles along the X-axis indicate the position of markers used in the mapping procedure. Downward arrows mark the location of vQTL peaks. The solid black line indicates the genome-wise threshold value for significance at the = 0.05 level (17.5 for 46×10, 18.2 for 51×35) and the dotted black line indicates the chromosome-wise threshold value for = 0.05; all chromosomes have a threshold of 13 or lower, except for Chr. 10, which is 14.2 for 46×10 and 14.4 for 51×35. Figure shown on next page. 41 Figure 2 (Continued) 42 Selection and dominance of vQTL Gene action was inferred from results of the two-locus model and inspection of segregation ratios at markers nearest to vQTL (Table 2). Selection coefficients (s), for the 11 markers that could be fit to the two-locus model, ranged from 0.77 to 1.0, with six above 0.9. Nine markers yielded mid- to low-dominance (h) estimates, with 95% confidence intervals that included zero (Table 2). For two markers (Cg198 and Cg178 in 46×10), 95% confidence intervals for h did not include zero, indicating partial dominance. Of the 13 cases that could not be fit to the two-locus model, six AB×AB cross types were significantly deficient for both homozygotes (Cg212 and Cg189 in 46×10; Cg126, Cg138, Cg141, and Cg184 in 51×35), suggesting potentially overdominant gene action. Overdominance was subsequently called into question for all but two of them, based on temporal data (see next section). Two markers with cross types AA×AB were significantly deficient for homozygous genotypes; the deleterious alleles at the linked vQTL were provisionally classified as recessive. Finally, five markers could not be fit to either the two-locus model or the chi-square framework, because they were not deficient for homozygous genotypes (cmrCgi005, and Cg162 in 46×10; Cg156, Cg197, and Cg212 in 51×35); however, in these cases, heterozygotes sharing an allele were deficient, suggesting a dominant viability effect of the shared allele. We found no evidence for epistasis among viability loci, using the regression method of Fu and Ritland (1996). Linear regression fit the observed data well in both families, (r 2 = 0.999, P = 0, for family 46×10; r 2 = 1.0, P = 0, for family 51×35), and 43 Table 2 Viability QTL , selection, dominance, and timing of selection at nearest markers, in families 46×10, and 51×35. Order of progeny genotypes for cross type AA×AB is AA, AB; for AB×AB is AA, AB, BB; for AB×AC is AA, AB, AC, BC; for AB×CD is AC, AD, BC, BD; and for ØA×ØB is ØØ, ØB, ØA, AB. ‗VL‘ is viability locus number, ‗LG‘ is linkage group. Degrees of dominance in the ‗Inf. (inference) of dominance‘ column: Recessive (R), Dominant (D), Overdominant (OD), Partially dominant (PD). Additional viability loci not detected by the QTL model were added based on their distinct timing of segregation distortion, relative to the marker nearest to the QTL peak; a um2L48, b Cg156, c Cg126 and Cg208, and d Cg186. Table shown on the next page. 44 Table 2 (Continued) Family VL LG Loc. (cM) LRT Nearest Marker Cross type Progeny genotypes Two-locus model results Infer. domi- nance Timing 1 2 3 4 s (95% CI) h (95% CI) LOD 46 10 1 1 0 27.5 cmrCg005 AB×CD 7 32 19 29 – – – – d18-30 2 3 33 48.3 Cg162 AB×AC 46 29 3 12 – – – – d12-13, d18-30 a 3 4 0 31 Cg198 AB×AC 11 9 31 35 0.77 (0.42-0.95) 0.65 (0.02-1.0) 2.94 PD d18-30 4 4 20 23 cmrCg001 AB×AB 10 46 33 1.0 (0.82-1.0) 0.29 (0.0-0.63) 2.78 R d18-30 5 4 70 23 Cg178 ØA×ØB 7 15 28 38 0.83 (0.61-0.94) 0.53 (0.1-0.83) 5.10 PD d15-16 6 5 5 25 Cg164 AB×AB 5 55 21 0.89 (0.6-1.0) 0.01 (0.0-0.33) 4.25 R <d18 7 6 4 15.8 Cg14 ØA×ØB 8 29 27 29 0.79 (0.47-0.96) 0.01 (0.0-0.53) 3.60 R d18-30 8 7 0 28 Cg131 AA×AB 32 60 – – – – d14, d18- 30 b 9 7 18 30 Cg155 AB×AB 37 52 3 0.92 (0.76-0.98) 0.33 (0.0-0.6) 7.75 R <d1 Table continues next page 44 45 Table 2 (Continued) Family VL LG Loc. (cM) LRT Nearest Marker Cross type Progeny genotypes Two Locus model results Inf. domi- nance Timing 1 2 3 4 s (95% CI) h (95% CI) LOD 10 10 7 34 Cg129 AB×AB 36 51 2 1.0 (0.88-1.0) 0.32 (0.0-0.61) 8.27 R d18-30 11 10 43 72 Cg212 AB×AB 17 76 0 – – – OD/R d8-9 12 10 45 70 Cg189 AB×AB 8 73 10 – – – OD/R >d30 51 35 1 1 4 59 Cg126 AB×AB 0 42 6 – – – R d18-30, d5-7, >d30 c 2 3 33 23.9 Crgi26 AB×AB 0 35 13 0.99 (0.85-0.99) 0.01 (0.0-0.35) 5.91 R d17 3 3 67 23.9 Cg148 AB×AC 0 18 16 11 1.0 (0.85-1.0) 0.01 (0.0-0.32) 5.58 R d10-12 4 5 1 16 Cg112 AA×AB 11 36 – – – – <d2 5 5 79 13 Cg138 AB×AB 7 34 6 – – – OD/R d18-30 6 6 0 28 Cg141 AB×AB 0 40 8 – – – R d18-30, <d4 d 7 7 5 36 Cg156 AB×AC 3 13 22 8 – – – – d18-30 Table continues next page 45 46 Table 2 (continued) Family VL LG Loc. (cM) LRT Nearest Marker Cross type Progeny genotypes Two Locus model results Inf. domi- nance Timing 1 2 3 4 s (95% CI) h (95% CI) LOD 8 7 24 15 Cg197 AB×AC 3 20 13 9 – – – – d18-30 9 8 7 15.4 Cg196 AB×AB 3 26 18 1.0 (0.85-1.0) 0.35 (0.0-0.71) 4.96 R d18-30 10 9 6 28.5 Cg184 AB×AB 0 39 7 – – – OD/R NA 11 10 39 19.1 Cg140 AB×AC 3 14 9 20 0.86 (0.53-0.98) 0.5 (0.0-0.58) 3.05 R d18-30 12 10 48 21.3 Cg212 AB×AC 14 0 12 19 – – – – d18-30 46 47 adding a quadratic term did not significantly improve the model for either family (F-test, P = 0.212 for 46×10, P = 0.20 for 51×35). Tests of epistasis between all pairwise combinations of closely linked markers showed significant contingency chi-square values, as expected; otherwise, only a handful of interactions among unlinked markers (three of 879 combinations for 46×10 and four of 1192 combinations for 51×35) were nominally significant, none after correction by the Benjamini and Hochberg (1995) method. Temporal changes in marker genotypic proportions Temporal segregation data were obtained for 21 of 29 adult-distorted markers in family 46×10 and for 28 of 37 adult-distorted markers in family 51×35 (supplementary Tables S3, S4). Altogether, 10,017 genotypes were determined in this portion of the study (i.e. not counting d700); on average, 2.8 and 2.4 temporal samples per locus, with average sizes of 79.4 and 78.4 individuals, were examined for families 46×10 and 51×35, respectively. With the exception of two markers (Cg155 in 46×10 and Cg112 in 51×35), all marker segregation ratios became distorted after d2. Within the same linkage group, markers separated by large distances displayed different temporal patterns; however, tightly linked markers often showed the same or very similar timing of segregation-ratio distortion (e.g. Cg162 and Cg160, LG 3 in 46×10, Table S3). For all three of the two- allele cases (AB×AB) in 46×10, for which both homozygotes were deficient and temporal information was available, all on LG 10, the two homozygous genotypes became 48 deficient at different time points (Table S3). Likewise, for four of seven such cases in 51×35, all on LG 1, the two homozygous genotypes became deficient at different time points (e.g. at Cg126, in 51×35, AA was deficient after d5, but BB became deficient only after d18; Table S4). However, temporal data for Cg138 were inconclusive and both homozygous genotypes became deficient at the same time for Cg141 and Cg184 (Table S4). Statistical tests of heterogeneity among segregation ratios of pre- and post- distortion samples were generally not significant. Only eight of 30 tests of post-distortion samples (both families) demonstrated significant heterogeneity among time points at the nominal = 0.05 level, but after correction for multiple tests, none of these cases remained significant. Similarly, only one of the 13 pre-distortion cases was significant at the nominal = 0.05 level (Crgi26, P = 0.025), but it was not significant after correction for multiple tests. In none of the nine nominally significant cases, in which there was some heterogeneity among segregation ratios, did the ratios switch from being distorted to fitting Mendelian expectation or vice-versa. Stability of genotypic proportions, before and after the onset of segregation distortion, is illustrated by chi-square P-values for markers on LG 3, in both families (Fig. 3). The timing of marker segregation-ratio distortions (Tables S3, S4) implied more viability loci than were detected by the QTL model (Table 2). For example, on LG 7 in family 46×10, three markers were significantly distorted, each with a distinct onset: day 0-1 (Cg155), day 14 (Cg131), and day 18 to day 30 (Cg156), suggesting three different viability loci (Table 2). These three viability loci fall, however, under two peaks in the 49 Figure 3 Goodness-of-fit chi-square P-values vs. time for markers on linkage group 3. (A) 46×10 and (B) 51×35, horizontal Dashed grey line represents the α=0.05 threshold. 50 LRT profile based on QTL analysis of adult data alone (Fig. 2A, LG 7). Altogether, we resolved five more viability loci than the QTL model for the two families, based on distinct timing of marker segregation-ratio distortion. All five fall under significant peaks in the LRT profile. For 46×10, these additional viability loci were suggested by markers um2L48 on LG 3 (distorted at days 18-30) and Cg156 on LG 7 (days 18-30) and for 51×35, by LG 1 markers Cg126 (two instead of one viability locus; day 5-7 and day 18-30) and Cg208 (day 30-700) and LG 6 marker Cg186 (before day 4; Tables 2, S3, S4). Thus, the total numbers of viability loci detected were 14 and 15 for families 46×10 and 51×35, respectively. These additional viability loci were incorporated with identified vQTL (except for Cg164 in family 46×10 and Cg184 in family 51×35, which lacked sufficient temporal data) into a plot of stage-specific expression of viability loci (Fig. 4). The distribution of viability loci across life stages did not differ between families (Fisher‘s exact test, P = 0.30), so family results were pooled to test whether viability loci were evenly distributed across life stages. Viability loci were not evenly distributed in expression across life stages (chi- square goodness-of-fit, 4 d.f., P = 0.0007). About half of the viability loci were detected around metamorphosis, between the day-18 larval sample and the day-30 juvenile (spat) sample, and another quarter were expressed at the veliger stage. Only a few viability loci (< 10%) appeared during the longest stage, from juvenile to adult, between days 30 and 700. 51 Figure 4 Distribution of viability loci expression across life stages of the Pacific oyster. Bars represent the proportion of viability loci expressed at each stage and sum to 1.0 for each family (black bars correspond to family 46×10; white bars to family 51×35). Stages are approximately, 0-24 hours (trochophore); 1-14 days (veliger); 15-18 days (pediveliger); 18-30 days (spat/metamorphosis) and 30-700 days (juvenile/adult). 51 52 Genotype-dependent mortality The selective mortality required to explain distorted genotypic proportions was calculated from the relative fitnesses of genotypes at each vQTL. Because no evidence of epistasis was found, we assumed that vQTL separated by >50cM were independent and had multiplicative effects on survival. When two QTL were separated by ≤50 cM, we took the more lethal of the two to estimate mortality. We calculated combined mortality attributable to i vQTL in each family as ∏ ̅ (Table 3). Average relative survival in this calculation was adjusted for sampling error through simulation, as described in Materials and Methods. Relative survival in simulations with no selection averaged 0.80 and 0.68 for families 46×10 and 51×35, respectively (cf. the theoretical expectation of 1.0), leading to substantial upward adjustments of average relative survival at vQTL. Still, selective mortality at individual vQTL ranged from 1% (vQTL 9, LG 8, 51×35) to 67% (vQTL 1, LG 1, 51×35). Combined mortality from the multiplicative fitness effects of viability QTL was estimated as 96.4% in both families (Table 3). 1.5 Discussion In this study, we found substantial distortion of segregation ratios—at over half of the microsatellite markers tested in adult F 2 oysters—in agreement with previous documentation of high genetic load in the Pacific oyster (Launey and Hedgecock 2001, Bucklin 2003). This load largely comprises recessive mutations under strong viability selection, having large effects on early mortality. Specific aspects of selection, 53 Table 3 Selective mortality estimates for each family based on viability QTL results Estimates of cumulative survival use only QTL that are unlinked (>50cM apart) or the more deleterious of QTL that are linked (≤50cM apart; survival at the less deleterious QTL is shown in parentheses). Selective mortality is calculated as one minus the adjusted survival described in eq. 2; w max is the maximum genotype frequency at each QTL (cf. to the theoretical 0.25). Table is shown on the following page. 54 Table 3 (Continued) Family Viability QTL Linkage group Position (cM) W max Adjusted survival Selective mortality 46×10 1 1 0 0.377 0.795 0.205 2 3 33 0.506 0.564 0.436 3 4 0 0.429 (0.682) 0.318 4 4 20 0.459 0.631 0.369 5 4 70 0.536 0.528 0.472 6 5 5 0.418 0.703 0.297 7 6 4 0.322 0.963 0.037 8 7 0 0.401 0.739 0.261 9 7 18 0.397 (0.747) 0.253 10 10 7 0.585 0.478 0.522 11 10 43 0.531 (0.534) 0.466 12 10 45 0.525 (0.541) 0.459 Cumulative 0.036 0.964 51×35 1 1 4 0.871 0.333 0.667 2 3 33 0.385 (0.943) 0.057 3 3 67.2 0.394 0.912 0.088 4 5 1 0.386 0.940 0.060 5 5 79 0.397 0.903 0.097 6 6 0 0.408 0.868 0.132 7 7 5.6 0.752 0.396 0.604 8 7 24.6 0.449 (0.760) 0.240 9 8 7 0.370 0.990 0.010 10 9 6 0.408 0.868 0.132 11 10 50 0.651 0.471 0.529 12 10 60 0.432 (0.801) 0.199 Cumulative 0.036 0.964 55 dominance, and early mortality will be addressed later; we turn, first, to the novel documentation of the life stage at which genotype-dependent mortality appears. Stage-specific expression of genetic load Temporal segregation data and QTL mapping revealed a large number of viability loci, 90% of which were ―expressed‖ during the larval stage, over half around metamorphosis and another quarter during the veliger stage. (We infer that the stage- specific appearance of genotype-dependent mortality reflects the expression or, more likely, lack of expression of a mutation in a functionally critical gene that is normally expressed at this point in development.) Genotype dependent mortality occurred over a short period of time, generally a matter of days, and produced grossly distorted genotypic proportions that remained stable thereafter. All but two viability loci (one in each family) were expressed after day two, which demonstrates that viability selection occurs predominately during the zygotic stage, not during the gametic stage. These results falsify meiotic drive or gametic selection as an explanation for segregation distortion in the Pacific oyster (cf. McCune et al. 2002) and affirm that this bivalve mollusc has many early-acting mutations that are lethal or highly deleterious. Few studies have documented the ontogeny of genetic load in animals. Rizski (1952) and Seto (1954, 1961) showed that selection on most lethal or semi-lethal chromosomes acted during the larval stages in Drosophila. Bierne et al. (1998) were the first to show that viability selection occurred at the larval stage of inbred crosses of the European flat oyster, and Launey and Hedgecock (2001) confirmed these results in F 2 56 crosses of the Pacific oyster. Escobar et al. (2008) demonstrated, for a freshwater snail, that inbreeding depression for early survival was substantial. Almost all other studies of stage-specific inbreeding depression have been carried out on plants. These studies also generally find substantial, early-stage, inbreeding depression in outcrossing species, in line with theoretical predictions (Charlesworth et al.1990, Husband and Schemske 1996). Interestingly, results from studies in outcrossing plants, particularly conifers (e.g. Savolainen 1992, Remington and O‘Malley 2000), show substantial embryonic-stage inbreeding depression, while we find evidence for only one or two mutations affecting survival during the embryonic stage of the oyster, from fertilization through the non- feeding trochophore stage (~0-24 hours). We hypothesized that viability loci would affect fitness disproportionately around times of developmental transitions, when essential genes are turned on and expressed. For example, in plants one might expect to observe inbreeding depression during embryogenesis because of the many genes first expressed at this critical stage (e.g. Thomas 1993, Meinke 1995). In Drosophila, significant changes in gene expression are observed across different life stages, with spikes in gene expression leading up to or during the transition from one pupal stage to the next and during metamorphosis (White et al. 2000, Arbeitman et al. 2002). We did observe a disproportionate number of viability loci expressed around metamorphosis, which highlights this stage as a potentially important developmental transition in terms of selection and mortality. The induction of metamorphosis in invertebrates is well known to involve a complex array of signaling mechanisms and morphogenetic changes, which radically alter the biology and 57 ecology of individuals as they transition to the juvenile stage (Williams et al. 2009). For a marine bivalve, like the Pacific oyster, the transition from a free-swimming larva, with a ciliated velum, to a sessile, filter-feeding juvenile, with gills, involves a complete rearrangement of the body plan and feeding apparatus, during which many new genes are most likely expressed for the first time (Kennedy et al. 1996, Heyland and Moroz 2006). We infer that mutations in these genes cause the many segregation-ratio distortions observed between day 18 and day 30 of development in the Pacific oyster. Ultimately, families, in which such "natural knockouts" of critical genes are segregating, could facilitate genomic dissection of oyster metamorphosis. Estimating the number of viability loci Temporal observations of segregation ratios and phase information suggested the presence of more viability loci than the 12 vQTL identified in each family by the QTL mapping procedure. Viability loci are not fully counted by the QTL model on linkage groups that have many severely distorted markers and high LRT values (>30) across most of their length (e.g. the one, broad LRT peak on linkage group 1, in 51×35). In these cases, markers are often deficient for both homozygous genotypes (AB×AB cross-type; marker proportions of AA:AB:BB are expected to be 1:2:1 in progeny, but ratios approaching 0:1:0 are observed), and the A and B alleles are linked to mutations that exhibit different stage-specific patterns of selection (e.g. at Cg126, in 51×35, marker proportions are ~0:2:1 at day 7 but are ~0:1:0 at day 700, Table S4). Furthermore, phase analysis reveals that, across markers, the A alleles reside on one haplotype, while the B 58 alleles all reside on the other haplotype (i.e. alleles are linked to mutations in repulsion- phase; Fig. 5). The most parsimonious explanation for the combined QTL, temporal, and phase information is two viability loci, each with a deleterious recessive mutation that was expressed at a specific time during development. While the QTL viability model marks an advance from the two-locus model used previously to map deleterious loci (e.g. Hedrick and Muona 1990, Launey and Hedgecock 2001), it clearly underestimates the number of viability loci, when multiple viability loci are linked in repulsion-phase. Our estimate of 14–15 viability loci is slightly higher than the previous estimate for the Pacific oyster (~12; Launey and Hedgecock 2001) and similar to molecular marker-based estimates from plants (e.g. 19 viability loci in Loblolly pine, Remington and O‘Malley 2000; 17 loci in Rice, Xu et al. 1997). Genome-wide, marker-based estimates of the number of deleterious loci are rare for other animals (an early estimate of 15–38 detrimental genes for the European flat oyster was based on only four microsatellite markers; Bierne et al. 1998, and three transmission ratio distortions (TRD) were found in inter-specific backcrosses of mice; Eversley et al. 2010). On the other hand, there are many estimates of lethal equivalents (LE), using the method of Morton et al. (1956), and these average 3–4 LE per genome (Lynch and Walsh 1998, Table 10.4). The number of lethal equivalents for egg to adult viability in the Pacific oyster is likely to be close to the tally of viability loci detected by the marker-based method, since many marker genotypes appear to be lethal, and results of the two-locus model show high selection coefficients (Table 2, discussed below). Thus, the Pacific oyster likely harbors a greater genetic load than any animal species thus far studied. 59 Figure 5 Explanation of two linked viability loci, underlying a single QTL peak. (A) Distribution of distorted markers underlying the QTL peak on Linkage Group 1 for F 2 family 51×35. The dashed line indicates the genome-wide threshold of significance. (B) The proportions of affected homozygous genotypes, relative to their expectations, at distorted markers on linkage group 1, 51×35, arranged by phase and sampling time. Expected ratios are 1:2 for AA or BB homozygotes relative to AB for all markers (cross type AB×AB), except for Cg153, which has an expected ratio of 1:1 for AA relative to AB (cross type AA×AB; phase information available for the ―A‖ haplotype only at Cg153). Dashes represent missing data. Frequencies in bold represent significant differences from expected; a value close to 1.0 indicates the genotype is close to its Mendelian expectation. The timing of selection against individuals homozygous for the ―A‖ phase alleles (~day 5) relative to the timing of ―B‖ phase alleles (day 18-700) suggests two different deleterious alleles acting on this linkage group, although the QTL model shows only one broad peak indicating one viability QTL. Figure shown on next page. 60 Figure 5 (continued) 60 61 Selection, dominance, and epistasis of viability loci Estimates of selection coefficients at most vQTL-associated markers are high, while estimates of dominance are low (Table 2). These results are consistent with previous studies in the Pacific oyster (Launey and Hedgecock 2001, Bucklin 2003). Two viability loci show strong evidence of partial dominance (Cg178 and Cg198), a finding supported by significant deviations from expected 1:1:1 ratios for non-homozygous genotypes at these markers (AB×AC cross type; Tables 2, S1, and S2). Again, this is consistent with the results of Launey and Hedgecock (2001), who attributed three of 20 cases to partially dominant viability loci. Heterozygote deficiencies and segregation distortion in molecular markers have been widely reported in pair crosses of wild bivalves (e.g. Mallet et al. 1985, see references in Launey and Hedgecock 2001), and partially dominant deleterious mutations may in part account for these observations. For a minority of vQTL, genotype deficiencies do not fit any of the classic gene-effect models and are not interpretable. Some vQTL appear to be overdominant, but most of these are resolved by temporal and phase data into separate, recessive deleterious alleles (see above; Fig. 5). Thus, the dominance theory continues to be the best explanation of inbreeding depression and genetic load for the Pacific oyster. Remington and O‘Malley (2000) also found that most viability loci were recessive or nearly recessive, finding only one potentially overdominant locus in a selfed family of loblolly pine. However, in another study of a selfed family of Loblolly pine, Williams et al. (2003) found overdominance for four out of seven lethal factors, using 62 single-marker and two-locus interval mapping procedures. Williams et al. (2003) used only 17 microsatellite markers, so it is possible that their inference of overdominance at individual loci may be biased by the presence of multiple lethal factors linked in repulsion-phase to these markers (i.e. pseudo-overdominance). Higher density mapping with linkage phase information may, therefore, provide better resolution of gene action of viability loci. No evidence of epistatic interactions between viability loci was detected in our experiments, using the regression method of Fu and Ritland (1996), the results of which fit the linear expectation of non-interacting loci extremely well. There were, moreover, no significant contingency chi-square tests between any pair of markers, except those that were closely linked. These results support those from a previous study of Pacific oysters that looked at transmission of microsatellite marker alleles linked to lethal loci across two generations, and showed that deleterious recessive alleles were heritable and were not ―synthetic lethals‖ (Bucklin 2003). Mukai et al. (1972) also found synergism to be relatively weak or nonexistent in fitness-based inbreeding studies of Drosophila, and Remington and O‘Malley (2000) found no evidence for epistasis, also using the regression method of Fu and Ritland (1996). Other molecular studies of the genetic basis of heterosis and inbreeding depression QTL in rice and Arabidopsis have found epistasis to be common (e.g. Li et al. 2001, Melchinger et al. 2007). Epistasis has also been demonstrated through non-linear decreases in log fitness with increasing inbreeding level (e.g. Whitlock and Bourguet 2000, Salathe and Ebert 2003); though in many cases not all fitness traits exhibit epistasis within an experiment (Willis 1993, Carr and Dudash 1997, 63 Kelly 2005). The marker-based regression analysis and the pairwise chi-square tests are perhaps more direct methods to evaluate epistasis compared with biometrical approaches that analyze trait means across levels of inbreeding. Epistasis may still play a role in the expression of inbreeding depression in the Pacific oyster, but we did not detect it using molecular marker-based methods. Genetic load and its role in early mortality We show that viability selection at deleterious loci is stage-specific and strong enough to require that 96% of the individuals in our experimental families must have died to generate the severe distortions of Mendelian ratios observed at marker loci. Do the magnitude and pattern of genetic mortality fit with actual mortality in experimental oyster cultures? Oysters, in general, show high early mortality (Type-III survival) both in the wild (Korringa 1946) and in culture (Guo and Allen 1994). In an experiment with a 51×35 F 2 family related to the one reared in this study, observed mortality, from fertilization through day 60 (early juvenile or spat stage), was 98% (Fig. 6; Plough, unpubl.); this level of mortality is typical of commercial hatchery cultures (J. P. Davis, Taylor Shellfish Farms, Quilcene, WA. pers. comm.). Thus, the genotype-dependent mortality measured here is nearly as high as observed mortality to the juvenile stage. Moreover, the stage-specific pattern of genetic mortality is similar to the actual survival curve, at least after day two. In particular, the substantial decrease in observed culture survival during the metamorphosis/early spat stage is matched by the cumulative 64 Figure 6 Relative survival vs. genetic mortality during the life cycle of the Pacific oyster. Mean relative log survival to day 60 of a replicated F 2 cross of Pacific oysters (filled circles correspond to primary Y-axis; error bars represent one standard error) and cumulative genetic mortality inferred from relative fitness estimates at viability QTL, averaged over families 51×35 and 46×10 (diamonds and dashed line correspond to the secondary Y-axis, also log transformed). The lower X-axis is broken between day 18 and day 30 and points are not to scale thereafter. The top X-axis displays the life history stages (Tr, trochophore; Veliger; Pediveli, pediveliger; M/S, metamorphosis/setting stage). 65 mortality attributable to vQTL (Fig. 6). Thus, the magnitude and pattern of cumulative genetic mortality do indeed closely match those observed in larval cultures, so that nearly all of the post-day-two mortality in F 2 populations may be explained by genetic load. Since, as we have shown, genetic load is mostly caused by recessive deleterious mutations, much of it would probably not be expressed in natural populations, unless population sizes were reduced—owing, for example, to overfishing or environmental degradation—and the probability of inbreeding thereby increased. Genetic mortality cannot account for the 50-80% mortality observed between day 0 and day 2 (Fig. 6), since very few viability loci are observed during that time period in our study (Fig. 4). One possible explanation is that fertilization success is much lower than we expect, perhaps around 50%, so that much of the apparent mortality between fertilization and day two might be attributable to fertilization failure. However, fertilization was monitored closely in our experiments, and there was no indication of substantial fertilization failure for either of the crosses in this study. Another explanation could be dominant de-novo germ-line mutations, which were predicted by Williams (1975) as a by-product of highly fecundity and would not be detected by linkage to marker loci. More data on the causes of early larval mortality, particularly at metamorphosis, should shed light on the role of genetic mortality over the entire Pacific oyster life cycle. 66 Chapter 1 References Anderson, D., and D. Hedgecock, 2010 Inbreeding depression and growth heterosis in larvae of the purple sea urchin Stronglyocentrotus purpuratus (Stimpson). J. Exp. Mar. Biol. Ecol. 384(1-2): 68-75. Arbeitman, M. N., E. E. M. Furlong, F. Imam, E. Johnson, B. H. Null, et al., 2002 Gene expression during the life cycle of Drosophila melanogaster. Science 297(5590): 2270-2275. Benjamini, Y., and Hochberg, Y. 1995 Controlling the false discovery rate: a practical and powerful approach to multiple testing. J. Roy. Stat. Soc. Ser. B 57: 289–300. Bierne, N., S. Launey, Y. Naciri-Graven, et al., 1998 Early effect of inbreeding as revealed by microsatellite analyses on Ostrea edulis larvae. Genetics 148: 1893- 1906. Bishop, D. T., C. Cannings, M. Skolnick, and J. A. Williamson, 1983 The number of polymorphic DNA clones required to map the human genome. In: Statistical Analysis of DNA Sequence Data, Weir, B. S. (ed.). New York, N.Y., Marcel Dekter, pp. 181–200. Breese, W. P. and R. E. Malouf, 1975 Hatchery manual for the Pacific Oyster, Crassostrea gigas (Gould). Oregon Agricultural Experiment Station Special Report No, 443, and Oregon State University Sea Grant College Program Publ. No. ORESU-H-75-002, 22 pp. Bucklin, K. A., 2003 Analysis of the Genetic Basis of Inbreeding Depression in the Pacific Oyster Crassostrea gigas. PhD Dissertation, University of California at Davis. Carr, D. E, and M. R. Dudash, 1997 The effects of five generations of enforced selfing on potential male and female function in Mimulus guttatus. Evolution 51:1797- 1807. Charlesworth, B. and D. Charlesworth, 1999 The genetic basis of inbreeding depression. Genet. Res. 74: 329-340. Charlesworth, D., and B. Charlesworth, 1987 Inbreeding depression and its evolutionary consequences. Annu. Rev. Ecol. Syst. 18: 237–268. 67 Charlesworth, D., M. T. Morgan, and B. Charlesworth, 1990 Inbreeding depression, genetic load, and the evolution of out-crossing rates in a multi-locus system with no linkage. Evolution 44:1469-1489. Charlesworth, D., and J. H. Willis, 2009 The genetics of inbreeding depression. Nat. Rev. Genet. 10:783-796. Cohen, C. S., 1992 Population Biology of two species of Corella: Mating systems and demography (Ascidians). PhD dissertation, University of Washington, Dept. of Zoology. Coon, S. L., D. B. Bonar, and R. M.Weiner, 1986 Chemical production of cultchless oyster spat using epinephrine and norepinephrine. Aquaculture 58: 255-262. Darwin, C. R., 1876 The effects of cross and self fertilization in the vegetable kingdom. John Murray, London. DeWoody, J. A., J. Schupp, Kenefic, L., J. Busch, L. Murfitt, et al. 2004 Universal method for producing ROX-labeled size standards suitable for automated genotyping. BioTechniques 37: 348-352. Dobzhansky, Th. 1946 Genetics of natural populations XIII. Recombination and variability in populations of Drosophila pseudoobscura. Genetics 31: 269-290. Escobar, J. S., A. Nicot, and P. David, 2008 The different sources of variation in inbreeding depression, heterosis and outbreeding depression in a metapopulation of Physa acuta. Genetics 180: 1593-1608. Eversley, C. D., T. Clark, Y. Xie, J. Steigerwalt, T. A. Bell et al., 2010 Genetic mapping and developmental timing of transmission ration distortion in a mouse interspecific backcross. BMC Genet. 11:98. Fu, Y. B, and K. Ritland, 1996 Marker-based inferences about epistasis for genes influencing inbreeding depression. Genetics 144: 339-348. Guo, X., and S. K. Allen, Jr., 1994 Reproductive potential and genetics of triploid Pacific oysters, Crassostrea gigas (Thunberg). Biol Bull. 187: 309-318. Hedgecock, D., D. J. McGoldrick, D. T. Manahan, J. Vavra, N. Appelmans and B. L. Bayne, 1996 Quantitative and molecular genetic analyses of heterosis in bivalve molluscs. J. Exp. Mar. Biol. Ecol. 203: 49-59. 68 Hedgecock, D., G. Li, S. Hubert, K. Bucklin, and V. Ribes, 2004 Widespread null alleles and poor cross-species amplification of microsatellite DNA loci cloned from the Pacific oyster, Crassostrea gigas. J. Shellfish Res. 23: 379-385. Hedgecock, D., and J. P. Davis, 2007 Heterosis for yield and crossbreeding of the Pacific oyster Crassostrea gigas. Aquaculture 272S: S17-S29. Hedrick P. W., and S. T. Kalinowski, 2000 Inbreeding depression in conservation biology. Annu. Rev. Ecol. Evol. Syst. 31:139-162. Hedrick, P. W., and O. Muona, 1990 Linkage of viability genes to marker loci in selfing organisms. Heredity 64: 67-72. Heyland, A., and L. L. Moroz, 2006 Signaling mechanisms underlying metamorphic transitions in animals. Integr. Comp. Biol. 46: 743-759. Hu, Z., and S. Xu, 2009 PROC QTL—A SAS procedure for mapping quantitative trait loci. Int. J. Plant Genomics 2009: 1-3. doi:10.1155/2009/141234. Hubert, S., E. Cognard, and D. Hedgecock, 2009 Centromere-mapping in triploid families of the Pacific oyster Crassostrea gigas (Thunberg). Aquaculture 288: 172–183. Hubert, S., and D. Hedgecock, 2004 Linkage maps of microsatellite DNA markers for the Pacific oyster Crassostrea gigas. Genetics 168: 351-362. Husband, B. C., and D. W. Schemske, 1996 Evolution of the magnitude and timing of inbreeding depression in plants. Evolution 50 (1): 54-70. Huvet, A., P. Boudry, M. Ohresser, C. Delsert, and F. Bonhomme, 2000 Variable microsatellites in the Pacific oyster Crassostrea gigas and other cupped oyster species. Anim. Genet. 31: 71-72. Kelly, J. K., 2005 Epistasis in monkeyflowers. Genetics 171:1917-1931. Koelewijn, H. P., V. Koski, and O. Savolainen, 1999 Magnitude and timing of inbreeding depression in Scots pine (Pinus sylvestris l.). Evolution 53: 758-768. Korringa, P., 1946 A revival of natural oyster beds? Nature 158: 586-587. Lander, E. S., and D. Botstein, 1989 Mapping Mendelian factors underlying quantitative traits using RFLP linkage maps. Genetics 121: 185-199. 69 Langdon, C., F. Evans, D. Jacobson and M. Blouin, 2003 Yields of cultured Pacific oysters Crassostrea gigas Thunberg improved after one generation of selection. Aquaculture 220(1-4): 227-244. Launey, S., and D. Hedgecock, 2001 High genetic load in the Pacific oyster Crassostrea gigas. Genetics 159: 255-265. Li, G., S. Hubert, K. Bucklin, V. Ribes, and D. Hedgecock, 2003 Characterization of 79 microsatellite DNA markers in the Pacific oyster Crassostrea gigas. Mol. Ecol. Notes 3: 228-232. Li, Z.-K., L. J. Luo, H. W. Mei, D. L. Wang, Q. Y. Shu, et al. 2001 Overdominant epistatic loci are the primary genetic basis of inbreeding depression and heterosis in rice. I. Biomass and grain yield. Genetics 158: 1737–1753. Lorieux M., X. Perrier, B. Goffinet, C. Lanaud, and D. González de León, 1995 Maximum-likelihood models for mapping genetic markers showing segregation distortion. 2. F 2 populations. Theor. Appl. Genet. 90: 81–89. Luo, L., and S. Xu, 2003 Mapping viability loci using molecular markers. Heredity 90: 459-467. Lynch, M., and B. Walsh, 1998 Genetics and Analysis of Quantitative Traits. Sinauer Associates, Inc. Sunderland Massachusetts, USA. Magoulas, A., B. Gjetvag, V. Tersoglou and E. Zouros, 1998 Three polymorphic microsatellites in the Japanese oyster, Crassostrea gigas (Thunberg). Anim. Genet. 29:69-70. Mallet, A. L., E. Zouros K. E. Gartner-Keykay, K. R. Freeman, and L. M. Dickie, 1985 Larval viability and heterozygote deficiency in populations of marine bivalves: evidence from pair matings of mussels. Mar. Biol. 87: 165-172. McCune, A. R., R. C. Fuller, A. A. Aquilina, R. M. Dawley, J. M. Fadool et al., 2002 A low genomic number of recessive lethals in natural populations of bluefin killifish and zebrafish. Science 296: 2398-2401. McGoldrick, D. J., D. Hedgecock, L. J. English, P. Baoprasertkul, and R. D. Ward, 2000 The transmission of microsatellite alleles in Australian and North American stocks of the Pacific oyster (Crassostrea gigas): Selection and null alleles. J. Shellfish Res. 19: 779-788. 70 Meinke, D. W., 1995 Molecular genetics of plant embryogenesis. Annu. Rev. Plant Physiol. Plant Mol. Biol. 46: 369-394. Melchinger, A. E., H. P. Piepho, H. F. Utz, J. Muminovic, T. Wegenast, et al. 2007 Genetic basis of heterosis for growth-related traits in Arabidopsis investigated by testcross progenies of near-isogenic lines reveals a significant role of epistasis. Genetics 177: 1827-1837. Mitchell-Olds, T., 1995 Interval mapping of viability loci causing heterosis in Arabidopsis. Genetics 140: 1105–1109. Morton, N. E., J. F. Crow, and H. J. Muller, 1956 An estimate of the mutational damage in man from data on consanguineous marriages. Proc. Natl. Acad. Sci. U.S.A. 42: 855. Mukai, T., S. I. Chigusa, L. E. Mettler, and J. F. Crow, 1972 Mutation rate and dominance of genes affecting viability in Drosophila melanogaster. Genetics 72:333-355. Piepho, H.-P., 2001 A quick method for computing approximate thresholds for quantitative trait loci detection. Genetics 157: 425-432. Remington, D. L., and D. M. O‘Malley, 2000 Whole genome characterization of embryonic stage inbreeding depression in a selfed loblolly pine family. Genetics 155: 337-348. Rice, W. R., 1989 Analyzing tables of statistical tests. Evolution 43: 223–225. Rizski, M. T. M., 1952 Ontogenetic distribution of genetic lethality in Drosophila willistoni. J. Exp. Zool. 121: 327-350. Roff, D. A., 2002 Inbreeding depression: Tests of the overdominance and partial dominance hypotheses. Evolution 56: 768-775. Salathe, P., and D. Ebert, 2003 The effects of parasitism and inbreeding on the competitive ability in Daphnia magna: evidence for synergistic epistasis. J. Evol. Biol. 16: 976-985. SAS Institute Inc., SAS 9.1.3 Help and Documentation, Cary, NC: SAS Institute Inc., 2000-2004. Savolainen, O., K. Karkkainen, and H. Kuittinen, 1992 Estimating numbers of embryonic lethals in conifers. Heredity 69: 308-314. 71 Sekino, M., M. Hamaguchi, F. Aranishi, and K. Okoshi, 2003 Development of novel microsatellite DNA markers from the Pacific oyster, Crassostrea gigas. Mar. Biotechnol. 5: 227-233. Seto, F., 1954 Time of action of a series of recessive lethal factors in Drosophila melanogaster. J. Exp. Zool. 126: 17-32. Seto, F., 1961 A developmental study of recessive lethals from wild populations of Drosophila melanogaster. Am. Nat. 95: 365-373. Shuelke, M., 2000 An economic method for the fluorescent labeling of PCR fragments. Nat. Biotechnol. 18: 233-234. Thomas, T. L., 1993 Gene expression during plant embryogenesis and germination: an overview. Plant Cell 5: 1401-1410. Thorson, G., 1950 Reproductive and larval ecology of marine bottom invertebrates. Biol. Rev. 25: 1–45. Van Ooijen, J.W., and R.E. Voorrips, 2001 JoinMap® 3.0, Software for the calculation of genetic linkage maps. Plant Research International, Wageningen, the Netherlands. Vogl, C., and S. Xu, 2000 Multipoint mapping of viability segregation distorting loci using molecular markers. Genetics 155: 1439-1447. White, K. P., S. A. Rifkin, P. Hurban, and Hogness, D. S., 1999 Microarray analysis of Drosophila development during metamorphosis. Science 286: 2179-2184. Whitlock M. C., and D. Bourguet, 2000 Factors affecting the genetic load in Drosophila: Synergistic epistasis and correlations among fitness components. Evolution 54: 1654-1660. Williams, E. A., B. M. Degnan, H. Gunter, D. J. Jackson, B. J. Woodcroft, et al., 2009 Widespread transcriptional changes pre-empt the critical pelagic–benthic transition in the vetigastropod Haliotis asinine. Mol. Ecol. 18: 1006-2025. Williams, G. C., 1975 Sex and Evolution. Princeton University Press, Princeton N.J. Williams, G. C., L. D. Auckland, M. M. Reynolds and K. A. Leach, 2003 Overdominant lethals as part of the conifer embryo lethal system. Heredity 91: 584-592. 72 Willis, J. H., 1993 Effects of different levels of inbreeding on fitness components in Mimulus guttatus. Evolution 47:864-876. Winemiller, K. O., and K.A. Rose, 1992 Patterns of life-history diversification in North American fishes: Implications for population regulation. Can. J. Fish. Aquat. Sci. 49: 2196-2218. Wright, S., 1977 Inbreeding in animals: differentiation and depression, pp.44-96, in Evolution and the genetics of populations, vol. 3, University of Chicago, Chicago. Xu, Y., L. Zhu, J. Xiao, N. Huang, and S. R. McCouch, 1997 Chromosomal regions associated with segregation distortion of molecular markers in F2, backcross, doubled haploid, and recombinant inbred populations in rice (Oryza sativa L.). Mol. Genet. Genomics 253: 535–545. Yamtich J., M. L. Voigt, G. Li, and D. Hedgecock, 2005 Eight microsatellite loci for the Pacific oyster Crassostrea gigas. Anim. Genet. 36: 524-526. Zaykin, D. V. and A. I. Pudovkin, 1993 Two programs to estimate significance of Chi- square values using pseudo-probability test. J. Hered. 84: 152. Zhu, C., C. Wang, and Y. M. Zhang, 2007 Modeling segregation distortion for viability selection 1. Reconstruction of linkage maps with distorted markers. Theor. Appl. Genet. 114: 295-305. 73 Chapter 2 Fine-scale temporal analysis of mortality and genetic load during metamorphosis in the Pacific oyster Crassostrea gigas 2.1 Abstract Settlement and metamorphosis is a critical period in the life cycle of marine invertebrates, during which larvae undergo substantial morphological, sensory, and genetic changes regulated by distinct developmental processes. High mortality during this transition has been well documented for a variety of marine invertebrates and is generally interpreted as occurring post-settlement; little is known about how mortality may occur during the process of metamorphosis itself. Previous genetic work with the Pacific oyster Crassostrea gigas has shown that the high genetic load present in this species is expressed predominately during metamorphosis; however, sparse temporal sampling during metamorphosis did not allow full characterization of the patterns and fine-scale timing of this genotype-dependent mortality. In this study, I followed segregation ratios of microsatellite markers during the settlement of an inbred F 2 cross of the Pacific oyster, sampling daily the new spat (post-metamorphic juveniles), as well as the larvae in the water column, to determine the timing and patterns of genotype dependent-mortality, i.e. whether mortality occurs before, during, or after metamorphosis. Settlement occurred over nine days, showing an early and a late peak 74 indicating a bimodal distribution of settlement timing; tracking the survival of settlers for 40 days after initial settlement revealed no post-settlement mortality. Examination of the temporal segregation data during metamorphosis revealed a marker linked to two deleterious mutations, one which appeared to act during metamorphosis, possibly affecting the morphogenetic pathway. The other linked mutation appeared to cause a delay in metamorphosis or prevented metamorphosis from beginning, suggestive of a defect in the competence pathway. Overall, selection during the larval-juvenile transition appears to be confined to the induction of metamorphosis and metamorphosis itself, which highlights the importance of understanding the developmental pathways associated with this critical transition. 2.2 Introduction Settlement and metamorphosis is a critical period in the life cycle of marine invertebrates. It is the final developmental hurdle to successful recruitment, wherein larvae go through the dramatic ecological and biological transformation from a free- swimming planktonic larval form to a sessile benthic juvenile. The successful completion of this process has profound implications for the evolution and ecology of species in the marine environment (Gaines and Roughgarden 1985, Hunt and Shiebling 1997, Roughgarden et al. 1988, Rodriquez et al. 1993). In marine molluscs, metamorphosis is well described, characterized by the complete rearrangement of the body plan, coinciding with the loss of larval features, such as the velum, and the emergence of juvenile or adult characteristics, such as the gills (e.g. Cole 1938, Hickman 75 and Gruffyd 1971, Bonar 1976, Degnan and Morse 1995). The process of settlement and metamorphosis comprises two distinct phases: 1) the attainment of ‗competency‘, the developmental capacity to respond to appropriate settlement cues and exhibit settlement behavior, and 2) the morphogenetic transformation that occurs when larvae attach to the substratum and complete metamorphosis. The role and variety of chemical cues in the induction of metamorphosis, the developmental attainment of competency, and the consequences of delayed metamorphosis have been the subjects of much research in the field of larval biology and ecology (Reviewed by Crisp 1974, Degnan and Morse 1995, Pechenik 1990, Rodriguez et al 1993, Hadfield 1986, Jackson 2002). Recently, developmental and gene expression studies have further revealed the complex nature of the metamorphic transition, showing that the competency and morphogenetic pathways display unique gene expression profiles (e.g. Jackson 2005, Williams et al. 2009). For example, patterns of increased gene expression late in the larval stages co-occur with the accumulation and deployment of a threshold level of receptors and signal transducers, which can then respond to inducible chemical cues that initiate metamorphosis (Degnan and Morse 1995, Jackson et al. 2005, Heyland and Moroz 2006). At metamorphosis, a suite of genes control an ‗anticipatory‘ pathway, which begins to form juvenile structures prior to metamorphosis (e.g. digestive and shell formation pathways; Jackson et al. 2007), and then the morphogenetic transformation is dictated by the up-regulation of genes related to apoptosis, cell cycling, protein synthesis, and calcium flux pathways, among others (Jackson et al. 2005, Jackson and Degnan 2006, Jackson et al. 2007, Williams et al. 2009). Though there appear to be clear patterns of gene classes expressed 76 across invertebrate taxa during competency and metamorphosis (e.g. Heylund and Moroz 2006), most studies have focused on only a few marine gastropods (Aplysia and Haliotus spp.) and ascidians (e.g. Degnan et al. 1997, Eri et al. 1999, Kawashima 2005, Jacobs et al. 2006), all of which have non-feeding or low-dispersal planktonic forms. Less is known about genetic processes at metamorphosis in other marine invertebrates, such as marine bivalves with feeding, widely dispersing, planktonic larvae. Perhaps the most striking ecological feature of metamorphosis in marine invertebrates, and marine bivalves in particular, is the substantial mortality that occurs at this stage in the life cycle. A high rate of mortality in recent settlers has been observed for a wide variety of invertebrate taxa, and, generally, survival curves of new settlers are type III: mortality is initially high but decreases rapidly after the first few days or weeks and then levels off (e.g. Rodriguez et al. 1993, Reviewed by Gosselin and Qian, 1997, Hunt and Sheibling 1997). Early mortality in invertebrates has largely been addressed in an ecological context, with field studies examining how recruitment rates and variation in settlement patterns affect population distribution and dynamics (e.g. Keogh and Downes 1982, Rodriguez et al. 1993, Gosselin and Qian 1996, Hunt and Shiebling 1997). For many of these studies it seems to be almost implicitly assumed that larvae survive through settlement but die at some point after and that type-III mortality occurs post- settlement (Keough and Downes 1982, Hunt and Shiebling 1997). Many of these studies suffer from ascertainment bias because they truly only measure ―recruitment,‖ survival of a population to a certain time point after settlement. Monitoring mortality immediately after settlement is difficult, if not impossible, for many species (e.g. Keough and Downes 77 1982). Few studies accurately measure larval supply, metamorphosis, and early and late post-settlement periods within the same experiment. Thus, a comprehensive understanding of when mortality occurs during settlement phases is lacking. Even less understood is how genetic processes and endogenous variation affect larval survival both during the process of metamorphosis and post-settlement. Experimental laboratory studies allow for more detailed analysis of mortality during settlement in marine invertebrates, but few such studies exist. Jones and Jones (1983) found that only a small fraction (10-30%) of pre-metamorphic oyster larvae successfully complete metamorphosis in hatchery conditions. Haws et al. (1993) also noted that substantial mortality occurred during the metamorphic transition in their detailed examination of biochemical and physiological changes during metamorphosis of the Pacific oyster Crassostrea gigas. Culture experiments have shown significantly greater post-metamorphosis survival associated with improved diet during the larval stages, suggesting that exogenous processes related to energy consumption during metamorphosis are important (e.g. Helm et al. 1991, Coutteau et al. 1994, Pernet and Tremblay 2004). Recent experimental genetic work with the Pacific oyster has confirmed high mortality at metamorphosis, finding that half of the deleterious recessive mutations uncovered in inbred crosses are expressed at metamorphosis and that this genetic mortality accounts for much of the substantial type-III culture mortality observed during the life cycle (Chapter 1). The stage-specific timing of inbreeding depression was determined by following the temporal onset of genotype deficiencies (primarily selection against identical-by-descent homozygotes) in daily larval samples, a post settlement 78 sample, and an adult sample. The finding of substantial, genetically determined mortality during metamorphosis, along with previous descriptions of novel gene expression pathways expressed leading up to and during metamorphosis (e.g. Degnan and Morse 1995, Williams et al. 2009), highlights the complex role of genetic variation during this critical stage and the consequences of the expression of deleterious mutations. Unfortunately, there were no temporal genetic data during metamorphosis in the previous experiment (Chapter 1). Understanding the fine-scale temporal pattern of selection against deleterious alleles during metamorphosis could provide important new information regarding mortality during this transition, and may highlight the importance of the two distinct phases of metamorphosis, competence and morphogenesis, . In the current study, I focused temporal sampling during the period of metamorphosis and settlement, following marker segregation distortion over the course the larval/juvenile transition. I allowed an inbred, F 2 family to set naturally on adult shell, and took daily samples of spat (recently settled juvenile oysters) and larvae still in the water column throughout the settlement period. Utilizing the method developed in Chapter 1, I first identified markers that were distorted at day 60 but were also in Mendelian proportions in pre-metamorphic, late larval samples. Then, I determined the segregation ratios and genotype deficiencies of these metamorphosis-distorted markers, in the late larval and spat populations, to explain the deficiency of marker genotypes observed post-metamorphosis. I propose three hypotheses for how selection and mortality might proceed: 1) larvae are able to go through metamorphosis and settle, but die soon after, which can be detected by high mortality of juveniles after settlement; 2) 79 larvae begin metamorphosis, but die sometime during this transition; thus, affected genotypes are present in the water column at time x, but at time x+1, they are missing from both the spat and larval pool; and 3) individuals with the affected genotypes are not able to begin metamorphosis and are thus are deficient in the spat population but remain and perhaps accumulate in the larval population, because settlement is delayed or fails to occur altogether. The detailed sampling and genotyping strategy carried out in this experiment should shed light on the fine-scale timing of genotype-dependent mortality, and determine the extent to which competency, morphogenesis, or post settlement processes are affected by the expression of genetic load. 2.3 Materials and Methods Crosses and Culturing methods Inbred lines 51 and 35 were derived from a naturalized population of C. gigas in Dabob Bay, WA, with initial families made from pair crosses of wild individuals in 1996 (Hedgecock and Davis 2007). These lines were inbred (full-sib mating) for four generations leading up to the F 1 hybrid cross that was made in 2007 (see Fig. 1, Introduction, for pedigrees). In 2009, the experimental F 2 family (f = 0.397) was created by mating a pair of male and female full-siblings from the 2007 51×35 F 1 hybrid cross. Crosses were performed at the University of Southern California (USC), Wrigley Marine Science Center (WMSC) on Catalina Island, CA. Pedigrees of parents were verified with microsatellite DNA markers (Hedgecock and Davis 2007) 60 80 The cross was performed by stripping the gonad from a single male and a single female, combining their gametes in a two-liter beaker of fresh seawater for fertilization (e.g. Breese and Malouf 1975, Hedgecock and Davis 2007). After a one hour incubation, one million zygotes were stocked in a 200-l vessel (5 larvae·ml -1 ) with fresh sea water and fed a diet of Isochrysis galbana (Tahitian Isolate) every two days, at a starting concentration of 30,000 cells·ml -1 , which was increased, as larvae grew, following standard larval rearing protocols for the Pacific oyster (e.g. Breese and Malouf 1975, Hedgecock et al. 1996, Chapters 2, 4). Deployment of shell and sampling protocols At day 18, a substantial number of larvae had developed eye spots and displayed probing with the larval foot, both characteristic of settlement behavior (Bonar 1976, Kennedy et al. 1996), so cured adult shell were deployed on a Vexar® mesh cylindrical harness for natural settlement (Fig. 7a). Adult shell was placed with the nacreous side facing both up and down and with shell overlapping to form random 3-D structures to give larvae a number of different substrates and angles to settle on regardless of their orientation to light or gravity, which affect settlement (Kennedy et al. 1996, Baker 1997, Baker and Mann 1998). Starting in the afternoon on day 18, and at about the same time each day thereafter, the harness was carefully pulled up, and the surface of shells were sprayed vigorously with seawater to return any probing or non-settled larvae back to the 81 Figure 7 Pictures of shell deployment and bags with spat for grow-out The collection of shells deployed each day on the Vexar® harness (A), and shells with spat separated in bags in the nursery system for grow-out (B). 81 82 200-l vessel and the larval pool. Larvae were filtered out of the 200-l vessel and concentrated for survival estimates each day during settlement. Survival was calculated by counting the number of larvae in four to six random sub-samples of known volume (50-100 μl) from homogeneously mixed, concentrated cultures (500-1000 ml). These counts were then multiplied by the inverse of the fraction of the subsample volume to the total volume ( , averaged to estimate the mean number of larvae in the concentrate, and then multiplied by the inverse of the concentration. The settled spat on each shell were counted, shell by shell, under a dissecting microscope at 8-48×, until no more could be found. This involved inspecting both sides of the shell under a range of magnifications and orientations. New shells that had been incubating for 24 hours in fresh, filtered seawater, were then placed on the harness, and lowered back into the 200-l tank, which had been rinsed and refilled with fresh seawater and algae. Finally, the counted larvae were placed back into the tank. This process was repeated each day until no larvae remained in the water column. From these two counts, the number of larvae at day x (L x ) and the number of settlers at day x (S x ), I was able to infer larval mortality between two adjacent days, M, with this equation: ( . Shells that had spat settling on a particular day were then collected and placed in a Vexar® mesh bag in the upwelling nursery system and grown out until day 60 to increase the amount of tissue for DNA extraction (Fig 7b). Spat were fed a mix of live Isochrysis galbana and Shellfish Diet 1800 (Reed Mariculture, Campbell, CA), twice daily. 83 To determine if mortality occurred post-settlement (after the day on which they were recorded as settling), a subset of metamorphosed spat from day 19, 21, and 24, were followed to day 60 to determine how many were still alive. This was done by marking an area around 50 spat on two shells from each of the three days and then counting the number of surviving spat at day 60. DNA extraction, PCR, and Electrophoresis At 60 days, spat were scraped off shell and placed directly in Qiagen tissue lysis buffer (Qiagen, Valencia CA) in individual tubes and stored at -80ºC until extraction using the Qiagen DNeasy blood and tissue kit. Live larvae were placed in small volumes of sterilized water with a drop of 70% ethanol to retard their movement, then they were pipetted directly into 96 well PCR plates with extraction buffer consisting of 1× PCR buffer (Promega, Madison,Wisconsin), 0.2 mM EDTA, 1.0 μg/μl Protienase K (Shelton Scientific, Connecticut) and purified H 2 O, which was frozen at -80ºC until extraction. Larvae were extracted in 40 μl volumes in a 96 well thermocycler (BioRad Tetrad, Hercules, CA) held at 56ºC for three hours followed by 15 minutes at 95ºC. Parent tissue was stored in 70% ethanol at 4ºC until extraction and DNA was extracted from 10-25 mg of tissue using either the Qiagen DNeasy animal tissue kit or the Gentra Puregene tissue kit (Qiagen, Valencia CA) following the manufacturer‘s protocols. Eighty-four markers cloned from the Pacific oyster were tested in this study (Magoulas et al. 1998, McGoldrick et al. 2000, Huvet 2000, Li et al. 2003, Sekino et al. 2003, Yamtich et al. 2005, Yu and Li 2007, Wang et al. 2007), many of which are on 84 published linkage maps (Hubert and Hedgecock 2004, Hubert et al. 2009). Markers were named as in the source publication, except for those developed in this laboratory (McGoldrick et al. 2000, Li et al. 2003, Yamtich et al. 2005), which were abbreviated from their original description, e.g. ucdCgi195 to Cg195. The polymerase chain reactions (PCR) were carried out in two phases; the first to amplify the target microsatellite marker, and the second to incorporate a fluorescently labeled dye attached to a ―zip‖ sequence that is complementary to the 3‘ tail of the forward primer (Shuelke 2000). PCR cycle conditions consisted of an initial denaturing step at 94°C for 2 minutes, then 20 phase-one cycles (30 sec at 94°C, 45 sec at T m , 45 sec at 72°C), followed by 10 ―zip‖ cycles (30 sec at 94°C, 45 sec at 59°C and 45 sec at 72°C) and a final 10-min elongation step at 72°C. The number of zip cycles was often increased to 30 or 40 when using larval DNA as template. T m (the optimum annealing temperature) and the MgCl2 concentration varied, depending on the locus (Magoulas et al. 1998, McGoldrick et al. 2000, Huvet 2000, Li et al. 2003, Sekino et al. 2003, Yamtich et al. 2005). PCR products were electrophoresed on a 4% denaturing PAGE gel (acrylamide: bisacrylamide 19:1, 3M Urea) using 1 TBE (Tris-base, boric acid, and EDTA di-sodium salt) on the ABI Prism 377 DNA Sequencer (Perkin Elmer, Waltham, Massachusetts). Gel images were used to score the fluorescently labeled microsatellites by eye, comparing progeny alleles to adult alleles along with fluorescently labeled size standards (DeWoody et al. 2004). 85 Genetic analysis of mortality during settlement To identify markers that became distorted during settlement, markers were, first, tested for significant deviations from expected Mendelian ratios in a sample of 60-day- old spat (n = 192) made up of individuals that settled either on day 19 or day 24. These two days were chosen because they represented the bulk of the settlement (68% of spat settled on day 19 or day 24; Fig. 8), making the pooled sample fairly representative of the total settled population that would have been sampled at day 60 without regard to settlement timing (as in Chapter 1). Markers were only considered for analysis through metamorphosis, if they exhibited segregation distortion in the pooled spat sample. Once markers exhibiting segregation distortion in the pooled spat sample were identified, they were genotyped in day-18 larval samples to eliminate markers that became distorted before metamorphosis. Markers that exhibited Mendelian ratios before metamorphosis were then genotyped in larvae and spat from each day throughout metamorphosis, in order to follow the pattern of selection and determine whether observed genotype deficiencies and occurred before, during, or after metamorphosis. 2.4 Results Settlement and mortality data Larvae showed the characteristic signs of settlement behavior (e.g. eyespots, and probing foot) on day 17 and adult shells were deployed starting on day 18. Around 50 larvae had already settled on day 17, sticking to the bottom of the poly-carbonate tank in 86 the absence of adult shell; these were not sampled. On the initial day of sampling, day 18, the number of surviving larvae was 132,000 ± 7,500 (from a starting stock of 1,000,000 embryos), which represents 13.2% survival to the end of the larval stage (Fig. 8). Settlement occurred over nine days, from day 18 to day 27, with settlement peaks at day 19 (1,084 counted spat) and day 24 (1,484 spat), indicating that settlement in this family followed a bimodal distribution. In total, 3,776 settlers were counted, representing 0.37 % survival over the 60 days of the experiment. Larval abundance appeared to drop sharply on the days when a substantial number of settlers were recorded (e.g. day 19 and day 24) and a corresponding spike in larval mortality was observed (Fig. 8). After day 24, the abundance of larvae steadily declined, despite few additional spat setting. Mortality of new settlers followed for approximately 40 days after initial settlement (examined at day 60) was minimal, with estimates of 2%, 0% and 1% for settlers on day 19, 21, and 24, respectively. In general, very few dead or empty shells from any settlement day were observed. Because no evidence of post-settlement mortality was found, segregation ratios at markers were not followed from initial settlement samples to day 60 settlement samples. Segregation analyses and segregation distortion during settlement Eighty-four microsatellite markers were tested, yielding 46 informative markers in the parents of the experimental progeny. A subset of 24 markers, comprising at least two markers on each of the 10 Pacific oyster linkage groups (LG), were tested for 87 Figure 8 Settlement and Mortality data Mortality is calculated as the difference in the abundance of larvae between days, accounting for the number new settlers. Error bars on the larval abundance estimates represent one standard error of the mean from volumetric counts. 87 88 distortion of segregation ratios in day 19 (n = 96) and day 24 (n = 96) settlers sampled at 60 days post fertilization. None of the three informative markers on LG 2 could be scored, so these markers and this linkage group were not included in the analysis. Significantly distorted segregation ratios at the = 0.05 level were found at 13 of 21 (62%) markers, Table 4. These distortions were mainly caused by deficiencies of homozygous genotypes or heterozygote carriers of parentally shared, presumably identical-by-descent, deleterious alleles. Markers with distorted segregation ratios were distributed broadly across the genome, on eight out of the nine linkage groups analyzed. Genotyping all but two (Cg162 and Cg156) post-settlement distorted markers in the day- 18 larval sample revealed that markers Cg109, Cg205, and Cg175 on LG 4, LG 6, and LG 8, respectively, exhibited Mendelian segregation ratios (three of 11 tested; Table 4), indicating that deleterious alleles linked to these markers were first expressed during settlement and metamorphosis. Difficulties in genotyping larval samples only allowed temporal analysis of Cg205 after day 19. In the pooled spat sample for Cg205, significant deficiencies in both homozygous genotypes were evident, suggesting that Cg205 might be linked to two deleterious alleles in repulsion phase. Starting with the inspection of genotypic data for day-19 spat, observed segregation ratios were highly distorted, with substantial deficiencies of the two homozygous genotypes (Table 5, Fig. 9). Segregation ratios in day-19 larvae did not deviate from their Mendelian expectation however, indicating that genotypes deficient in the spat sample were still present among larvae in the water column. At the next larval sample, day 21, segregation ratios were significantly 89 Table 4 Marker segregation data for day 18 larvae and day 60 spat Cross type LG a Marker Day Genotype numbers Total Chi- square P- Value AA×AB AA AB 5 Cg139 60 85 96 181 1.77 0.414 AB×AB AB AA BB 1 Cg200 60 100 53 22 175 14.55 0.001 18 61 19 11 91 11.97 0.003 Cg124 60 132 51 0 183 64.28 <0.001 18 60 2 0 62 54.39 <0.001 3 um2L48 60 109 52 13 174 28.61 <0.001 18 55 12 7 74 18.19 <0.001 Cg148 60 130 51 2 183 58.64 <0.001 18 85 4 0 89 74.08 <0.001 Cg162 60 114 23 41 178 17.69 <0.001 4 Cg198 60 85 56 38 179 4.07 0.131 5 Cg163 60 137 40 0 177 71.24 <0.001 18 79 3 0 82 70.66 <0.001 Cg138 60 91 38 37 166 1.55 0.460 6 Cg205 60 143 28 13 184 58.99 <0.001 18 41 28 20 89 1.99 0.370 Cg209 60 138 3 44 185 62.94 <0.001 18 52 3 17 72 19.67 <0.001 Table continues next page 90 Table 4 (continued) Cross type LG a Marker Day Genotype numbers Total Chi- square P- Value 7 Cg28 60 113 45 26 184 13.51 0.001 18 63 24 8 95 15.51 <0.001 8 um2L16 60 92 46 36 174 1.72 0.422 9 Cg183 60 82 51 31 164 4.88 0.087 Cg184 60 51 17 23 91 2.12 0.346 10 Cg140 60 99 32 41 172 4.87 0.088 AB×AC AA AC AB BC 1 cmrCg5 60 40 52 43 41 176 2.05 0.563 4 Cg109 60 31 66 46 34 177 17.10 0.001 18 18 28 17 26 89 4.17 0.244 7 Cg156 60 24 40 40 52 156 10.15 0.017 8 Cg175 60 27 27 50 51 155 14.26 0.003 18 16 19 28 17 80 4.50 0.212 10 Cg129 60 18 50 62 60 190 26.17 <0.001 18 4 22 19 40 85 30.81 <0.001 a linkage group. χ 2 represents the goodness of fit chi-square test to expected Mendelian inheritance ratios and P-values indicate significance of the chi-square test. 91 distorted, but only the AA genotype was deficient; the BB genotype did not deviate from its expected 1:2 ratio, relative to the frequency of AB. At day 24, segregation ratios in the larval sample looked dramatically different from the previous time point, with a much larger proportion of BB genotypes than expected (60% compared with the expected 25%), and significantly fewer AB genotypes, while the number (and relative frequency) of AA genotypes remained highly deficient (only five spat with AA genotypes were observed in both day 21 and day 24 larval samples; Fig. 9, Table 5). In the day-24 spat sample, both homozygous genotypes were again deficient, but segregation ratios were significantly different relative to the day 19 sample, owing to a slightly greater frequency of BB individuals settling (P<0.001, r×c contingency chi-square test; Zaykin and Pudovkin 1992). At the final larval time point, day 26, the BB genotype was in even greater excess (79% compared to an expected 25%), while the AA genotype was completely absent; the AB genotype was deficient at this time point as well (Table 5, Fig. 9). In the day-26 spat sample, a slight increase in BB genotypes was observed, but both the BB and AA genotypes were substantially deficient, a result similar to that for spat samples on days 19 and 24 and consistent with segregation data for the pooled or cumulative spat sample (days 19, 24, and 26 combined; Fig. 9). In summary, individuals with the AA genotype suffered swift mortality relatively early in settlement (day 19 to day 21) with few individuals surviving in the water column or present as spat after this point. In contrast, while BB individuals were also missing from the spat samples relatively early in settlement (day 19), they were present in the larval pool throughout settlement, becoming the dominant genotype remaining in the water column up to the last day in 92 Table 5 Genotype numbers for Cg205 during settlement Settlers Cg205 Larvae Cg205 Day AB AA BB Chi- Square P-Value Day AB AA BB Chi- Square P-Value d18 -- -- -- -- -- d18 41 20 28 1.99 0.370 d19 82 9 2 55.26 <0.0001 d19 36 14 17 0.64 0.725 d21 -- -- -- -- -- d21 42 5 24 12.55 0.002 d24 61 11 19 11.97 0.003 d24 27 5 47 52.57 <0.0001 d26 30 3 7 10.80 0.005 d26 7 0 27 54.65 <0.0001 Cum. 173 23 28 86.50 <0.0001 Chi-square represents the goodness-of-fit chi-square test to expected Mendelian inheritance ratios. 92 93 Figure 9 Cg205 genotype proportions in larvae and spat during settlement ns is non-significant, asterisks denote the level of significance for the goodness-of-fit chi-square test (cf. Mendelian expectations of 2:1:1, AB:AA:BB ) of genotype proportions at Cg205 (* P<0.05, ** P<.01, *** P<0.001, **** P<0.0001) 93 94 larval development, never settling in appreciable numbers even in the last days of settlement. The differing temporal patterns of mortality and segregation distortion in the two homozygous genotypes suggest that Cg205 is linked to two deleterious recessive mutations with very different patterns of selection. 2.5 Discussion In this study, daily sampling of larvae and spat during the settlement of an F 2 family of the Pacific oyster was carried out to examine, in detail, the patterns of settlement and mortality during metamorphosis. I first describe the general patterns of settlement and mortality from the observed counts of settlers and larvae, and then turn to the temporal genetic analysis of segregation distortion during metamorphosis, which addresses the biologically relevant hypotheses regarding patterns of mortality during settlement. Settlement patterns and post settlement mortality Mortality during settlement was very high, with only 0.37% of the initial larvae surviving as spat at day 60. Compared with survival up to the end of the larval stage (~13%), survival through settlement (2.8%) was almost five-times lower, which confirms settlement and metamorphosis as a critically important life history transition, during which substantial mortality occurs (e.g. Hunt and Sheibling 1996, Chapter 1). Comparing daily settlement and larval abundance, high rates of mortality in the larval pool appeared to coincide with the peaks of settlement, suggesting that larval mortality 95 was associated with some aspect of the metamorphic transition. Essentially, when large pulses of larvae attempted to go through metamorphosis, a substantial portion of them died. Another striking result was that larval settlement was concentrated on two days, producing a bimodal distribution. These two pulses of settlement accounted for 68% of all settlement in this family over the nine-day period. Field-based recruitment studies have shown that settlement during the course of a season is often non-random, with specific peaks of settlement and variable survival over time (e.g. Raimondi 1990, Pineda 1994, Balch and Sheibling 2000, Broitman et al. 2000, Pineda 2006). Temporal variation in settlement in the marine environment could be related to a number of factors, including reproductive condition, spawning timing, or oceanographic features, such as currents and food availability, none which were variable or relevant in this experiment. Instead, variation of settlement timing, particularly the early and late peaks of settlement, must somehow depend on genetic variability for settlement timing within the family. Only one family was studied in this experiment, thus, the observed settlement patterns may not be generalizable. Nevertheless, the settlement pattern is intriguing and poses the question: Is there genetic control of settlement timing? The reader is referred to the Appendix, where evidence for genetic difference is presented. Analysis of spat survival 40 days after initial settlement shows conclusively that very little mortality occurred post-settlement. Essentially, once larvae had gone through metamorphosis and appeared as spat on adult shell (time to census was 24 hours or less), individuals survived at a rate of almost 100%. These findings support previous observations of low genotype-dependent morality after metamorphosis, in two inbred 96 experimental crosses (Chapter 1). Settlement studies of natural populations do, however, find substantial post-settlement mortality in a variety of marine invertebrates (Hunt and Shiebling 1996, Gosselin and Qian 1997) and for oysters in particular (e.g. Kennedy et al. 1996, Newell et al. 2000). In this experiment, no external sources of mortality, such as predators or oceanographic stressors, were present. Thus, these results suggest only that genetic or endogenous causes of mortality do not greatly affect oyster survival once settlement is complete in inbred crosses. Temporal genetic analysis of mortality during settlement and metamorphosis Analysis of temporal segregation identified three of 11 markers (Cg205, Cg175, and Cg109) that became distorted during metamorphosis. Previously, I showed that half of the detected viability loci were expressed at metamorphosis (Chapter 1). Though three of 11 is lower than the expected half, it is not significantly different than the seven of 15 loci expressed at metamorphosis in the temporal study (P=0.428, Fishers exact test, 2×2 table). Three markers that were distorted prior to metamorphosis showed patterns of segregation distortion at day 18 that differed substantially from the day-60 spat sample (significant contingency chi-square test of heterogeneity, P<0.001 for Cg163, Cg124, and Cg148; Zaykin and Pudovkin 1992). This finding differs from previous results, in which genotypic proportions remained stable after viability selection had distorted them (Chapter 2). As mentioned in the settlement timing results, however, some individuals had settled but were not sampled before the day 18 larval sample was taken. Thus, some 97 markers, which were distorted at day 18 and were interpreted to be under pre- metamorphic selection, may actually have been undergoing viability selection during the earliest phases of metamorphosis. At the outset of this study, three hypotheses were proposed to explain the previously observed patterns of substantial genotype-dependent mortality at metamorphosis in inbred crosses of the Pacific oyster: 1) mortality occurs post-settlement due to genetic or developmental abnormalities affecting fitness just after the transition to the juvenile stage, 2) mortality occurs during the process of metamorphosis, because deleterious mutations in critical genes involved in morphogenesis are expressed and, 3) mortality occurs from the inability of larvae to commence settlement behavior or begin metamorphosis, because of mutations affecting the development of competency or delaying metamorphosis. The observation of very minimal mortality after initial settlement suggests that genetic load does not affect viability after settlement and, thus, hypothesis 1 is falsified. Results of the temporal genetic analysis address hypotheses two and three. Segregation ratios at Cg205 revealed the presence of two deleterious loci which exhibited very different temporal patterns of mortality. First, individuals with the AA genotype (the A allele linked to a highly deleterious recessive allele), suffered early (day 19 to day 21) and permanent mortality, disappearing from the water column and never appearing in appreciable numbers in the spat samples (Fig. 9). The relatively swift mortality of AA individuals during the settlement period suggests the expression of a linked deleterious recessive mutation affecting the morphogenetic transition (hypothesis 98 2). Close inspection of the genotype numbers in the day-19 larval and spat samples shows that there was a slight ‗delay‘ in mortality; larval genotypes remained in Mendelian proportions while the spat sample exhibited significant deficiencies of the AA genotype. This can likely be explained by the fact that fitness effects appeared during the metamorphic transition and thus, if not all AA larvae had begun to metamorphose at day 19, some AA individuals would appear to survive initially in the larval pool. Ultimately, deficiencies of the AA genotype were observed in the larval samples at day 21, indicating that mortality in the both the larval and spat samples did occur within a few days of initial settlement. Selection against the deleterious allele linked to the BB genotype at Cg205 produced a very different pattern of mortality. BB genotypes were deficient, relative to expected Mendelian ratios, in all spat samples, but increased slightly in later spat samples, and were in excess in late larval samples (day 26), when essentially all surviving larvae were BB homozygotes (Fig. 9, Table 5). These results suggest that the B allele may be linked to a deleterious recessive mutation causing either a delay in metamorphosis or the inability to develop competence and thus begin metamorphosis. Interestingly, BB larvae did not survive in the plankton very long after settlement was complete (only an estimated 300-500 larvae remained at day 28, on which no settlement was detected), so it is likely that the delay in metamorphosis or inability to develop competence may also have been accompanied by some other metabolic or genetic abnormality, such that larval life could only be prolonged for a finite amount of time. 99 Overall, with segregation data at Cg205, there is evidence both for a mutation expressed during the morphogenetic transition, and a mutation affecting either the development of competence or the initiation of metamorphosis. Thus, this single marker provides support for both hypothesis two and three. These findings highlight the important fitness effects of deleterious mutations that occur during the two distinct phases of metamorphosis, and demonstrate that the expression of genetic load during metamorphosis occurs during a relatively small window of time during settlement. Future studies, using greater marker coverage and examining transcriptional data during metamorphosis, will greatly enhance the identification of the putative function of genes affected by deleterious mutations, and provide insight into the critical genetic and physiological processes that occur during the settlement of Pacific oysters. 100 Chapter 2 References Baker, P, 1997 Settlement site selection by oyster larvae, Crassostrea virginica: Evidence for geotaxis. J. Shellfish Res. 16: 125-128 Baker P. and R. Mann, 1998 Response of settling oyster larvae, Crassostrea virginica, to specific portions of the visible light spectrum. J. Shellfish Res. 17: 1081-1083. Balch, T., and R. E. Scheibling, 2000 Temporal and spatial variability in settlement and recruitment of echinoderms in kelp beds and barrens in Nova Scotia. Mar. Ecol. Prog. Ser. 205: 139–154. Bierne, N., S. Launey, Y. Naciri-Graven, et al., 1998 Early effect of inbreeding as revealed by microsatellite analyses on Ostrea edulis larvae. Genetics 148: 1893- 1906. Bonar, D.B., 1976 Molluscan metamorphosis: A study in tissue transformation. Amer. Zool. 16: 573-591. Breese, W. P. and R. E. Malouf, 1975 Hatchery manual for the Pacific Oyster, Crassostrea gigas (Gould). Oregon Agricultural Experiment Station Special Report No, 443, and Oregon State University Sea Grant College Program Publ. No. ORESU-H-75-002, 22 pp. Broitman, B. R., C. A. Blachette, B. A. Menge, J. Lubchenko, C Krenz, M. Foley, P. T. Raimondi and D. Lohse, 2008 Spatial and temporal patterns of invertebrate recruitment along the west coast of the United States. Ecol. Mono. 78(3): 403– 421. Bucklin, K. A., 2003 Analysis of the Genetic Basis of Inbreeding Depression in the Pacific Oyster Crassostrea gigas. Ph.D. Dissertation, University of California at Davis. Crisp, D. J., 1974. Factors influencing the settlement of marine invertebrate larvae. In, Chemoreception in marine organisms, edited by P. T. Grant & A. M. Mackie. Academic Press, New York, pp. 177-265. Crisp, D. J., 1984 Overview of research on marine invertebrate larvae, 1940-1980. In , Marine Biodeterioration: An interdisiplinary study, edited by J. D. Costlow & R.C. Tipper. Naval Institiute Press, Annapolis, Maryland. pp 103-126. 101 Cole, H. A., 1938 The fate of the larval organs in the metamorphosis of Ostrea edulis. J. Mar. Bio. Assn., U.K. 23: 469-484. DeWoody, J. A., J. Schupp, L. Kenefic, J. Busch, L. Murfitt, et al., 2004 Universal method for producing ROX-labeled size standards suitable for automated genotyping. BioTechniques 37: 348-352. Degnan B. M., and Morse D. E., 1995 Developmental and morphogenetic gene regulation in Haliotis rufescens larvae at metamorphosis Amer. Zool. 35: 391-398. Eri, R., J. M. Arnold, V. F. Hinman, K. M. Green, M. K. Jones, B. M. Degnan, M. F. Lavin, 1999 Hemps, a novel EGF-like protein, plays a central role in ascidian metamorphosis. Development 126: 5809-5818. Gaines, S., and J. Roughgarden, 1985 Larval settlement rate: a leading determinant of structure in an ecological community of the marine intertidal zone. Proc. Natl. Acad. Sci. U.S.A. 82: 3707-3711 Gosselin, L. A. and P. Y. Qian, 1996 Early post-settlement mortality of an intertidal barnacle: A critical period for survival. Mar. Ecol. Prog. Ser. 135: 69-75. Gosselin L. A., and P. Y. Qian, 1997 Juvenile mortality in benthic marine invertebrates. Mar. Ecol. Prog. Ser. 146: 265-282. Hadfield, M. G., 1986 Settlement and recruitment of marine invertebrates: A perspective and some proposals. Bull. Mar. Sci. 39: 418-425. Haws, M. C., L. DiMichele and S. C. Hand, 1993 Biochemical changes and mortality during metamorphosis of the Eastern oyster, Crassostrea virginica, and the Pacific oyster, Crassostrea gigas. Mol. Mar. Biol. Biotechnol. 2: 207-217. Hedgecock, D., and J. P. Davis, 2007 Heterosis for yield and crossbreeding of the Pacific oyster Crassostrea gigas. Aquaculture 272S: S17-S29. Hedgecock, D., D. J. McGoldrick, D. T. Manahan, J. Vavra, N. Appelmans and B. L. Bayne, 1996 Quantitative and molecular genetic analyses of heterosis in bivalve molluscs. J. Exp. Mar. Biol. Ecol. 203: 49-59. Hedgecock, D., G. Li, S. Hubert, K. Bucklin, and V. Ribes, 2004 Widespread null alleles and poor cross-species amplification of microsatellite DNA loci cloned from the Pacific oyster, Crassostrea gigas. J. Shellfish Res. 23: 379-385. 102 Heylund, A., and L. L. Moroz, 2006 Signaling mechanisms underlying metamorphic transitions in animals. Integr. Comp. Biol. 46: 743-759. Hickman, R. W., and L. L. D. Gruffydd, 1971 The histology of the larva of Ostrea edulis during metamorphosis. In: E. J., Crisp (ed.) Fourth European marine biology symposium pp. 281 -294. University press, Cambridge. Hubert, S., and D. Hedgecock, 2004 Linkage maps of microsatellite DNA markers for the Pacific oyster Crassostrea gigas. Genetics 168: 351-362. Hubert, S., E. Cognard, and D. Hedgecock, 2009 Centromere-mapping in triploid families of the Pacific oyster Crassostrea gigas (Thunberg). Aquaculture 288: 172–183. Hunt, H. L., and R. E. Scheibling, 1997 Role of early post settlement mortality in recruitment of benthic marine invertebrates. Mar. Ecol.-Prog. Ser. 155: 269-301. Huvet, A., P. Boudry, M. Ohresser, C. Delsert, and F. Bonhomme, 2000 Variable microsatellites in the Pacific oyster Crassostrea gigas and other cupped oyster species. Anim. Genet. 31: 71-72. Jackson, D. J., G. Worheide, B. M. Degnan, 2007 Dynamic expression of ancient and novel molluscan shell genes during ecological transitions. BMC Evol. Biol. 7(160): 1-17. Jackson, D. J., B. M. Degnan, 2006 Expressed sequence tage analysis of genes expressed during development of the tropical abalone Haliotis asinine. J. Shell. Res. 25(1): 225-231. Jackson, D.J., N. E. Ellemor, and B. M. Degnan, 2005 Correlating gene expression with larval competence, and the effect of age and parentage on metamorphosis in the tropical abalone Haliotis asinine. Mar. Biol. 147: 681-697. Jackson, D., S. P. Leys, V. F. Hinman, R. Woods, M. F. Lavin, and B. M. Degnan, 2002 Ecological regulation of development: induction of marine invertebrate metamorphosis. Int. J. Dev. Biol. 46: 679-686. Jones, G., and B. Jones, 1983 Methods for setting hatchery produced oyster larvae. Marine Resources Branch, Ministry of Environment, BC, Canada. Information report No. 4., 94pp. 103 Jacobs, M. W., S. M. Degnan, R. Woods, E. Williams, K. Roper, K. Green., and B. M. Degnan, 2006 The effect of larval age on morphology and gene expression during ascidian metamorphosis. Int. Comp. Biol. 46(6):760-776 Kennedy, V. S., R. I. E. Newell, and A. F. Eble, 1996 The Eastern Oyster: Crassostrea virginica. Maryland Sea Grant press. Kawashima, T., Y. Satou, S. D. Murakami, and N. Satoh, 2005 Dynamic changes in developmental gene expression in the basal chordate Ciona intestinalis. Develop. Growth Differ. 47: 187-99. Keough, M. J. and B. J. Downes, 1982 Recruitment of marine invertebrates: the role of active larval choices and early mortality. Oecologia (Berl) 54:348-352. Launey, S., and D. Hedgecock, 2001 High genetic load in the Pacific oyster Crassostrea gigas. Genetics 159: 255-265. Li, G., S. Hubert, K. Bucklin, V. Ribes, and D. Hedgecock, 2003 Characterization of 79 microsatellite DNA markers in the Pacific oyster Crassostrea gigas. Mol. Ecol. Notes 3: 228-232. Magoulas, A., B. Gjetvag, V. Tersoglou and E. Zouros, 1998 Three polymorphic microsatellites in the Japanese oyster, Crassostrea gigas (Thunberg). Anim. Genet. 29:69-70. McGoldrick, D. J., D. Hedgecock, L. J. English, P. Baoprasertkul, and R. D. Ward, 2000 The transmission of microsatellite alleles in Australian and North American stocks of the Pacific oyster (Crassostrea gigas): Selection and null alleles. J. Shellfish Res. 19: 779-788. Newell, R. I. E., G. S. Alspach, Jr., V. S. Kennedy, and D. Jacobs, 2000 Mortality of newly metamorphosed eastern oysters (Crassostrea virginica) in mesohaline Chesapeake Bay. Mar. Biol. 136: 665-676. Pechenik, J. A., 1990 Delayed metamorphosis by larvae of benthic marine invertebrates: Does it occur? Is there a price to pay? Ophelia 32: 63-94. Pineda, J., 1994 Spatial and temporal patterns in barnacle settlement rate along a southern California rocky shore. Mar. Ecol. Prog. Ser. 107:125-138. 104 Pineda, J., V. Starczak, and T. A. Stueckle, 2006 Timing of successful settlement: demonstration of a recruitment window in the barnacle Semibalanus balanoides Mar. Ecol. Prog. Ser. 320: 233–237. Raimondi, P. T., 1990 Patterns, mechanisms, and consequences of variability in settlement and recruitment in an intertidal barnacle. Ecological Monographs 60:283-309. Rodriguez, S. R., F. Patricio-Ojedal, and N. C. Inestrosa, 1993 Settlement of benthic marine invertebrates Mar. Ecol. Prog. Ser. 97: 193-207. Roughgarden J., S. Gaines, and H. Possingham, 1988 Recruitment dynamics in complex life cycles. Science 241: 1460-1466. Sekino, M., M. Hamaguchi, F. Aranishi, and K. Okoshi, 2003 Development of novel microsatellite DNA markers from the Pacific oyster, Crassostrea gigas. Mar. Biotechnol. 5: 227-233. Shuelke, M., 2000 An economic method for the fluorescent labeling of PCR fragments. Nat. Biotechnol. 18: 233-234. Wang, Y., R. Ren, and Z. Yu, 2008 Bioinformatic mining of EST-SSR loci in the Pacific oyster, Crassostrea gigas. Animal Genetics 39: 287-289. Williams, E. A., B. M. Degnan, H. Gunter, D. J. Jackson, B. J. Woodcroft, et al., 2009 Widespread transcriptional changes pre-empt the critical pelagic–benthic transition in the vetigastropod Haliotis asinine. Mol. Ecol. 18: 1006-2025 Yamtich J., M. L. Voigt, G. Li, and D. Hedgecock, 2005 Eight microsatellite loci for the Pacific oyster Crassostrea gigas. Anim. Genet. 36: 524-526. Yu, H., and Q, Li, 2007 EST-SSR markers from the Pacific oyster, Crassostrea gigas. Molecular Ecology Notes, 7: 860-862. Zaykin, D.V. and A. I. Pudovkin, 1993 Two programs to estimate significance of Chi- square values using pseudo-probability test. J. Hered. 84: 152. 105 Chapter 3 Inbreeding-by-environmental interaction affects the expression of genetic load in the Pacific oyster Crassostrea gigas 3.1 Abstract Inbreeding depression is a widely observed evolutionary phenomenon and a major concern in the management and conservation of threatened and endangered species. Inbreeding generally reduces fitness, but its magnitude and effects are highly variable because they depend on the genetic constitution of a population and on interaction with the environment. In general, harsher environments significantly increase inbreeding depression; however, most studies have examined the interaction of environment and inbreeding depression at the population level and less is understood about its underlying genetic mechanisms. Using QTL mapping methods, I performed a genome scan for deleterious (viability) loci in two inbred (f = 0.352), F 2 families of the Pacific oyster Crassostrea gigas reared in two environments: nutrient-poor (1-algal) and nutrient-rich (mixed, 3-algal) diets, which had significant effects on growth rate and post metamorphosis survival. Thirteen and 10 viability loci were detected in the 1-algal treatment across the two families, causing 99.3 and 98.3% mortality up to the juvenile stage; in contrast, only 6 and 5 significant viability loci were detected in the 3-algal diet, causing substantially less total genetic mortality (93.6% and 84.1 % respectively). 106 Epistasis was not detected between viability loci within the treatments in either family. Estimates of selection at markers nearest to viability QTL were significantly greater in the harsher environment (mean selection was 0.86 vs. 0.54 in the 1-algal and 3- algal diets, respectively). Remarkably, dominance also increased significantly for most QTL (average dominance was 0.18 vs. 0.35 in the 3-algal and 1- algal diets, respectively). Individual QTL differed significantly across diets in selection only, dominance only, and in both selection and dominance. Some viability QTL were not greatly affected by diet, suggesting that they were lethal in any environment. These findings demonstrate that inbreeding-by-environment interaction is locus-specific, and they help to explain previously observed lineage-specific inbreeding-by-environment interaction and environmentally conditional purging, which are important considerations in the management of endangered or rare species. 3.2 Introduction Inbreeding depression, the reduction in fitness in progeny from mating of close relatives, has been observed for well over a century in domesticated plants and animals (e.g. Darwin 1876, Wright 1977, Hedrick and Kalinowski 2000, Charlesworth and Willis 2009). In outcrossing species, inbreeding reduces the mean of fitness traits such as litter size and survival and increases the probability of extinction in experimental lines (Charlesworth and Charlesworth 1987, Charlesworth and Willis 2009). Although the precise genetic basis of inbreeding depression is still debated, the majority of fitness effects are thought to be a result of increased homozygosity for deleterious recessive 107 alleles (Roff 2002, Charlesworth and Willis 2009). Small populations may be at increased risk of inbreeding depression, owing to genetic drift and the chance fixation of deleterious alleles; therefore, understanding the causes and consequences of inbreeding depression has become of primary concern for the conservation of rare and endangered species (e.g. Frankel and Soule 1981, Frankham 1995, Hedrick and Kalinowski 2000). However, the magnitude of inbreeding depression can vary dramatically across species (Byers and Waller 1999, Lynch and Walsh 1998, Hedrick et al. 1999), between fitness traits (e.g. Husband and Schemske 1996, Thiele et al. 2010), among lineages (Bijlsma et al. 1999, Kristensen et al. 2003), across years (e.g. Hayes et al. 2005), and even at different stages in the life cycle (Husband and Schemske 1996, Cohen 1992, Plough Chapter 2). Thus, predicting inbreeding depression for a given species or population is difficult. Recently, a substantial amount of effort in conservation biology has been focused on understanding how environmental variation may interact with inbreeding depression and affect fitness (reviewed by Armbruster and Reed 2005 and Fox et al. 2010, Cheptou and Donahue 2011). This is particularly important for threatened or endangered species because these populations may face a number of environmental stressors including reduction in suitable habitat, climatic changes, and habitat fragmentation, in addition to the genetic problems caused by small population sizes (e.g. Frankham 1995, Hedrick and Kalinowski 2000). How the environment affects the expression of genes, which in a statistical framework is called genotype-by-environment interaction (G×E; Falconer and McKay 1996), is a fundamental question in evolutionary biology, relevant for 108 understanding patterns of selection and maintenance of genetic variation across heterogeneous environments. Inbreeding-by-environment interaction (I×E) constitutes a special case of G×E and numerous studies show that the magnitude of inbreeding depression is affected by a wide range of environmental perturbations, and in particular, that inbreeding depression increases in stressful environments (e.g., Bijlsma et al. 1999, Cheptou et al. 2000, Kristensen et al. 2003, Hayes et al. 2005). Many experimental studies compare inbreeding depression in laboratory or ‗greenhouse‘ conditions to inbreeding depression in the field, finding increased inbreeding depression in natural or wild conditions, which are generally thought to be more stressful (e.g. Jimenez et al. 1994, Reviewed by Crnokrak and Roff 1999, but see Keller and Waller 2002). However, results over a broad range of taxa and experimental conditions have been somewhat inconsistent (Keller and Waller 2002, reviewed by Armbruster and Reed 2005) and experiments show that different inbred lineages often exhibit highly variable responses to inbreeding under stressful conditions (e.g. Bijmsla et al. 1999, Fowler and Whitlock 2002). Some authors have posited that the variability of I×E across lineages could be a consequence of different deleterious loci acting in an environment-specific fashion (Bijslma et al. 1999, Armbruster and Reed 2005). The environmental dependence of inbreeding depression may also explain the inconsistent purging of deleterious alleles in experimental studies (reviewed by Byers and Waller 1999, Crnokrak and Barret 2002, Boakes et al. 2007). Despite the large amount of scientific effort focused on inbreeding-by- environment interaction, there is a paucity of data on genetic mechanism. There are two, 109 general, not mutually exclusive explanations in the literature for why stressful environments might increase inbreeding depression. First, stressful environments may amplify the negative effects of deleterious mutations, exposed upon inbreeding, which disrupt important physiological pathways. Specific mutations may have only a mild effect in benign conditions, but with stress, a threshold may be reached at which their effects become more deleterious. Summed across loci, these individual effects can reduce overall fitness significantly. This can be considered a threshold, or locus-specific model in which selection against particular mutant loci increases in more stressful environments (Fox and Reed 2011). Another possible explanation is that stressful environments may compromise repair mechanisms that are already modified or diminished by inbreeding (e.g. immune response, heat-shock chaperones and DNA repair mechanism), which then has widespread, negative fitness effects on a variety of important physiological pathways. This would be considered a more general, inbreeding×stress response (Kristensen et al. 2006, Tomala and Korona 2008). At this point, there are few data to support or refute either explanation of inbreeding-by- environment interaction, because, with the exception of Kristensen et al. (2006), who analyzed changes in gene expression in inbred lines of Drosophila between control and stress treatments, most studies look at inbreeding depression at the level of the population (changes in mean trait values across stressors and levels of inbreeding), so the precise genetic basis of this interaction remains poorly understood. Recently, the use of quantitative-trait locus (QTL) mapping methods has allowed more detailed genomic investigation of the location, number and effects of these 110 deleterious mutations (e.g. Mitchell-Olds 1995, Remington and O‘Malley 2000, Luo and Xu 2003, Plough Chapter 2). Inbreeding-by-environment interaction can be identified by examining the fitness effects of deleterious alleles across multiple environments and observing: 1) significant viability effects in one environment but not others, or 2) shifts in selection or dominance of the deleterious alleles across environments, resulting in different patterns of segregation distortion. Examining the effect of stress on the fitness of individual viability loci, rather than the fitness of individuals, which is a complex, polygenic trait, allows investigation of the locus-specific effects of inbreeding-by- environment interactions across the genome. The Pacific oyster is a good model for studying inbreeding depression because it has a high genetic load (Williams 1975, Launey and Hedgecock 2001) and the biological and genomic resources to support experiments that can dissect the roles of genes and environment (Hedgecock et al. 2005). Previous work has shown that the Pacific oyster carries a large load of highly deleterious, recessive mutations, which is revealed as widespread distortion of microsatellite marker segregation ratios in inbred F 2 or F 3 families (Launey and Hedgecock 2001, Bucklin 2003, Chapter 1). Since standard rearing practices were employed in these studies, these previous results beg the question of whether these mutations would be deleterious in other environments, particularly the natural environment. To investigate how growth, survival, and the expression of deleterious loci might be affected by a stressful environment, I created two inbred, F 2 families of the Pacific oyster, the larvae of which were divided and reared in two treatments, differing 111 significantly in micro-algal diet (fed one algal species vs. a mix of three). Mono-algal diets generally lack essential fatty acids necessary for bivalve growth and survival, while a mix of algae provide better overall nutrition (e.g. Helm et al. 1991, Thomson and Harrison 1992), thus, the single-algal diet represents a stressful environment for the Pacific oyster. I used QTL mapping methods to examine the number, location and effect of deleterious loci across the genome, comparing the location and effect of deleterious loci between the two environments. I also investigated interaction between deleterious loci (epistasis) and how selection and dominance at markers nearest to viability QTL were altered by the stressful environment, using a modified version of the ‗two-locus model‘ (Hedrick and Muona 1990, Launey and Hedgecock 2001, Chapter 1). This study is the first to take a molecular marker-based approach to investigate inbreeding-by- environment interaction in marine animals and animals in general, providing insight into this important interaction and its underlying genetic mechanisms. 3.3 Materials and Methods Biological material and molecular methods Crosses Inbred lines 51 and 35 were derived from a naturalized population of C. gigas in Dabob Bay, WA, with initial families made from pair crosses of wild individuals in 1996 (Hedgecock and Davis 2007). These lines were inbred (full-sib mating) for two generations leading up to the F 1 hybrid cross that was made in 2004. The first 112 experimental F 2 family was created in 2007, by crossing a pair of male and female full- siblings from the 2004 51×35 F 1 hybrid families (family 07×5), and the second experimental family was made, in 2008, by a cross of full siblings from the same 2004 51×35 F 1 hybrids (family 08×3). Thus, the two F 2 families share grandparents and have the same inbreeding coefficient, f = 0.352 (see chapter 1 for pedigrees). Crosses were performed at the University of Southern California (USC), Wrigley Marine Science Center (WMSC) on Catalina Island, CA. Pedigrees of parents were verified with microsatellite DNA markers (Hedgecock and Davis 2007, Curole and Hedgecock 2007). Culturing methods and larval rearing in two feeding environments Each family was created by stripping the gonad from a single male and a single female, combining their gametes in a two-liter beaker of fresh seawater for fertilization (Breese and Malouf 1975). After one hour of incubation, fertilized embryos from a single family were then divided up between the two diet treatments and stocked, at 20 larvae per milliliter, in three, replicate 20-liter vessels (total of six, 20-l vessels per family) with fresh, 0.2 micron filtered seawater. The diet for the first treatment consisted of Isochrysis galbana only, named the 1-algal diet hereafter. The second treatment consisted of a mixed algal diet of Isochrysis galbana, Tetraselmis suecica and Chaetoceros calcitrans, which is named the 3-algal diet hereafter. Because T. suecica is a relatively large alga, suitable only for feeding larvae with an average size of 120μm or greater, it was not included in the 3-algal diet until day 10. Oyster trocophore larvae are non-feeding for the first 24-48 hours after fertilization, so feeding treatments did not 113 Table 6 Algal feeding treatments for the two experiments Day 1-Algal Diet 3- Algal diet Concentration 0-2 no food no food -- 2-6 100 % I. galb. 25% C. calc. 75% I. galb. 30,000 cells ml -1 6-10 100 % I. galb. 50% C. calc. 50% I. galb. 50,000 cells ml -1 10-14 100 % I. galb. 50% C. calc. 25% I. galb., 25% T. suec. 50,000 cells ml -1 14-20 100 % I. galb. 50% C. calc. 50% T. suec. 100,000 cells ml -1 I.galb is Isochrysis galbana, C. calc is Cheotocerus calcitrans, and T. suec. is Tetraselmis suecica. 114 commence until day two post-fertilization, and the two treatments experienced identical culture conditions up to this point. On day two, the 3-algal treatment was fed a mix of I. galbana and C. calcitrans, while the 1-algal treatment was fed only I. galbana. Both treatments were fed at an initial concentration of 30,000 cells ml -1 , which was increased, as larvae grew, to a concentration of 100,000 cells ml -1 by the end of the larval period, following standard larval culturing conditions for C. gigas (e.g. Hedgecock et al. 1996). Algal diet composition (% of each species) and cell concentrations over the duration of the experiment are presented in Table 6. When competent for settling, based on observations of morphology (presence of eyespots), behavior (probing substrate with their foot), and size (~250 microns), larvae were pooled across replicates within each treatment, screened, treated with epinephrine (Coon et al. 1986), and set at low density in a down-welling system (Hedgecock and Davis 2007) with a mesh size of ~200 microns, where they were grown for approximately 40 more days. DNA extraction, PCR, and Electrophoresis Juvenile oysters (spat) from the 1-algal and 3-algal treatments were sampled at day 60 and immediately frozen at -80C in Qiagen DNeasy tissue lysis buffer (Valencia, CA). DNA was extracted from 96 spat per treatment, using the Qiagen DNeasy animal tissue kit following the manufacturer‘s protocol. Over 90 microsatellite markers cloned from the Pacific oyster were tested for use in this study (Magoulas et al. 1998, McGoldrick et al. 2000, Huvet 2000, Li et al. 2003, Sekino et al. 2003, Yamtich et al. 2005, Wang et al. 2005, Yu et al. 2007), most of which 115 are on published linkage maps (Hubert and Hedgecock 2004, Hubert et al. 2009). Markers were named as in the source publication, except for those developed in this laboratory (McGoldrick et al. 2000, Li et al. 2003, Yamtich et al. 2005), which were abbreviated from their original description; e.g ucdCgi195 to Cg195. A new microsatellite marker developed in this laboratory, Cg213, was used here for the first time (5‘ F-GGATCATACCACATTGTCCAGA, R-TTCGGGCTTTATTGGATTTG, 21 ‗CT‘ repeats in the isolated clone, MgCl 2 is 2 mM, annealing temp is 55ºC). Polymerase chain reactions were carried out in two phases; the first to amplify the target microsatellite marker, and the second to incorporate a fluorescently labeled dye attached to a ―zip‖ sequence that is complementary to the 3‘ tail of the forward primer (Shuelke 2000). PCR cycle conditions consisted of an initial denaturing step at 94°C for 2 minutes, then 20 phase-one cycles (30 sec at 94°C, 45 sec at T m , 45 sec at 72°C), followed by 10 ―zip‖ cycles (30 sec at 94°C, 45 sec at 59°C and 45 sec at 72°C) and a final 10-min elongation step at 72°C. The number of zip cycles was often increased to 30 or 40 when using larval DNA as template. T m (the optimum annealing temperature) and the MgCl 2 concentration varied, depending on the locus (Magoulas et al. 1998, McGoldrick et al. 2000, Huvet 2000, Li et al. 2003, Sekino et al. 2003, Yamtich et al. 2005, Wang et al. 2005, Yu et al. 2007). PCR products were electrophoresed on a 4% denaturing PAGE gel (acrylamide: bisacrylamide 19:1, 3M Urea) using 1 TBE (Tris(base), boric acid, and EDTA di-sodium salt) on the ABI Prism 377 DNA Sequencer (Perkin Elmer, Waltham Massachusetts). Gel images were used to score the 116 fluorescently labeled microsatellites by eye, comparing progeny alleles to adult alleles along with fluorescently labeled size standards (DeWoody et al. 2004). Data analyses Survival and growth measurements in larvae Survival was calculated for each of the three replicate culture vessels across the two treatments by counting the number of larvae in four to six random sub-samples of known volume (50-100 μl) from homogeneously mixed, concentrated cultures (500-1000 ml). These counts were then multiplied by the inverse of the fraction of the subsample volume to the total volume ( and averaged to estimate the mean number of larvae in a replicate. Total population size was then calculated by multiplying the mean number of larvae per replicate by the inverse of the concentration factor. Relative survival was calculated as the fraction of larvae remaining from the original stocking of 400,000 larvae per culture vessel. Survival was estimated at days 2, 4, 6, 12 and 18 and for family 07×5 and days 2, 6, 10, and 18 for family 08×3. Repeated measures ANOVA over all days (Survival = day + day×diet) and nested ANOVA at each day (tank replicate nested within diet; Survival = diet(replicate)), was performed to examine the effect of diet on survival. A direct count of the number of spat in each diet treatment surviving to day 60 was also made for family 08×3. The average rate of larval growth across the two diet treatments was determined by measuring shell length for random samples of 50 larvae taken every four days through 117 the larval period (to day 18), from each tank. Each sample of 50 larvae was photographed under the microscope at 100× magnification, using a calibrated micrometer slide. ImageJ image analysis software (v1.44o, National Institutes of Health, Bethesda, MD) was used to measure length from the anterior to the posterior edge of the shell. Rates of shell growth were determined from the slope of shell length vs. age and analysis of growth differences among the diets was performed using a general linear model, shell- length = day + diet + diet×day. Segregations and null alleles Because the inbred lines used in this experiment experienced only two generations of inbreeding (full-sib mating), many of the loci exhibit segregation types other than the classical F 2 cross (e.g. AB×AB producing progeny genotypes AA, AB, BB). The types of segregation observed in this study include cases with two alleles (AB AB or AA AB), three alleles (AB AC), and four alleles (AB CD). Non-amplifying or null alleles, common in the Pacific oyster (McGoldrick et al. 2000, Launey and Hedgecock 2001, Hedgecock et al. 2004), require modification of classical segregation ratios only in cross type AØ AB, where the AA genotype cannot be distinguished from AØ genotype and the resultant progeny genotypes, A-, AB, and BØ have expected frequencies of 2:1:1, respectively. Under the null hypothesis of no viability selection, progeny genotypes should conform to the expected Mendelian ratios (either 1:1, 1:2:1, 1:1:1:1, or 2:1:1 for the null allele case). Deviations from expected Mendelian proportions were tested with goodness-of-fit chi-square tests, with the level of significance adjusted for multiple 118 simultaneous tests within each type of segregation (Launey and Hedgecock 2001). Adjustment for multiple tests may increase type II error, so statistical significance at both the nominal =0.05 and Bonferroni adjusted levels are reported. In cases of significant deviations from Mendelian ratios attributable to deficiencies of homozygous genotypes, we further tested for the effect of deleterious recessive alleles on heterozygotes, by means of secondary goodness-of-fit chi-square tests. If deleterious alleles are recessive, we expect a segregation ratio of 2:1 for counts of heterozygotes to the unaffected homozygotes (two allele cases) and a 1:1:1 ratio among the three heterozygotes (three- allele cases). Statistically significant deviation from these expectations indicates deleterious effects on carrier heterozygotes. Though the QTL mapping of viability selection is the focus of this study, because it integrates all marker information in the analysis simultaneously, it is still instructive to examine segregation patterns at the single marker level to observe trends, which should be reflected in the QTL analyses. Mapping viability QTL (vQTL) in two environments Linkage maps were constructed separately for the two experimental families with day 60 oyster genotype data pooled between treatments (n = 192), using the CP (cross pollinator) population type in Joinmap 3.0 (VanOoijen and Voorrips 2001). The Kosambi mapping function with a minimum likelihood of the odds (LOD) score of 2.0 was used for linkage group assignments. Deviations from Mendelian segregation ratios may affect linkage mapping (Lorieux et al. 1995, Zhu et al. 2007), so we compared our marker order and distances with previously published linkage maps constructed using 119 larval samples with little segregation distortion (Hubert and Hedgecock 2004, Hubert et al. 2009). If markers failed to map, we used locations from published maps (Hubert and Hedgecock 2004). Phase was determined by Joinmap 3.0, from the frequencies of parental and recombinant types. Phase information, marker locations, and parent and progeny genotypes were input for the vQTL model of Luo and Xu (2003), which locates and characterizes loci under viability selection. This model, which is implemented in PROC QTL (Hu and Xu 2009), a user defined procedure for SAS (version 9.2, SAS Institute, Inc., Cary, NC), scans the genome in 1 cM increments for the presence of loci associated with viability selection, under a maximum-likelihood QTL framework. The model produces a likelihood ratio test (LRT) statistic and estimates the proportions of genotypes for each 1 cM increment (Luo and Xu 2003). The model interprets all cross types as AB CD, yielding progeny genotypes AC, AD, BC, BD and in the absence of viability selection, the genotype frequencies at any point in the genome are expected to be 0.25: 0.25: 0.25: 0.25; viability selection alters these relative proportions. The vQTL model was run on the day 60 genotype data from 1-algal and 3-algal treatments separately (n = 96), and the genomic positions with significant viability selection represented by QTL peaks were then compared between the two treatments. Since we could not use permutation tests to set significance thresholds, we used an approximate method based on the LRT profile to establish thresholds at the = 0.05 level (Piepho 2003). QTL were identified if the LRT profile was above the genome-wide threshold for significance, and QTL that were only above the chromosome-wide significance thresholds were considered suggestive. 120 Multiple QTL on a single linkage group were identified when the LRT statistic fell by at least 4.60 (~2 LOD) between two QTL peaks (Lander and Botstein 1989). Mortality from viability QTL Relative fitnesses of genotypes at QTL peaks were used to estimate survival and mortality attributable to each viability locus, within the two diet treatments. The relative fitness of each genotype at a QTL is max w w ij where w max is the frequency of the genotype in highest proportion (Luo and Xu 2003). Average survival, S, at a QTL is therefore; 4 max 22 max 21 max 12 max 11 w w w w w w w w S = max 4 1 w (1) and average mortality, M, is simply 1- S or max 4 1 1 w . In finite samples, chance fluctuations in genotype proportions will produce a non-zero estimate of mortality by this equation, even in the absence of selection, so to correct for this sampling effect, we randomly generated 1000 datasets of progeny genotypes (i.e. no selection), using the parental genotypes and sample sizes appropriate for each family. We then ran each of the 1000 simulated datasets through the viability QTL model to determine the average relative survival expected at each 1 cM interval of the genome, in the absence of selection. These relative survival estimates were averaged over the whole genome and then over the 1000 simulations to obtain a grand average relative survival, in the absence of selection. Finally, the average maximum genotype frequency over the simulations 121 without selection, w maxO , was calculated by substituting the simulated grand average relative survival into eq. 1. The correction for sampling error is applied to the estimate of average survival at a vQTL by adjusting w max , the maximum frequency of the four genotypes: 25 . 0 4 1 max max O V adj w w S , (2) where w maxV is estimated from the data by the QTL model and 0.25 is the Mendelian expectation in a sufficiently large sample without selection. Estimating dominance and selection for vQTL in two environments To compare how selection and dominance at viability QTL were affected by algal diet, we used a two-locus selection model (Hedrick and Muona 1990, Launey and Hedgecock 2001, Chapter 1) to examine genotype arrays at markers linked most closely to significant viability QTL. We adapted this model to estimate jointly the selection coefficient (s) and the dominance deviation (h) for each viability locus, given the recombination distance between a vQTL peak and the nearest microsatellite DNA marker (c), which we calculated from the QTL results. In each case, we took the estimates of s and h that maximized the ratio of the likelihood of genotype data, with selection, to the likelihood of the data, without selection. Maximum likelihood estimates (MLE) were obtained using the ‗optim‘ function in R (v 6.2.2) with box constraints of 0.0-1.0 for selection coefficients and -1.0 – 1.0 for dominance coefficients. Maximum likelihood estimates of s and h were then tested for significant differences between the two 122 environments, using Welch‘s t-test (assuming unequal variances and unequal sample sizes), with the variances of MLE‘s obtained from the hessian matrix extracted from the maximization procedure. R code for the two-locus model program and t-test is available in the supplementary information, File S10. Testing for epistasis Interaction (epistasis) between viability QTL was assessed by chi-square tests for associations between the genotypes at the two markers nearest to each pair of QTL peaks. Epistasis was also tested for all markers within each family and diet. Tests were adjusted for multiple comparisons using the Benjamini and Hochberg method to control false discovery rate, which is less conservative than the Bonferroni correction (Benjamini and Hochberg 1995). R code for the epistasis test is available in the supplementary information, File S11. 3.4 Results Survival and growth Survival during the larval stages was highly variable across time points but in general, no clear effect of diet was observed (Fig 10 a,b). For family 07×5, results of the repeated measures ANOVA showed a significant day*diet interaction (P <0.05) driven by lower survival in some of the 3-algal replicates at day four and six (Fig. 10a). Nested 123 Figure 10 Survival data for 07×5 and 08×3 Box and whisker plot of survival over replicate cultures and time in the two families, 07×5 (a) and 08×3 (b). Grey boxes represent the 1-algal diet, white boxes represent the 3-algal diet. The edges of the box represent the 25 % and 75 % quartiles, the solid line in the middle of the box is the median, and end of the whiskers represent 1.5 * the inter quartile range (IQR). Outliers (open circles) fall outside 1.5 *IQR. 123 124 ANOVA, by day, using all replicate counts, showed significant heterogeneity among tanks within diet at all time points and only the day 6 day×diet interaction remained significant after taking into account the error variance among tank replicates (P <0.0374). In general there was higher inter-tank variation in the 3-algal diet, and no significant differences in survival were found at the end of the larval phase. In family 08×3, no statistically significant differences in survival between the diets were found at any point during the larval stages, and significant heterogeneity was observed among tanks for all but day 2. Direct counts of the number of spat surviving at 60 days (pooled across replicates) were substantially different between the two treatments for family 08×3: 12,183 spat survived in the 3-algal diet vs. 3,086 spat in the 1-algal diet. Growth rate of larvae varied significantly between the two diet treatments for both families (significant diet×day interaction in ANOVA, P <0.00001, P=0.0156, for families 07×5 and 08×3, respectively; Fig. 11). Growth rate was almost two-fold greater for larvae in the 3-algal compared with the 1-algal diet in family 07×5 (12.1 ± 0.76 µm per day vs. 7.64 ± 0.233 µm per day, respectively; ± standard error of the mean; Fig. 11). Growth rate was higher for the 3-algal treatment in family 08×3 as well, but not by as much (13.95 ± 0.916 µm per day vs. 10.65 ± 0.717 in the 3-algal and 1-algal diets, respectively). Single marker analyses of segregation distortion Of the 90 microsatellite markers tested in the F 1 parents of families 07×5 and 08×3, 41 and 39, respectively, were informative or heterozygous in at least one parent 125 Figure 11 Growth data for family 07x5 and 08x3 Growth rate results for families 07×5 and 08×3. Regression line of mean shell length vs. day and the growth rate (μm*day -1 , ± 1 standard error of the mean) are displayed in each panel. 126 Table 7 Counts of distorted markers at the α= 0.05 level arranged by cross types and diet (1-algal or 3-algal) in the two families. 07×5 08×3 Cross type 1- Algal diet 3-Algal diet Total scored 1- Algal diet 3- Algal diet Total scored AA×AB 0 0 1 1 (0) 0 5 AB×AB 16 (13) 8 (5) 20 8 (7) 7 (7) 10 AB×AC 11 (8) 11 (2) 20 10 (10) 7 (3) 18 AB×CD -- -- -- 1 (0) 1 (0) 6 Totals 27 (21) 19 (7) 41 20 (17) 15 (10) 39 Numbers in parentheses represent markers significantly distorted after a Bonferroni correction within segregation type. ‗Total scored‘ are the total number of markers scored within a segregation type, regardless of segregation distortion. 127 (Supplementary information, Tables S6-S9). In family 07×5, a greater number of significantly distorted segregation ratios were found in the 1-algal diet (27/41) versus the 3-algal diet (19/41; Table 7). After adjustment for multiple tests within segregation types, only 21 and seven markers remained significant in the 1-algal and 3-algal diets respectively (Table 7). In family 08×3, a similar pattern was observed. In the 1-algal diet, 20 out of 39 markers were significantly distorted, compared with 15/39 in the 3- algal diet, and after corrections for multiple comparisons, 17 and 10 markers remained significant in the 1-algal and 3-algal diets, respectively. Distorted markers were distributed broadly across the genome, on nine of nine linkage groups (LG‘s) in family 07×5 and seven of 10 LG‘s in family 08×3, with many instances of multiple markers exhibiting distorted segregation ratios on the same linkage group (Tables S6-S9). Inferences about the mode of action of these distortions were made after the vQTL analysis (see below), but inspection of marker segregations shows that the majority of segregation-ratio distortions (23 of 27 for 07×5 and 16 of 20 for 08×3) were attributable to deficiencies of homozygous genotypes in the 1-algal treatment (Supplemental tables S6-S9). For some markers with AB AB cross types (three in 07×5 and one in 08×3), both homozygotes were deficient in the progeny, which suggests overdominance or the presence of two deleterious mutations linked in repulsion-phase. Most markers that were cross type AA×AB or AB×CD exhibited Mendelian segregation ratios, but a minority of distortions from three- or four-allele cross types yielded deficiencies of heterozygous but not homozygous genotypes (Cg157, Cg12 family 08×3 ; Tables S8, S9). For markers with significant deficiencies of homozygous genotypes, 128 goodness-of-fit chi-square tests for the dominant effect of deleterious alleles in heterozygous genotypes were significant for 21 and 10 markers in the 1-algal diet for families 07×5 and 08×3 respectively, while in the 3-algal diet, tests at only four and five markers were significant for families 07×5 and 08×3, respectively, suggesting that the effects of deleterious alleles on heterozygous carriers were stronger in the 1-algal treatments. Mapping viability loci in two environments Linkage maps were constructed from 41 and 39 markers scored in the day 60 spat samples for families 07×5 (n = 192) and 08×3 (n = 192), respectively. Six previously unmapped markers (Crgi26, Cg9, Crgi10, Cg213, E004,and E214) were added to the map for family 07×5 and E007 was added to the map for family 08×3 (Tables S6, S8). For both families, JoinMap grouped markers previously assigned to linkage group 1A and 10 (Hubert and Hedgecock 2004) into one linkage group (designated as LG 1 in this study), a finding supported by gene-centromere mapping (Hubert et al. 2009) and linkage mapping results from Chapter 1. For both families, markers assigned to linkage group 10 failed to form one continuous map and were split into two different segments (phase was successfully estimated for all markers on the linkage group). Marker orders and distances between markers at the ends of the two segments were taken from the consensus map in Hubert and Hedgecock (2004). LG 9 was only used in the QTL mapping for family 08×3, because there were no informative markers in family 07×5 for this linkage group. Finally, LG 7 had marker orders that differed substantially from published linkage maps; 129 therefore, distances and marker orders for this LG were taken from Hubert and Hedgecock (2004) for both families. Viability QTL (vQTL) were identified in separate runs of the vQTL model for the two treatments in each cross (n = 96 for each treatment, for both 07×5 and 08×3). QTL results from the 1-algal diet were used as the baseline for comparison between the two environments, because the 1-algal diet had a significantly negative effect on growth rate (shown above); thus, vQTL were expected to have their greatest effect in the more stressful environment. For family 07×5, 13 significant vQTL were identified in the 1-algal diet, on all linkage groups (LG) except for LG 2 (Fig. 12, Table 8). Each of these QTL was highly significant, above the genome-wide threshold of 16.5 (α = 0.05) except for vQTL 5 on LG 4, which was well above the chromosome-wide threshold, (13.2) and is, therefore, considered suggestive (Table 8). In the more benign, 3-algal treatment, only six significant vQTL were detected; the other six QTL, which were present in the 1-algal diet, did not reach the genome-wide or chromosome-wide level of significance (Table 8, Fig. 12). For the six vQTL that were significant and identified in either environment, the magnitude of segregation distortion, as represented by the likelihood ratio test (LRT) statistic, was reduced for four of those QTL (vQTL 1, 2, 4, and 8; Fig. 12, Table 8). Overall, the magnitude of the LRT for 10 of the 12 vQTL identified was reduced in the 3- algal diet compared to the LRT in the 1-algal diet. In family 08×3, 10 vQTL were identified on seven of 10 linkage groups in the 1- algal diet, all of which were well above the genome-wide significance threshold of 16.2, 130 Figure 12 QTL mapping results in the 1-algal and 3-algal diets for family 07×5 Likelihood ratio test statistic vs. map distance (cM) for 1-algal diet (light grey area, background) and 3-algal diet (dark grey area, foreground). Long vertical lines mark the ends of linkage groups which are numbered 1-10 in Roman numerals. Small triangles along the X-axis indicate the position of markers used in the mapping procedure. Arabic numerals mark the number and location of vQTL peaks. The dashed black line indicates the genome-wise threshold value for significance at the = 0.05 level (16.5). Figure shown on next page. 131 Figure 12 (Continued) 131 132 except for vQTL 9 (LG 8), which was significant at the chromosome-wide level of 13.8 only, and is thus considered suggestive (Fig. 13, Table 9). Five of these vQTL (vQTL 1, 2, 4, 5, and 8) were not significant in the 3-algal diet, falling well below the genome-wide threshold for significance (Table 9, Fig. 13). Of the four vQTL present in both environments, the LRTs for three were slightly reduced in magnitude in the 3-algal diet; however, the magnitude of the LRT for vQTL 9 actually increased substantially in the 3- algal diet. No QTL were detected on LG 2 in family 08×3 (similar to 07×5) or on LG10. Comparing the location of QTL between these related F 2 families, some QTL appear to be in similar locations (LG 1, QTL 1 and 2, LG 5, and LG 6), suggesting that the same identical-by-descent deleterious loci are segregating in both families, while others are present in only one family (e.g. LG 3, LG 4,and LG 10). Epistasis Pairwise tests of epistasis between markers closely linked to significant QTL revealed no cases of significant interaction, at the uncorrected threshold of α =0.05. Pairwise epistasis tests among all mapped markers (the full marker segregation data set) showed only six potentially significant interactions, all but one of which was non- significant after correction for the false discovery rate. The interaction between Cg108 and Cg139 in family 07×5 on the 3-algal diet was highly significant (P <0.00001, after correction for false discovery rate or Bonferroni correction). However, neither of these markers showed significant distortion of segregation ratios, nor were they linked to vQTL. 133 Figure 13 QTL results for family 08×3 in two environments QTL mapping results in the 1-algal and 3-algal diets for family 07×5. Likelihood ratio test statistic vs. map distance (cM) for 1-algal diet (light grey area, background) and 3- algal diet (dark grey area, foreground). Long vertical lines mark the ends of linkage groups which are numbered 1-10 in Roman numerals. Small triangles along the X-axis indicate the position of markers used in the mapping procedure. Arabic numerals mark the number and location of vQTL peaks. The dashed black line indicates the genome- wise threshold value for significance at the = 0.05 level (16.2). Figure shown on next page. 134 Figure 13 (continued) 134 135 Mortality The selective mortality attributable to viability QTL was calculated from the relative fitnesses of genotypes at each significant vQTL. Because no interactions among deleterious viability loci were observed in either of the diets across the two families, we calculated the multiplicative effects of mortality attributable to i unlinked vQTL, for each diet within a family, as ∏ ̅ (Tables 8, 9). Average relative survival in this calculation was adjusted for sampling error through simulation, as described in Materials and Methods. Relative survival in simulations with no selection averaged 0.82 for both families 07×5 and 08×3 (cf. to the expected 1.0), leading to small upward adjustments of average relative survival at vQTL. For family 07×5, cumulative mortality measured at day 60 in the 1-algal diet was estimated at 98.76 percent, with mortality at individual QTL ranging from 0.104 to 0.672 (Table 8). In the 3-algal diet, cumulative mortality was estimated at only 92.9%, an almost six-fold increase in survival, with mortality at individual QTL ranging from 0.0 to 0.516. Mortality was lower at every vQTL in the 3- algal diet except for vQTL 9. In family 08×3, cumulative mortality at day 60 was 98.3 % in the 1-algal diet with mortality at individual QTL ranging from 0.286 to 0.430 (Table 9). Cumulative mortality in the 3-algal diet was 84.1 % with five QTL exhibiting less than 10 % mortality. Mortality was lower at every vQTL in the 3-algal diet except for vQTL 6 and 9, and cumulative survival was over nine-fold higher in the 3-algal diet. 136 Table 8 Mortality and QTL results for 07×5 Estimates of cumulative survival use only QTL that are unlinked (>50cM apart) or the more deleterious of QTL that are linked (≤50cM apart; survival at the less deleterious QTL is shown in parentheses). Selective mortality is calculated as one minus the adjusted survival described in eq. 2; w max is the maximum genotype frequency at each QTL (cf. to the theoretical 0.25). Table shown on next page. 137 Table 8 (Continued) Diet vQTL LG position LRT Survival (adj) Mortality (adj) 1-algal 1 1 6 104.27 0.328 0.672 2 1 37 69.68 (0.442) 0.558 3 3 2 35.02 (0.638) 0.362 4 3 38 69.46 0.467 0.533 5 4 105 14.56 0.761 0.239 6 5 39 40.43 0.642 0.358 7 6 2 38.24 (0.578) 0.422 8 6 14 32.46 (0.541) 0.459 9 6 30 26.66 0.506 0.494 10 7 35 47.97 0.551 0.449 11 8 26 15.72 0.896 0.104 12 10 0 62.28 (0.668) 0.332 13 10 17 52.55 0.661 0.339 Cum. Survival 0.012 3-algal 1 1 0 63.23 0.504 0.496 2 1 43 44.97 (0.725) 0.275 3 3 0 2.34 (0.966) 0.034 4 3 38 52.03 0.588 0.412 5 4 105 7.93 0.98 0.02 6 5 40 53.13 0.674 0.326 7 6 0 31.79 (0.849) 0.151 8 6 14 15.59 (0.818) 0.182 9 6 30 6.63 (0.771) 0.229 10 7 36 50.56 0.484 0.516 11 8 26 1.76 1.0 0 12 10 0 10.73 (1.0) 0 13 10 17 9.08 0.97 0.03 Cum. Survival 0.071 138 Table 9 Mortality and QTL results for 08×3 Estimates of cumulative survival use only QTL that are unlinked (>50cM apart) or the more deleterious of QTL that are linked (≤50cM apart; survival at the less deleterious QTL is shown in parentheses). Selective mortality is calculated as one minus the adjusted survival described in eq. 2; w max is the maximum genotype frequency at each QTL (cf. to the theoretical 0.25). Table is shown on next page. 139 Table 9 (Continued) Diet vQTL LG Position (cM) LRT Survival (adj.) Mortality (adj.) 1-Alga 1 1 11.6 28.92 0.691 0.309 2 1 59.3 26.41 0.674 0.326 3 3 26 67.92 0.603 0.397 4 4 0 21.86 0.64 0.36 5 5 50.2 35.53 0.57 0.43 6 6 8 33.54 (0.714) 1 7 6 38 30.87 0.681 0.319 8 7 88 26.12 0.602 0.398 9 8 0 14.10 0.685 0.315 10 8 89 31.70 0.596 0.404 Cumulative Survival 0.017 3-alga 1 1 11.6 6.55 1.009 0 2 1 59.3 13.15 0.918 0.082 3 3 26 60.88 0.669 0.331 4 4 0 4.06 0.951 0.049 5 5 50.2 2.48 0.975 0.025 6 6 8 36.07 (0.685) 0.315 7 6 38 32.69 0.688 1 8 7 88 9.11 0.779 0.221 9 8 0 3.00 0.978 0.022 10 8 89 48.79 0.53 0.47 Cumulative Survival 0.159 140 Selection and dominance of vQTL in the two diets Genotype arrays at markers nearest to significant vQTL were analyzed with the modified two-locus model to evaluate selection and dominance coefficients of deleterious alleles at viability loci in the two environments. Across vQTL, selection coefficients averaged 0.89 and 0.82 in the 1-algal diet versus 0.51 and 0.53 in the 3-algal diet, in families 07×5 and 08×3, respectively (Table 10). Selection coefficients were significantly higher in the 1-algal treatment for seven out of eight QTL tested in 07×5 and six out of eight QTL in 08×3 (Welch‘s t-test P < 0.0001; Table 10). At vQTL 3 in family 07×5 (Crgi26), the selection coefficient was reduced from 0.85 in the 1-algal diet to zero in the 3-algal diet. Dominance coefficients were also higher in the 1-algal treatment, averaging 0.46 and 0.31 versus 0.18 and 0.167 in the 3-algal diet, in families 07×5 and 08×3, respectively. Dominance was significantly higher for seven of nine and for six of eight vQTL in family 07×5 and 08×3, respectively (Welch‘s t-test, P<0.0001; Table 10). A few markers could not be tested in the two-locus model framework, because alleles appeared either to be linked to two viability loci in repulsion phase (Cg186 in 07×5 and Cg 186, Cg141 in 08×3) or showed peculiar selection patterns and deficiencies of genotypes other than the identical-by-descent homozygote (Cg21, Cg14, and Cg131 in 07×5; Cg12 in 08×3). These cases were the minority, however, as 16 of 23 QTL were tested in this framework. 141 Table 10 Two-locus model results for families 07×5 and 08×3 1 algal diet 3- algal diet Fam. vQTL LG Marker AA AB AC/ BB BC AA AB AC/ BB BC s-1 algal s-3 algal P-value h-1 algal h-3 algal P-value 07×5 1 1 Cg126 4 18 15 46 13 20 34 28 0.978 0.660 <0.0001 0.753 -0.040 <0.0001 2 1 Cg09 1 31 43 10 53 28 1.000 0.710 <0.0001 0.688 0.030 <0.0001 3 3 Crgi26 6 42 37 24 45 20 0.857 0.000 0.009 0.510 0.000 4 3 Cg01 0 38 55 2 43 46 0.981 0.956 0.116 0.666 0.557 0.0002 5 4 Cg109 11 28 19 36 13 22 29 30 0.695 0.567 0.0002 0.500 0.265 0.0002 6 5 Cg141 5 28 25 30 14 25 23 32 0.833 0.562 <0.0001 0.140 0.440 <0.0001 7 6 Cg186 4 69 18 3 64 26 8 6 Cg14 8 24 44 16 6 28 34 24 9 6 Cg21 16 44 20 12 14 32 25 18 10 7 Cg131 15 28 38 12 19 21 41 10 11 8 Cg196 8 52 32 18 48 24 0.750 0.250 <0.0001 0.250 0.000 0.1 12 10 Crgi10 0 13 41 41 10 25 29 29 0.976 0.655 <0.0001 0.350 0.110 <0.0001 13 10 Cg129 1 41 14 38 12 29 20 29 0.974 0.586 <0.0001 0.284 0.265 0.26 Average 0.894 0.549 0.460 0.181 Table continues next page 141 142 Table 10 (Continued) 1-algal diet 3-algal diet Fam. vQTL LG Marker AA AB AC/ BB BC AA AB BB/ AC BC s-1 algal s-3 algal P-value h-1 algal h-3 algal P-value 08×3 1 1 Cg126 5 28 17 33 10 28 27 27 0.954 0.746 <0.0001 0.351 -0.030 <0.0001 2 1 Cg181 4 42 38 10 53 32 0.895 0.687 <0.0001 0.499 0.249 <0.0001 3 3 CgE07 0 56 39 0 53 37 0.998 0.987 0.860 0.285 0.287 0.949 4 4 Cg109 10 42 21 16 21 31 23 17 0.533 0.151 <0.0001 5 5 Cg117 4 25 19 28 16 25 27 24 0.931 0.374 <0.0001 0.222 -0.300 0.0001 6 6 Cg21 4 62 21 2 58 24 7 6 Cg141 9 51 8 7 68 18 8 7 Cg131 6 14 16 33 15 18 26 33 0.841 0.568 <0.0001 0.683 0.600 0.25 9 8 Cg175 10 51 23 20 47 25 0.565 0.200 <0.0001 - 0.192 0.299 0.007 10 8 Cg12 20 16 39 4 21 20 50 4 Average 0.817 0.530 0.308 0.184 Marker genotypes and their respective numbers are given in columns AA, AB, AC, BB, and BC for each diet. LG represents linkage group, c, the recombination fraction between the nearest marker and the viability QTL (vQTL). s-1 algal, h-1 algal and s-3 algal, h-3 algal are the 2-locus model maximum likelihood estimates of selection and dominance at the nearest markers to viability QTL in the 1-algal and 3-algal diets, respectively. 142 143 3.5 Discussion In this study, I examined the effect of two micro-algal diets, which differed substantially in nutritional quality, on the growth, survival, and expression of genetic load in inbred families of the Pacific oyster. The overarching goal was to elucidate the genetic details underlying the interaction between the expression of genetic load (inbreeding depression) and the environment, which has been widely observed. Food quality as an environmental stress The lower nutritional profile of a diet consisting of only Isochrysis galbana, relative to that of the 3-algal, mixed diet (e.g. Nasciamento 1981, Thomspon and Harrison 1992, Coutteau et al. 1994, Caers et al. 1998) clearly reduced larval growth in the experiment. The 1-algal diet did not clearly reduce survival, however. At most time points, the two diet treatments showed no significant differences; for day six in family 07×5 larval cultures, survival was significantly higher for the 1-algal diet, though this was not as noteworthy in the face of substantial inter-tank variation within treatments and no significant survival differences between the treatments late in the larval period in either family. Survival data taken during the larval phase, in general, lack precision because of high inter-replicate variation among volumetric survival counts (e.g. family 07×5, Fig. 10a) and afford little power for detecting treatment effects. Manipulating diet did appear to affect settlement and metamorphosis success more strongly, as there was a four-fold increase in the number of surviving spat at 60 144 days, in the 3-algal diet compared with the 1-algal diet. This agrees well with results from Chapter 1 that show substantial genetic mortality during metamorphosis from the expression of half of the genetic load in the F 2 families examined. Studies from a number of other marine bivalves have similarly showed significant differences in survival among micro-algal diet treatments after metamorphosis (e.g. Coutteau et al. 1994, Pernet and Tremblay 2004). Accumulation of lipid reserves and the quality of poly unsaturated fatty acids (PUFA‘s) consumed are critical factors for successful development and survival through metamorphosis in marine bivalves (e.g. Helm et al. 1991, Couteau et al. 1994, Thompson et al. 1996, Pernet and Trembley 2004). The major difference between the two diets used in this experiment was the relative concentrations of two PUFA‘s, eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA), which are produced in substantially different amounts by the algal species that comprised the two diets (Helm et al. 1991, Thompson and Harrison 1992). In a study of the hard clam Mercenaria mercenaria, larvae demonstrated significantly better growth and survival through metamorphosis when reared on a diet supplemented with DHA, relative to being fed Isochrysis galbana alone, which generally exhibits low levels of DHA (Coutteau et al. 1994). Based on the algal species we chose for our 3-algal treatment, this diet should have provided high levels of EPA and DHA and overall, a fairly balanced profile of other essential fatty acids, while the 1-algal diet most likely lacked EPA and had low levels of DHA. Though a clear effect of diet on growth and possibly metamorphosis success was observed, the questions remains, how do these algal treatments compare with what larvae 145 would face in the natural environment (i.e. are the estimates of inbreeding depression exaggerated in the 1-algal diet)? On the one hand, the stressful, 1-algal diet may be more extreme (nutrient deficient) than what oyster larvae would experience in coastal oceanic environment, where there is a mix of numerous algal species, which should supply a balanced level of PUFA‘s and other critical nutrients (e.g. Kennedy et al. 1996, Bos et al. 2006). This would suggest that our estimates of inbreeding depression may be artificially high in this highly stressful treatment. On the other hand, larvae fed either experimental diet had almost constant, saturated levels of food for the duration of the experiment, a situation not likely to be sustained in the ocean environment over long periods of time (Bos et al. 2006). Inbreeding depression may be even higher for (inbred) oyster larvae reared in the wild (as commonly observed; Crnokrak and Roff 2002), where phytoplankton availability is patchy and can be limiting (e.g. Fenaux et al 2004, Bos et al. 2006). Inbreeding × environment interaction at viability loci QTL mapping of viability loci across the two environments demonstrated unequivocally that inbreeding depression in the Pacific oyster increases significantly in the more stressful, 1-algal diet environment. The magnitude of marker segregation distortion and viability selection in the nutritionally stressed environment increased for almost every identified viability QTL across the two families. The multiplicative effects of vQTL in the 1-algal diet caused six- and nine-fold decreases in survival, compared to the 3-algal diet, for families 07×5 and 08×3, respectively. These results agree well with 146 the multitude of studies finding that, in general, inbreeding depression increases in more stressful environments (reviewed by Armbruster and Reed 2005, Fox and Reed 2011, Cheptou and Donahue 2011). Comparing the change in inbreeding depression between the two environments, I find approximately two times the number of statistically significant viability loci in the harsher, 1-algal environment. This two-fold increase in the number of viability loci agrees well with the overall conclusions of Armbruster and Reed (2005), whose review of the subject showed an average 1.7-fold increase in lethal equivalents in more stressful environments. Direct comparisons of lethal equivalents (Morton et al. 1956) to number of significant viability loci determined from QTL mapping have not been reported. This study is the first to examine, at the molecular- marker level, how the environment affects the expression of genetic load. Results, in this instance, are relatively consistent with the biometric studies that make up most of the current scientific literature. A key finding from this study is that the increase in viability selection and mortality in the stressful environment was due to an increase in both the selection coefficient and the dominance of deleterious alleles, as inferred from results of the two locus model (Table 9). Estimates of selection coefficients (s) and dominance (h) were significantly higher in the more stressful environment for almost all identified vQTL. At the outset of the experiment, I predicted that selection coefficients of deleterious alleles would increase under stress, a hypothesis that resembles the ―threshold effect‖ or individual locus model described generally by Fox and Reed (2010) as the increase in selection against deleterious loci under stress (see also Cheptou and Donahue 2011). 147 While we did observe an increase in selection at most vQTL in the 1-algal diet, the corresponding increase in dominance to almost additive levels is perhaps more remarkable and important because it exposes more of the population, i.e. the heterozygous carriers of deleterious alleles, to selection. To my knowledge, no studies have documented the interaction between inbreeding depression and the environment at the level of the individual locus; however, there is some support in the literature for increased dominance of deleterious mutations in more stressful conditions. In a study of new mutations generated in diploid yeast with their mutation repair system knocked-out, Szafraniec et al. (2000) found that mutations had little effect on heterozygotes in benign conditions, but were highly deleterious when exposed to thermal shock. Tobari (1965) similarly showed that the degree of dominance of lethal chromosomes carried in heterozygous individuals of Drosophila was significantly dependent on temperature. Directional dominance is necessary for inbreeding depression (Falconer and Mackay 1996), so perhaps it is not surprising that changes in dominance underlie stress-induced responses to inbreeding depression. Nevertheless, the finding, here, that a poor quality diet increased dominance at most viability loci is a remarkable result. How do our estimates of dominance compare to those in the literature? The best estimates of the average dominance of new mutations come from large-scale studies of mutation accumulation in Drosophila, which show that most deleterious mutations have relatively low dominance, averaging ~ 0.1- 0.2, similar to our average estimates of dominance (~0.18 for the 3-algal diet in both families; Mukai 1972, Simmons and Crow 1977, but see Garcia-Dorado and Caballero 2000). Highly deleterious mutations in genes 148 coding for important enzymatic pathways are expected to be recessive or partially recessive under models of metabolic control theory (e.g. Kacser and Burns 1981, Phadnis and Fry 2005). Under the Kacser and Burns model of metabolic flux, mutations at enzyme encoding genes that have large fitness effects as homozygotes should have relatively small effects in heterozygotes because the wild type levels of enzyme activity are usually at or near the plateau of the flux curve. Reducing enzyme activity by as much as 50% (e.g. heterozygous for a mutation causing a non-functioning enzyme) would have a negligible effect on flux through the pathway (Kacser and Burns 1981, Fig 14a). In contrast, under a stressful environment such as the 1-algal diet, this curve shifts towards diminished flux for the heterozygote, displaying intermediate or low levels of flux (Fig. 14b). Again, the shift in the curve points to a threshold situation, in which even the compensating effect of one wild-type copy of an enzyme may not be enough to maintain 100% function, when the physiological limits of the enzyme are being approached. Of course, this simplistic model of metabolic flux assumes that flux is correlated with viability and that the viability loci we identified in this study code for enzymes. One of the criticisms of the Kacser-Burns model is that it does not directly address non-enzyme loci, though Omholt et al. (2000) showed that simple extensions of metabolic flux to network models captures most molecular mechanisms causing dominance (Phadnis and Fry 2005). Thus, the shift of dominance in the flux curve serves as an initial model for a genetic mechanism underlying the increase in inbreeding depression with stress. 149 Figure 14 Model of enzyme flux with dominance in normal (A) and stressful (B) environments. 150 Inbreeding depression is due to different loci in different environments Another clear pattern to emerge from this study is that the effect of a more stressful environment was not consistent across all viability loci. The change in selection coefficients and dominance differed among viability loci across the two diets, which suggests that different deleterious alleles responded individually to the stressful environment. For example, the selection coefficients and dominance of deleterious alleles at vQTL 4, in 07×5 and vQTL 3, in 08×3, were virtually unchanged across the two diets, while at most other vQTL, selection and dominance were increased significantly in the 1-algal diet (Table 10). At two QTL (vQTL 11, 07×5 and vQTL 9, 08×3), selection increased significantly in the 1-algal environment, while dominance did not, indicating a change only in the selection against the deleterious allele and no effect on the fitness of the heterozygote. For the remainder of the QTL identified, selection and dominance were greater in the 1-algal diet, but the magnitude of this change varied substantially at each locus. These results highlight the variable response of viability loci to the stressful environment and suggest that inbreeding depression in different environments is caused by locus-specific fitness responses. The individual and varied responses of loci to the more stressful diet generally support the ―threshold‖ effect model in which inbreeding-by-environment interaction causes shifts in the fitness of specific alleles (e.g. Fox and Reed 2011). This contrasts with the hypothesis that inbreeding-by- environment interactions are caused by general, directional responses of gene expression to stress (e.g. Kristensen et al. 2006). However, there was an overall response of greater 151 selection and more significant segregation distortion at almost every viability QTL in the more stressful, 1-algal environment. Whether or not this reflects a general stress response mechanism is unclear. It would be interesting to examine whether the vQTL identified only in the 1-algal diet would be responsible for inbreeding by environment interactions across a range of environmental stressors within these families. If so, this would suggest that a general stress mechanism is involved or that these loci are somehow involved in a range of pathways that respond to stress. Despite the observation that most viability QTL were in some way affected by diet, it should also be noted that around half of the vQTL in each family remained significant in both the stressful and the benign environment (i.e. reduced, but strong selection against deleterious alleles) and thus were not altered in overall effect by diet. Put another way, these QTL were highly lethal in either environment. While this does not discount the substantial interaction between most vQTL and dietary stress, these loci potentially have mutations that cannot carry out critical functions needed for survival in any environment. This would suggest that there may be two classes of viability loci— those that are conditionally lethal or deleterious and those that are lethal in any environment, constituting a baseline level of genetic load for the species or population. More studies of inbreeding and stress, examining changes at the genetic and transcriptomic level, across multiple magnitudes of stress and using multiple stressors, will help to disentangle the locus-specific responses from a possible general stress response. 152 Evolutionary and conservation implications of inbreeding-by- environment interaction The finding of inbreeding-by-environment interaction at the genomic level, and for the first time, genetic or locus-specific evidence of the independent response of viability loci to stress or environmental change has evolutionary and conservation implications. First, these results help to explain the widespread observations of lineage- specific interactions between stress and inbreeding depression: different alleles/loci are determining the susceptibility to stress in different environments (e.g. Fowler and Whitlock 2002, Armbruster and Reed 2005). This result emphasizes the difficulty in predicting the response of managed populations to stress. For captive breeding programs, in which individuals have become adapted to artificial rearing conditions that are more benign than their natural habitat, populations may experience substantially reduced fitness upon reintroduction to the wild, from the expression of deleterious alleles unique to the natural or novel environment. Moreover, attempts to purge inbreeding depression, under particular environmental conditions, is unlikely to significantly improve fitness (e.g. Hedrick 1994, Bijslmsa et al. 1999, Bijlsma et al. 2000, Dahlgaard and Hoffman 2000). Therefore, avoidance of inbreeding is still the best practice for maintaining or improving the viability of threatened populations. Populations likely harbor a suite of deleterious alleles that reduce fitness only in certain environments, and predicting the fitness effects of deleterious mutations will be difficult in a constantly shifting environment. 153 Chapter 3 References Armbruster, P. and D. H. Reed, 2005 Inbreeding depression in benign and stressful environments. Heredity 95: 235-242. Benjamini, Y., and Hochberg, Y., 1995 Controlling the false discovery rate: a practical and powerful approach to multiple testing. J. R. Statist. Soc. B 57: 289–300. Bijlsma, R., J. Bundgaard, and W. F. Van Putten, 1999 Environmental dependence of inbreeding depression and purging in Drosophila melanogaster. J. Evol. Biol. 12: 1125-1137. Bijlsma R., J. Bundgaard , and A. C. Boerema, 2000 Does inbreeding affect the extinction risk of small populations?: predictions from Drosophila. J. Evol. Biol. 13: 502–514. Breese, W. P. and R. E. Malouf, 1975 Hatchery manual for the Pacific Oyster, Crassostrea gigas (Gould). Oregon Agricultural Experiment Station Special Report No, 443, and Oregon State University Sea Grant College Program Publ. No. ORESU-H-75-002, 22 pp. Boakes, E. H., J. Wang, and W. Amos, 2007 An investigation of inbreeding depression and purging in captive pedigreed populations. Heredity 98:172-182. Bos, O. G., I. E. Hendriks, M. Strasser, P. Dolmer, P. Kamermans, 2006 Estimation of food limitation of bivalve larvae in coastal waters of northwestern Europe. J. Sea Res. 55: 191-206. Bucklin, K. A., 2003 Analysis of the Genetic Basis of Inbreeding Depression in the Pacific Oyster Crassostrea gigas. PhD Dissertation, University of California at Davis. Byers, D. L., and D. M. Waller, 1999 Do plant populations purge their genetic load? Effects of population size and mating history on inbreeding depression. Annu. Rev. Ecol. Syst. 30: 479-513. Caers, M., P. Coutteau, P. Lombeida, and P. Sorgeloos, 1998 The effect of lipid supplementation on growth and fatty acid composition of Tapes philippinarum spat. Aquaculture 162: 287-299. Carr, D.E., and M.R. Dudash, 2003 Recent approaches into the genetic basis of inbreeding depression in plants. Proc. R. Soc. B-Biol. Sci. 358: 1071-1084. 154 Charlesworth, D., and B. Charlesworth, 1987 Inbreeding depression and its evolutionary consequences. Annu. Rev. Ecol. Syst. 18: 237–268. Charlesworth, D., and J. H. Willis, 2009 The genetics of inbreeding depression. Nat. Rev. Genet. 10:783-796. Cheptou, P–O., E. Imbert, J. LePart, and J. Escarre, 2000 Effects of competition on lifetime estimates of inbreeding depression in the outcrossing plant Crepis santa (Asteraceae). J. Evol. Biol. 13: 522-531. Cheptou, P-O., K. Donohue, 2011 Environment-dependent inbreeding depression: its ecological and evolutionary significance. New Phyt. 189(2):395-407. Cohen, C. S., 1992 Population Biology of two species of Corella: Mating systems and demography (Ascidians). PhD dissertation, University of Washington, Dept. of Zoology. Coon, S. L., D. B. Bonar, and R. M.Weiner, 1986 Chemical production of cultchless oyster spat using epinephrine and norepinephrine. Aquaculture 58: 255-262. Coutteau, P, M. Caers, A. Mallet, W. Moore, J. J. Manzi, and P. Soorgeloos, 1994 Effect of lipid supplementation on growth, survival, and fatty acid composition of bivalve larvae. In: Measures for Success. Eds. Kestemont, P. Muir, J., and Williot, P. Proceedings of Bordeaux Aquaculture 1994, Mar. 23-25, Bordeaux, France. Crnokrak, P., and S. C. H. Barrett, 2002 Purging the genetic load: A review of the experimental evidence. Evolution 56(12): 2347–2358 Crnokrak, P., and D. A. Roff, 1999 Inbreeding depression in the wild. Heredity 83: 260- 270 Curole, J. P., and D. Hedgecock, 2007 Chapter 29. Bivalve genomics: Complications, challenges, and future perspectives. In: Aquaculture Genome Technologies, Liu, Z. (editor), Blackwell Publishing, Ames, Iowa, 525-543. Dahlgaard J. and A. A. Hoffmann, 2000 Stress resistance and environmental dependency of inbreeding depression in Drosophila melanogaster. Conserv. Biol 14: 1187– 1192. Darwin, C. R., 1876 The effects of cross and self-fertilization in the vegetable kingdom. John Murray, London. 155 DeWoody, J. A., J. Schupp, L. Kenefic, J. Busch, L. Murfitt, et al., 2004 Universal method for producing ROX-labeled size standards suitable for automated genotyping. BioTechniques 37: 348-352. Falconer D. S., and Mackay T. F. C. 1996 Introduction to Quantitative Genetics, 4th edn. Longman: Harlow. Fenaux, L., M. F., Strathmann, and R. R. Strathmann, 1994 Five tests of food-limited growth of larvae in coastal waters by comparisons of rates of development and form of echinoplutei. Limnol. Oceanogr. 39: 84–98. Fowler K., Whitlock M. C., 2002 Environmental stress, inbreeding, and the nature of phenotpyic and genetic variance in Drosophila melanogaster. Proc. Roy. Soc. Lond. B 269: 677–683. Fox C. W., and D. H. Reed, 2011 Inbreeding depression increases with environmental stress: an experimental study and meta-analysis. Evolution 65(1): 246-258. Frankham, R., 1995 Conservation genetics. Ann. Rev. Genetics 29: 305-327. Frankel, O. H. and M. E. Soulé, 1981 Conservation and Evolution. Cambridge University Press, Cambridge. Garcia-dorado, A., and A. Caballero, 2000 On the average coefficient of dominance of deleterious spontaneous mutations. Genetics 155: 1991-2001. Hayes, C. N., J. A. Winsor, and A. G. Stephenson, 2005 Environmental variation influences the magnitude of inbreeding depression in Cucurbita pepo spp. texana (Cucurcbitaceae). J. Evol. Biol. 18: 147-155. Helm, M. M., D. L. Holland, S. D., Utting, and J. East, 1991 Fatty acid composition of early non-feeding larvae of the European flat oyster, Ostrea edulis. J. Mar. Biol. Ass. U.K. 71: 691-705. Hedgecock, D., and J. P. Davis, 2007 Heterosis for yield and crossbreeding of the Pacific oyster Crassostrea gigas. Aquaculture 272S: S17-S29. Hedgecock, D., P. M. Gaffney, P. Goulletquer, X. Guo, K. Reece, G. Warr, 2005 The case for sequencing the Pacific oyster genome. J. Shellfish Res. 24(2): 429–441. Hedgecock, D., G. Li, S. Hubert, K. Bucklin, and V. Ribes, 2004 Widespread null alleles and poor cross-species amplification of microsatellite DNA loci cloned from the Pacific oyster, Crassostrea gigas. J. Shellfish Res. 23: 379-385. 156 Hedgecock, D., D. J. McGoldrick, D. T. Manahan, J. Vavra, N. Appelmans and B. L. Bayne, 1996 Quantitative and molecular genetic analyses of heterosis in bivalve molluscs. J. Exp. Mar. Biol. Ecol. 203: 49-59. Hedrick P. W., 1994 Purging inbreeding depression and the probability of extinction: full-sib mating. Heredity 73: 363–372. Hedrick P. W., and S. T. Kalinowski, 2000 Inbreeding depression in conservation biology. Annu. Rev. Ecol. Evol. Syst. 31:139-162. Hedrick, P. W., and O. Muona, 1990 Linkage of viability genes to marker loci in selfing organisms. Heredity 64: 67-72. Hedrick, P. W., O. Savolainen, and K. Karkkainen, 1999 Factors influenceing the extent of inbreeding depression: an example from Scots pine. Heredity 82: 441-450. Hu, Z., and S. Xu, 2009 PROC QTL—A SAS procedure for mapping quantitative trait loci. Int. J. Plant Genomics 2009: 141234. Hubert, S., E. Cognard, and D. Hedgecock, 2009 Centromere-mapping in triploid families of the Pacific oyster Crassostrea gigas (Thunberg). Aquaculture 288: 172–183. Hubert, S., and D. Hedgecock, 2004 Linkage maps of microsatellite DNA markers for the Pacific oyster Crassostrea gigas. Genetics 168: 351-362. Husband, B. C., and D. W. Schemske, 1996 Evolution of the magnitude and timing of inbreeding depression in plants. Evolution 50 (1): 54-70. Huvet, A., P. Boudry, M. Ohresser, C. Delsert, and F. Bonhomme, 2000 Variable microsatellites in the Pacific oyster Crassostrea gigas and other cupped oyster species. Anim. Genet. 31: 71-72. Jimenez, J. A., K. A. Hughes, G. Alaks, L. Graham, & R. C. Lacy, 1994 An experimental study of inbreeding depression in a natural habitat. Science 266: 271–273. Kacser, H., and J. A. Burns, 1981 The molecular basis of dominance. Genetics 97(3-4): 639-666. Keller, L. F., P. R. Grant, B. R. Grant, and K. Petren, 2002 Environmental conditions affect the magnitude of inbreeding depression in survival of Darwin‘s finches. Evolution 56(2): 1229-1239. 157 Keller, L. F. and D. M. Waller, 2002 Inbreeding effects in wild populations. Trends Ecol. Evol. 17: 230–241. Kennedy, V. S., R. I. E. Newell, and A. F. Eble, 1996 The Eastern Oyster: Crassostrea virginica. Maryland Sea Grant press. Kristensen, T. N., J. Dahlgaard, and V. Loeschcke, 2003 Effects of inbreeding and environmental stress on fitness – using Drosophila buzzatii as a model organism. Cons. Genet. 4: 453-465. Kristensen, T. N., P. Sørensen, K. S. Pedersen, M. Kruhøffer, and V. Loeschcke, 2006 Inbreeding by environmental interactions affect gene expression in Drosophila melanogaster. Genetics 173(3): 1329-1336. Lander, E.S., and D. Botstein, 1989 Mapping Mendelian factors underlying quantitative traits using RFLP linkage maps. Genetics 121: 185-199. Langdon, C. J. and M. J. Waldcock, 1981 The effect of algal and artificial diets on the growth and fatty acid composition of Crassostrea gigas spat. J. Mar. Biol. Ass. U. K. 61: 431-448. Launey, S., and D. Hedgecock, 2001 High genetic load in the Pacific oyster Crassostrea gigas. Genetics 159: 255-265. Li, G., S. Hubert, K. Bucklin, V. Ribes, and D. Hedgecock, 2003 Characterization of 79 microsatellite DNA markers in the Pacific oyster Crassostrea gigas. Mol. Ecol. Notes 3: 228-232. Lorieux M., X. Perrier, B. Goffinet, C. Lanaud, and D. González de León, 1995 Maximum-likelihood models for mapping genetic markers showing segregation distortion. 2. F 2 populations. Theor. Appl. Genet. 90: 81–89. Luo, L., and S. Xu, 2003 Mapping viability loci using molecular markers. Heredity 90: 459-467 Lynch, M. and B. Walsh, 1998 Genetics and analysis of quantitative traits. Sinauer Associates, Sunderland Massachusetts. Magoulas, A., B. Gjetvag, V. Tersoglou and E. Zouros, 1998 Three polymorphic microsatellites in the Japanese oyster, Crassostrea gigas (Thunberg). Anim. Genet. 29:69-70. 158 McGoldrick, DJ., D. Hedgecock, L. J. English, P. Baoprasertkul, and R. D. Ward, 2000 The transmission of microsatellite alleles in Australian and North American stocks of the Pacific oyster (Crassostrea gigas): Selection and null alleles. J. Shellfish Res. 19: 779-788. Mitchell-Olds, T., 1995 Interval mapping of viability loci causing heterosis in Arabidopsis. Genetics 140: 1105–1109. Morton, N. E., J. F. Crow, and H. J. Muller, 1956 An estimate of the mutational damage in man from data on consanguineous marriages. Proc. Natl. Acad. Sci. U.S.A. 42: 855. Mukai, T., S. I. Chigusa, L. E. Mettler, and J. F. Crow, 1972 Mutation rate and dominance of genes affecting viability in Drosophila melanogaster. Genetics 72:333-355. Nascimento, I. A., 1980 Growth of the larvae of Crassostrea gigas Thunberg, fed with different algal species at high cell concentration. J. Cons. Int. Explor. Mer. 39(2): 134-139. Omholt, S. W., E. Plahte, L. Øyehaug, K. Xiang, 2000 Gene Regulatory Networks Generating the Phenomena of Additivity, Dominance and Epistasis. Genetics 155: 969-980. Pernet, F., an R. Tremblay, 2004 Effect of varying levels of dietary essential fatty acid during early onteogeny of the sea scallop Placopecten magellanicus. J. Exp. Mar. Bio. Ecol. 310(1): 73-86. Phadnis, N. & Fry, J.D., 2005 Widespread correlations between dominance and homozygous effects of mutations: Implications for theories of dominance. Genetics 171: 385-392. Piepho, H-P., 2001 A quick method for computing approximate thresholds for quantitative trait loci detection. Genetics 157: 425-432. Remington D.L., and D. M. O‘Malley, 2000 Whole genome characterization of embryonic stage inbreeding depression in a selfed loblolly pine family. Genetics 155: 337-348. Rice, W. R., 1989 Analyzing tables of statistical tests. Evolution 43: 223–225. Roff, D. A., 2002 Inbreeding depression: Tests of the overdominance and partial dominance hypotheses. Evolution 56: 768-775. 159 SAS Institute Inc., SAS 9.1.3 Help and Documentation, Cary, NC: SAS Institute Inc., 2000-2004. Sekino, M., M. Hamaguchi, F. Aranishi, and K. Okoshi, 2003 Development of novel microsatellite DNA markers from the Pacific oyster, Crassostrea gigas. Mar. Biotechnol. 5: 227-233. Shuelke, M., 2000 An economic method for the fluorescent labeling of PCR fragments. Nat. Biotechnol. 18: 233-234. Simmons, M. J. and Crow J.F., 1977 Mutations affecting fitness in Drosophila populations. Ann. Rev. Genet. 11: 49-78. Szafraniec, K., R. H. Borts, and R. Korona, 2001 Environmental stress and mutational load in diploid strains of the yeast Saccharomyces cerevisiae. Proc. Natl. Acad. Sci. USA 98: 1107–1112. Thiele, J., T. Hansen, H.R. Siegismund, and T. P. Hauser, 2009 Genetic variation of inbreeding depression among floral and fitness traits in Silene nutans. Heredity 104: 52-60. Thomspon, P. A., M-X Guo and P. J. Harrison, 1996 Nutritional value of diets that vary in fatty acid composition for larval Pacific oysters (Crassostrea gigas). Aquaculture 143(3-4): 379-391. Thompson, P. A., and P. J. Harrison, 1992 Effects of mono-specific algal diets of varying biochemical composition on the growth and survival of Pacific oyster (Crassostrea gigas) larvae. Mar. Biol. 113: 645-654. Tobari, I., 1966 Effects of Temperature on the Viability of Heterozygotes of Lethal Chromosomes in Drosophila melanogaster. Genetics 53(2): 249-259. Tomala, K., and R. Korona, 2008 Alleviation of deleterious effects of protein mutation through inactivation of molecular chaperones. Mol. Genet. Genom. 280: 409-417. Van Ooijen, J.W. and R.E. Voorrips, 2001 JoinMap® 3.0, Software for the calculation of genetic linkage maps. Plant Research International, Wageningen, the Netherlands. Wang, Y., R. Ren, and Z. Yu, 2008 Bioinformatic mining of EST-SSR loci in the Pacific oyster, Crassostrea gigas. Animal Genetics 39: 287-289. 160 Yu, H., and Q. Li, 2007 EST-SSR markers from the Pacific oyster, Crassostrea gigas. Molecular Ecology Notes, 7: 860-862. Wright, S., 1977 Inbreeding in animals: differentiation and depression, pp.44-96, in Evolution and the genetics of populations, vol. 3, University of Chicago, Chicago. Yamtich J, M. L. Voigt, G. Li and D. Hedgecock, 2005 Eight microsatellite loci for the Pacific oyster Crassostrea gigas. Anim. Genet. 36: 524-526. Zhu, C., C. Wang, and Y. M. Zhang, 2007 Modeling segregation distortion for viability selection 1. Reconstruction of linkage maps with distorted markers. Theor. Appl. Genet. 114: 295-305. 161 Chapter 4 Mutations cause massive mortality in wild full-sib families of the Pacific oyster Crassostrea gigas 4.1 Abstract High mortality during the early life-history stages of highly fecund marine invertebrates has been widely observed and is generally attributed to environmental sources; less is known about how endogenous variation in larvae may affect mortality. Evidence of substantial genetic variation in traits affecting early survival and growth, combined with observations of segregation distortion in wild full sib families of marine bivalves, suggests that genetic differences among larvae may play an important role in mortality, but no study has investigated viability selection and the cumulative effect of deleterious alleles at the genome-wide level. In this study, four outbred families were created from wild oysters in Dabob Bay WA, and a genome-wide scan for viability loci, using quantitative trait locus (QTL) mapping methods, was performed to examine how viability selection contributed to early life-history mortality. An average of nearly 8 viability QTL (vQTL) was found across families, each causing relative mortality averaging 41%. As no evidence for epistasis was found, calculation of the multiplicative effects of vQTL yielded cumulative genetic mortalities ranging from 87.4% to 99% among families. Patterns of selection at most vQTL indicated that two deleterious alleles, one from each parent, were segregating in the progeny, causing an additive 162 pattern of genotype deficiencies. Viability QTL generally appeared to be in different genomic locations in the four families, suggesting that viability loci were primarily family-specific and, therefore, likely not related to a common genetic pathway or selective process. Overall, data from the four families show that viability selection explains much of the observed type-III mortality. Moreover, the finding of large numbers of additive or partially dominant deleterious alleles, which should be quickly eliminated by natural selection, suggest that oysters have a high mutation rate, possibly as a by-product of gametogenesis for high fecundity. 4.2 Introduction The majority of marine invertebrates and many teleost marine fish are highly fecund (>10 6 eggs) and exhibit a bi-phasic life history which involves a larval stage that suffers very high mortality, followed by a more sedentary or sessile adult stage of much lower and relatively constant mortality (type III survivorship; Thorson 1950, Roughgarden et al. 1988, Winemiller and Rose 1992, Llodora 2002). The evolution of this life history strategy, the sources of such substantial mortality, and the relationship between the quality and quantity of offspring have been topics of much research in marine ecology and fisheries science (Hjort 1914, Thorson 1950, Cushing 1971, Strathmann 1985, Winemiller and Rose 1992, Caley et al. 1996, Gosselin and Qian 1997, Hunt and Schiebling 1997, Llodora 2002). One consequence of this high and seemingly stochastic early mortality is that many marine animals exhibit high variability in population abundance and recruitment from year to year (e.g. Hjort 1914, Thorson 1950, Cushing 1971, Fogarty et al. 1991). In particular, fecundity in marine fish was shown to 163 be positively correlated with high variability in recruitment (Rickman et al. 2000) and negatively correlated with stock abundance (Cushing 1971). High variation in recruitment and stock abundance has made the management of commercially important species difficult (e.g. herring and cod, Hjort 1914) and researchers have generally proposed oceanographic or ecological (exogenous) factors (i.e. food availability, oceanographic currents, and predation) as explanations for substantial larval mortality and recruitment variation (e.g. Hjort 1914, Thorson 1950, Cushing 1971, Roughgarden et al. 1988, Fogarty et al. 1991, Rickman et al. 2000). Less attention has been paid to the possibility that genetic factors could play a role in explaining these phenomena. Two lines of evidence suggest that genetic variation may be important in shaping the recruitment dynamics and survival variability in the early life history stages of high fecundity marine animals. First, experimental studies of high fecundity marine bivalves show that larvae have substantial genetic variance in growth, which is strongly associated with survival through the larval stage (e.g. Innes and Haley 1977, Losee 1979, Hillbish et al. 1993, Hedgecock et al. 1995, Ernande et al. 2003). In the Pacific oyster, Crassostrea gigas, Lannon (1980a,b) showed that variation in larval survival in diallel cross experiments had a strong genetic component and was attributable to variance in reproductive responses of broodstock. Significant paternal effects on growth and survival in highly fecund marine fish larvae have been documented as well (e.g. Panagiotaki and Geffen 2002, Probst et al. 2006, Rideout et al. 2006), and in general, great variation in growth, behavior, and mortality is observed in the larvae of such species (e.g. Crowder et al. 1992, Houde 1996, Heath and Gallego 1997, Cowan et al. 1997, Hare and Cowan 1997, Houde 2002, Fuiman and Cowan 2003). Additionally, experimental studies of 164 physiology in Pacific oyster larvae show that larvae have substantial genetic variation in physiological processes that regulate growth in the larval stages, such as protein depositional efficiency (Pace et al. 2006, Meyer and Manahan 2010). Overall, these studies suggest that endogenous biological factors influence a larva‘s ability to grow, survive, and therefore, potentially to recruit. The second, perhaps more direct, line of evidence for the role of genetic variation in larval viability comes from studies showing genotype deficiencies and genotype- dependent mortality in larval stages of marine bivalves. Studies of full-sib families from wild pair-crosses of a number of marine bivalves show significant deficiencies of allozyme and DNA marker genotypes, often heterozygous genotypes, which can largely be explained by differential viabilities of the genotypes (e.g. Wada 1975, Wilkins 1976 Beaumont et al. 1983, Mallet et al. 1985, Foltz 1986a,b, Beaumont 1991, Hu and Foltz 1996). These findings agree well with predictions made by Williams (1975) in his Elm- Oyster Model, in which he suggested that a consequence of a high fecundity life history strategy would be intense selection on sub-vital genotypes during the larval stage, contributing to the dramatic variation in recruitment and abundance observed in highly fecund marine fish (Fogarty et al. 1991, Rickman et al. 2000). Genotype dependent mortality of highly fecund marine bivalves during the larval stages may indeed be an important factor shaping recruitment patterns and abundances in these species. While experimental findings of genotype deficiencies and segregation distortion suggest an underlying genetic basis to differential viability and mortality in wild larvae, these studies used very few markers and thus they did not provide a genomic understanding of selection and mortality at the larval stage. Whether or not viability 165 selection occurs at a few loci or across the genome, and how much mortality this actually causes during the life cycle of a population of larvae are important questions that will remain unanswered until larger, genome-wide studies are carried out. Fortunately, quantitative trait locus (QTL) and linkage-mapping approaches now permit genome-wide analyses of genotype dependent mortality in non-model organisms. These approaches have been used, for example, to examine segregation distortion and identify the genomic regions implicated in adaptive population divergence and reproductive isolation in divergent hybrid crosses (e.g. Harushima et al. 2001, Yin et al. 2004, Hall and Willis 2005, Harrison and Edmands 2006, Rogers et al. 2007, Pritchard et al. 2011). Transmission ratio distortion of markers in hybrid crosses is common, and distorted loci characterize regions involved in the evolution of barriers to gene flow and hybrid lethality (e.g. Reiseberg et al. 1995, Reiseberg 1998, Fishman et al. 2001, Janczewski et al. 2007, Willet et al. 2007, Pritchard et al. 2011). Genome-wide studies of genotype-dependent mortality in inbred crosses, using QTL mapping methods, have demonstrated that the Pacific oyster carries 14-15 highly deleterious mutations affecting viability, most of which are recessive or partially recessive (Chapter 1). Recessive mutations are not expected to cause selection in wild crosses, because their effects would be masked by dominant, wild-type alleles. However, a few partially dominant loci were detected in experimental studies, and these are expected to affect the viability of heterozygous individuals in natural populations. Applying the same QTL methods to intra-population crosses of wild-caught Pacific oysters, I examine genotype-dependent mortality across the genomes of full-sib families from random pair matings like those that would occur in nature. This study, the first of 166 its kind for any marine invertebrate or fish, sheds new light on the degree to which the substantial type-III mortality observed in early life history stages is caused by endogenous differences in larval viability, as predicted by Williams (1975) for highly fecund organisms. More specifically, I use the QTL viability model (Luo and Xu 2003) to integrate segregation data for microsatellite DNA markers and single-nucleotide polymorphisms (SNPs), thereby characterizing the number, location, and effect of mutations causing viability selection in four full-sib families produced by randomly chosen wild oysters. Linkage maps were constructed separately for each of the four families, QTL scans compared the locations and effects of viability selection across the four families, and tests of epistasis revealed the extent of interaction among markers and viability QTL. Finally, the magnitude and patterns of genotypic selection in these outbred crosses were compared with previous results from inbred crosses. 4.3 Materials and Methods Biological materials and molecular methods Crosses and culturing methods in the wild crosses In July 2006, four outbred (G 0 ) families (12, 20, 24, and 45), were created by pair- crossing wild oysters sampled from Dabob Bay, WA. Crosses were performed at the Taylor Shellfish Farms hatchery in Quilcene, WA, and larvae from each family were cultured in separate, 100-l vessels of fresh seawater, stocked at a density of 10 larvae/ml, which was thinned to 3 larvae ˑml -1 at day 2. Larvae were fed a mixture of algae: 1/3 Isochrysis galbana (Tahitian isolate), 1/3 Chaetocerus calcitrans, and 1/3 Pavlova sp. 167 (CCMP 459) at 30,000 cellsˑml -1 starting at day 2, which was increased to 60,000 cellsˑml -1 at day 8, following standard larval rearing procedures for the Pacific oyster (e.g. Breese and Malouf 1975, Hedgecock et al. 1996, Hedgecock and Davis 2007). When competent for settlement—based on morphology (the presence of eye spots), behavior (probing the substrate with their foot), and size (>250 μm)—larvae holding on a 254 μm mesh screen were treated for three hours with a 0.1 mM solution of epinephrine- bitartate, in a 1-l beaker of fresh seawater (Coon et al. 1986). Metamorphosed individuals (those that did not swim when re-suspended after 3-hours of incubation) were placed on a 200 μm mesh screen in a down-welling indoor nursery, fed a constant but low-density mix of algae from outdoor algal ponds. Settlement treatments began on day 20, and larvae that were still swimming after the 3-h incubation were placed back in the culture and treated again on days 22, 24, and 26. Up to 20,000 spat were initially set on the mesh for each family and after a few weeks the density was reduced to 4000 per screen, and the food supply was changed to an upwelling regime, in which oysters grew until large enough to be placed in intertidal bags in Willipa Bay, WA or Thorndyke Bay, WA. DNA extraction, PCR, and Electrophoresis Adductor muscles from adult oysters were sampled in December of 2007, at 528 to 533 days post fertilization (~1.45 years old) and stored in 70% ethanol at 4ºC until extraction. Additional spat from the same families were sampled in June of 2009 (~2.9 years old) to increase sample sizes. DNA was extracted from 10-25 mg of tissue using 168 either the Qiagen DNeasy Animal Tissue or the Gentra Pure-gene tissue kit (Qiagen, Valencia, CA), following manufacturers‘ protocols. Over 80 microsatellite markers cloned from the Pacific oyster were tested for this study (Magoulas et al. 1998, McGoldrick et al. 2000, Huvet 2000, Li et al. 2003, Sekino et al. 2003, Yamtich et al. 2005, Yu and Li 2007, Wang et al. 2007), many of which are on published linkage maps (Hubert and Hedgecock 2004, Hubert et al. 2009). Markers were named as in the source publication, except for those developed in this laboratory (McGoldrick et al. 2000, Li et al. 2003, Yamtich et al. 2005), which were abbreviated from their original description, e.g. ucdCgi195 to Cg195. Polymerase chain reactions (PCR) were carried out in two phases; the first to amplify the target microsatellite marker, and the second to incorporate a fluorescently labeled dye attached to a ―zip‖ sequence that is complementary to the 3‘ tail of the forward primer (Shuelke 2000). PCR cycles consisted of an initial denaturing step, at 94°C, for 2 min, then 20, phase-one cycles (30 sec at 94°C, 45 sec at T m , 45 sec at 72°C), followed by 10 ―zip‖ cycles (30 sec at 94°C, 45 sec at 59°C and 45 sec at 72°C) and a final 10-min elongation step at 72°C. T m (the optimum annealing temperature) and the MgCl 2 concentration varied, depending on the locus (Magoulas et al. 1998, McGoldrick et al. 2000, Huvet 2000, Li et al. 2003, Sekino et al. 2003, Yamtich et al. 2005). PCR products were electrophoresed on a 4% denaturing PAGE gel (acrylamide: bisacrylamide 19:1, 3M Urea) using 1 TBE (Tris- base, boric acid, and EDTA di-sodium salt) on the ABI Prism 377 DNA Sequencer (Perkin Elmer, Waltham Massachusetts). Gel images were used to score the fluorescently labeled microsatellites by eye, comparing progeny alleles to adult alleles along with fluorescently labeled size standards (DeWoody et al. 2004). 169 Table 11 Information for 11 novel SNP markers SNP name a Contig name b SNP c Genbank accession d Possible function e SNP1 contig724_514 C/A HS162916.1 ras-related protein Rab-39B SNP2 contig12a_431 G/C HS219155.1 heat shock 70kDa protein SNP3 contig383a_347 G/T HS160566.1 none found SNP4 contig221_614 T/C HS152932.1 Actin-related protein 3 SNP5 contig1054_375 G/A HS225868.1 chaperone protein dnaJ SNP6 contig358_216 C/T HS152106.1 BTB/POZ domain-containing protein KCTD21 SNP7 contig613_432 C/T HS136333.1 Solute carrier family SNP8 contig508_396 G/A HS139539.1 none found SNP9 contig495_545 T/C HS161495.1 none found SNP10 contig58a_415 G/A HS204620.1 headcase protein homolog SNP11 contig353a_388 C/T HS152106.1 major egg antigen a name used in this study, b name of the contig, in which the SNP was identified; in the process of submission to GenBank SNPdb; c allelic state of the SNP polymorphism; d GenBank EST database accession for one of multiple cDNA clones sequenced by JGI, which went into the contig; e potential function determined by blastx of contig nucleotide sequences against the GenBank non-redundant protein database. 169 170 For family 45, 12 single nucleotide polymorphism markers (SNP‘s) were isolated from expressed sequence tags (EST) for the Pacific oyster and genotyped at the Roswell Park Cancer Institute, Buffalo, NY, using the iPLEX ® Gold mass array platform (Sequenom, San Diego, CA.). SNP‘s were named consecutively for simplicity, SNP1-SNP11; SNP marker information is presented in Table 11. These markers have been submitted to Genbank. Data analyses Analysis of segregation data All types of segregations can occur in microsatellite markers in an outbred cross, including segregation of two alleles (AB×AA or AB×AB), three alleles (AB×BC), and four alleles (AB×CD), though the majority of segregations are likely to be four allele cases. Non-amplifying or null alleles are common in microsatellite segregations in the Pacific oyster (e.g. McGoldrick and Hedgecock 1997, McGoldrick et al. 2000, Hedgecock et al. 2004) but genotypes can be inferred without error for most null segregations (e.g. AØ×BØ and AØ×BC; Chapter 1). As a first examination of segregation at markers across the genome, genotype data for each marker was tested individually for deviations from expected Mendelian ratios, using goodness-of-fit chi- square tests, with significance reported at both the α =0.05 level. To assess genotyping error (Pompanon et al. 2005), genotypes were rescored in at least 5% of individuals in each family. 171 Linkage map construction Linkage maps were constructed separately for the four experimental families (Family 12, n = 158; Family 20, n = 136; Family 24, n = 96; Family 45, n = 169), using the CP (cross pollinator) population type in JoinMap 3.0 (VanOoijen and Voorrips 2001). The Kosambi mapping function with a minimum likelihood of the odds (LOD) score of 2.0 was used for linkage group (LG) assignments. Deviations from Mendelian segregation ratios may affect linkage mapping (Lorieux et al. 1995, Zhu et al. 2007), so we compared our marker order and distances with previously published linkage maps constructed from outbred larval samples with little segregation distortion (Hubert and Hedgecock 2004). If markers failed to map, we used locations from published maps (Hubert and Hedgecock 2004, Hubert et al. 2009). Phase was determined by JoinMap 3.0 from the frequencies of parental and recombinant types. Phase information, marker locations, and parent and progeny genotypes were input for the vQTL model of Luo and Xu (2003), which locates and characterizes loci under viability selection. This model, which is implemented in PROC QTL (Hu and Xu 2009), a user defined procedure for SAS (version 9.2, SAS Institute, Inc., Cary, NC), scans the genome in 1 cM increments for the presence of loci associated with viability selection, under the maximum-likelihood QTL framework. The model produces a likelihood ratio test (LRT) statistic and estimates the proportions of genotypes for each 1 cM increment (Luo and Xu 2003). The model interprets all cross types as AB CD (sire×dam), yielding progeny genotypes AC, AD, BC, BD. In the absence of viability selection, the genotype frequencies at any point in the genome are expected to be 0.25: 0.25: 0.25: 0.25; sampling error and viability selection 172 alter these expected proportions. The vQTL model was run on the adult genotype data from each of the four families separately and significance thresholds were set using an approximate method based on the LRT profile rom each family, establishing genome- wide thresholds at the = 0.01 level (Piepho 2003). QTL were identified if the LRT profile was above the genome-wide threshold for significance. Multiple QTL on a single linkage group were identified when the LRT statistic fell by at least 4.60 (~1 LOD) between two QTL peaks (Lander and Botstein 1989). Genetic effects from viability QTL associated markers Once significant QTL were found, genotype arrays at markers closest to the QTL were analyzed to examine genetic effects. A series of nested chi-square tests (e.g. Foltz 1986a, Bechsgaard et al. 2004) were employed to distinguish statistically between the fitness effects of the individual paternal alleles (the sire effect), maternal alleles (the dam effect) and the residual, the fitness effects of the interaction between parental alleles, as the cause of segregation distortion and selection. These tests could only be performed for markers in which all four segregating alleles were distinguishable in the progeny genotypes (e.g. three and four allele cross types), so most, but not all vQTL were examined. The first test was the deviation from expected 1:1 segregation in the paternal alleles (AC and AD vs. BC and BD), the second was the deviation from the expected 1:1 segregation in the maternal alleles (AC and BC vs. AD and BD), and the third was the residual component, representing the possible interaction between the female and male gametes. Each test had one degree of freedom and was independent of the other tests. 173 Tests were considered significant if their P-values fell below the threshold level of α = 0.05. These genetic contrasts provide insight into the mode of action at vQTL, comparing the substitution effect of sire and dam alleles to the interaction of alleles. Testing for epistasis Interaction between viability QTL was assessed by contingency chi-square tests of genotypic associations between markers nearest to QTL peaks (see supplementary information, file S11 for R Code). Interaction was also tested for all markers within each family. Tests were adjusted for multiple comparisons, using the Benjamini and Hotchberg method to control false discovery rate (Benjamini and Hotchberg 1995). Power analyses were conducted to determine an adequate sample size for detecting a significant interaction between loci that also showed the apparent effects of individual deleterious alleles (see supplementary information, File S12 for details). This analysis revealed excellent power with both the smallest family sample size, n = 96 (0.90) and the largest family sample size, n = 169 (>0.99) to reject the null hypothesis of bi-locus independence, in favor of the alternative hypothesis of bi-locus interaction. The detection of interaction between viability loci is critical for appropriate calculation of mortality (S12 and see below). Estimating mortality at viability QTL Relative fitnesses of the estimated genotype frequencies at QTL peaks were used to estimate survival and mortality attributable to each viability locus. The relative fitness 174 of each genotype at a QTL is max w w ij where w max is the frequency of the genotype in highest proportion (Luo and Xu 2003). Average survival, S, at a QTL is therefore; 4 max 22 max 21 max 12 max 11 w w w w w w w w S = max 4 1 w (1) and average mortality, M, is simply 1- S or max 4 1 1 w . In finite samples, chance fluctuations in genotype proportions will produce a non-zero estimate of mortality by this equation, even in the absence of selection, so to correct for this sampling effect, we randomly generated 1000 datasets of progeny genotypes (i.e. no selection), using the parental genotypes and sample sizes appropriate for each family. We then ran each of the 1000 simulated datasets through the viability QTL model to determine the average relative survival expected at each 1 cM interval of the genome, in the absence of selection. These relative survival estimates were averaged over the whole genome and then over the 1000 simulations to obtain a grand average relative survival, in the absence of selection. Finally, the average maximum genotype frequency over the simulations without selection, w maxO , was calculated by substituting the simulated grand average relative survival into eq. 1. The correction for sampling error is applied to the estimate of average survival at a vQTL by adjusting w max , the maximum frequency of the four genotypes: 25 . 0 4 1 max max O V adj w w S , (2) 175 where w maxV is estimated from the data by the QTL model and 0.25 is the Mendelian expectation in a sufficiently large sample without selection. 4.4 Results Segregation results and genotyping error A total of 80 microsatellite markers were tested in the four families, of which 39, 39, 32 and 36 were informative and successfully genotyped for linkage mapping and segregation analyses in families 12, 20, 24 and 45, respectively (Table 12). Fifty SNPs were tested in family 45, yielding 10 informative loci for a total of 46 markers segregating in this family. For the microsatellite markers, the majority of segregation types (73%) were four allele cases (AB×CD), although two allele and three allele cross types were observed (Table 12). Nine of 10 SNP markers were cross type AB×AA and only one was cross type AB×AB (Table S7). Across all four families, SNP and microsatellite markers exhibited high levels of significant segregation distortion. Fifty, 71, 40 and 58 percent of markers deviated significantly from expected Mendelian ratios in families 12, 20, 24 and 45, respectively (Table 12). SNP‘s genotyped in family 45 showed a similar frequency of segregation distortion (5 of 10). Of the distorted markers, 94% displayed deficiencies of heterozygous genotypes, most of which were due to the low fitness of genotypes sharing an allele apparently linked to a deleterious allele, or more commonly, three of the four genotypes were deficient, suggesting two deleterious 176 alleles, one from each parent, segregating in the progeny (supplementary information, Tables S13-S16; see ―Genetic effects‖ section below). For example, at Cg109 in family 45, one genotype was observed to be in the highest relative frequency (no linked deleterious alleles), while two of the genotypes were at modest frequency (each with one deleterious allele), and the other genotype was in the lowest frequency (two deleterious alleles; Table S16), illustrating the apparently ―additive‖ effect of two deleterious alleles. Only one distorted marker exhibited deficiencies of just one heterozygous genotype, a result not consistent with selection against a dominant deleterious allele that necessarily would be transmitted to two genotypes in the progeny (Cg142, family 12; Table S13). Distorted markers were distributed broadly across all linkage groups over the four families, although, never on all 10 linkage groups within a family (Tables S13-S16). An average of 17 markers was examined for genotype error in each family. Genotyping error was low, averaging 2.51% across all families, with a maximum error rate of 4.55% in family 12. QTL mapping results Assignment of microsatellite markers to linkage groups was largely in accord with previously published maps (Hubert and Hedgecock 2004, Hubert et al. 2009, Chapter 1). Total map lengths were estimated at 412, 478, 345 and 422 cM, for families 12, 20, 24, and 45, respectively, reflecting family and marker sample sizes. All 10 linkage groups were constructed successfully in family 45, with the 10 new SNP markers 177 Table 12 General segregation results in the progeny of the four wild crosses Family Cross type 12 20 24 45 Totals AA×AB -- 3(1) -- 10 (3) 13(4) AA×BC 3(0) 2(1) 2(0) 3(2) 10(3) AB×AB 1(1) -- -- 1(0) 2(1) AB×AC 2(1) 4(3) 2(1) 2(1) 10(6) AB×CD 24(13) 21(16) 20(9) 27(19) 92(57) Totals 30(15) 30(21) 24(10) 43(25) 127(71) General segregation results across the four wild families, arranged by cross type and family. Numbers in parentheses represent the numbers of markers with significant deviations of genotypic proportions from expected Mendelian ratios at the nominal α=0.05 level. 178 mapping across seven of the 10 linkage groups (Table S16). Families 20, 24, and 12 had at least two representative markers on nine linkage groups, but markers expected on LG 4, LG 5, and LG 9, in these families, respectively, failed to link to each other. Only for LG 6 in families 24 and 45 did marker orders differ substantially from published linkage maps; the observed order was used, since heterogeneity in marker order was documented for this linkage group among non-distorted larval samples (Hubert and Hedgecock 2004). Overall, segregation distortion appeared to have only a marginal influence on marker order and map distances. QTL analyses of viability selection across the genome was carried out for all 10 linkage groups in family 45, and 9 of 10 linkage groups for families 12, 20, and 24 (Fig. 15-18, Table 13). Seven viability QTL (vQTL) were detected in families 12 and 24, eight vQTL were detected in family 20, and nine vQTL were detected in family 45, all of which were above the α = 0.01 genome-wise significance threshold of 22, for families 12, 20, and 24, and 19.1, for family 45 (Table 13, Fig. 15-18). QTL were broadly distributed across the genome in each of the families, but no more than eight linkage groups exhibited significant vQTL within a family. Comparing across families, positions of QTL showed no clear patterns that would implicate particular regions of the genome as being consistently under viability selection. QTL peaks ranged substantially in significance as well as effect. For example, the likelihood ratio test (LRT) ranged to a high of 180 for a QTL peak on LG 1 in family 45; LRT values for most QTL were below 60, with a mode in the 30s (Table 13). 179 Figure 15 QTL mapping results for family 12 Likelihood ratio test statistic vs. map position (cM) for family 12. QTL are numbered consecutively next their peaks. Long vertical lines mark the ends of linkage groups which are numbered 1-10. Small black triangles along the X-axis indicate the position of markers used in the mapping procedure. The dotted black line indicates the genome-wise threshold value for significance at the = 0.05 level, 22.0. 179 180 Figure 16 QTL mapping results for family 20 Likelihood ratio test statistic vs. map position (cM) for family 20. QTL are numbered consecutively next their peaks. Long vertical lines mark the ends of linkage groups which are numbered 1-10. Small black triangles along the X-axis indicate the position of markers used in the mapping procedure. The dotted black line indicates the genome-wise threshold value for significance at the = 0.01 level, 22. 180 181 Figure 17 QTL mapping results for family 24 Likelihood ratio test statistic vs. map position (cM) for family 24. QTL are numbered consecutively next their peaks. Long vertical lines mark the ends of linkage groups which are numbered 1-10. Small black triangles along the X-axis indicate the position of markers used in the mapping procedure. The dotted black line indicates the genome-wise threshold value for significance at the = 0.01 level, 22.0. 181 182 Figure 18 QTL mapping results for family 45 Likelihood ratio test statistic vs. map position (cM) for family 45. QTL are numbered consecutively next their peaks. Long vertical lines mark the ends of linkage groups which are numbered 1-10. Small black triangles along the X-axis indicate the position of markers used in the mapping procedure. The dotted black line indicates the genome-wise threshold value for significance at the = 0.01 level, 19.1. 182 183 Table 13 Viability QTL results, genetic mortality, and genetic effects QTL, survival and genetic effects results for the four wild families. Frequencies of the genotypes estimated by the model at QTL are displayed in columns AC, AD, BC, and BD. Genetic effects are estimated from the nested chi-square tests of markers nearest to QTL, and the S (sire effect) D (dam effect) and I (interaction effect) are listed if they are significant at the α = 0.05 level. Distance is the centi-morgan (cM) map distance between the QTL peak and the nearest marker used for the nested chi-square tests. Survival (equation 1, ―materials and methods‖) and adjusted survival (corrected for sample size through simulation, equation 2) at each QTL are reported, along with with v max , the maximum genotype frequency at a given QTL. Estimates of cumulative survival use adjusted survival estimates and only QTL that are unlinked (>50cM apart) or the more deleterious of QTL that are linked (survival at the less deleterious QTL is shown in parentheses). Table shown on next page. 184 Table 13 (Continued) Family vQTL LG Marker Dist. Pos. LRT AC AD BC BD Genetic effects W max Survival adj S 12 1 2 Cg157 33 98.6 0.01 0.53 0.18 0.27 D,I 0.535 0.468 0.525 2 3 Cg148 31 139.7 0.02 0.3 0.03 0.65 S 0.654 0.382 0.420 3 4 imb49 4cm 6 88.4 0.2 0.09 0.65 0.06 S,D,I 0.652 0.383 0.421 4 6 Cg142 12cm 31 35.7 0.3 0.24 0.05 0.42 D,I 0.417 0.599 0.697 5 8 Cg155 3 25.4 0.08 0.45 0.3 0.17 I 0.452 0.553 0.636 6 10 Cg129 0 93.1 0.28 0.04 0.61 0.07 S,D,I 0.613 0.408 0.451 7 10 Cg140 4cm 21 45.9 0.2 0.07 0.48 0.25 S,D 0.476 0.525 0.598 Cum. Surv 0.019 Avg. Surv 0.474 20 1 1 cmrCg61 9cm 9 28.6 0.45 0.23 0.232 0.09 S,D 0.453 0.552 0.629 2 3 Cg150 7cm 7 65.3 0.06 0.24 0.104 0.6 S,D,I 0.595 0.42 0.463 3 4 Cg109 46 50 0.37 0.06 0.42 0.15 D 0.42 0.595 0.684 4 4 Cg2 71 43.7 0.47 0.1 0.298 0.13 S,D,I 0.468 0.534 0.605 5 8 Cg131 0 23.9 0.13 0.18 0.27 0.42 S,D 0.591 (0.660) 6 8 Cg28 17 44.6 0.05 0.49 0.31 0.16 -- 0.486 0.515 0.580 7 9 Cg167 28 52.5 0.36 0.43 0.045 0.17 S,D 0.425 0.588 0.675 8 10 L10 16cM 42 26.8 0.13 0.16 0.232 0.48 S,I 0.485 0.516 0.582 Cum Surv 0.040 Avg. S 0.61 Table continues next page 185 185 Table 13 (Continued) Family vQTL LG Marker Dist. Pos. LRT AC AD BC BD Genetic effects W max Survival adj S 24 1 2 Cg145 3 cM 8 25.5 0.37 0.26 0.24 0.12 -- 0.373 0.67 0.781 2 5 Cg128 7cm 11 25.6 0.23 0.12 0.4 0.26 S,D,I 0.398 0.628 0.724 3 6 Cg14 56 51 0.13 0.23 0.1 0.54 S,D 0.542 0.462 0.512 4 8 Cg28 0 78.6 0.24 0.16 0.4 0.19 S,D,I 0.403 0.62 0.714 5 8 Cg108 16cm 46 34.9 0.24 0.12 0.4 0.24 D 0.404 0.618 0.712 6 10 Cg129 5cm 5 30.8 0.3 0.25 0.21 0.25 NS 0.296 0.844 1 7 10 Cg189 32 34.2 0.26 0.14 0.26 0.34 D,I 0.344 0.726 0.858 Avg. S 0.757 Cum. Surv 0.126 45 1 2 Cg145 59 177.1 0.13 0.74 0.07 0.06 S,D 0.743 0.32 0.343 2 3 Cg01 15 53.1 0.22 0.09 0.52 0.17 S,D 0.519 0.481 0.494 3 3 Cg162 27 89.2 0.21 0.08 0.62 0.09 S,D,I 0.621 0.402 0.411 4 4 Cg165 (2cm) 21 29.2 0.11 0.23 0.25 0.42 S,D 0.416 0.601 0.622 5 6 Cg130 (5cm) 10 24.5 0.12 0.42 0.17 0.28 -- 0.425 0.589 0.608 6 7 Cg131 M34 30 34.3 0.28 0.13 0.43 0.16 S,D 0.429 0.583 0.602 7 8 Cg196 (2cm) 18 31.4 0.34 0.16 0.38 0.11 S 0.381 0.656 0.681 8 9 Cg183 (11cm) 26 46.1 0.51 0.17 0.24 0.09 S,D,I 0.507 0.493 0.507 9 10 SNP3 (7cm) 14 26 0.38 0.14 0.34 0.14 -- 0.385 0.65 0.674 0 Avg. S 0.549 Cum. Surv. 0.008 186 186 Genetic effects at vQTL associated markers Genetic effects of vQTL were evaluated by nested chi-square tests of nearest- marker genotype arrays. Of the 31 QTL detected across the four families, 27 were testable in the nested chi square framework (Table 13). Allele-substitution effects (either sire or dam) were the major effects detected and were significant at 23 of the 27 vQTL tested. Interaction effects were significant at 14 of 27 QTL but were the sole significant effect for only four vQTL (Table 13). Significant sire and dam allelic substitution effects were detected in almost equal frequency, 21 sire vs. 22 dam, indicating no parent-of- origin effect. Most vQTL-associated markers had both significant sire and dam effects (18 of 27), which suggested the segregation of one deleterious allele from each parent in the progeny, a result consistent with patterns of genotype deficiencies observed at the majority of markers (Tables S13-S16). Epistasis Pairwise tests of epistasis between markers closely linked to significant QTL showed no cases of significant interaction across the four families. Pairwise epistasis tests, using all mapped markers (the full data set), showed a very similar result. Linked markers showed significant contingency chi-square tests, as expected from linkage (Fig. 19); otherwise, only a handful of interactions among unlinked markers (8 of 1825 combinations) were nominally significant, none after correction for false discovery rate (Table 14). 187 Figure 19 Heat map of epistasis results for family 45 Pairwise chi-square tests of the bi-locus association of genotypes using all markers. Each block represents the chi-square test P-value (uncorrected) with warm colors (red) indicating significance and cold colors (blue) indicating non-significance. Markers are arranged in a pairwise fashion by ascending linkage group, which are marked with black lines; the diagonal of the triangular matrix has markers from the same linkage group side by side, and significant tests represent expected linkage associations. All red blocks (low P-values) off the diagonal were either above the α = 0.05 level or were not significant after adjustment for multiple comparisons (see Table 12). 188 Table 14 Epistasis results Family Marker 1 marker 2 χ 2 P-value 24 Cg28 Cg194 0.058 Cg177 Cg108 0.020 20 Cg141 Cg167 0.098 Cg141 Cg183 0.045 45 Cg165 Cg175 0.014 Cg164 SNP3 0.049 Cg160 Cg196 0.059 Cg162 Cg14 0.012 The most significant results (χ 2 p-value) from the contingency chi-square test of marker associations across the genome. 189 Mortality The average relative survival calculation (eq. 1), which is used to estimate genotype dependent mortality at a QTL, was adjusted for sampling error through simulation, as described in ―Materials and Methods.‖ Relative survival in simulations with no selection averaged 0.87, 0.85, 0.85 and 0.81 for families 45, 12, 20 and 24, respectively (cf. the theoretical expectation of 1.0), leading to upward adjustments of average relative survival at vQTL (eq. 2). Mortality at individual vQTL was generally quite high, ranging from 0.25 to 0.6 (i.e. a survival of 0.75 to 0.4; Table 14). Across the four families, average mortality at a QTL was 0.41. No evidence of epistasis was found in any family (Table 12), thus, viability QTL separated by >50 cM were assumed to be independent, having multiplicative effects on survival. The cumulative mortality attributable to i vQTL in each family was calculated as ∏ ̅ (Table 14). Cumulative genotype-dependent mortality was estimated at 99.2 % and 98.1% in families 45 and 12, respectively, and 96% and 87.4 % in in family 20 and 24, respectively (Table 14). 4.5 Discussion It has been widely observed that highly fecund marine animals suffer substantial rates of larval mortality early in the life-cycle. The causes of this mortality are generally attributed to non-selective deaths (Hjort 1914, Thorson 1950, Cushing 1971, Roughgarden et al. 1988, Fogarty et al. 1991, Hunt and Shiebling 1997), but few studies 190 have directly examined how genetic variation in the early life history stages might be associated with mortality. The goals of this study were to examine genotype-dependent mortality in the progeny of pair crosses of wild Pacific oysters, using QTL methods to identify viability loci in the genome, and to determine how much of the high mortality observed in culture experiments was due to viability selection. Though I did not study larval samples, an average of 93.8% survival was observed for the four oyster families from 4/23/07 to 4/7/08 (~nine to 21 months post-fertilization), and massive mortality in the seed stage (recently settled juveniles) prior to this time period would have been unusual, and was not noted (D. Hedgecock and J. Davis, personal communication). Thus, in accord with data from inbred families (Chapter 1,2), the genetic mortality detected in this study likely occurred during the early life-history stages. Substantial genotype deficiencies and segregation distortion were observed throughout the genome, and an average of 7.75 viability QTL was identified in each family. Across the genome, selection at these vQTL caused an astonishing mortality of 87-99% mortality within a family, indicating that viability loci have substantial effects on fitness in the early life history stages of oysters. The results of this genome-wide analysis of viability selection agrees well with previous studies of wild crosses of other bivalve species that showed similar patterns of segregation distortion and heterozygote deficiencies in allozyme loci (e.g. Beaumont et al. 1983, Mallet et al. 1985, Foltz 1986a,b). Mallet et al. (1985) and others concluded that heterozygote deficiencies could only be explained by selection on genotypes with differential viability, and that deleterious alleles were segregating in the progeny of wild crosses. 191 Are there alternative explanations to viability selection for the observed distortions of segregation ratios observed in these families? Inbreeding and the expression of deleterious recessive alleles can be ruled out, because we chose parents at random from a natural population of wild oysters. The preponderance of four-allele segregation types in these families (73%) is very different from their low frequency in inbred F 2 families (e.g. 6% in family 46×10 and 8.9% in family 51×35, Chapter 1; also see Chapters 2 and 3) and is not compatible with wild parents being closely related. Meiotic drive is another possible cause of distortion, implicated in patterns of transmission ratio distortion in Drosophila, mice and plants (e.g. Lyttle et al. 1991). Though we do not have temporal genetic data for these families, meiotic drive was conclusively ruled out in studies of segregation distortion in inbred families of the Pacific oyster (Launey and Hedgecock 2001, Chapter 1), inbred families of the European flat oyster Ostrea edulis (Bierne et al. 1998), and in outbred families of the Pacific oyster, in which genotyping of early larval samples revealed very low levels or the absence of segregation distortion (Hubert and Hedgecock 2004). Finally, epistatic interactions could provide an explanation for observed heterozygote deficiencies in pair crosses, since, under fairly simple genetic models, they can give the appearance of intra-locus dominance, overdominance, and under-dominance (heterozygote deficiencies; e.g. Wade 2001). I found little evidence for di-genic interactions in genome-wide tests of marker associations (Table 14, Fig. 19), which agrees with previous findings in inbred crosses (Bucklin 2003, Chapters 1, 3). Overall, my results suggest that the most likely explanation of genotype deficiencies in wild families is selection acting independently 192 against genotypes or deleterious alleles segregating at different loci in the progeny of crosses between wild individuals. Patterns of segregation distortion and genotype dependent mortality Across the four families, 40-70% of the microsatellite or SNP markers scored showed severe deficiencies of heterozygous genotypes. Patterns of segregation distortion at markers consistently showed strong selection against heterozygous genotypes that appeared to share a partially dominant deleterious allele inherited from either the sire or dam, and in some cases, progeny appeared to inherit deleterious alleles from both parents. Inspection of genotype numbers revealed one genotype that was in much higher frequency than the rest, two at intermediate frequencies, and one genotype was at low frequency (Fig. 20), suggesting inheritance of no deleterious alleles (unaffected heterozygote), one deleterious allele inherited from either the male or female parent (heterozygotes affected by one deleterious allele), and both male and female deleterious alleles (doubly affected heterozygotes), respectively. At the genome-wide level, results from the nested-chi-square tests at viability QTL-associated markers show a very similar pattern of significant sire and dam effects at most viability QTL, thus confirming single marker evidence of two deleterious alleles segregating at most viability loci. Most cases, in which marker segregation appears to reflect two deleterious alleles, are identified under one vQTL peak (Table 13), even though it is unlikely that the deleterious alleles coming from two, unrelated wild parents would be at the same locus. Deleterious alleles from each parent might, instead, represent different, closely linked 193 Figure 20 Genotype numbers at two markers illustrating the effects of two deleterious alleles Genotype numbers and proportions for two linked markers, Cg109 and Cg165 (linkage group 4, family 45), showing the effect of linked deleterious mutations causing distortion of segregation ratios. Panel A shows the genotype frequencies at each locus, first for Cg165 (AB×CD, blue bars) and second for Cg109 (AB×AC, red bars-- a the A alleles are likely only identical in state and not identical by descent, thus, AA is considered a heterozygote). Displayed below each genotype are the parental haplotypes of the two-marker region (Cg 165 first locus, Cg109 second locus) with the linked deleterious alleles (l), inferred from the phase information of the marker alleles and the genotype frequencies at each marker, demonstrating the negative fitness effects of two deleterious alleles. 194 viability loci. For example, in LG 4, family 45, where markers appear to be affected by a deleterious allele from each parent, nested chi-square tests show that sire and dam allelic effects are maximal at opposite ends of the linkage group (and minimal where the other effect is high), suggesting that the sire and dam alleles causing these effects are located at different positions within the linkage group (Fig. 21). Likewise, the sire and dam allelic substitution effects, which are calculated from contrasts of model-estimated genotype frequencies every 1 cM, are also greatest on opposite ends of the linkage group, shifting rank around the midpoint. The likelihood ratio test (LRT) profile for this linkage group does not show two clearly defined peaks with a 4.60 LRT drop between them, however. At this and a number of linkage groups with similar sire and dam effect patterns, there are often ―shoulders‖ suggestive of separate peaks (e.g. LG 3, Fig. 16 ), thus, the 4.60 LRT criteria may be too conservative. Nonetheless, similar difficulty resolving multiple QTL on linkage groups with large numbers of distorted markers was encountered with the mapping of deleterious alleles linked tightly in repulsion-phase in inbred families (Chapter 1). Thus, where viability loci are numerous or relatively closely linked, the QTL model may not identify all of them. Genomic architecture of viability selection Genomic locations of individual vQTL were different across families, which suggests that many of the deleterious alleles at viability QTL were unique to each family and do not underlie a common process or function that is the target of selection. This is consistent with the hypothesis that the source of viability selection may be random 195 Figure 21 Allelic effects on linkage group 4, family 45 Plot of sire allelic substitution effects (solid line; AB+AC - BC+BD, sire alleles in bold) and dam allelic substitution effects (dashed line; AC+BC – AD+BD, dam alleles in bold) at each cM across LG 4 of family 45 (genotype frequencies at each cM are produced by the QTL model inferring the most probable genotype based on the flanking marker data). Nested chi-square test results for sire and dam effects at markers Cg02 and Cg109 are overlaid on the figure with ―S‖, sire, or ―D‖, dam, together with the significance (*,P < 0.05; **, P < 0.01; ***, P < 0.001; ns, non-significant). 196 mutations generated during gametogenesis (Williams 1975, Launey and Hedgecock 2001), as opposed to loci that are associated with specific genetic pathways, the genes of which should localize to particular regions of the genome. The QTL results in this study contrast with other QTL studies of segregation distortion in hybrid crosses that find common genomic signatures of adaptive divergence across families and experiments (e.g. Hall and Willis 2005, Rogers and Bernatchez 2007, Pritchard et al. 2011). For example, Rogers and Bernatchez (2007) found homologous regions of segregation distortion in divergent hybrid crosses of the Whitefish (Coregonus clupeaformis), which implicated specific genomic regions involved in adaptive divergence and hybrid breakdown. Another study comparing regions of transmission ratio distortion (TRD) in an inter- specific hybrid cross and a phenotypically divergent, intra-population cross of Mimulus spp, found that six of the 11 homologous linkage groups showed TRD in both crosses (Hall and Willis 2005). Finally, in the intertidal copepod Tigriopus californicus, a homologous region of the genome exhibited strong segregation distortion in different divergent hybrid crosses and a non-recombinant backcross family (Pritchard et al. 2011). The finding of homologous regions of selection in divergent crosses most likely reflects underlying genes that are important in adaptive divergence and post-zygotic isolation, processes that would be operating at a population level, and would not be unique to a particular cross (Reiseberg 1998). In the intra-population crosses used in this study, parents were chosen randomly from within the same population, and presumably were not substantially genetically divergent from each other. The selection observed in these 197 crosses likely resulted, instead, from large numbers of deleterious mutations that were segregating in the progeny and were unique to each family (discussed below). Though the locations of QTL differed across families, the overall number of viability QTL (seven for families 12, 20, and 24, and nine for family 45) was fairly consistent, which suggests a common level of selection and genetically determined mortality in the progeny of outbred crosses. Some variability was observed in estimated cumulative mortality (87 % to 99%) however, which may be associated with differences in map coverage and variation in sample sizes among the families. For example, the largest number of vQTL, as well as the highest cumulative genetic mortality, was detected in family 45, which had the largest number of markers (covering all 10 linkage groups) and the largest sample size (n = 169). On the other end of the spectrum, the lowest estimated cumulative mortality of the four families (87.4%) was found for family 24,which had the lowest sample size (n = 96), the fewest number of markers (22), and the smallest map coverage (345 cM). Nevertheless, families 12, 20 and 24 did have at least two markers on nine of 10 linkage groups, providing coverage for most of the genome. Experimental results from all four families support the same general conclusion that viability selection caused substantial mortality in the wild crosses. Comparing levels of genetic load between wild and inbred families At the outset of this study, it was predicted that both the magnitude and frequency of segregation distortion would be reduced in outbred families relative to inbred families (Launey and Hedgecock 2001, Chapters 1,3), because the expression of deleterious 198 recessive alleles is much less likely in progeny from a wild cross. Compared to viability QTL results from inbred crosses reared in a similar mixed-algal diet (Chapter 3), the numbers of vQTL detected in the wild families was actually quite similar (average of 7.75 in the wild families vs. ~6 in the 3-algal, inbred families). Patterns of segregation distortion and selection in the progeny of inbred and outbred families were quite different, however. Almost all cases of segregation distortion in the inbred F 2 families were caused by deficiencies of homozygous genotypes (selection against identical-by- descent deleterious recessive alleles), many of which were completely absent (selection coefficient of 1.0; Chapters 1, 2, and 3). In contrast, segregation distortion in the wild crosses resulted from the deficiencies of heterozygous genotypes (most of which were not totally absent; Tables S3-S16), indicating that the mixed diet did not suppress the deleterious effects of viability mutations (Chapter 3). Though the magnitude of selection on any one genotype was less in the wild crosses, more genotypes were affected, which resulted in cumulative mortality estimates comparable to or exceeding those for inbred families, in which homozygotes were often lethal, but fewer genotypes were affected (87 to 99% mortality in family 12 and 45 compared with ~84 and 92% in the 3-algal diet; Chapter 3). Overall, it appears that selection and mortality are high in the progeny of both inbred and wild families; the real difference between them appears to be the average dominance of the deleterious alleles, which may be because they represent different classes of mutations. 199 Genetic mortality and the cost of selection in the marine environment The finding of substantial, genetically determined mortality caused by deleterious mutations segregating in the progeny of wild crosses of the Pacific oyster raises intriguing questions regarding the origin of these deleterious alleles and the evolution of high fecundity life-history strategies in the marine environment. Simulations using a simple stochastic population genetic model (Selection 3.0; gsoftnet.us) of selection against a partially dominant deleterious mutant allele (such as the ones identified in this study), assuming a starting frequency of 0.125 (one mutant allele in the eight wild parents; probably a gross overestimate), an average selection coefficient of 0.41 (estimated from the average fitness at viability QTL in this study), and a population with an effective size (N e ) of 100 (N e of ~100 in wild populations of Dabob Bay, WA Pacific oysters; Hedgecock 1994, Li and Hedgecock 1998), show that these alleles would be rapidly purged from the population within 20 generations. Thus, the high frequency of deleterious mutations segregating in the progeny of the four wild families suggests that these mutations are relatively recent in origin, and that the mutation rate in this species may be quite high. Though few studies have examined the molecular genetic details of mutation in the Pacific oyster, the possibility of a high mutation rate has some empirical support stemming from observations of high polymorphism and high nucleotide diversity (e.g. Beckenbach 1994), high rates of null alleles (McGoldrick et al. 2000, Hedgecock et al. 2004), and a high genetic load (Launey and Hedgecock 2001, Bucklin 2003, Chapter 1). One explanation for these observations could be the presence of large numbers of transposable elements that are active in the Pacific oyster genome, which have been 200 documented in gastropod marine molluscs (McInerny et al. 2011) and in the closely related Crossostrea virgninica (Gaffney et al. 2003), as well as the Pacific oyster (P. M. Gaffney, University of Delaware, personal communication). Another source of deleterious mutations could be replication errors generated during oyster gametogenesis, in which 100‘s of millions of gametes are created from possibly only a few starting germ cells each year (Fabioux et al. 2004), a process characterized by the rapid expansion of the germ cell line, over the course of a few months, to a gonad that comprises up to two- thirds of the body mass by the height of the reproductive season. In his Elm-oyster model, G.C. Williams (1975) suggested that a characteristic biological feature of high fecundity species is the generation of a great diversity of genotypes during each reproductive cycle, which would result in substantial selective mortality during the early life-history stages. The significance of selective deaths in high fecundity species has been largely overlooked in the literature, with the general argument that high mortality in the early stages of reproductively prolific organisms is mostly non- selective genetically, because of environmental or density-dependent mortality. Specifically, Barber (1965) argued that if the non-selective or environmental causes of early mortality were density-dependent than it was probable that selection occurring early in development would have little effect on the number of parents surviving to produce the next generation. In other words, the cost of selection would be ―debited against non- selective deaths‖ which would occur regardless of selection pressures during the same life-stage (Barber 1965). The data generated in the current study demonstrates that genetically selective deaths do occur in the offspring of crosses of wild oysters and that 201 these deaths may be able to explain a large percentage of the overall mortality in marine bivalves observed in culture experiments (e.g. Mallet et al. 1985) and, possibly, in the wild (e.g. Korringa 1941). These results do not discount the fact that non-selective deaths do occur during the vulnerable early life history stages of the oyster. Settlement and subsequent growth of juvenile marine bivalves in the natural environment is certainly limited by space, competition, and density-dependent effects (e.g. Gaines and Roughgarden 1985, Kennedy et al. 1996, Sietz et al. 2001, Miron 2002). Environmental sources of mortality in the ocean are very significant as well. For example, post- settlement predation has been reported to cause high rates of mortality in spat of the Eastern oyster (Kennedy et al. 1996, Newell et al 2000). Wind-driven turnover of the water column has been implicated as the cause of recruitment failure of the Pacific oyster in Dabob Bay, WA (Packer 1980, Packer and Matthews 1980, Hedgecock 1994). I argue that alongside these environmental or exogenous sources of mortality, selection is playing an equally important role, removing a large proportion of the next generation of recruits because of their genotypes. What is the benefit of creating so many maladaptive genotypes? In the highly variable ocean environment, if one can suppose that certain genotypes are more fit in some years, while others less fit, the generation of many different gene combinations could be a strategy for dealing with spatial and temporal environmental heterogeneity, a bet-hedging strategy to generate as much diversity as possible to fit variable environmental conditions (Williams 1975, Winemiller and Rose 1992, Winemiller and Rose 1993). Though most of the genotypes, and thus the individuals that carry them, 202 may fail in a given reproductive cycle, the next season or generation may be more successful, and marine animals might be able to afford the high levels of mortality early in the life cycle, when life is cheap, and when a large proportion of individuals would have died in any case from environmental causes. Another explanation is that the diversity of genotypes in the progeny of highly fecund marine animals is merely an unintended by-product of mistakes made during gametogenesis; mutational errors that may cluster or cascade down the germ line during the many germ-line mitoses required to make hundreds of millions of eggs or billions of sperm (e.g. Woodruff and Thompson 1996, Stienberger et al. 1996, Jones et al. 1999). In this scenario, most zygotes produced are actually unfit and only individuals at the high end of the fitness distribution have the chance to survive the sweepstakes world of the marine environment. Future studies documenting mutational processes during gametogenesis and examining mutational change over multiple reproductive cycles will provide insight into the patterns of genetic diversity generated during gametogenesis. This will require scanning the genome or transcriptome for single nucleotide or insertion/deletion polymorphisms in the progeny that are absent in the parents (e.g. Roach et al. 2010). This type of study may soon be feasible with the assembly of the Pacific oyster genome nearing completion. 203 Chapter 4 References Barber, H. N., 1965 Selection in natural populations. Heredity 20: 551–572. Beaumont, A.R., 1991 Genetic studies of laboratory reared mussels, Mytilus edulis: heterozygote deficiencies, heterozygosity and growth. Biol. J. Linn. Soc. 44: 273-285. Beaumont , A. R., C.M. Beveridge, and M. D. Budd, 1983 Selection and heterozygosity within single families of the mussel, Mytilus edulis (L.) Mar. Biol. 4: 151-161. Beckenbach, A. T., 1994 Mitochondrial haplotype frequencies in oysters: neutral alternatives to selection models. pp. 188–198 in Non-neutral evolution, B. Golding, ed., Chapman & Hall New York. Bechsgaard, J., T. Bataillon, and M. H. Schierup, 2004 Uneven segregation of sporophytic self-incompatibility alleles in Arabidopsis lyrata. J. Evol. Biol. 17: 554–561. Benjamini, Y., and Hochberg, Y., 1995 Controlling the false discovery rate: a practical and powerful approach to multiple testing. Journal of the Royal Statistical Society Series B, 57: 289–300. Bierne, N., S. Launey, Y. Naciri-Graven, and F. Bonhomme, 1998 Early effect of inbreeding as revealed by microsatellite analyses on Ostrea edulis larvae. Genetics 148: 1893-1906. Breese, W. P. and R. E. Malouf, 1975 Hatchery manual for the Pacific Oyster, Crassostrea gigas (Gould). Oregon Agricultural Experiment Station Special Report No, 443, and Oregon State University Sea Grant College Program Publ. No. ORESU-H-75-002, 22 pp. Bucklin, K. A., 2003 Analysis of the Genetic Basis of Inbreeding Depression in the Pacific Oyster Crassostrea gigas. PhD Dissertation, University of California at Davis. Caley, M., M. Carr, M. Hixon, T. Hughes, G. Jones and B. Menge, 1996 Recruitment and the local dynamics of open marine populations. Ann. Rev. Ecol. Syst. 27: 477 – 500. Coon, S. L., D. B. Bonar, and R. M.Weiner, 1986 Chemical production of cultchless oyster spat using epinephrine and norepinephrine. Aquaculture 58: 255-262. 204 Cowan, J. H., Jr., K. A. Rose, and E. D. Houde, 1997 Size based foraging success and models of larval fish populations. Pages 357-386 in R. C. Chambers and E. A. Trippel, editors. Early life history and recruitment in fish populations. Chapman and Hall, London, UK. Crowder, L. W., J. A. Rice, T. J. Miller, and E. A. Marschall, 1992 Empirical and theoretical approaches to size-based interactions and recruitment variability in fishes. Pages 237-255 in D. L. DeAngelis and L. J. Gross, editors. Individual- based models and approaches in ecology. Chapman and Hall, New York, New York, USA. Cushing, D. H., 1971 The dependence of recruitment on parent stock in different groups of fishes. J. du Conseil Intern. Expl. Mer. 33: 340-362. DeWoody, J. A., J. Schupp, L. Kenefic, J. Busch, L. Murfitt, et al., 2004 Universal method for producing ROX-labeled size standards suitable for automated genotyping. BioTechniques 37: 348-352. Ernande, B., J. Clobert, H. Mccombie, and P. Boudry, 2003 Genetic polymorphisms and trade-offs in the early life-history strategy of the Pacific oyster, Crassostrea gigas (Thunberg, 1795): A quantitative genetic study. J. Evol. Biol. 16: 399-414 Fabioux, C., A. Huvet, C. Lelong, R. Robert, S. Pouvreau, J.Y. Daniel et al., 2004 Oyster vasa-like gene as a marker of the germline cell development in Crassostrea gigas. Biochem. Biophys. Res. Comm. 320: 592-598. Fishman L, A. J. Kelly, E. Morgan, and J. H. Willis, 2001 A genetic map in the Mimulus guttatus species complex reveals transmission ratio distortion due to heterospecific interactions. Genetics 159: 1701-1716. Fogarty, M. J., M. P. Sissenwine, and E. B. Cohen, 1991 Recruitment variability and the dynamics of exploited marine populations. Trends Ecol. Evol. 6:241–246. Foltz, D. W., 1986a Null alleles as a possible cause of heterozygote deficiencies in the oyster Crassostrea virginica and other bivalves. Evolution 40: 869-870. Foltz, D. W., 1986b Segregation and linkage studies of allozyme loci in Pair crosses of the oyster Crassostrea virginica. Biochem. Genet. 24: 941-956. Fuiman, L. and J. H. Cowan, Jr., 2003 Behavior and recruitment success in fish larvae: repeatability and covariation of survival skills. Ecology 84(1): 53-67. 205 Gaffney P. M, J. C. Pierce, A. G. MacKinley, D. A. Titchen, W. K. Glenn, 2003 Pearl, a novel family of putative transposable elements in bivalve molluscs. J. Mol. Evol. 56: 308–316. Gosselin, L. A., and P. Y. Qian, 1997 Juvenile mortality in benthic marine invertebrates. Mar. Ecol. Prog. Ser. 146: 265–282. Hall, M. C., and J. H. Willis, 2005 Transmission ratio distortion in intraspecific hybrids of Mimulus guttatus. Genetics 170: 375-386. Hare, J. A., and R. K. Cowen, 1997 Size, growth, development, and survival of the planktonic larvae of Pomatomus saltatrix (Pisces: Pomatomidae). Ecology 78: 2415- 2431. Heath, M., and A. Gallego, 1997 From the biology of the individual to the dynamics of the population: bridging the gap in fish early life studies. J. Fish Biol. 51: 1-29. Harrison J. S. and S. Edmands, 2006 Chromosomal basis of viability differences in Tigriopus californicus interpopulation hybrids. J. Evol. Biol. 19: 2949-2051. Harushima, Y., M. Nakagahra, M. Yano, T. Sasaki, and K . Kurata, 2001 A genome-wide survey of reproductive barriers in an intraspecific hybrid. Genetics 159: 883-892. Hedgecock, D., 1994 Does variance in reproductive success limit effective population sizes of marine organisms? In: A. R. Beaumont,editor. Genetics and evolution of aquatic organisms. London, Chapman Hall. pp. 122–134. Hedgecock, D., and J. P. Davis, 2007 Heterosis for yield and crossbreeding of the Pacific oyster Crassostrea gigas. Aquaculture 272S: S17-S29. Hedgecock, D., P. M. Gaffney, P. Goulletquer, X. Guo, K. Reece, G. Warr, 2005 The case for sequencing the Pacific oyster genome. J. Shellfish Res. 24(2): 429–441. Hedgecock, D., G. Li, S. Hubert, K. Bucklin, and V. Ribes, 2004 Widespread null alleles and poor cross-species amplification of microsatellite DNA loci cloned from the Pacific oyster, Crassostrea gigas. J. Shellfish Res. 23: 379-385. Hedgecock, D., D. J. McGoldrick, D. T. Manahan, J. Vavra, N. Appelmans and B. L. Bayne, 1996 Quantitative and molecular genetic analyses of heterosis in bivalve molluscs. J. Exp. Mar. Biol. Ecol. 203: 49-59. 206 Hilbish, T. J., E. P. Winn, and P. D. Rawson, 1993 Genetic variation and covariation during larval and juvenile growth in Mercenaria mercenaria. Marine Biology 115: 94-107.Hu, Z., and S. Xu, 2009 PROC QTL—A SAS procedure for mapping quantitative trait loci. Int. J. Plant Genomics 2009: 141234. Hu, Y. P. and D. W. Foltz, 1996 Genetics of scnDNA polymorphisms in juvenile oysters Crassostrea virginica Part 1: Characterizing the inheritance of polymorphisms in controlled crosses. Mol. Mar. Biol. Biotechnol. 5(2): 123-129. Hubert, S., E. Cognard, and D. Hedgecock, 2009 Centromere-mapping in triploid families of the Pacific oyster Crassostrea gigas (Thunberg). Aquaculture 288: 172–183. Hubert, S., and D. Hedgecock, 2004 Linkage maps of microsatellite DNA markers for the Pacific oyster Crassostrea gigas. Genetics 168: 351-362. Huvet, A., P. Boudry, M. Ohresser, C. Delsert, and F. Bonhomme, 2000 Variable microsatellites in the Pacific oyster Crassostrea gigas and other cupped oyster species. Anim. Genet. 31: 71-72. Hjort, J., 1914 Fluctuations in the great fisheries of Northern Europe. Rapports et Proce`s-Verbaux des Re´unions du Conseil International pour l‘Exploration de la Mer, 20:1–228. Houde, E. D., 1996 Evaluating stage-specific survival during the early life of fish. Pages 51-66 in Y. Watanabe, Y. Ya- mashita, and Y. Oozeki, editors. Survival strategies in early life stages of marine resources. A. A. Balkema, Rotterdam, The Netherlands. Houde, E. D., 2002 Mortality. Pages 64-87 in L. A. Fuiman and R. G. Werner, editors, Fishery science: the unique contributions of early life stages. Blackwell Science, Oxford, UK. Hu, Z., and S. Xu, 2009 PROC QTL—A SAS procedure for mapping quantitative trait loci. Int. J. Plant Genomics 2009: 141234. Hunt, H. L., and R. E. Scheibling, 1997 Role of early post-settlement mortality in recruitment of benthic marine invertebrates. Mar. Ecol. Prog. Ser. 155: 269–301. Innes, D. J., and L. E. Haley, 1977 Genetic aspects of larval growth under reduced salinity in Mytilus edulis. Biol. Bull. 153: 312-321. 207 Janczewski E, M. Gherardi, I. Bonnin, J. M. Prosperi, I, Olivierei, T. Huguet, 1997 Insight on segregation distortions in two intraspecific crosses between annual species of Medicago (Leguminosae). Theor. Appl. Genet. 94: 682-691. Jones, A. G., G. Rosenqvist, A. Berglund, and J. C. Avise, 1999 Clustered microsatellite mutations in the pipefish Syngnathus typhle. Genetics 152: 1057-1063. Kennedy, V. S., R. I. E. Newell, and A. F. Eble, 1996 The Eastern Oyster: Crassostrea virginica. Maryland Sea Grant press. Korringa, P., 1941 Experiments and observations on swarming, pelagic life and setting in the European flat oyster, Ostrea edulis L. Arch. neerl. Zool. 5: 1-249 [Thesis, Univ. Amsterdam, 1940]. Launey, S., and D. Hedgecock, 2001 High genetic load in the Pacific oyster Crassostrea gigas. Genetics 159: 255-265. Lander, E.S., and D. Botstein, 1989 Mapping Mendelian factors underlying quantitative traits using RFLP linkage maps. Genetics 121: 185-199. Li, G. and D. Hedgecock, 1998 Genetic heterogeneity, detected by PCRSSCP, among samples of larval Pacific oysters (Crassostrea gigas) supports the hypothesis of large variance in reproductive success. Can. J. Fish. Aquat. Sci. 55:1025–1033. Li, G., S. Hubert, K. Bucklin, V. Ribes, and D. Hedgecock, 2003 Characterization of 79 microsatellite DNA markers in the Pacific oyster Crassostrea gigas. Mol. Ecol. Notes 3: 228-232. Lannan, J.E., 1980a Broodstock management of Crassostrea gigas I.Genetic variation in in the larval rearing system. Aquaculture, 21: 323-336. Lannan, J. E., 1980b Broodstock management of Crassostrea gigas: III. Selective breeding for improved larval survival. Aquaculture 21: 347-351. Llodra, E. R., 2002 Fecundity and life-history strategies in marine invertebrates. Adv. Mar. Bio. 43: 87-170. Lorieux M., X. Perrier, B. Goffinet, C. Lanaud, and D. González de León,1995 Maximum-likelihood models for mapping genetic markers showing segregation distortion. 2. F 2 populations. Theor. Appl. Genet. 90: 81–89. Losee, E., 1979 Relationship between larval and spat growth rates in the oyster Crassostrea virginica. Aquaculture 16: 123-126. 208 Luo, L. and S. Xu, 2003 Mapping viability loci using molecular markers. Heredity 90: 459-467. Lyttle, T. W., 1991 Segregation distorters. Ann. Rev. Genet. 25: 511–557. Magoulas, A., B. Gjetvag, V. Tersoglou and E. Zouros, 1998 Three polymorphic microsatellites in the Japanese oyster, Crassostrea gigas (Thunberg). Anim. Genet. 29:69-70. Mallet, A. L., E. Zouros K. E. Gartner-Keykay, K. R. Freeman, and L. M. Dickie, 1985 Larval viability and heterozygote deficiency in populations of marine bivalves: evidence from pair matings of mussels. Mar. Biol. 87: 165-172. McGoldrick, D.J., Hedgecock, D., 1997 Fixation, segregation and linkage of allozyme loci in inbred families of the Pacific oyster Crassostrea gigas (Thunberg): implications for the causes of inbreeding depression. Genetics 146: 321-334. McGoldrick D. J., D. Hedgecock, L. J. English, P. Baoprasertkul, and R. D. Ward, 2000 The transmission of microsatellite alleles in Australian and North American stocks of the Pacific oyster (Crassostrea gigas): selection and null alleles. J. Shellfish Res. 19: 779–788. McInerney, C. E., A. L. Allcock, M. P. Johnson, D. A. Bailie, and P. A. Prodöhl, 2011 Comparative genomic analysis reveals species-dependent complexities that explain difficulties with microsatellite marker development in molluscs. Heredity 106: 78-87. Meyer, E., and D. T. Manahan, 2010 Gene expression profiling of genetically determined growth variation in bivalve larvae (Crassostrea gigas). J. Exp. Biol. 213: 749- 758. Miron, G., B. Boudreau, and E. Bourget, 1999 Intertidal barnacle distribution: a case study using multiple working hypotheses. Mar. Ecol. Prog. Ser. 189: 205–219. Newell, R. I. E., G. S. Alspach, V. S. Kennedy and D. Jacobs, 2000 Mortality of newly metamorphosed eastern oysters (Crassostrea virginica) in mesohaline Chesapeake Bay. Mar. Biol. 136: 665-676. Packer, J. F., 1980 Prediction of Pacific oyster spatfall intensity in Dabob Bay. Wash. St. Dept. Fish., Prog. Rep., 126: 1-80. Packer, J. F. and S. B. Mathews, 1980 Vertical movement, behavior, growth, and prediction of spatfall for Pacific oyster larvae in Dabob Bay. Wash. St. Dept. Fish., Prog. Rep., 97. 209 Pace, D. A., A. G. Marsh, P. K. Leong, A. J. Green, D. Hedgecock, and D. T. Manahan, 2006 Physiological bases of genetically determined variation in growth of marine invertebrate larvae: A study of growth heterosis in the bivalve Crassostrea gigas. J. Exp. Mar. Biol. Ecol. 335: 188-209. Panagiotaki, P, and A. J. Geffen, 1992 Parental effects on size variation in fish larvae. J. Fish Biol. 41(B):37-42. Piepho, H-P., 2001 A quick method for computing approximate thresholds for quantitative trait loci detection. Genetics 157: 425-432. Pompanon, F., A. Bonin, E. Bellemain, P. Taberlet, 2005 Genotyping errors: causes, consequences and solutions. Nat. Rev. Genet. 6(11): 84. Pritchard, V. L., L. Dimond, J. S. Harrison, C. C. S. Velasquez, J. T. Zieba, R. S. Burton, and S. Edmands, 2011 Interpopulation hybridization results in widespread viability selection across the genome in Tigriopus californicus. BMC Genetics 12: 54. Probst, W. N., G. Krausa, R. M. Rideout, and E. A. Trippel, 2006 Parental effects on early life history traits of haddock Melanogrammus aeglefinus. ICES J. Mar. Sci. 63: 224-234. Reiseberg, L. H., C. R. Linder, and G. J. Seiler, 1995 Chromosomal and genic barriers to introgression in Helianthus. Genetics 141: 1163-1171. Reiseberg, L. H., 1998 Genetic mapping as a tool for studying speciation, pp. 459-487 in Molecular Systematics of Plants, edited by P.S. Soltis and J. J. Doyle. Chapman & Hall, New York. Rickman, S., Dulvy, N., Jennings, S., and Reynolds, J., 2000 Recruitment variation related to fecundity in marine fishes. Can. Jour. Fish. Aquat. Sci. 57: 116–124. Rideout, R. M, E. A. Trippel, and M. K. Litvak, 2006 Paternal effects on haddock early life history traits. J. Fish Biol 64: 695-701. Roache, J. C., G. Glusman, A. F. A. Smit, C. D. Huff, R. Hubley, et al., 2010 Analysis of genetic inheritance in a family quartet by whole-genome sequencing. Science 328: 636-639. Rogers, S. M., N. Isabel, and L. Bernatchez, 2007 Linkage maps of the dwarf and normal lake whitefish (Coregonus cleaformis) species complex and their hybrids reveal the genetic architecture of population divergence. Genetics 175: 375-398. 210 Roughgarden, J., S. D. Gaines, and H. Possingham, 1988 Recruitment dynamics in complex life-cycles. Science, 241: 1460–1466. Seitz, R. D., R. N. Lipcius, A. H. Hines and D. B. Eggleston, 2001 Density-Dependent predation, habitat variation, and the persistence of marine bivalve prey. Ecology 82: 2435-2451. Sekino, M., M. Hamaguchi, F. Aranishi, and K. Okoshi, 2003 Development of novel microsatellite DNA markers from the Pacific oyster, Crassostrea gigas. Mar. Biotechnol. 5: 227-233. Shuelke, M., 2000 An economic method for the fluorescent labeling of PCR fragments. Nat. Biotechnol. 18: 233-234. Steinberg, E. K., K .R. Lindner, J Gallea, A Maxwell, J. Meng, et al. 2002 Rates and patterns of microsatellite mutations in Pink salmon. Mol. Biol. Evol 19(7): 1198- 1202. Strathmann, R. R., 1985 Feeding and nonfeeding larval development and life-history evolution in marine invertebrates. Ann. Rev. Ecol. Syst. 16: 339. Thorson, G., 1950 Reproductive and larval ecology of marine bottom invertebrates Biological Reviews 25(1):1-45. Van Ooijen, J.W. and R.E. Voorrips, 2001 JoinMap® 3.0, Software for the calculation of genetic linkage maps. Plant Research International, Wageningen, the Netherlands. Wada, K. T., 1975 Electrophoretic variants of leucine aminopeptidase of the Japanese pearl oyster Pinctada fucata (Gould). Bull. Natl. Pearl Res. Lab. Jpn. 19: 2152– 2156. Wade, M. J., 2001 Epistasis, complex traits, and mapping genes Genetica 112–113: 59– 69. Wang, Y., R. Ren, and Z. Yu, 2008 Bioinformatic mining of EST-SSR loci in the Pacific oyster, Crassostrea gigas. Animal Genetics 39: 287-289. Willett, C. S. and J. N. Berkowitz, 2007 Viability effects and not meiotic drive cause dramatic departures from Mendelian inheritance for malic enzyme in hybrids of Tigriopus californicus populations. J. Evol. Biol. 20: 1196-1205. 211 Wilkins, N. P., 1976 Genic variability in marine Bivalvia: implications and applications in molluscan mariculture. 10 th European Symposium on Marine Biology, Ostend Belgium Sept. 17-23, 1: 549-563. Williams, G. C., 1975 Sex and Evolution. Princeton University Press, Princeton, NJ. Winemiller, K. O., 1992 Life history strategies and the effectiveness of sexual selection. Oikos 62: 318-327. Winemiller, K. O., and K. A. Rose, 1992 Patterns of life-history diversification in North American fishes: Implications for population regulation. Can. J. Fish. Aquat. Sci. 49: 2196-2218. Winemiller, K. O., and K. A. Rose, 1993 Why do most fish produce so many tiny offspring? Amer. Nat. 142: 585-603. Woodruff R. C., H. Huai, J. N. Thompson Jr., 1996 Clusters of identical new mutations in the evolutionary landscape. Genetica 98:149-160. Yamtich J, M. L. Voigt, G. Li and D. Hedgecock, 2005 Eight microsatellite loci for the Pacific oyster Crassostrea gigas. Anim. Genet. 36: 524-526. Yin, T. M., S. P. DiFazio, L. E. Gunter, D. Riemenschneider, and G. A. Tuskan, 2004 Large-scale heterospecific segregation distortion in Populus revealed by a dense genetic map. Theor Appl Genet. 109: 451–463 Yu, H., and Q, Li, 2007 EST-SSR markers from the Pacific oyster, Crassostrea gigas. Molecular Ecology Notes, 7: 860-862. Zhu, C., C. Wang, and Y. M. Zhang, 2007 Modeling segregation distortion for viability selection 1. Reconstruction of linkage maps with distorted markers. Theor. Appl. Genet. 114: 295-305. 212 Bibliography Anderson, D., and D. Hedgecock, 2010 Inbreeding depression and growth heterosis in larvae of the purple sea urchin Stronglyocentrotus purpuratus (Stimpson). J. Exp. Mar. Biol. Ecol. 384(1-2): 68-75. Arbeitman, M. N., E. E. M. Furlong, F. Imam, E. Johnson, B. H. Null, et al., 2002 Gene expression during the life cycle of Drosophila melanogaster. Science 297(5590): 2270-2275. Armbruster, P. and D. H. Reed, 2005 Inbreeding depression in benign and stressful environments. Heredity 95: 235-242. Baker, P., 1997 Settlement site selection by oyster larvae, Crassostrea virginica: Evidence for geotaxis. J. Shellfish Res. 16: 125-128 Baker P., and R. Mann, 1998 Response of settling oyster larvae, Crassostrea virginica, to specific portions of the visible light spectrum. J. Shellfish Res. 17: 1081-1083. Balch, T., and R. E. Scheibling, 2000 Temporal and spatial variability in settlement and recruitment of echinoderms in kelp beds and barrens in Nova Scotia. Mar. Ecol. Prog. Ser. 205: 139–154. Bannister, M. H., 1965 Variation in the breeding system of Pinus radiata. In: The Genetics of Colonizing Species (pp.353-374), H. G. Baker and G. L. Stebbins, editors, New York, Acadmic, xv + 588 pp. Barber, H. N., 1965 Selection in natural populations. Heredity 20: 551–572. Beaumont, A. R., C. M. Beveridge, and M. D. Budd, 1983 Selection and heterozygosity within single families of the mussel, Mytilus edulis (L.) Mar. Biol. 4: 151-161. Beaumont, A. R., 1991 Genetic studies of laboratory reared mussels, Mytilus edulis: heterozygote deficiencies, heterozygosity and growth. Biol. J. Linn. Soc. 44: 273-285. Bechsgaard, J., T. Bataillon, and M. H. Schierup, 2004 Uneven segregation of sporophytic self-incompatibility alleles in Arabidopsis lyrata. J. Evol. Biol. 17: 554–561. Beckenbach, A. T., 1994 Mitochondrial haplotype frequencies in oysters: neutral alternatives to selection models. pp. 188–198 in Non-neutral evolution, B. Golding, ed., Chapman & Hall New York. 213 Benjamini, Y., and Hochberg, Y., 1995 Controlling the false discovery rate: a practical and powerful approach to multiple testing. J. Roy. Stat. Soc. Ser. B 57: 289–300. Bierne, N., S. Launey, Y. Naciri-Graven and F. Bonhomme, 1998 Early effect of inbreeding as revealed by microsatellite analyses on Ostrea edulis larvae. Genetics 148: 1893-1906. Bijlsma R., J. Bundgaard , and A. C. Boerema, 2000 Does inbreeding affect the extinction risk of small populations?: predictions from Drosophila. J. Evol. Biol. 13: 502–514. Bijlsma, R., J. Bundgaard, and W. F. Van Putten, 1999 Environmental dependence of inbreeding depression and purging in Drosophila melanogaster. J. Evol. Biol. 12: 1125-1137. Bishop, D. T., C. Cannings, M. Skolnick, and J. A. Williamson, 1983 The number of polymorphic DNA clones required to map the human genome. In: Statistical Analysis of DNA Sequence Data, Weir, B. S. (ed.). New York, N.Y., Marcel Dekter, pp. 181–200. Boakes, E. H., J. Wang, and W. Amos, 2007 An investigation of inbreeding depression and purging in captive pedigreed populations. Heredity 98:172-182. Bogdan, M., and R. W. Doerge, 2005 Biased estimators of genetic effects and heritability in interval mapping. Heredity 95:476-484. Bonar, D. B., 1976 Molluscan metamorphosis: A study in tissue transformation. Amer. Zool. 16: 573-591. Bos, O. G., I. E. Hendriks, M. Strasser, P. Dolmer, P. Kamermans, 2006 Estimation food limitation of bivalve larvae in coastal waters of northwestern Europe. J. Sea Res. 55: 191-206. Breese, W. P. and R. E. Malouf, 1975 Hatchery manual for the Pacific Oyster, Crassostrea gigas (Gould). Oregon Agricultural Experiment Station Special Report No, 443, and Oregon State University Sea Grant College Program Publ. No. ORESU-H-75-002, 22 pp. Broitman, B. R., C. A. Blachette, B. A. Menge, J. Lubchenko, C Krenz, M. Foley, P. T. Raimondi and D. Lohse, 2008 Spatial and temporal patterns of invertebrate recruitment along the west coast of the United States. Ecol. Mono. 78(3): 403– 421. Broman, K. W., H. Wu, S. Sen, and G. A. Churchill, 2003 R/qtl: QTL mapping in experimental crosses. Bioinformatics 19: 889-90. 214 Bucklin, K. A., 2003 Analysis of the Genetic Basis of Inbreeding Depression in the Pacific Oyster Crassostrea gigas. PhD Dissertation, University of California at Davis. Byers, D. L., and D. M. Waller, 1999 Do plant populations purge their genetic load? Effects of population size and mating history on inbreeding depression. Annu. Rev. Ecol. Syst. 30: 479-513. Caers, M., P. Coutteau, P. Lombeida, and P. Sorgeloos, 1998 The effect of lipid supplementation on growth and fatty acid composition of Tapes philippinarum spat. Aquaculture 162: 287-299. Caley, M., M. Carr, M. Hixon, T. Hughes, G. Jones and B. Menge, 1996 Recruitment and the local dynamics of open marine populations. Annual Review of Ecology and Systematics, 27: 477 –500. Carr, D. E, and M. R. Dudash, 1997 The effects of five generations of enforced selfing on potential male and female function in Mimulus guttatus. Evolution 51:1797- 1807. Carr, D.E., and M. R. Dudash, 2003 Recent approaches into the genetic basis of inbreeding depression in plants. Phil. Trans. Roy. Soc. L. B 358:1071-1084. Charlesworth, B. and D. Charlesworth, 1999 The genetic basis of inbreeding depression. Genet. Res. 74: 329-340. Charlesworth, D., and B. Charlesworth, 1987 Inbreeding depression and its evolutionary consequences. Annu. Rev. Ecol. Syst. 18: 237–268. Charlesworth, D., and J. H. Willis, 2009 The genetics of inbreeding depression. Nat. Rev. Genet. 10:783-796. Charlesworth, D., M. T. Morgan, and B. Charlesworth, 1990 Inbreeding depression, genetic load, and the evolution of out-crossing rates in a multi-locus system with no linkage. Evolution 44:1469-1489. Cheng, R., A. Saito and Y. Ukai, 1996 Estimation of the position and effect of a lethal factor locus on a molecular marker linkage map. Theor. Appl. Genet. 93: 494–502 Cheptou, P–O., E. Imbert, J. LePart, and J. Escarre, 2000 Effects of competition on lifetime estimates of inbreeding depression in the outcrossing plant Crepis santa (Asteraceae). J. Evol. Biol. 13: 522-531. Cheptou, P-O., and K. Donohue, 2011 Environment-dependent inbreeding depression: its ecological and evolutionary significance. New Phyt. 189(2):395-407. 215 Cohen, C. S., 1992 Population Biology of two species of Corella: Mating systems and demography (Ascidians). PhD dissertation, University of Washington, Dept. of Zoology. Cole, H. A., 1938 The fate of the larval organs in the metamorphosis of Ostrea edulis. J. Mar. Bio. Assn., U.K. 23: 469-484. Connell J. H., 1985 The consequences of variation in initial settlement vs. post-settlement mortality in rocky intertidal communities. J. Exp. Mar. Biol. Ecol. 93: 11–45. Coon, S. L., D. B. Bonar, and R. M.Weiner, 1986 Chemical production of cultchless oyster spat using epinephrine and norepinephrine. Aquaculture 58: 255-262. Coutteau, P, M. Caers, A. Mallet, W. Moore, J. J. Manzi, and P. Soorgeloos, 1994 Effect of lipid supplementation on growth, survival, and fatty acid composition of bivalve larvae. In: Measures for Success. Eds. Kestemont, P. Muir, J., and Williot, P. Proceedings of Bordeaux Aquaculture 1994, Mar. 23-25, Bordeaux, France. Cowan, J. H. Jr., K. A. Rose, and E. D. Houde, 1997 Size based foraging success and models of larval fish populations. Pages 357-386 in, Early life history and recruitment in fish populations, R. C. Chambers and E. A. Trippel, editors Chapman and Hall, London, UK. Cowan, J. H., Jr,, E. D. Houde, and K. A. Rose, 1996 Size dependent vulnerability of marine fish larvae to predation: an individual-based numerical experiment. ICES J. Mar. Sci. 53: 23-37. Cowen, R. K., G. Gawarkiewicz, J. Pineda, S. R. Thorrold, F. E. Werner, 2007 Population connectivity in marine systems: An Overview. Oceanography 20(3):14-21. Crisp, D. J., 1974 Factors influencing the settlement of marine invertebrate larvae. In, Chemoreception in marine organisms, edited by P. T. Grant & A. M. Mackie. Academic Press, New York, pp. 177-265. Crisp, D. J., 1984 Overview of research on marine invertebrate larvae, 1940-1980. In , Marine Biodeterioration: An interdisiplinary study, edited by J. D. Costlow & R.C. Tipper. Naval Institiute Press, Annapolis, Maryland. pp 103-126. Crnokrak, P., and D. A. Roff, 1999 Inbreeding depression in the wild. Heredity 83: 260- 270. Crnokrak, P., and S. C. H. Barrett, 2002 Purging the genetic load: A review of the experimental evidence. Evolution 56(12): 2347–2358 216 Crowder, L. W., J. A. Rice, T. J. Miller, and E. A. Marschall, 1992 Empirical and theoretical approaches to size-based interactions and recruitment variability in fishes. Pages 237-255 in, Individual-based models and approaches in ecology, D. L. DeAngelis and L. J. Gross, editors. Chapman and Hall, New York, New York, USA. Curole, J. P., and D. Hedgecock, 2007 Chapter 29. Bivalve genomics: Complications, challenges, and future perspectives. In: Aquaculture Genome Technologies, Liu, Z. (editor), Blackwell Publishing, Ames, Iowa, 525-543. Cushing, D. H., 1971 The dependence of recruitment on parent stock in different groups of fishes. J. du Conseil Intern. Expl. Mer. 33: 340-362. Dahlgaard J. and A. A. Hoffmann, 2000 Stress resistance and environmental dependency of inbreeding depression in Drosophila melanogaster. Conserv. Biol 14: 1187– 1192. Darwin, C. R., 1876 The effects of cross and self-fertilization in the vegetable kingdom. John Murray, London. Degnan B. M., and D. E. Morse, 1995 Developmental and morphogenetic gene regulation in Haliotis rufescens larvae at metamorphosis Amer Zool 35: 391-398. DeWoody, J. A., J. Schupp, Kenefic, L., J. Busch, L. Murfitt, et al. 2004 Universal method for producing ROX-labeled size standards suitable for automated genotyping. BioTechniques 37: 348-352. Dobzhansky, Th. 1946 Genetics of natural populations .XIII. Recombination and variability in populations of Drosophila pseudoobscura. Genetics 31: 269-290. Eri, R., J. M. Arnold, V. F. Hinman, K. M. Green, M. K. Jones, B. M. Degnan, M. F. Lavin, 1999 Hemps, a novel EGF-like protein, plays a central role in ascidian metamorphosis. Development 126: 5809-5818. Ernande, B., J. Clobert, H. Mccombie, and P. Boudry, 2003 Genetic polymorphisms and trade-offs in the early life-history strategy of the Pacific oyster, Crassostrea gigas (Thunberg, 1795): A quantitative genetic study. J. Evol. Biol. 16: 399-414. Escobar, J. S., A. Nicot, and P. David, 2008 The different sources of variation in inbreeding depression, heterosis and outbreeding depression in a metapopulation of Physa acuta. Genetics 180: 1593-1608. Eversley, C. D., T. Clark, Y. Xie, J. Steigerwalt, T. A. Bell et al., 2010 Genetic mapping and developmental timing of transmission ration distortion in a mouse interspecific backcross. BMC Genet. 11: 98. 217 Fabioux, C., A. Huvet, C. Lelong, R. Robert, S. Pouvreau, J.Y. Daniel et al., 2004 Oyster vasa-like gene as a marker of the germline cell development in Crassostrea gigas. Biochem. Biophys. Res. Comm. 320: 592-598. Falconer D. S. and T. F. C. Mackay, 1996 Introduction to Quantitative Genetics. 4th edn. Longman: Harlow. Fenaux, L., M. F., Strathmann, and R. R. Strathmann, 1994 Five tests of food-limited growth of larvae in coastal waters by comparisons of rates of development and form of echinoplutei. Limnol. Oceanogr. 39: 84–98. Fishman L, A. J. Kelly, E. Morgan, and J. H. Willis, 2001 A genetic map in the Mimulus guttatus species complex reveals transmission ratio distortion due to heterospecific interactions. Genetics 159: 1701-1716. Fogarty, M. J., M. P. Sissenwine, and E. B. Cohen, 1991 Recruitment variability and the dynamics of exploited marine populations. Trends Ecol. Evol. 6:241–246. Foltz, D. W., 1986a Null alleles as a possible cause of heterozygote deficiencies in the oyster Crassostrea virginica and other bivalves. Evolution 40: 869-870. Foltz, D. W., 1986b Segregation and linkage studies of allozyme loci in Pair crosses of the oyster Crassostrea virginica. Biochem. Genet. 24: 941-956. Fowler K., and M. C. Whitlock, 2002 Environmental stress, inbreeding, and the nature of phenotpyic and genetic variance in Drosophila melanogaster. Proc. Roy. Soc. Lond. B 269: 677–683. Fox C. W., and D. H. Reed, 2011 Inbreeding depression increases with environmental stress: an experimental study and meta-analysis. Evolution 65(1): 246-258. Frankel, O. H., and M. E. Soulé, 1981 Conservation and Evolution. Cambridge University Press, Cambridge. Frankham, R., 1995 Conservation genetics. Ann. Rev. Genetics 29: 305-327. Franklin, E. C., 1972 Genetic load in the loblolly pine. Amer. Naturalist 106: 262-265. Fu, Y. B, and K. Ritland, 1996 Marker-based inferences about epistasis for genes influencing inbreeding depression. Genetics 144: 339-348. Fu, Y. B. and K. Ritland, 1994 On estimating linkage of marker genes to viability genes controlling inbreeding depression. Theoretical and Applied Genetics 88: 925-93. Fuiman, L. and J. H. Cowan, Jr., 2003 Behavior and recruitment success in fish larvae: repeatability and covariation of survival skills. Ecology 84(1): 53-67. 218 Fuiman, L. and J. H. Cowan, Jr., 2003 Behavior and recruitment success in fish larvae: repeatability and covariation of survival skills. Ecology 84(1): 53-67. Gaffney P. M, J. C. Pierce, A. G. MacKinley, D. A. Titchen, W. K. Glenn, 2003 Pearl, a novel family of putative transposable elements in bivalve molluscs. J. Mol. Evol. 56: 308–316. Gaines, S., and J. Roughgarden, 1985 Larval settlement rate: a leading determinant of structure in an ecological community of the marine intertidal zone. Proc. Natl. Acad. Sci. U.S.A. 82: 3707-3711. Garcia-dorado, A., and A. Caballero, 2000 On the average coefficient of dominance of deleterious spontaneous mutations. Genetics 155: 1991-2001. Giagrande, A., S. Geraci, and G. Belmonte, 1994 Life-cycle and life-history diversity in marine invertebrates and the the implications in community dynaimcs. Oceanography and Marine Biology, an Annual review 32: 305-333. Göring, H. H. H., J. D. Terwilliger, and J. Blangero, 2001 Large upward bias in estimation of locus-specific effects from genomewide scans. Am. J. Hum. Genet. 69: 1357-1369. Gosselin, L. A., and P. Y. Qian, 1996 Early post-settlement mortality of an intertidal barnacle: A critical period for survival. Mar. Ecol. Prog. Ser. 135: 69-75. Gosselin, L. A., and P. Y. Qian, 1997 Juvenile mortality in benthic marine invertebrates. Mar. Ecol. Prog. Ser. 146: 265-282. Guo, X., and S. K. Allen, Jr., 1994 Reproductive potential and genetics of triploid Pacific oysters, Crassostrea gigas (Thunberg). Biol Bull. 187: 309-318. Hadfield, M. G., 1977 Chemical interactions in larval settling of a marine gastropod. In: Faulkner, D. J., Fenical, W. H. (eds.) Marine natural products chemistry. Plenum Publishing Corporation, New York, pp. 403-413. Hadfield, M. G., 1984 Settlement requirements of molluscan larvae: New data on chemical and genetic roles. Aquaculture 39: 283-298. Hadfield, M. G., 1986 Settlement and recruitment of marine invertebrates: A perspective and some proposals. Bull. Mar. Sci. 39: 418-425. Hall, M. C., and J. H. Willis, 2005 Transmission ratio distortion in intraspecific hybrids of Mimulus guttatus. Genetics 170: 375-386. 219 Hare, J. A., and R. K. Cowen, 1997 Size, growth, development, and survival of the planktonic larvae of Pomatomus saltatrix (Pisces: Pomatomidae). Ecology 78: 2415- 2431. Harrison J. S. and S. Edmands, 2006 Chromosomal basis of viability differences in Tigriopus californicus interpopulation hybrids. J. Evol. Biol. 19: 2949-2051. Harushima, Y., M. Nakagahra, M. Yano, T. Sasaki, and K . Kurata 2001 A genome-wide survey of reproductive barriers in an intraspecific hybrid. Genetics 159: 883-892. Haws, M. C., L. DiMichele and S. C. Hand, 1993 Biochemical changes and mortality during metamorphosis of the Eastern oyster, Crassostrea virginica, and the Pacific oyster, Crassostrea gigas. Mol. Mar. Biol. Biotechnol. 2: 207-217. Hayes, C. N., J. A. Winsor, and A. G. Stephenson, 2005 Environmental variation influences the magnitude of inbreeding depression in Cucurbita pepo spp. texana (Cucurcbitaceae). J. Evol. Biol. 18: 147-155. Heath, M., and A. Gallego, 1997 From the biology of the individual to the dynamics of the population: bridging the gap in fish early life studies. J. Fish Biol. 51(Supplement A): 1-29. Hedgecock, D., 1994 Does variance in reproductive success limit effective population sizes of marine organisms? In: A. R. Beaumont,editor. Genetics and evolution of aquatic organisms. London, Chapman Hall. pp. 122–134. Hedgecock, D., and J. P. Davis, 2007 Heterosis for yield and crossbreeding of the Pacific oyster Crassostrea gigas. Aquaculture 272S: S17-S29. Hedgecock, D., D. J. McGoldrick, and B. L. Bayne, 1995 Hybrid vigor in Pacific oysters: an experimental approach using crosses among inbred lines. Aquaculture 137(1- 4): 285-298. Hedgecock, D., D. J. McGoldrick, D. T. Manahan, J. Vavra, N. Appelmans and B. L. Bayne, 1996 Quantitative and molecular genetic analyses of heterosis in bivalve molluscs. J. Exp. Mar. Biol. Ecol. 203: 49-59. Hedgecock, D., G. Li, S. Hubert, K. Bucklin, and V. Ribes, 2004 Widespread null alleles and poor cross-species amplification of microsatellite DNA loci cloned from the Pacific oyster, Crassostrea gigas. J. Shellfish Res. 23: 379-385. Hedgecock, D., P. M. Gaffney, P. Goulletquer, X. Guo, K. Reece, G. Warr, 2005 The case for sequencing the Pacific oyster genome. J. Shellfish Res. 24(2): 429–441. Hedrick P. W., 1994 Purging inbreeding depression and the probability of extinction: full-sib mating. Heredity 73: 363–372. 220 Hedrick P. W., and S. T. Kalinowski, 2000 Inbreeding depression in conservation biology. Annu. Rev. Ecol. Evol. Syst. 31:139-162. Hedrick, P. W., and O. Muona, 1990 Linkage of viability genes to marker loci in selfing organisms. Heredity 64: 67-72. Hedrick, P. W., O. Savolainen, and K. Karkkainen, 1999 Factors influenceing the extent of inbreeding depression: an example from Scots pine. Heredity 82: 441-450. Helm, M. M., D. L. Holland, S. D. Utting, and J. East, 1991 Fatty acid composition of early non-feeding larvae of the European flat oyster, Ostrea edulis. J. Mar. Biol. Ass. U.K. 71: 691-705. Heylund, A., and L. L. Moroz, 2006 Signaling mechanisms underlying metamorphic transitions in animals. Integr. Comp. Biol. 46: 743-759. Hickman, R. W., and L. L. D. Gruffydd, 1971 The histology of the larva of Ostrea edulis during metamorphosis. In: E. J., Crisp (ed.) Fourth European marine biology symposium pp. 281 -294. University press, Cambridge. Hilbish, T. J., E. P. Winn, and P. D. Rawson, 1993 Genetic variation and covariation during larval and juvenile growth in Mercenaria mercenaria. Marine Biology 115: 94-107.Hu, Z., and S. Xu, 2009 PROC QTL—A SAS procedure for mapping quantitative trait loci. Int. J. Plant Genomics 2009: 141234. Hjort, J., 1914 Fluctuations in the great fisheries of Northern Europe. Rapports et Proce`s-Verbaux des Re´unions du Conseil International pour l‘Exploration de la Mer, 20:1–228. Houde, E. D., 2002 Mortality. Pages 64-87 in Fishery science: the unique contributions of early life stages, L. A. Fuiman and R. G. Werner, editors. Blackwell Science, Oxford, UK. Houde, E. D., 1996 Evaluating stage-specific survival during the early life of fish. Pages 51-66 in, Survival strategies in early life stages of marine resources, Y. Watanabe, Y. Ya-mashita, and Y. Oozeki, editors. A. A. Balkema, Rotterdam, The Netherlands. Hu, Y. P. and D. W. Foltz, 1996 Genetics of scnDNA polymorphisms in juvenile oysters Crassostrea virginica Part 1: Characterizing the inheritance of polymorphisms in controlled crosses. Mol. Mar. Biol. Biotechnol. 5(2): 123-129. Hu, Z., and S. Xu, 2009 PROC QTL—A SAS procedure for mapping quantitative trait loci. Int. J. Plant Genomics 2009: 1-3. doi:10.1155/2009/141234. 221 Hubert, S., and D. Hedgecock, 2004 Linkage maps of microsatellite DNA markers for the Pacific oyster Crassostrea gigas. Genetics 168: 351-362. Hubert, S., E. Cognard, and D. Hedgecock, 2009 Centromere-mapping in triploid families of the Pacific oyster Crassostrea gigas (Thunberg). Aquaculture 288: 172–183. Hunt, H. L., and R. E. Scheibling, 1997 Role of early post settlement mortality in recruitment of benthic marine invertebrates. Mar. Ecol.-Prog. Ser. 155: 269-301. Husband, B. C., and D. W. Schemske, 1996 Evolution of the magnitude and timing of inbreeding depression in plants. Evolution 50 (1): 54-70. Huvet, A., P. Boudry, M. Ohresser, C. Delsert, and F. Bonhomme, 2000 Variable microsatellites in the Pacific oyster Crassostrea gigas and other cupped oyster species. Anim. Genet. 31: 71-72. Innes, D. J., and L. E. Haley, 1977 Genetic aspects of larval growth under reduced salinity in Mytilus edulis. Biol. Bull. 153: 312-321. Jackson, D. J., B. M. Degnan, 2006 Expressed sequence tage analysis of genes expressed during development of the tropical abalone Haliotis asinine. J. Shell. Res. 25(1): 225-231. Jackson, D. J., N. E. Ellemor, and B. M. Degnan, 2005 Correlating gene expression with larval competence, and the effect of age and parentage on metamorphosis in the tropical abalone Haliotis asinine. Mar. Biol. 147: 681-697. Jackson, D., S. P. Leys, V. F. Hinman, R. Woods, M. F. Lavin, and B. M. Degnan, 2002 Ecological regulation of development: induction of marine invertebrate metamorphosis. Int. J. Dev. Biol. 46: 679-686. Jackson, D. J., G. Worheide, B. M. Degnan, 2007 Dynamic expression of ancient and novel molluscan shell genes during ecological transitions. BMC Evol. Biol. 7(160): 1-17. Jacobs, M. W., S. M. Degnan, R. Woods, E. Williams, K. Roper, K. Green., and B. M. Degnan, 2006 The effect of larval age on morphology and gene expression during ascidian metamorphosis. Int. Comp. Biol. 46(6):760-776 Janczewski E, M. Gherardi, I. Bonnin, J. M. Prosperi, I, Olivierei, T. Huguet, 1997 Insight on segregation distortions in two intraspecific crosses between annual species of Medicago (Leguminosae). Theor. Appl. Genet. 94: 682-691. Jarrett, J. N., and Pechenik J. A., 1997 Temporal variation in cyprid quality and juvenile growth capacity for an intertidal barnacle. Ecology 78: 1262–1265 222 Jimenez, J. A., K. A. Hughes, G. Alaks, L. Graham, & R. C. Lacy, 1994 An experimental study of inbreeding depression in a natural habitat. Science 266: 271–273. Jones, G., and B. Jones, 1983 Methods for setting hatchery produced oyster larvae. Marine Resources Branch, Ministry of Environment, BC, Canada. Information report No. 4., 94pp. Jones, A. G., G. Rosenqvist, A. Berglund, and J. C. Avise, 1999 Clustered microsatellite mutations in the pipefish Syngnathus typhle. Genetics 152: 1057-1063. Kacser, H., and J. A. Burns, 1981 The molecular basis of dominance. Genetics 97(3-4): 639-666. Kawashima, T., Y. Satou, S. D. Murakami, and N. Satoh, 2005 Dynamic changes in developmental gene expression in the basal chordate Ciona intestinalis. Develop. Growth Differ. 47: 187-99. Keller, L. F., P. R. Grant, B. R. Grant, and K. Petren, 2002 Environmental conditions affect the magnitude of inbreeding depression in survival of Darwin‘s finches. Evolution 56(2): 1229-1239. Keller, L. F. and D. M. Waller, 2002 Inbreeding effects in wild populations. Trends Ecol. Evol. 17: 230–241. Kelly, J. K., 2005 Epistasis in monkeyflowers. Genetics 171: 1917-1931. Kennedy, V. S., R. I. E. Newell, and A. F. Eble, 1996 The Eastern Oyster: Crassostrea virginica. Maryland Sea Grant press. Keough, M. J. and B. J. Downes, 1982 Recruitment of marine invertebrates: the role of active larval choices and early mortality. Oecologia (Berl) 54:348-352. Koehn, R. K, and P. M. Gaffney, 1984 Genetic heterozygosity and growth rate in the Mytilus edulis. Mar. Biol. 82: 1-7. Koelewijn, H. P., V. Koski, and O. Savolainen, 1999 Magnitude and timing of inbreeding depression in Scots pine (Pinus sylvestris l.). Evolution 53: 758-768. Korringa, P., 1941 Experiments and observations on swarming, pelagic life and setting in the European flat oyster, Ostrea edulis L. Arch. neerl. Zool. 5: 1-249 [Thesis, Univ. Amsterdam, 1940]. Korringa, P., 1946 A revival of natural oyster beds? Nature 158: 586-587. 223 Kristensen, T. N., J. Dahlgaard, and V. Loeschcke, 2003 Effects of inbreeding and environmental stress on fitness – using Drosophila buzzatii as a model organism. Cons. Genet. 4: 453-465. Kristensen, T. N., P. Sørensen, K. S. Pedersen, M. Kruhøffer, and V. Loeschcke, 2006 Inbreeding by environmental interactions affect gene expression in Drosophila melanogaster. Genetics 173(3): 1329-1336. Lander, E. S., and D. Botstein, 1989 Mapping Mendelian factors underlying quantitative traits using RFLP linkage maps. Genetics 121: 185-199. Langdon, C. J. and M. J. Waldcock, 1981 The effect of algal and artificial diets on the growth and fatty acid composition of Crassostrea gigas spat. J. Mar. Biol. Ass. U.K. 61: 431-448. Langdon, C., F. Evans, D. Jacobson and M. Blouin, 2003 Yields of cultured Pacific oysters Crassostrea gigas Thunberg improved after one generation of selection. Aquaculture 220(1-4): 227-244. Lannan, J.E., 1980a Broodstock management of Crassostrea gigas I.Genetic variation in in the larval rearing system. Aquaculture, 21: 323-336. Lannan, J. E., 1980b Broodstock management of Crassostrea gigas: III. Selective breeding for improved larval survival. Aquaculture 21: 347-351. Launey, S., and D. Hedgecock, 2001 High genetic load in the Pacific oyster Crassostrea gigas. Genetics 159: 255-265. Li, G. & D. Hedgecock, 1998 Genetic heterogeneity, detected by PCRSSCP, among samples of larval Pacific oysters (Crassostrea gigas) supports the hypothesis of large variance in reproductive success. Can. J. Fish. Aquat. Sci. 55:1025–1033. Li, G., S. Hubert, K. Bucklin, V. Ribes, and D. Hedgecock, 2003 Characterization of 79 microsatellite DNA markers in the Pacific oyster Crassostrea gigas. Mol. Ecol. Notes 3: 228-232. Li, Z.-K., L. J. Luo, H. W. Mei, D. L. Wang, Q. Y. Shu, et al. 2001 Overdominant epistatic loci are the primary genetic basis of inbreeding depression and heterosis in rice. I. Biomass and grain yield. Genetics 158: 1737–1753. Llodra, E. R., 2002 Fecundity and life-history strategies in marine invertebrates. Adv. Mar. Bio. 43: 87-170. Lorieux M., X. Perrier, B. Goffinet, C. Lanaud, and D. González de León, 1995 Maximum-likelihood models for mapping genetic markers showing segregation distortion. 2. F 2 populations. Theor. Appl. Genet. 90: 81–89. 224 Losee, E., 1979 Relationship between larval and spat growth rates in the oyster Crassostrea virgininca. Aquaculture 16: 123-126. Luo, L, Y. M. Zhang and S. Xu, 2005 A quantitative genetics model for viability selection. Heredity 94: 347-355. Luo, L. and S. Xu, 2003 Mapping viability loci using molecular markers. Heredity 90: 459-467. Lynch, M. and B. Walsh, 1998 Genetics and analysis of quantitative traits. Sinauer Associates, Sunderland Massachusetts. Lyttle, T. W., 1991 Segregation distorters. Ann. Rev. Genet. 25: 511–557. MacArthur, R. H., and E. O. Wilson, 1967 ―Theory of Island Biogeography‖. Princeton University Press, Princeton. Magoulas, A., B. Gjetvag, V. Tersoglou and E. Zouros, 1998 Three polymorphic microsatellites in the Japanese oyster, Crassostrea gigas (Thunberg). Anim. Genet. 29:69-70. Maldonado M., and C. M. Young, 1999 Effects of the duration of larval life on postlarval stages of the demosponge Sigmadocia caerulea. J. Exp. Mar. Biol. Ecol. 232: 9– 21. Mallet, A. L., E. Zouros K. E. Gartner-Keykay, K. R. Freeman, and L. M. Dickie, 1985 Larval viability and heterozygote deficiency in populations of marine bivalves: evidence from pair matings of mussels. Mar. Biol. 87: 165-172. McCune, A. R., R. C. Fuller, A. A. Aquilina, R. M. Dawley, J. M. Fadool et al., 2002 A low genomic number of recessive lethals in natural populations of bluefin killifish and zebrafish. Science 296: 2398-2401. McGoldrick D. J., D. Hedgecock, L. J. English, P. Baoprasertkul, and R. D. Ward, 2000 The transmission of microsatellite alleles in Australian and North American stocks of the Pacific oyster (Crassostrea gigas): selection and null alleles. J. Shellfish Res. 19: 779–788. McGoldrick, D.J., Hedgecock, D., 1997 Fixation, segregation and linkage of allozyme loci in inbred families of the Pacific oyster Crassostrea gigas (Thunberg): implications for the causes of inbreeding depression. Genetics 146: 321-334. McInerney, C. E., A. L. Allcock, M. P. Johnson, D. A. Bailie, and P. A. Prodöhl, 2011 Comparative genomic analysis reveals species-dependent complexities that explain difficulties with microsatellite marker development in molluscs. Heredity 106: 78-87. 225 Meinke, D. W., 1995 Molecular genetics of plant embryogenesis. Annu. Rev. Plant Physiol. Plant Mol. Biol. 46: 369-394. Melchinger, A. E., H. P. Piepho, H. F. Utz, J. Muminovic, T. Wegenast, et al. 2007 Genetic basis of heterosis for growth-related traits in Arabidopsis investigated by testcross progenies of near-isogenic lines reveals a significant role of epistasis. Genetics 177: 1827-1837. Meyer, E., and D. T. Manahan, 2010 Gene expression profiling of genetically determined growth variation in bivalve larvae (Crassostrea gigas). J. Exp. Biol. 213: 749- 758. Miron, G., B. Boudreau, and E. Bourget. 1999 Intertidal barnacle distribution: a case study using multiple working hypotheses. Mar. Ecol. Prog. Ser. 189: 205–219. Mitchell-Olds, T., 1995 Interval mapping of viability loci causing heterosis in Arabidopsis. Genetics 140: 1105–1109. Morton, N. E., J. F. Crow, and H. J. Muller, 1956 An estimate of the mutational damage in man from data on consanguineous marriages. Proc. Natl. Acad. Sci. U.S.A. 42: 855. Mukai, T., S. I. Chigusa, L. E. Mettler, and J. F. Crow, 1972 Mutation rate and dominance of genes affecting viability in Drosophila melanogaster. Genetics 72:333-355. Nascimento, I. A., 1980 Growth of the larvae of Crassostrea gigas Thunberg, fed with different algal species at high cell concentration. J. Cons. Int. Explor. Mer. 39(2): 134-139. Newell, R. I. E., G. S. Alspach, Jr., V. S. Kennedy, and D. Jacobs, 2000 Mortality of newly metamorphosed eastern oysters (Crassostrea virginica) in mesohaline Chesapeake Bay. Mar. Biol. 136: 665-676. Omholt, S. W., E. Plahte, L. Øyehaug, K. Xiang, 2000 Gene Regulatory Networks Generating the Phenomena of Additivity, Dominance and Epistasis. Genetics 155: 969-980. Pace, D. A., A. G. Marsh, P. K. Leong, A. J. Green, D. Hedgecock, and D. T. Manahan, 2006 Physiological bases of genetically determined variation in growth of marine invertebrate larvae: A study of growth heterosis in the bivalve Crassostrea gigas. J. Exp. Mar. Biol. Ecol. 335: 188-209. Packer, J. F. and S. B. Mathews, 1980 Vertical movement, behavior, growth, and prediction of spatfall for Pacific oyster larvae in Dabob Bay. Wash. St. Dept. Fish., Prog. Rep., 97. 226 Packer, J. F., 1980 Prediction of Pacific oyster spatfall intensity in Dabob Bay. Wash. St. Dept. Fish., Prog. Rep., 126: 1-80. Palumbi, S., 1994 Genetic divergence, reproductive isolation, and marine speciation. Ann. Rev. Ecol. Syst. 25: 547 –572. Palumbi, S., 1995 Using genetics as an indirect estimator of larval dispersal. Pages 369 – 387, in Ecology of marine invertebrate larvae, L. McEdward, ed., CRC Press, Boca Raton, Florida. Pampanon, F., A. Bonin, E. Bellemain, P. Taberlet, 2005 Genotyping errors: causes, consequences and solutions. Nat. Rev. Genet. 6(11): 84. Panagiotaki, P, and A. J. Geffen, 1992 Parental effects on size variation in fish larvae. J. Fish Biol. 41(B):37-42. Pechenik, J. A., 1990 Delayed metamorphosis by larvae of benthic marine invertebrates: Does it occur? Is there a price to pay? Ophelia 32: 63-94. Pechenik, J. A., and W. D. Heyman, 1987 Using KCl to determine size at competence for larvae of the marine gastropod Crepidula fornicata (L.), J. Exp. Mar. Biol. Ecol. 112: 27–38 Pernet, F., an R. Tremblay, 2004 Effect of varying levels of dietary essential fatty acid during early onteogeny of the sea scallop Placopecten magellanicus. J. Exp. Mar. Bio. Ecol. 310(1): 73-86. Peterman, R. M, and M. J. Bradford, 1987 Wind-speed and mortality-rate of a marine fish, the northern Anchovy (Engraulis-Mordax). Science 23 (4786): 354-356. Phadnis, N., and J. D. Fry, 2005 Widespread correlations between dominance and homozygous effects of mutations: Implications for theories of dominance. Genetics 171: 385-392. Pianka, E. R., 1970 On ―r‖ and ―k‖ selection . Amer. Nat. 104: 592-597. Piepho, H.-P., 2001 A quick method for computing approximate thresholds for quantitative trait loci detection. Genetics 157: 425-432. Pineda, J., 1994 Spatial and temporal patterns in barnacle settlement rate along a southern California rocky shore. Mar. Ecol. Prog. Ser. 107: 125-138. Pineda, J., V. Starczak, and T. A. Stueckle, 2006 Timing of successful settlement: demonstration of a recruitment window in the barnacle Semibalanus balanoides Mar. Ecol. Prog. Ser. 320: 233–237. 227 Pogson, G. H., and E. Zouros, 1994 Allozyme and RFLP heterozygosities as correlates of growth rate in the scallop Placopecten magellanicus: a test of the associative overdominance hypothesis. Genetics 137: 221-231 Pritchard, V. L., L. Dimond, J. S. Harrison, C. C. S. Velasquez, J. T. Zieba, R. S. Burton, and S. Edmands, 2011 Interpopulation hybridization results in widespread viability selection across the genome in Tigriopus californicus. BMC Genetics 12: 54. Probst, W. N., G. Krausa, R. M. Rideout, and E. A. Trippel, 2006 Parental effects on early life history traits of haddock Melanogrammus aeglefinus. ICES J. Mar. Sci. 63: 224-234. Qian P. Y., and J. A. Pechenik, 1998 Effects of larval starvation and delayed metamorphosis on juvenile survival and growth of the tube-dwelling polychaete Hydroides elegans (Haswell). J. Exp. Mar. Biol. Ecol. 227: 169–185. Raimondi, P. T., 1990 Patterns, mechanisms, and consequences of variability in settlement and recruitment in an intertidal barnacle. Ecological Monographs 60:283-309. Rankin T. L., Sponaugle S., 2011 Temperature influences selective mortality during the early life stages of a coral reef fish. PLoS ONE 6(5): e16814. Reiseberg, L. H., 1998 Genetic mapping as a tool for studying speciation , pp. 459-487 in Molecular Systematics of Plants, edited by P.S. Soltis and J. J. Doyle. Chapman & Hall, New York. Reiseberg, L. H., C. R. Linder, and G. J. Seiler, 1995 Chromosomal and genic barriers to introgression in Helianthus. Genetics 141: 1163-1171. Remington D. L., and D. M. O‘Malley, 2000 Whole genome characterization of embryonic stage inbreeding depression in a selfed loblolly pine family. Genetics 155: 337-348. Rice, W. R., 1989 Analyzing tables of statistical tests. Evolution 43: 223–225. Rickman, S., N. Dulvy, S. Jennings, and J. Reynolds, 2000 Recruitment variation related to fecundity in marine fishes. Can. Jour. Fish. Aquat. Sci. 57: 116–124. Rideout, R. M, E. A. Trippel, and M. K. Litvak, 2006 Paternal effects on haddock early life history traits. J. Fish Biol 64: 695-701. Ritland, K., 1996 Inferring the genetic basis of inbreeding depression in plants. Genome 39: 1-8. 228 Rizski, M. T. M., 1952 Ontogenetic distribution of genetic lethality in Drosophila willistoni. J. Exp. Zool. 121: 327-350. Roache, J. C., G. Glusman, A. F. A. Smit, C. D. Huff, R. Hubley, et al., 2010 Analysis of genetic inheritance in a family quartet by whole-genome sequencing. Science 328: 636-639. Rodriguez, S. R., F. Patricio-Ojedal, and N. C. Inestrosa, 1993 Settlement of benthic marine invertebrates Mar. Ecol. Prog. Ser. 97: 193-207. Roff, D. A., 2002 Inbreeding depression: Tests of the overdominance and partial dominance hypotheses. Evolution 56: 768-775. Rogers, S. M., N. Isabel, and L. Bernatchez, 2007 Linkage maps of the dwarf and normal lake whitefish (Coregonus cleaformis) species complex and their hybrids reveal the genetic architecture of population divergence. Genetics 175: 375-398. Roughgarden, J., S. D. Gaines, and H. Possingham, 1988 Recruitment dynamics in complex life-cycles. Science, 241: 1460–1466. Salathe, P., and D. Ebert, 2003 The effects of parasitism and inbreeding on the competitive ability in Daphnia magna: evidence for synergistic epistasis. J. Evol. Biol. 16: 976-985. SAS Institute Inc., SAS 9.1.3 Help and Documentation, Cary, NC: SAS Institute Inc., 2000-2004. Savolainen, O., K. Karkkainen, and H. Kuittinen, 1992 Estimating numbers of embryonic lethals in conifers. Heredity 69: 308-314. Seitz, R. D., R. N. Lipcius, A. H. Hines and D. B. Eggleston, 2001 Density-Dependent predation, habitat variation, and the persistence of marine bivalve prey. Ecology 82: 2435-2451. Sekino, M., M. Hamaguchi, F. Aranishi, and K. Okoshi, 2003 Development of novel microsatellite DNA markers from the Pacific oyster, Crassostrea gigas. Mar. Biotechnol. 5: 227-233. Semlitsch, R. D., D. E. Scott, and J. H. K. Pechmann, 1988 Time and size at metamorphosis related to adult fitness in Ambystoma talpoideum. Ecology 69: 184–192. Seto, F., 1954 Time of action of a series of recessive lethal factors in Drosophila melanogaster. J. Exp. Zool. 126: 17-32. 229 Seto, F., 1961 A developmental study of recessive lethals from wild populations of Drosophila melanogaster. Amer. Nat. 95: 365-373. Shuelke, M., 2000 An economic method for the fluorescent labeling of PCR fragments. Nat. Biotechnol. 18: 233-234. Simmons, M. J. and Crow J.F., 1977 Mutations affecting fitness in Drosophila populations. Ann. Rev. Genet. 11: 49-78. Sørensen, F. C., 1967 Linkage between marker genes and lethal factors may cause distrubed segregation rations. Silvae. Genet. 16: 132–134 Stearns, S. C., 1980 A new view of life history evolution. Oikos 35: 266-281 Stearns, S. C., 1992 The Evolution of Life Histories. Oxford University Press, Oxford. Steinberg, E. K., K .R. Lindner, J Gallea, A Maxwell, J. Meng, et al. 2002 Rates and patterns of microsatellite mutations in Pink salmon. Mol. Biol. Evol 19(7): 1198- 1202. Strathmann, R. R., 1985 Feeding and nonfeeding larval development and life-history evolution in marine invertebrates. Ann. Rev. Ecol. Syst. 16: 339. Szafraniec, K., R. H. Borts, and R. Korona, 2001 Environmental stress and mutational load in diploid strains of the yeast Saccharomyces cerevisiae. Proc. Natl. Acad. Sci. USA 98: 1107–1112. Taris, N., B. Ernande, H. McCombie, and P. Boudry, 2005 Phenotypic and genetic consequences of size selection at the larval stage in the Pacific oyster Crassostrea gigas. J. Exp. Mar. Biol. Ecol. 333: 147-158. Thiele, J., T. Hansen, H.R. Siegismund, and T. P. Hauser, 2009 Genetic variation of inbreeding depression among floral and fitness traits in Silene nutans. Heredity 104: 52-60. Thomas, T. L., 1993 Gene expression during plant embryogenesis and germination: an overview. Plant Cell 5: 1401-1410. Thompson, P. A., and P. J. Harrison, 1992 Effects of mono-specific algal diets of varying biochemical composition on the growth and survival of Pacific oyster (Crassostrea gigas) larvae. Mar. Biol. 113: 645-654. Thomspon, P. A., M-X Guo and P. J. Harrison, 1996 Nutritional value of diets that vary in fatty acid composition for larval Pacific oysters (Crassostrea gigas). Aquaculture 143(3-4): 379-391. 230 Thorson, G., 1950 Reproductive and larval ecology of marine bottom invertebrates. Biol. Rev. 25: 1–45. Tobari, I., 1966 Effects of Temperature on the Viability of Heterozygotes of Lethal Chromosomes in Drosophila melanogaster. Genetics 53(2): 249-259. Todd, C. D., and R. W. Doyle, 1981 Reproductive strategies of marine benthic invertebrates: a settlement-timing hypothesis. Mar. Ecol. Prog. Ser. 4: 75-83. Tomala, K., and R. Korona, 2008 Alleviation of deleterious effects of protein mutation through inactivation of molecular chaperones. Mol. Genet. Genom. 280: 409-417. Van Ooijen, J. W. and R. E. Voorrips, 2001 JoinMap® 3.0, Software for the calculation of genetic linkage maps. Plant Research International, Wageningen, the Netherlands. Vogl, C. and S. Xu, 2000 Multipoint mapping of segregation distorting loci using molecular markers. Genetics 155: 1439-1447 Wada, K. T., 1975 Electrophoretic variants of leucine aminopeptidase of the Japanese pearl oyster Pinctada fucata (Gould). Bull. Natl. Pearl Res. Lab. Jpn. 19: 2152– 2156. Wade, M. J. 2001 Epistasis, complex traits, and mapping genes. Genetica 112–113: 59– 69. Wang, Y., R. Ren, and Z. Yu, 2008 Bioinformatic mining of EST-SSR loci in the Pacific oyster, Crassostrea gigas. Animal Genetics 39: 287-289. Wethey, D. S., 1986 Local and regional variation in settlement and survival in the littoral barnacle Semibalanus balanoides (L.): patterns and consequences. In: Moore, P. G., and R. Seed (eds) The ecology of rocky coasts. Columbia University Press, New York, p 194–202 White, K. P., S. A. Rifkin, P. Hurban, and Hogness, D. S., 1999 Microarray analysis of Drosophila development during metamorphosis. Science 286: 2179-2184. Whitlock M. C., and D. Bourguet, 2000 Factors affecting the genetic load in Drosophila: Synergistic epistasis and correlations among fitness components. Evolution 54: 1654-1660. Wilkins, N. P., 1976 Genic variability in marine Bivalvia: implications and applications in molluscan mariculture. 10 th European Symposium on Marine Biology, Ostend Belgium Sept. 17-23, 1: 549-563. 231 Willett, C. S. and J. N. Berkowitz, 2007 Viability effects and not meiotic drive cause dramatic departures from Mendelian inheritance for malic enzyme in hybrids of Tigriopus californicus populations. J. Evol. Biol. 20: 1196-1205. Williams, E. A., B. M. Degnan, H. Gunter, D. J. Jackson, B. J. Woodcroft, et al., 2009 Widespread transcriptional changes pre-empt the critical pelagic–benthic transition in the vetigastropod Haliotis asinine. Mol. Ecol. 18: 1006-2025. Williams, G. C., 1975 Sex and Evolution. Princeton University Press, Princeton N.J. Williams, G. C., L. D. Auckland, M. M. Reynolds and K. A. Leach, 2003 Overdominant lethals as part of the conifer embryo lethal system. Heredity 91: 584-592. Willis, J. H., 1993 Effects of different levels of inbreeding on fitness components in Mimulus guttatus. Evolution 47:864-876. Winemiller, K. O., and K. A. Rose, 1993 Why do most fish produce so many tiny offspring? Amer. Nat. 142: 585-603. Winemiller, K. O., and K. A. Rose, 1992 Patterns of life-history diversification in North American fishes: Implications for population regulation. Can. J. Fish. Aquat. Sci. 49: 2196-2218. Winemiller, K. O. 1992 Life history strategies and the effectiveness of sexual selection. Oikos 62: 318-327. Woodruff R. C., H. Huai, J. N. Thompson Jr., 1996 Clusters of identical new mutations in the evolutionary landscape. Genetica 98:149-160. Wright, S., 1977 Inbreeding in animals: differentiation and depression, pp.44-96, in Evolution and the genetics of populations, vol. 3, University of Chicago, Chicago. Xu, S., 2008 Quantitative trait locus mapping can benefit from segregation distortion. Genetics 180: 2201-2208. Xu, Y., L. Zhu, J. Xiao, N. Huang, and S. R. McCouch, 1997 Chromosomal regions associated with segregation distortion of molecular markers in F2, backcross, doubled haploid, and recombinant inbred populations in rice (Oryza sativa L.). Mol. Genet. Genomics 253: 535–545. Yamtich J, M. L. Voigt, G. Li and D. Hedgecock, 2005 Eight microsatellite loci for the Pacific oyster Crassostrea gigas. Anim. Genet. 36: 524-526. Yin, T. M., S. P. DiFazio, L. E. Gunter, D. Riemenschneider, and G. A. Tuskan, 2004 Large-scale heterospecific segregation distortion in Populus revealed by a dense genetic map. Theor Appl Genet. 109: 451–463 232 Yu, H., and Q, Li, 2007 EST-SSR markers from the Pacific oyster, Crassostrea gigas. Molecular Ecology Notes, 7: 860-862. Zaykin, D. V. and A. I. Pudovkin, 1993 Two programs to estimate significance of Chi- square values using pseudo-probability test. J. Hered. 84: 152. Zhang, L., S. Want, H. Li, Q. Deng, A. Zheng, S. Li, P. Li, Z. Li, and J. Wang, 2010 Effects of missing marker and segregation distortion on QTL mapping in F 2 populations. Theor. Appl. Genetic. 121: 1071-1082. Zhu, C., C. Wang, and Y. M. Zhang, 2007 Modeling segregation distortion for viability selection 1. Reconstruction of linkage maps with distorted markers. Theor. Appl. Genet. 114: 295-305. Zouros, E., S. M. Singh, and H. E. Miles, 1980 Growth rate in oysters: an overdominant phenotype and its possible explanations. Evolution 34: 856–867. 233 Appendix Quantitative trait locus analysis of settlement timing in the Pacific oyster reveals significant genetic variance in early vs. late settlement A.1 Abstract Settlement timing plays a critical role in recruitment success of marine invertebrates, but its genetic control remains poorly understood. In this study, I performed a quantitative trait locus (QTL) mapping analysis of settlement timing in the progeny of an inbred F 2 cross of the Pacific oyster Crassostrea gigas. Spat were sampled from an early and a late settlement peak, days 19 and 24 post-fertilization, and examined for genetic differences at 24 markers selected to represent minimally the 10 linkage groups of the oyster genome. Three significant, unlinked QTL and one marginally significant, epistatic QTL were found, explaining 29% of the variance in settlement timing. The strongest QTL explained almost 9% of the overall variance, and demonstrated additive gene effects. These results suggest that there is significant genetic variance for settlement timing, but the correlated effect of growth rate on settlement timing and genotype-dependent mortality during metamorphosis may confound the identification of QTL. Additive genetic variance for settlement timing has important implications for the improvement of hatchery traits related to spat production and the evolution of larval life history traits. 234 A.2 Introduction Variability in rates and patterns of recruitment of invertebrates in the marine environment is an important focus of research in marine larval ecology and is determined by the interaction of biotic and abiotic factors that operate at different temporal and spatial scales (e.g. Keough and Downes 1982, Wethey 1986, Gaines and Roughgarden 1985, Rodriguez et al. 1993). Differential larval settlement appears to play a major role in the variability in recruitment (Gaines and Roughgarden 1985), and a number of factors have been shown to affect settlement, including appropriate chemical and physical cues, environmental sources of mortality, competition, and the delay of metamorphosis (Reviewed by Crisp 1974, Keough and Downes 1982, Pechenik 1990, Degnan and Morse 1995, Rodriguez et al. 1993, Jackson 2002). Settlement timing can vary dramatically for species within a reproductive season (e.g. Pineda 1994, Balch and Shiebling 2000, Broitman et al. 2008), and some studies have even found that individuals that settle early suffer less mortality than those that settle later (Connell 1965, Todd and Doyle 1981, Raimondi 1990), have overall better growth at the spat stage (Losee 1979), and that settlement during a particular temporal window is correlated with greatest juvenile success (Jarett and Pechanik 1997, Pineda 2006). Timing of settlement can influence fitness because larvae that settle more quickly spend less time in the water column, which reduces their exposure to predation and perhaps provides them a head start on growth in the juvenile stage. Earlier settlement may also ensure greater availability of endogenous energy reserves in non-feeding species and, thus, greater survival during metamorphosis (Pechanik 1990, Crisp 1974). Given this evidence for strong associations between settlement timing and survival, timing of settlement should have adaptive significance, 235 but little is known about how settlement timing may be influenced by heritable genetic variation. Numerous laboratory based observations of marine invertebrate larvae have revealed substantial intra-specific variation in the age at which competence develops (e.g. Hadfield 1977, Pechenik and Heyman 1987, Pechenik and Qian 1998), suggesting that variables other than the environment are affecting competency and settlement. Only a few studies have directly examined the role of genetic variation in these processes, however. One of the earliest studies to examine genetic effects on settlement timing found that 27 generations of selection for early development of competence did not significantly alter settlement timing in the nudibranch Phestilla sibogae, which suggests that variance in competence in this species is under epigenetic control, balancing selection, or purely environmental control (Hadfield 1984). In contrast, generations of culling and size selection at the larval stage in hatchery production of the Pacific oyster Crassostrea gigas have produced strong phenotypic effects on time to settlement, as a likely by-product of selection for high growth rate (Taris et al. 2005). More concrete evidence for the role of genetic variation in settlement timing comes from a cross- breeding study of the vetigastropod Haliotus refenscens, which revealed significant paternal effects on settlement timing (Jackson et al. 2005). This lone study suggests a genetic basis for variance in settlement timing, but no study has directly examined the genetic basis of variation in this trait, at either the organismal or genetic marker level. Knowledge of the genetic basis of settlement timing has implications both for the evolution of larval life history traits and for improvement of larval traits in marine bivalve culture. 236 Figure 22 Settlement timing data for the F 2 family Observed numbers of new settled juveniles counted each day, over the course of metamorphosis. Arrows mark the early and late peaks of settlement (accounting for 68% of all settlement), from which spat were sampled for the genetic mapping analysis. 236 237 In this study, I performed quantitative-trait locus (QTL) analysis of settlement timing in an inbred F 2 cross sampled daily during the period of metamorphosis and settling (Chapter 2). Settlement in this family was observed to be bi-modal (Fig. 22, see Chapter 2), therefore, settlement timing was treated as binary character, with spat settling on the early and late ―peaks‖ of settlement (individuals settling on these two days made up 68% of all settlement) sampled for genetic analysis. In addition to the search for individual QTL, gene effects and epistatic interactions were also assessed, in order to elucidate the components of genetic variance in this important larval trait. A.3 Materials and Methods Biological materials Crosses and culturing methods The data for this study were obtained from the same experiment described in Chapter 2; thus, the details of the cross, culturing methods, and sampling, which are fully described in the Materials and Methods of Chapter 2, will only be summarized here. The experimental F 2 family (inbreeding coefficient, f = 0.397) was created in 2009, by mating a pair of male and female full-siblings from the 2007 51×35 F 1 hybrid cross (see Fig. 1, in the Introduction, for pedigrees). Crosses were performed at the University of Southern California (USC), Wrigley Marine Science Center (WMSC) on Catalina Island, CA, and reared according to standard protocols for Crassostrea gigas (Chapters 1, 2, 3 and 4). Pedigrees of parents were verified with microsatellite DNA markers (Hedgecock and Davis 2007). 238 Deployment of shell and sampling protocols At day 18 post-fertilization, when a substantial number of larvae had developed eye spots and displayed probing of the substrate with the larval foot, both precursors to settlement (Bonar 1976), cured adult shells were deployed on a Vexar® mesh cylindrical harness for natural settlement (Chapter 2, Fig. 7). The settled spat on each shell were counted, shell by shell, under a dissecting microscope at 8-48×, until no more could be found, and larvae were placed back into the culture tank. This process was repeated each day until no larvae remained in the water column and no additional spat settled. Shells that had spat settling on a particular day were then collected and placed in a Vexar® mesh bag in the upwelling nursery system and grown out until day 60 to increase the amount of tissue for DNA extraction. Microsatellite marker genotyping Eighty-four markers cloned from the Pacific oyster were tested in this study (Magoulas et al. 1998, McGoldrick et al. 2000, Huvet 2000, Li et al. 2003, Sekino et al. 2003,Yamtich et al. 2005, Yu and Li 2007, Wang et al. 2007), and as described in Chapter 2, 46 markers were informative, all found on published linkage maps (Hubert and Hedgecock 2004, Hubert et al. 2009). From these 46 markers, a subset of 24 were chosen for the mapping analysis, with the aim of having at least two markers on each of the 10 linkage groups, thereby maximizing genomic coverage with the fewest markers (Hubert and Hedgecock 2004, Hubert et al. 2009, Chapters 1, 3, and 4). Details of DNA extraction, PCR amplification, electrophoresis, and genotyping of these markers are presented in the Materials and Methods, Chapter 2. 239 Data analysis Single marker analysis Marker genotypes were tested for deviation from expected Mendelian segregation ratios, using goodness-of-fit chi square tests (e.g. Chapter 1, 3). To test for heterogeneity between the two settlement samples (day 19 and day 24), contingency chi-square tests were performed, using the program ‗Chirxc‘ (Zaykin and Pudovkin 1992) QTL analysis of settlement timing Linkage maps were constructed with day 19 and day 24 spat samples pooled (n = 192), using the CP (cross pollinator) population type in JoinMap 3.0 (VanOoijen and Voorrips 2001). The Kosambi mapping function with a minimum likelihood of the odds (LOD) score of 2.0 was used for linkage group assignments, though most linkage groups had LOD scores well above 4.0. Deviations from Mendelian segregation ratios may affect linkage mapping (Lorieux et al. 1995, Zhu et al. 2007), so I compared the marker order and distances with previously published linkage maps, which were constructed using larval samples with little segregation distortion (Hubert and Hedgecock 2004). If markers that should have mapped failed to map, I used locations from published maps (Hubert and Hedgecock 2004, Hubert et al. 2009). Phase was determined by JoinMap 3.0, from the frequencies of parental and recombinant types. Genomic coverage was estimated with the equation where d = mean inter-marker distance and n = total number of markers assigned to linkage groups, and L is the estimated map length (Bishop et al. 1983). 240 QTL analysis was performed in R/qtl (v.1.1; Broman et al. 2003), under a binary trait model, (early vs. late settlement, day 19 vs. day 24), using the outcross settings. A one-dimensional scan for single QTL was first performed (‗scan-one‘ module) and 5% significance thresholds at the genome-wide and chromosome-wide level were calculated with 1000 permutations. Next, genetic interaction was tested at every position in the genome, using a two-QTL model (‗scan-two‘ module; significance adjusted with 1000 permutations). Once significant QTL were identified, they were fit into an overall multiple regression ANOVA model, using a backwards selection procedure to find the best model with lowest model comparison criterion. A.4 Results Of the 24 informative markers selected for this study, 21 were successfully typed in spat settling on days 19 and 24 (n = 192). These markers were assigned to linkage groups, as expected from previous crosses and studies (Hubert and Hedgecock 2004, Hubert et al. 2009, Chapters 1,3, and 4), and phase was successfully estimated within linkage groups for all 21 markers. None of the three informative markers on linkage group two (LG2) could be scored, so these markers and this linkage group were not included in the analysis. Map distances for the two markers on LG7 were obtained from previously published maps because this linkage group had only two genotyped markers, which were separated by more than 40 cM. Total map length was 369 cM, calculated by summing all inter-marker distances, with a mean inter-marker distance of 17.65 cM (maximum 45 cM; LG 7). Estimated genomic coverage for this map was 86% (Bishop et al. 1983). Single-marker analysis revealed that four of the 21 markers, Cg205, Cg162, 241 Table 15 Segregation data for day 19 and day 24 spat samples a Goodness-of-fit chi-square test of segregation ratios to Mendelian expectation; b Linkage group; c r×c contingency chi-square test for heterogeneity in segregation ratios between day 19 and day 24 spat samples, P-value in parentheses(Zaykin and Pudovkin 1992). Table shown next page. 242 Table 15 (Continued) Cross type Marker LG a Genotype numbers total χ 2 Value b P-Value Heterogeneity χ 2 (P-value) c AA×AB AA AB d19 Cg139 5 38 48 86 2.54 0.281 1.04(0.546) d24 Cg139 5 47 48 95 0.17 0.918 85 96 181 1.77 0.414 AB×AB AB AA BB d19 Cg200 1 48 28 9 85 9.92 0.007 0.914(0.623) d24 Cg200 1 52 25 13 90 5.38 0.068 100 53 22 175 <0.001 d19 Cg124 1 66 24 0 90 32.40 <0.001 0.12(0.879) d24 Cg124 1 66 27 0 93 32.03 <0.001 132 51 0 183 64.28 <0.001 d19 L48 3 58 24 4 86 19.77 <0.001 2.66(0.252) d23 L48 3 51 28 9 88 10.43 0.005 109 52 13 174 <0.001 d19 Cg148 3 64 23 1 88 29.18 <0.001 0.25(0.834) d24 Cg148 3 66 28 1 95 29.76 <0.001 130 51 2 183 58.64 <0.001 Table continues next page 242 243 Table 15 (Continued) Cross type Marker LG a Genotype numbers total χ 2 Value b P-Value Heterogeneity χ 2 (P-value) c d19 Cg162 3 63 13 12 88 16.43 <0.001 8.68(0.017) d24 Cg162 3 51 10 29 90 9.62 0.008 114 23 41 178 17.69 <0.001 d19 Cg198 4 41 24 23 88 0.43 0.806 2.88(0.301) d24 Cg198 4 44 32 15 91 6.45 0.040 85 56 38 179 4.07 0.131 d19 Cg163 5 68 18 0 86 36.60 <0.001 1.39(0.654) d24 Cg163 5 69 22 0 91 34.91 <0.001 137 40 0 177 71.24 <0.001 d19 Cg138 5 42 20 19 81 0.14 0.934 0.57(0.781) d24 Cg138 5 49 18 18 85 1.99 0.370 91 38 37 166 1.55 0.460 d19 Cg205 6 82 9 2 93 55.26 <0.001 12.87(<0.001) d24 Cg205 6 61 19 11 91 11.97 0.003 143 28 13 184 58.99 <0.001 d19 Cg209 6 65 1 27 93 29.26 <0.001 3.06(0.339) d24 Cg209 6 73 2 17 92 36.59 <0.001 138 3 44 185 62.94 <0.001 Table continues next page 243 244 Table 15 (Continued) Cross type Marker LG a Genotype numbers total χ 2 Value b P-Value Heterogeneity χ 2 (P-value) c d19 Cg28 7 51 24 14 89 4.15 0.126 1.23(0.532) d24 Cg28 7 62 21 12 95 10.56 0.005 113 45 26 184 13.51 0.001 d19 L16 8 42 25 20 87 0.68 0.712 1.66(0.427) d23 L16 8 50 21 16 87 2.52 0.284 92 46 36 174 1.72 0.422 d19 Cg183 9 37 19 15 71 0.58 0.749 1.20(0.553) d24 Cg183 9 45 32 16 93 5.60 0.061 82 51 31 164 4.88 0.087 d24 Cg184 9 23 10 13 46 0.39 0.822 1.40(0.516) d19 Cg184 9 28 7 10 45 3.09 0.213 51 17 23 91 2.12 0.346 d19 Cg140 10 48 6 28 82 14.20 0.001 17.75(<0.001) d24 Cg140 10 51 26 13 90 5.36 0.069 99 32 41 172 4.87 0.088 Table continues next page 244 245 Table 15 (Continued) Cross type Marker LG a Genotype numbers total χ 2 Value b P-Value Heterogeneity χ 2 (P-value) c AB×AC AA AC AB BC d19 cmrCg5 1 25 25 17 17 84 3.05 0.384 5.30(0.159) d24 cmrCg5 1 15 27 26 24 92 3.91 0.271 40 52 43 41 176 2.05 0.563 d19 Cg109 4 19 22 24 21 86 0.60 0.895 10.75(0.02) d24 Cg109 4 12 44 22 13 91 29.13 0.000 31 66 46 34 177 17.10 0.001 d19 Cg156 7 9 17 21 24 71 7.14 0.068 1.56(0.673) d24 Cg156 7 15 23 19 28 85 4.36 0.225 24 40 40 52 156 10.15 0.017 d19 Cg175 8 7 11 20 29 67 17.24 0.001 7.43(0.068) d23 Cg175 8 20 16 30 22 88 4.73 0.193 27 27 50 51 155 14.26 0.003 d19 Cg129 10 9 22 27 38 96 18.08 0.000 5.99(0.108) d24 Cg129 10 9 28 35 22 94 15.53 0.001 18 50 62 60 190 26.17 0.000 245 246 Cg140 and Cg109, had significant heterogeneity between days 19 and 24, indicating that these markers might be associated with settlement timing (Table 15). Significantly distorted segregation ratios at the = 0.05 level were found at 13 of 21 (61%) markers (day 19 and day 24 pooled). The one-dimensional QTL scan identified three QTL on linkage groups 4, 6, and 10 (QTL 1-3; Table 16). The most significant QTL, on linkage group 10 (close to Cg140), was above the genome-wide threshold LOD score of 2.99 (QTL 3, LOD=3.794; Table 16). The other two QTL, one on LG 4 at Cg109 (QTL 1, LOD=2.567) and one on LG 6 at Cg205 (QTL 2, LOD=2.714), were slightly below the genome-wide threshold but well above the chromosome-wide thresholds of 2.36 and 1.60 for LG 4 and LG 6, respectively, and were therefore considered suggestive QTL (Table 16). The two- dimensional scan for interacting QTL identified one significant interaction between genomic regions on LG 5 and LG 8 (QTL 4; LG 5 at 10cM, and LG 8 at 40cM, LOD = 5.642), which was above the α = 0.05 genome-wide interaction threshold LOD score of 5.46. Fitting a model for the four identified QTL (from the single QTL scan and the epistasis scan), I performed the ‗drop-one‘ analysis of variance (ANOVA) to examine the fit of the model with each QTL included and then dropped in sequential order. In the drop-one analysis, all QTL remained significant except for the interaction QTL, which was marginally significant (P=0.057). Results of the ANOVA with four QTL indicated that the model was highly significant (P<0.00001), with the four QTL explaining almost 30% of the variance in settlement time (Table 16). The strongest effect was at QTL3 (Cg140), which alone explained 8.7 % of the variance. Examining the gene effects of the 247 Table 16 QTL and ANOVA model results QTL LG Position (cM) a Marker b LOD % Var. d P-Value (χ 2 ) 1 4 0.0 Cg109 2.567 3.84 0.030 2 6 1.0 Cg205 2.714 7.085 0.001 3 10 1.0 Cg140 3.794 8.7 <0.001 5 10.0 e -- -- 6.66 0.210 8 40 e -- -- 5.18 0.079 4 5:8 10:40 -- 5.642 6.97 0.052 Full model 29.9 <0.00001 a linkage group number @ position along linkage group, in centimorgans (cM), the colon represents interaction between the two positions for QTL 4; b markers closely linked to QTL, if any; LOD represents the log of the odds (LOD) score for each identified QTL; d percent variance explained by the individual QTL or the full model; e these genomic positions were not significant QTL in the 1-dimensional scan and, thus, were not assigned numbers; because a significant interaction between them was detected in the two-dimensional scan, R/qtl includes them in the hierarchical ANOVA model by convention, thus their effects are reported. 248 single QTL at the nearest markers, Cg140 (QTL 3) demonstrated an additive pattern; settlement on day 24 (late settlement) was observed for 83, 49, and 33% of individuals with genotypes AA, AB, and BB, respectively (Fig. 23). Patterns of gene effects at Cg205 and Cg109 were less clear but appeared to be non-additive (Fig. 24). A.5 Discussion In the F 2 cross of the Pacific oyster examined in this study, 68% of all settlement occurred on two days, corresponding to an early and a late peak in settlement. Thus, time of settlement was not normally distributed in this cross. QTL mapping analysis of settlement timing identified three significant QTL linked to those markers and one marginally significant epistatic QTL, all of which explained an estimated 29.9% of the variance in settlement time. Settlement timing appears, therefore, to have significant genetic variance, results that strongly agree with Jackson et al. (2005) who also found significant genetic (paternal) effects on the development of competency and early settlement in crosses of Haliotus spp. The current study is the first to associate specific genetic markers with settlement timing variation, and even with relatively few molecular markers (21), there is evidence of loci with major effects on variation in this important larval trait. Cg140 had the largest effect of the QTL identified, explaining nearly 9% of the variance, and demonstrating apparently additive gene effects (Fig 23). That this QTL may be additive is especially important for selective breeding programs, in which selection to improve traits is only possible if these traits demonstrate additive 249 Figure 23 Plot of genotype effects on settlement timing for Cg140 Values (0 or 1, corresponding to early and late settlement, respectively) are jittered and represented by open circles. Large blue filled circles represent mean p (value of p displayed below filled circles), the binomial probability of late settlement for individuals of each genotype, AA, AB, and BB, which shows an additive pattern of gene effects at this marker. 250 Figure 24 Plot of genotype effects on settlement timing for Cg205 and Cg109 Values (0 or 1, corresponding to early and late settlement, respectively) are jittered and represented by open circles for Cg205 (panel A) and Cg209 (panel B). Large blue filled circles represent mean p, the binomial probability of late settlement for individuals of each genotype. 250 251 genetic variance (Falconer and Mackay 1996). With evidence for at least one additive QTL in the rather small study reported here, hatchery managers could theoretically select for either early settlement or for reduced variance in time of settlement, which are important husbandry traits. Domestication selection for early settlement time may have already occurred in many hatchery stocks if hatchery operators simply take the first settlers and discard the rest of the population. Individual QTL effects and estimates of heritability have been shown to be upwardly biased for small sample sizes, when heritability is low (Göring et al. 2001, Bogdan and Doerge 2005). Interpretation of these effects is, thus, approached with caution. However, the estimates of effect size and heritability may actually be deflated in this study, because segregation distortion, which was observed at markers closely associated with settlement-time QTL, is known to substantially reduce power in QTL studies (e.g. Broman et al. 2003, Zhang et al. 2010). Segregation distortion is ubiquitous in both inbred and outbred crosses of the Pacific oyster (Launey and Hedgecock 2001, Bucklin 2003, Chapters 2, 4, 5) and is unlikely to be avoided in experimental families. Recently, methods have been developed to account for segregation distortion during the mapping analysis; accounting for distortion may actually improve power to detect QTL, but at present, these models are not available for analysis of binary traits (Hu and Xu 2008). The finding of significant genetic variance for settlement timing raises the question of whether other correlated traits are involved that might explain the results. Previous studies in the Pacific oyster have shown that hatchery selection for growth rate 252 in larvae has resulted in a simultaneous decrease in the average time to settlement, suggesting that growth rate and settlement timing may be correlated (Taris et al. 2005). Studies in a variety of marine invertebrates, marine fish, and vertebrate species also support the idea that time to settlement or metamorphosis is correlated with other traits, showing, for example, that early time to metamorphosis is associated with faster growth, greater size-at-age in the juvenile stages, and generally, greater survival (e.g. Losee 1978, Semlitsch et al. 1988, Hare and Cowen 1997, Qian and Pechanek 1998, Maldonaldo and Young 1999, Rainkin and Sponaugle 2011). Unfortunately, larval growth was not measured in the current study, so that a direct comparison of genetic variance for growth and settlement timing is not possible. However, previous studies have shown that individual size is positively associated with multi-locus heterozygosity for a number of marine bivalves, including the Pacific oyster (e.g. Zouros et al. 1980, Koehn and Gaffney 1984, Pogson and Zorous 1994). If growth differences underlie variation in settlement timing, we might expect to see a difference in heterozygosity between the two settlement samples, but there was no difference in mean multi-locus heterozygosity between the early and late settlement samples (0.66 vs 0.67, respectively; t-test of arcsine square-root transformed values, P<0.42). To the extent that growth and MLH are correlated, growth rate may not have much association with settlement timing in this cross. Of course, the proportions of genotypes in spat samples during and after metamorphosis are significantly affected by segregation distortion (Chapter 1 and 2), which results primarily in the absence of homozygous genotypes and the excess of heterozygotes, possibly obscuring the heterozygosity/growth relationship in this family. Ultimately, an analysis 253 of settlement timing with growth accounted for either as a covariate or mapped simultaneously, would allow for a more complete understanding of the relationship between the two traits. For example, if significant QTL for the two traits are found in different regions of the genome, that would suggest separate genes or genetic pathways that underlie the two traits. Alternatively, if QTL co-localize or if significant QTL for settlement timing are reduced or eliminated with the addition of growth as a covariate, variation in settlement timing would be interpreted as a by-product of variance in growth rate. Finally, genotype-specific mortality during settlement could also explain differentiation in segregation ratios between the early and late settlement samples, rather than genotype-dependent differences in settlement time. The fact that two of the QTL (Cg205 and Cg109) were also associated with deleterious mutations acting during metamorphosis supports this hypothesis (Chapter 2). On the other hand, not all QTL may have been confounded by segregation distortion. For example, Cg140 (closely linked to QTL 3) is not distorted in the pooled spat sample (chi-square goodness-of-fit test, P=0.088; Table 1), which suggests it is not linked to a viability mutation, or not closely linked. Thirty centi-morgans (cM) away on the same linkage group (LG 10), Cg129 is significantly distorted, but prior to metamorphosis (P<0.0001, goodness-of-fit chi-square test for the day 18 larval sample; Table 1, Chapter 3). Thus, genotype-dependent mortality during metamorphosis cannot explain genotype-dependent differences between early and late settlement for Cg140. The epistatic QTL identified in this study (QTL 4) is also not likely to be explained by segregation distortion because epistatic interactions 254 between viability QTL were virtually absent in all crosses examined (Chapters 1, 3, and 4). Overall, the confounding effects of growth rate and segregation distortion on genotype-dependent differences in settlement timing warrant serious consideration, but likely do not completely account for the QTL results in this preliminary study. Future work should include the construction of a more complete genetic map, the simultaneous mapping of size QTL, and genetic analysis of samples from all settlement days, so that settlement timing can be mapped as a continuous quantitative trait. 255 Appendix References Balch, T., and R. E. Scheibling, 2000 Temporal and spatial variability in settlement and recruitment of echinoderms in kelp beds and barrens in Nova Scotia. Mar. Ecol. Prog. Ser. 205: 139–154. Bishop, D.T., C. Cannings, M. Skolnick, and J. A. Williamson, 1983 The number of polymorphic DNA clones required to map the human genome. In: Statistical Analysis of DNA Sequence Data, Weir, B.S. (ed.). New York, N.Y., Marcel Dekter, pp. 181–200. Bogdan, M., and R. W. Doerge, 2005 Biased estimators of genetic effects and heritability in interval mapping. Heredity 95:476-484. Bonar, D.B., 1976 Molluscan metamorphosis: A study in tissue transformation. Amer. Zool. 16: 573-591. Broitman, B. R., C. A. Blachette, B. A. Menge, J. Lubchenko, C. Krenz, M. Foley, P. T. Raimondi and D. Lohse, 2008 Spatial and temporal patterns of invertebrate recruitment along the west coast of the united states. Ecol. Mono. 78(3): 403–421. Broman, K. W., H. Wu, S. Sen, and G. A. Churchill, 2003 R/qtl: QTL mapping in experimental crosses. Bioinformatics 19: 889-90. Bucklin, K. A., 2003 Analysis of the Genetic Basis of Inbreeding Depression in the Pacific Oyster Crassostrea gigas. Ph.D. Dissertation, University of California at Davis. Connell J. H., 1985 The consequences of variation in initial settlement vs. post-settlement mortality in rocky intertidal communities. J. Exp. Mar. Biol. Ecol. 93: 11–45. Crisp, D. J., 1974 Factors influencing the settlement of marine invertebrate larvae. In, Chemoreception in marine organisms, edited by P. T. Grant & A. M. Mackie. Academic Press, New York, pp. 177-265. Degnan B. M., and D. E. Morse, 1995 Developmental and morphogenetic gene regulation in Haliotis rufescens larvae at metamorphosis Amer Zool 35: 391-398. Falconer D. S. and T. F. C. Mackay, 1996 Introduction to Quantitative Genetics. 4th edn. Longman: Harlow. 256 Gaines, S., and J. Roughgarden, 1985 Larval settlement rate: a leading determinant of structure in an ecological community of the marine intertidal zone. Proc. Natl. Acad. Sci. U.S.A. 82: 3707-3711. Göring, H. H. H., J. D. Terwilliger, and J. Blangero, 2001 Large upward bias in estimation of locus-specific effects from genomewide scans. Am. J. Hum. Genet. 69: 1357-1369. Hadfield, M. G., 1977 Chemical interactions in larval settling of a marine gastropod. In: Faulkner, D. J., Fenical, W. H. (eds.) Marine natural products chemistry. Plenum Publishing Corporation, New York, pp. 403-413. Hadfield, M. G., 1984 Settlement requirements of molluscan larvae: New data on chemical and genetic roles. Aquaculture 39: 283-298. Hare, J. A., and R. K. Cowen, 1997 Size, growth, development, and survival of the planktonic larvae of Pomatomus saltatrix (Pisces: Pomatomidae). Ecology 78: 2415- 2431. Hedgecock, D., and J. P. Davis, 2007 Heterosis for yield and crossbreeding of the Pacific oyster Crassostrea gigas. Aquaculture 272S: S17-S29. Hu, Z., and S. Xu, 2009 PROC QTL—A SAS procedure for mapping quantitative trait loci. Int. J. Plant Genomics 2009: 141234. Hubert, S., E. Cognard, and D. Hedgecock, 2009 Centromere-mapping in triploid families of the Pacific oyster Crassostrea gigas (Thunberg). Aquaculture 288: 172–183. Hubert, S., and D. Hedgecock, 2004 Linkage maps of microsatellite DNA markers for the Pacific oyster Crassostrea gigas. Genetics 168: 351-362. Huvet, A., P. Boudry, M. Ohresser, C. Delsert, and F. Bonhomme, 2000 Variable microsatellites in the Pacific oyster Crassostrea gigas and other cupped oyster species. Anim. Genet. 31: 71-72. Jackson, D. J., N. E. Ellemor, and B. M. Degnan, 2005 Correlating gene expression with larval competence, and the effect of age and parentage on metamorphosis in the tropical abalone Haliotis asinine. Mar. Biol. 147: 681-697. Jackson, D., S. P. Leys, V. F. Hinman, R. Woods, M. F. Lavin, and B. M. Degnan, 2002 Ecological regulation of development: induction of marine invertebrate metamorphosis. Int. J. Dev. Biol. 46: 679-686. 257 Jarrett, J. N., and Pechenik J. A., 1997 Temporal variation in cyprid quality and juvenile growth capacity for an intertidal barnacle. Ecology 78: 1262–1265 Keough, M. J., and B. J. Downes, 1982 Recruitment of marine invertebrates: The role of active larval choices and early mortality. Oecologia (Berl) 54: 348-352 Koehn, R. K, and P. M. Gaffney, 1984 Genetic heterozygosity and growth rate in the Mytilus edulis. Mar. Biol. 82: 1-7. Launey, S., and D. Hedgecock, 2001 High genetic load in the Pacific oyster Crassostrea gigas. Genetics 159: 255-265. Li, G., S. Hubert, K. Bucklin, V. Ribes, and D. Hedgecock, 2003 Characterization of 79 microsatellite DNA markers in the Pacific oyster Crassostrea gigas. Mol. Ecol. Notes 3: 228-232. Lorieux M., X. Perrier, B. Goffinet, C. Lanaud, and D. González de León, 1995 Maximum-likelihood models for mapping genetic markers showing segregation distortion. 2. F 2 populations. Theor. Appl. Genet. 90: 81–89. Maldonado M., and C. M. Young, 1999 Effects of the duration of larval life on postlarval stages of the demosponge Sigmadocia caerulea. J. Exp. Mar. Biol. Ecol. 232: 9– 21. Lorieux M., X. Perrier, B. Goffinet, C. Lanaud, and D. González de León, 1995 Maximum-likelihood models for mapping genetic markers showing segregation distortion. 2. F 2 populations. Theor. Appl. Genet. 90: 81–89. Losee, E., 1979 Relationship between larval and spat growth rates in the oyster Crassostrea virgininca. Aquaculture 16: 123-126. Magoulas, A., B. Gjetvag, V. Tersoglou and E. Zouros, 1998 Three polymorphic microsatellites in the Japanese oyster, Crassostrea gigas (Thunberg). Anim. Genet. 29:69-70. McGoldrick D. J., D. Hedgecock, L. J. English, P. Baoprasertkul, and R. D. Ward, 2000 The transmission of microsatellite alleles in Australian and North American stocks of the Pacific oyster (Crassostrea gigas): selection and null alleles. J. Shellfish Res. 19: 779–788. Pechenik, J. A., 1990 Delayed metamorphosis by larvae of benthic marine invertebrates: Does it occur? Is there a price to pay? Ophelia 32: 63-94. 258 Pechenik, J. A., and W. D. Heyman, 1987 Using KCl to determine size at competence for larvae of the marine gastropod Crepidula fornicata (L.), J. Exp. Mar. Biol. Ecol. 112: 27–38 Pineda, J., 1994 Spatial and temporal patterns in barnacle settlement rate along a southern California rocky shore. Mar. Ecol. Prog. Ser. 107: 125-138. Pineda, J., V. Starczak, and T. A. Stueckle, 2006 Timing of successful settlement: demonstration of a recruitment window in the barnacle Semibalanus balanoides Mar. Ecol. Prog. Ser. 320: 233–237. Pogson, G. H., and E. Zouros, 1994 Allozyme and RFLP heterozygosities as correlates of growth rate in the scallop Placopecten magellanicus: a test of the associative overdominance hypothesis. Genetics 137: 221-231 Qian P. Y., and J. A. Pechenik, 1998 Effects of larval starvation and delayed metamorphosis on juvenile survival and growth of the tube-dwelling polychaete Hydroides elegans (Haswell). J. Exp. Mar. Biol. Ecol. 227: 169–185. Rankin T. L., Sponaugle S., 2011 Temperature influences selective mortality during the early life stages of a coral reef fish. PLoS ONE 6(5): e16814. Raimondi, P. T., 1990 Patterns, mechanisms, and consequences of variability in settlement and recruitment in an intertidal barnacle. Ecological Monographs 60: 283-309. Rodriguez, S. R., F. Patricio-Ojedal, and N. C. Inestrosa, 1993 Settlement of benthic marine invertebrates Mar. Ecol. Prog. Ser. 97: 193-207. Sekino, M., M. Hamaguchi, F. Aranishi, and K. Okoshi, 2003 Development of novel microsatellite DNA markers from the Pacific oyster, Crassostrea gigas. Mar. Biotechnol. 5: 227-233. Semlitsch, R. D., D. E. Scott, and J. H. K. Pechmann, 1988 Time and size at metamorphosis related to adult fitness in Ambystoma talpoideum. Ecology 69: 184–192. Taris, N., B. Ernande, H. McCombie, and P. Boudry, 2005 Phenotypic and genetic consequences of size selection at the larval stage in the Pacific oyster Crassostrea gigas. J. Exp. Mar. Biol. Ecol. 333: 147-158. Todd, C. D., and R. W. Doyle, 1981 Reproductive strategies of marine benthic invertebrates: a settlement-timing hypothesis. Mar. Ecol. Prog. Ser. 4: 75-83. 259 Wang, Y., R. Ren, and Z. Yu, 2008 Bioinformatic mining of EST-SSR loci in the Pacific oyster, Crassostrea gigas. Anim. Genet. 39: 287-289. Van Ooijen, J. W. and R. E. Voorrips, 2001 JoinMap® 3.0, Software for the calculation of genetic linkage maps. Plant Research International, Wageningen, the Netherlands. Wethey, D. S., 1986 Local and regional variation in settlement and survival in the littoral barnacle Semibalanus balanoides (L.): patterns and consequences. In: Moore, P. G., and R. Seed (eds) The ecology of rocky coasts. Columbia University Press, New York, p 194–202 Xu, S., 2008 Quantitative trait locus mapping can benefit from segregation distortion. Genetics 180: 2201-2208. Yamtich J, M. L. Voigt, G. Li and D. Hedgecock, 2005 Eight microsatellite loci for the Pacific oyster Crassostrea gigas. Anim. Genet. 36: 524-526. Yu, H., and Q. Li, 2007 EST-SSR markers from the Pacific oyster, Crassostrea gigas. Molecular Ecology Notes, 7: 860-862. Zaykin, D. V. and A. I. Pudovkin, 1993 Two programs to estimate significance of χ 2 values using pseudo-probability tests. J. Hered. 84:152. Zhang, L., S. Want, H. Li, Q. Deng, A. Zheng, S. Li, P. Li, Z. Li, and J. Wang, 2010 Effects of missing marker and segregation distortion on QTL mapping in F 2 populations. Theor. Appl. Genetic. 121: 1071-1082. Zhu, C., C. Wang, and Y. M. Zhang, 2007 Modeling segregation distortion for viability selection 1. Reconstruction of linkage maps with distorted markers. Theor. Appl. Genet. 114: 295-305. Zouros, E., S. M. Singh, and H. E. Miles, 1980 Growth rate in oysters: an overdominant phenotype and its possible explanations. Evolution 34: 856–867.
Abstract (if available)
Abstract
Most marine fish and shellfish are very fecund, with females often producing many millions of eggs per spawn (Thorson 1950, Winemiller and Rose 1992). Their larval offspring characteristically suffer high (type-III) mortality, while developing for days to weeks in the plankton, but this mortality is largely attributed by most marine scientists to environmental factors. The biological and ecological characteristics of this life history mode have profound implications for population structure, adaptation, and recruitment variation, yet relatively little is known about the role of endogenous variation during the early life stages. Experimental inbred crosses of the highly fecund Pacific oyster Crassostrea gigas have previously revealed a high genetic load, which causes substantial genotype-dependent mortality, explains widespread observations of segregation distortion in crosses of other highly fecund marine bivalves, and fits predictions (Williams 1975) that high fecundity species should suffer substantial inbreeding depression (Launey and Hedgecock 2001, Bucklin 2003). Though these previous studies provided important experimental verification of high genetic load in the oyster, they left unanswered a number of questions about the genetic basis of this load and whether it could explain the high, early life-history mortality observed in the oyster. These questions can now be addressed, thanks to the advent of genome-wide statistical approaches, such as quantitative-trait locus (QTL) mapping. In this dissertation, a series of experiments were performed to address these and other questions, by creating experimental inbred and outbred families and analyzing molecular marker segregation data across the genome to detect the number, location, and effect of deleterious mutations in the Pacific oyster. There were four major findings. First, inbred families exhibited 10-15, mainly recessive or partially dominant, deleterious mutations, with little evidence for epistatic interaction between them. Genotype-dependent mortality calculated from the multiplicative fitness effects of these deleterious loci was ~96%, accounting for nearly all observed actual mortality. Second, the expression of genetic load occurred primarily during the larval stages in inbred crosses, with half of all deleterious mutations expressed during settlement and metamorphosis; further analysis revealed that mutations expressed at metamorphosis occurred over a relatively small temporal window, just prior to and during metamorphosis, but not after. Third, the environment significantly affected the fitness of deleterious alleles; dominance of and selection against some deleterious alleles increased significantly in a stressful, nutrient-deficient algal diet, while other alleles were un-affected, suggesting that inbreeding depression is caused by different loci in different environments. Fourth, seven to nine deleterious mutations were detected in wild families, causing an estimated 87-98% cumulative genetic mortality. Overall, there appears to be remarkable genetic variation in viability during the early life-stages of the Pacific oyster, causing much of the mortality that is observed in wild and inbred cultures. These results suggest that larvae of highly fecund marine animals in the natural environment may also be subject to substantial genotype-dependent mortality, which possibly contributes to the high recruitment variation and fluctuations in abundance of fished populations. The results of the dissertation also have broad implications for understanding the genetic basis of inbreeding depression, conservation genetics, and shellfish aquaculture.
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Genetic architecture underlying variation in different traits in the Pacific oyster Crassostrea gigas
PDF
Stage-specific transcriptomic analyses of reproductive tissue in the Pacific oyster, Crassostrea gigas
PDF
Physiological strategies of resilience to environmental change in larval stages of marine invertebrates
PDF
Dynamics of protein metabolism in larvae of marine invertebrates
PDF
Physiological rates during development of marine invertebrates in temperate and polar oceans
PDF
Phylogeography, reproductive isolation, and the evolution of sex determination mechanisms in the copepod Tigriopus californicus
PDF
Genetic analysis of population structure in striped marlin, Tetrapturus audax, in the Pacific Ocean
PDF
Applications of next generation sequencing in sessile marine invertabrates
PDF
Transgenerational inheritance of thermal tolerance in two coral species in the Florida Keys
Asset Metadata
Creator
Plough, Louis Valentine
(author)
Core Title
Genome-wide analysis of genetic load and larval mortality in a highly fecund marine invertebrate, the Pacific Oyster Crassostrea gigas
School
College of Letters, Arts and Sciences
Degree
Doctor of Philosophy
Degree Program
Biology
Degree Conferral Date
2011-12
Publication Date
10/12/2011
Defense Date
08/05/2011
Publisher
Los Angeles, California
(original),
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
gene action,genetic load,high fecundity,inbreeding,larval mortality,life-history,metamorphosis,OAI-PMH Harvest,oyster,QTL mapping
Format
theses
(aat)
Language
English
Contributor
Electronically uploaded by the author
(provenance)
Advisor
Conti, David V. (
committee member
), Edmands, Suzanne (
committee member
), Hedgecock, Dennis (
committee member
), Manahan, Donal (
committee member
)
Creator Email
louis.plough@gmail.com,lplough@usc.edu
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-oUC1350469
Unique identifier
UC1350469
Identifier
etd-PloughLoui-333.pdf (filename)
Legacy Identifier
etd-PloughLoui-333
Dmrecord
660319
Document Type
Dissertation
Format
theses (aat)
Rights
Plough, Louis Valentine
Internet Media Type
application/pdf
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Access Conditions
The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the author, as the original true and official version of the work, but does not grant the reader permission to use the work if the desired use is covered by copyright. It is the author, as rights holder, who must provide use permission if such use is covered by copyright.
Repository Name
University of Southern California Digital Library
Repository Location
USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA
Repository Email
cisadmin@lib.usc.edu
Tags
gene action
genetic load
high fecundity
inbreeding
larval mortality
life-history
metamorphosis
oyster
QTL mapping