The Pennsylvania State University

The Graduate School

Eberly College of Science

Diversity, Stability and Connectivity of Symbiodinium

Populations at Various Spatial Scales

A Dissertation in

Biology

by

Daniel T. Pettay

© 2011 Daniel T. Pettay

Submitted in Partial Fulfillment

of the Requirements

for the Degree of

Doctor of Philosophy

August 2011

The dissertation of Daniel T. Pettay was reviewed and approved* by the following:

Todd C. LaJeunesse Assistant Professor of Biology Dissertation Adviser

Stephen W. Schaeffer Associate Professor of Biology Chair of Committee

Andrew F. Read Professor of Biology

Timothy C. Reluga Assistant Professor of Mathematics

Douglas R. Cavener Professor of Biology Department Head of Biology

*Signatures are on file in the Graduate School

ii ABSTRACT

Coral-algal symbioses construct and maintain entire ecosystems, making symbiotic dinoflagellates (Symbiodinium) among the most abundant microbial eukaryotes found on coral reefs. Despite decades of research on these associations the genotypic diversity, dispersal and population differentiation of Symbiodinium, along with the fine-scale dynamics between populations of hosts and their endosymbionts, remain unclear. Modern population genetic approaches can elucidate these patterns and substantially advance our understanding of these associations. Coral bleaching is the loss of endosymbiotic dinoflagellates due to physiological stress, such as changes temperature or light. Bleaching can lead to disease outbreaks and mass mortality and has caused widespread degradation of many reef coral communities. However, some host-symbiont combinations show higher resistance to these perturbations and their proliferation has the potential to influence the response of reef communities to climate change. Symbionts of the phylogenetic grouping Clade D represent the best known and most geographically widespread example, where host colonies associating with these symbionts exhibit tolerance to physiological stress. Given the predicted changes in the global climate, a better understanding these stress-tolerant symbionts is imperative to understanding the future of coral reef ecosystems. From a population perspective, the migration and dispersal of thermally tolerant species may influence how algal-coral symbioses respond to climate warming. To address our lack of knowledge the population processes of two species belonging to clade D were studied, the pan-Pacific species Symbiodinium glynii (D1) and the globally distributed Symbiodinium trenchii (D1a). S. glynni associates exclusively with Pocillopora spp. (i.e., host specialist) ranging from the Eastern Pacific, where the symbiosis is regionally abundant and the predominant reef- builder, to the Western Pacific and Gulf of Thailand. In contrast, S. trenchi associates with a large diversity of invertebrates in the Western Atlantic, Indian and Pacific Oceans. The intra-colony diversity, prevalence and stability of S. glynni multilocus genotypes in association with populations of Pocillopora from two sites in the Gulf of California was investigated by applying microsatellite markers developed for both host and symbiont. The genetic diversity and allelic frequencies in reef populations of S. glynni remained stable over time, with common clone lineages persisting and no temporal population subdivision (ΦPT = 0.021 & -0.003) found over three years. Collections from circular plots showed no statistical correlation between related Pocillopora individuals and their associations with particular S. glynni genotypes, with no spatial structuring or clonal aggregation across a reef for the symbiont. From permanent linear transects, samples were analyzed from multiple locations within a colony and some were re-sampled ~1 year later. Many colonies of Pocillopora (~ 53%) were dominated by a single S. glynni genotype and tended to associate with the same symbiont genotype(s) temporally, while colony ramets often possessed unrelated symbiont genotypes. The lack of correlation between host and symbiont lineages and the possibility for temporal turnover of symbiont genotypes suggests that these genotypic combinations are potentially highly flexible. Expanding to S. glynni populations distributed throughout the Eastern Pacific, a total of four hundred and two multilocus haploid genotypes were acquired, of which two hundred and fifteen were distinct. Clone, or strain, diversity was high at most collection sites, with a

iii single genotype detected in most samples (> 80%). Minimal population subdivision was detected throughout the Eastern Tropical Pacific (ETP) encompassing 1000’s of square kilometers from mainland Mexico and Clipperton Atoll to the Gulf of Panama and the Galapagos Islands. The general lack of population differentiation over these distances appears to correspond with recent findings of extensive host genetic connectivity and indicates Pocillopora larvae, which maternally inherit their symbionts, aid in the dispersal of this symbiont. Unlike a confluent host population, high latitude populations in the subtropical Gulf of California (Sea of Cortez) were strongly differentiated from the ETP. The tropical (ETP) and subtropical (Gulf California, GoC) Eastern Pacific populations of S. glynni were compared to Palau in the west Pacific, and the Gulf of Thailand the Indo-Pacific. Bayesian clustering resolved several distinct populations (ΦRT = 0.280, p = 0.0001) including the previously described latitudinal separation between the GoC and ETP, which were well differentiated from populations in the west Indo-Pacific region. Additional structuring between Palau, the Gulf of Thailand and the Andaman Sea was statistically significant (ΦPR = 0.179, p = 0.0001). Latitudinal environments and large oceanic expanses appear to have structured S. glynni into genetically differentiated populations. However, several genotypes from Palau were similar in allelic composition to genotypes found in the GoC, suggesting that occasional dispersal may occur between east and west Pacific regions. While S. glynni associates with Pocillopora type 1 across the GoC and ETP, it associates with a different species of Pocillopora (genetically-defined as type 5) in western regions and this may in part explain the longitudinal differentiation. Contrasting the host specificity and vertical transmission of S. glynni, Symbiodinium trenchi (D1a) is a host generalist and horizontally transmitted. This symbiont differs ecologically from other genetically defined types within Clade D in that it is the most geographically widespread, associates with a diversity of host taxa and in the Caribbean S. trenchi has increased in prevalence and abundance in coral communities under prolonged exposure to raised sea surface temperatures. The unique ecology and distribution of this symbiont prompted an investigation into its genotypic diversity and population structuring in representative reefs from the Indo-Pacific and throughout the Atlantic using twelve diallelic microsatellite loci. Populations of S. trenchi from the Indo-Pacific were genetically diverse (I > 1.14) with most genotypes the products of sexual recombination (R > 0.80). In contrast, populations throughout the greater Caribbean were limited in genetic diversity (I = 0.67) and excessively clonal (R = 0.28). While no multilocus genotypes (MLGs) were shared among reefs in the Indo-Pacific, one unusual MLG was found forty-four times in the Caribbean and from every sample location except the Flower Garden Banks. Genotypic diversity is typically high among other Caribbean Symbiodinium spp. suggesting something is unusual about the population dynamics and dispersal of S. trenchi in this region. Severe thermal anomalies including recent events in 2005 and 2009 possibly facilitated expansion of a particularly aggressive clone lineage and/or populations of this symbiont, which may stem from a small number of opportunistic introductions originating from the Indo-Pacific. Data such as these on the population structure, genetic diversity and stability of coral/algal symbioses are necessary for inferring the processes by which these symbioses evolve and respond to changes in environmental conditions. Most symbiont populations are comprised of genotypically diverse and divergent individuals, with some symbiont genotypes found numerous times on a reef and as a result of asexual reproduction. The one extreme exception was the Caribbean S. trenchi population, which contained low allelic diversity and

iv is dominated by a single MLG or symbiont clone. The stability of a population’s genotypic diversity appears stable and the dispersal of this diversity controlled by environmental conditions and oceanic currents. While both symbiont species show the propensity for long- distance dispersal, it is believed that co-dispersing with its Pocillopora host larvae enhances S. glynni’s dispersal, while S. trenchi must rely on its own intrinsic dispersal ability. Lastly, comparisons between the genetic identity of Pocillopora and S. glynni demonstrate flexibility at the individual and population level. Although these symbionts are vertically transmitted, the symbiosis is not fixed and individual associations can change in response to environmental conditions and chance events, providing the mechanism for symbiosis evolution. These findings highlight the complex nature and the species-specific patterns of population structuring and dispersal for coral-algal symbioses.

v

TABLE OF CONTENTS

List of Tables ...... vii List of Figures...... ix Acknowledgements ...... x

Introduction...... 1 Background Information...... 1 Statement of Purpose ...... 6 References...... 8

Chapter 1. GENOTYPIC DIVERSITY AND SPATIAL-TEMPORAL DISTRIBUTION OF SYMBIODINIUM CLONE LINEAGES IN AN ABUNDANT REEF CORAL ...... 14 Abstract...... 14 Introduction...... 15 Materials & Methods ...... 18 Results...... 26 Discussion ...... 39 Conclusions ...... 46 References...... 47

Chapter 2. LONG-RANGE DISPERSAL AND HIGH-LATITUDE ENVIRONMENTS INFLUENCE THE POPULATION STRUCTURE OF A “STRESS-TOLERANT” ENDOSYMBIONT HARBORED BY POCILLOPORA IN THE EASTERN PACIFIC ...... 55 Abstract...... 55 Introduction...... 56 Materials & Methods ...... 59 Results...... 67 Discussion ...... 78 References...... 86

Chapter 3. POPULATION STRUCTURE OF THE PAN-PACIFIC POCILLOPORA SYMBIONT, SYMBIODINIUM GLYNNI...... 94 Abstract...... 94 Introduction...... 95 Materials & Methods ...... 97 Results...... 103 Discussion ...... 112 References...... 119

vi Chapter 4. HIGH CLONALITY AND LOW GENETIC DIVERSITY CHARACTERIZE POPULATIONS OF SYMBIODINIUM TRENCHI IN THE CARIBBEAN RELATIVE TO THE INDO-PACIFIC...... 126 Abstract...... 126 Introduction...... 127 Materials and Methods...... 129 Results...... 136 Discussion ...... 144 Conclusions ...... 150 References...... 151

Appendices ...... 159 Appendix A. Allele frequencies for S. glynni by reef in the GoC...... 159 Appendix B. Plots of genotypic diversity w/in colonies based on sampling strategy....161 Appendix C. Allele frequencies for S. glynni by location in the Eastern Pacific...... 162 Appendix D. Summary plots from the method by Evanno et al. for S. glynni ...... 165 Appendix E. Principal coordinate analysis of S. glynni populations in the E. Pacific ...167 Appendix F. Allele frequencies for S. glynni by location across the Pacific...... 168 Appendix G. Allele frequencies for S. trenchi by location ...... 171

vii LIST OF TABLES

Table 1. Description of microsatellite loci used in Chapter 1 ...... 27

Table 2. Summary of AMOVA for S. glynni temporal populations in the GoC ...... 28

Table 3. Summary statistics for S. glynni linear transects and circular plots in the GoC ...... 29

Table 4. Within colony diversity of S. glynni MLGs in Pocillopora from the GoC ...... 33

Table 5. Description of microsatellite loci used in Chapter 2 ...... 68

Table 6. Summary statistics for S. glynni MLGs by location across the Eastern Pacific ...... 70

Table 7. Pairwise comparisons (ΦPT) between S. glynni populations by host morphospecies...75

Table 8. Summary of AMOVA for S. glynni populations in the Eastern Pacific...... 76

Table 9. Pairwise comparisons (ΦPT) between S. glynni populations in the Eastern Pacific.....76

Table 10. Description of microsatellite loci used in Chapter 3 ...... 94

Table 11. Summary statistics for S. glynni MLGs by location across the Pacific...... 107

Table 12. Summary of AMOVA for S. glynni populations in the Pacific ...... 111

Table 13. Pairwise comparisons (ΦPT) between S. glynni populations across the Pacific...... 111

Table 14. Description of microsatellite loci used in Chapter 4 ...... 126

Table 15. Summary statistics for S. trenchi MLGs by location...... 137

Table 16. Summary of AMOVA for S. trenchi populations ...... 143

Table 17. Pairwise comparisons (ΦPT) between S. trenchi populations...... 143

viii LIST OF FIGURES

Figure 1. Description of sampling locations and strategies for S. glynni in the GoC...... 19

Figure 2. An example of artificial mixtures of S. glynni genotypes for one locus ...... 22

Figure 3. The distribution of S. glynni clone diversity and stability in Pocillopora...... 34

Figure 4. Representative circular plots from Punta Galeras and Isla Gaviotas ...... 36

Figure 5. Correlograms showing the spatial correlation for circular plots in the GoC...... 37

Figure 6. The distribution of 15 different S. glynni genotypes in a single Pocillopora clone....38

Figure 7. The distribution of genetic distances between individuals for ISLG and PGAL ...... 40

Figure 8. Sampling locations of S. glynni populations in the Eastern Pacific...... 60

Figure 9. Structure plots of S. glynni populations in the Eastern Pacific...... 72

Figure 10. Biogeographic map of Eastern Pacific showing S. glynni regional populations ...... 73

Figure 11. Structure plots of S. glynni populations according to host morphospecies ...... 74

Figure 12. Graphs depicting SSTs and PAR for all sampling locations in the Eastern Pacific ..77

Figure 13. Sampling locations of S. glynni populations in the Pacific ...... 98

Figure 14. Structure plots of S. glynni populations in the Pacific ...... 109

Figure 15. Sampling locations of S. trenchi populations ...... 131

Figure 16. Occurrence of increasing clone sizes for S. trenchi based on location ...... 138

Figure 17. Distribution and abundance of S. trenchi clones across the Caribbean...... 139

Figure 18. The distribution of genetic distances between S. trenchi individuals by location .....140

Figure 19. Principal coordinate analysis of S. trenchi MLGs ...... 141

Figure 20. UPGMA tree based on genetic distance between S. trenchi MLGs...... 142

ix ACKNOWLEDGEMENTS

I would like to acknowledge my dissertation advisor, Todd LaJeunesse, for all of his help and instruction throughout this entire process. Similarly, I would like to acknowledge my committee (S. Schaeffer, A. Read and T. Reluga) who provided valuable advice and assistance during the planning, implementation and writing of my dissertation. All of the current and past members of the LaJeunesse Lab (R. Smith, J. Pinzón, E. Sampayo, and D.

Wham) contributed to the successful completion of my dissertation in numerous ways.

Several faculty members external to my committee from Florida International University (D.

Mills), the Pennsylvania State University (I. Baums, T. Ridgway) and the College of

Charleston (A. Strand) provided assistance and advice during the research and development and statistical analysis portion of my research. This research would not have been possible without the help of numerous collaborators including: H. Reyes-Bonilla, A. L. Cupul-

Magaña, M. Walther and A. López-Pérez for sample collections and logistics in Mexico; P.

Medina-Rosas and Kristie Kaiser provided samples from the Clipperton Atoll; M. Torchin,

C. Schloeder (Smithsonian Tropical Research Institute-Panama) and J. Pinzón for the collection of samples in Panama; I. Baums and A. Baker provided samples from the

Galapagos Islands; N. Kongjandtre collected samples in the Andaman Sea and Gulf of

Thailand; B. Fitt, N. Phongsuwan, B. Brown, D. Obura and O. Hoegh-Guldberg for sample collections and logistics in Tanzania and Thailand; C. Finney and H. Oxenford for sample collections in Barbados; D. Kemp for sample collection in Mexico; R. Smith for sample collections in the Flower Garden Banks. I would like to thank Drew Wham for assistance with the development of additional Clade D microsatellite loci. The significant help and advice provided by the DNA Core Facility at FIU and the Genomics Core Facility at PSU

x was greatly appreciated. This research was funded in part by Florida International University,

The Pennsylvania State University, the National Science Foundation (IOB 544854 and OCE-

09287664), DHHS/NIH/NIEHS (ES11181) the PADI Foundation and a PSU Braddock

Award.

xi DEDICATION

There are so many people that have influenced and inspired me during my life, that it would be impossible to list them all.

First and foremost, I would like to dedicate the completion of my dissertation to my wife, Marcia, who has survived nearly a decade of my graduate research. She has been particularly patient with me and wonderfully supportive throughout this long ordeal. I definitely could not have made it through it without her. Likewise, I would like to dedicate this dissertation to the brown boys (Grover and Stagger Lee) who never failed to keep me company and provide entertainment when I needed to take my mind off things. While

Grover patiently stayed by my side during the writing, I know he is happy it is finally over.

I also dedicate this work to all of my family and friends who supported me over the years. Specifically, my parents gave me the initial push to pursue a career in marine biology and have supported my decisions the entire way. Growing up, they instilled in me an interest in the natural world and biology that has persisted through my life. Together with my brother Scott, we developed an appreciation for the natural world as the family traveled around some of the country’s most inspirational locations. Similarly, my grandmother’s love of the earth sciences sparked my interest in the sciences and inspired me to make a career in science.

Lastly, I especially dedicate this dissertation to Popeye and Stagger Lee, who passed away during its completion. Both were inspirations to me (in their own ways) during their lives and provided unending support. I wish they could have been here with me to see this degree come to fruition.

xii EPIGRAPH

“The night was clear and the moon was yellow, and the leaves came tumbling down.”

Stagger Lee, written by Lloyd Price

xiii Introduction

Diversity, Stability and Connectivity of Symbiodinium Populations at Various Spatial Scales

Background Information

Coral/algal symbiosis

The mutualistic symbiosis between scleractinian coral and photosynthetic dinoflagellates within the genus Symbiodinium is a critical component of shallow tropical marine ecosystems (Trench 1993). These dinoflagellates have become the primary producers of an ecosystem characterized by high light, warm temperatures and oligotrophic waters

(Muscatine & Porter 1977). This productivity is made possible by the fact the coral host not only provides the symbiont with the nutrients necessary for photosynthesis, but also provides a “habitat” which allows these dinoflagellates to reach large population sizes (millions per cm2 host tissue) (Drew 1972, Muscatine & Porter 1977). In return, the symbionts provide the host with nutrients in the form photosynthates that may cover up to 100% of the host’s metabolic needs (Muscatine & Porter 1977, Muscatine et al. 1981). In addition to the nutritional benefits provided by this symbiosis, the holobiont (combination of symbiont and host) is capable of constructing massive skeletal frameworks of calcium carbonate that form the structural basis for the entire ecosystem (Hoegh-Gulberg 1999). These ecosystems provide nursery grounds and safe habitat for numerous commercially import fisheries, along with billions of dollars in economic revenue worldwide (Wilkinson 2004).

1 Symbiodinium diversity, host specificity and biogeography

Symbiodinium spp., also referred to as zooxanthellae, were first cultured and verified as being gymnodinoid dinoflagellates in 1944 by Kawaguti (1944). The taxonomy, morphology and life cycle was later formally described as a new genus by Freudenthal

(1962). Originally it was believed that one pandemic species, Symbiodinium microadriaticum Freudenthal, occurred in all host taxa (Taylor 1974). Morphological, biochemical, physiological and behavioral differences between cultured isolates of

Symbiodinium originating from various hosts indicated that S. microadriaticum was in fact composed of multiple species (Schoenberg & Trench 1980a, b & c). The use of molecular techniques later showed that the genus could be divided into distinct phylogenetic lineages or clades (based on sequence variation of the small ribosomal subunit, 18S rDNA, and named A through H) and that the divergence between these clades was comparable to Orders among free-living dinoflagellates (Rowan & Powers 1991, 1992, reviewed by Coffroth & Santos

2005). Subsequent research using chloroplast and mitochondrial markers has confirmed these broad taxonomic groupings (Santos et al. 2002, Takabayashi et al. 2004), and the use of more variable regions, such as the internal transcribed spacer (ITS) regions 1 & 2 of the ribosomal array, has revealed numerous species within each of these clades (LaJeunesse

2001). The use of the ITS-DGGE fingerprinting to distinguish between species is supported by clear host specific (i.e. habitat), geographic and bathymetric patterns (LaJeunesse 2001,

LaJeunesse et al. 2003, Sampayo et al. 2007). With the use of microsatellite loci, these species have been shown to be reproductively isolated, further validating the use of the ITS region as a species level marker (Santos et al. 2004, Pettay & LaJeunesse 2007, Sampayo et al. 2009, LaJeunesse et al. 2010a).

2 Initial research using coral colonies within the Montastrea annularis complex indicated the possibility of a large degree of flexibility between host and symbiont associations (Rowan & Knowlton 1995). This species complex is capable of associating with all four clades (A, B, C & D) known to associate with scleractinian coral (Rowan &

Knowlton 1995, Toller et al. 2001). While these results demonstrated flexibility, the associations correlated with geographic and bathymetric factors (Toller et al. 2001). Further research of the symbiont at the species level has indicated that specificity between partners seems to be the rule and not the exception. Of course, there is a range of specificity displayed by both partners, with some hosts and symbionts having strict specificity for the other partner and others appearing much more flexible (reviewed by Baker 2003). This range in flexibility has led to the use of the terms specialists and generalists to describe symbiont species that are found to associate with only one host or multiple hosts, respectively (Toller et al. 2001, LaJeunesse 2002, LaJeunesse et al. 2003). Recent research comparing microsatellite flanker sequences indicate that some of these generalists may in fact be a species complex and that the ITS region may be too conserved to distinguish certain symbiont species (Santos et al. 2004, Finney et al. 2010).

While each geographic region has its own compliment of Symbiodinium species, large regions such as oceans or seas tend to show an overall pattern of distribution based on clade.

For example, the Caribbean Sea is dominated by and possesses the largest diversity of Clade

B Symbiodinium spp., although there are members of the other clades throughout the

Caribbean (LaJeunesse et al. 2003). Similar patterns exist on the Great Barrier Reef (clade

C), the Indo-Pacific (clade C and D) and the Red Sea (clade A and C) (Baker 2003,

LaJeunesse et al. 2004, Baker et al. 2008, LaJeunesse unpublished). These large

3 biogeographic patterns are most likely the result of historical contingencies and long standing environmental conditions in each region (LaJeunesse 2005, LaJeunesse et al. 2010a).

Species diversification within an ocean or sea, however, appears to be more the result of local environmental conditions (light, temperature and circulation) and strong ecological specialization to particular host taxa (LaJeunesse 2005, LaJeunesse et al. 2010a). These patterns of distribution indicate speciation by isolation, both by distance and habitat

(primarily the host).

Coral bleaching and the future of coral reefs

While tropical coral reefs have been extremely successful in the face of harsh environmental conditions in the geologic past, coral/dinoflagellate symbioses are adapted to the narrow environmental range of tropical climates (Hoegh-Gulberg 1999). Sudden or prolonged shifts in one of these environmental parameters can have devastating effects on the function of the symbiosis and, consequently, the ecosystem as a whole (Hoegh-Gulberg

1999). One consequence of this stress is coral bleaching. Coral bleaching is described as either a reduction in concentration of the photosynthetic pigments of the symbionts or a reduction in the number of symbionts themselves within the tissues of the host (Brown 1997,

Fitt et al. 2001). The causes of coral bleaching, while numerous, are typically the increase of sea surface temperatures and/or UV light (Brown 1997, Fitt et al. 2001). The consequences of thermal bleaching are oxidative damage to both the host and symbiont and a reduction of host nutrition, and ultimately a decline in host health (Brown 1997, Hoegh-Guldberg 1999).

While corals can recover from a bleaching event, a prolonged event can cause coral mortality and the collapse of an entire reef ecosystem (Hoegh-Guldberg 1999).

4 Over the past several decades, the severity and frequency of thermal bleaching events have increased (Hoegh-Guldberg 1999). This fact has led to the question of whether coral reefs are capable of surviving as the planet continues to warm. The current trajectory of global temperature change predicts sea surface temperatures to increase as much as 2°C over the next 100 years, with bleaching occurring if temperatures rise at little as 1°C over the local seasonal maxima (Hoegh-Guldberg 1999). In response, much research has focused on the ability of coral and their symbionts to adapt (or acclimate) to increasing temperatures.

Increasing evidence suggests that certain species of Symbiodinium within Clade D are more thermally and stress tolerant and those colonies hosting D symbionts are better able to survive and recover from bleaching (Glynn et al. 2001, Jones et al. 2008, LaJeunesse et al.

2009, 2010b). This tolerance was first suggested when it was discovered that Clade D symbionts are sometimes common in western Caribbean Montastraea colonies located in marginal habitats and at the boundaries of their species distribution (Toller et al. 2001). At the same time, the first observation of differential bleaching and colony mortality (between clade C & D) corresponding to symbiont clade identity was reported for Pocillopora colonies in the Eastern Pacific (Glynn et al. 2001). Given the genetic and physiological diversity within the genus, the possibility that there are some symbionts that can tolerate thermal stresses is not surprising, and if corals can acquire these symbionts it may provides a mechanism for coral reefs to survive (Adaptive Bleaching Hypothesis; Buddemeier & Fautin

1993).

Clade D symbionts have now been shown to have a global distribution and evidence has mounted that some species within the clade do in fact appear to be stress and/or thermally tolerant (Rowan 2004). Similar observations of differential bleaching based on the presence

5 or absence of D have now been reported for reefs worldwide (Jones et al. 2008, LaJeunesse et al. 2010b). Various Clade D species are now known to be ecologically important to reefs in the Sea of Cortez, Persian Gulf and around Thailand, all of which contain reefs exposed to extreme environmental conditions such as large temperature fluctuations, air-exposed reef flats and high turbidity (LaJeunesse et al. 2007, LaJeunesse et al. 2008, Mostafavi et al.

2007). Given their potential ability to provide host resilience in the face of stress, these symbionts are of major interest. Characterizing the stability and flexibility of these associations and the population structure and dispersal of these stress tolerant symbionts is therefore imperative to fully understand the ability of corals, and the reef ecosystems they construct, to survive the predicted warming of the planet.

Statement of Purpose

A comprehensive understanding of the mechanisms driving change and/or maintaining stability among coral symbioses is imperative given the predictions that coral bleaching events will increase in frequency and severity in the coming decades causing significant degradation of tropical near-shore coral reef communities (Hoegh-Guldberg et al.

2007). Particular Symbiodinium spp. appear to be more stress tolerant, and this tolerance has allowed these species and their host to survive while others have perished in the face of environmental change (Jones et al. 2008, Sampayo et al. 2008, LaJeunesse et al. 2009,

2010b). Continued research on these stress tolerant species is critically important, since these species may provide one mechanism for coral reefs to adapt to climate change.

The maintenance of genetic diversity through connectivity between populations is another mechanism important in how corals and their symbionts adapt to a changing

6 environment. Genetic diversity improves a population’s chances to adapt to changing environmental conditions, as it may equate to physiological diversity and allow particular individual or clone genotypes to survive while others do not. Since Symbiodinium spp. periodically undergo sexual reproduction (Santos et al. 2003), the migration of genetically unique individuals between local populations could result in new genetic combinations and it is within these interbreeding units that adaptive evolution occurs (Hartl & Clark 1997).

Lastly, connectivity between reef populations may provide the opportunity to replenish severely impacted reefs and understanding the potential for dispersal of these organisms is particularly important in the design of effective marine protected areas (Palumbi 2003).

The ability of coral symbioses to adapt to climate change may also reside in the degree of specificity and stability between individual host and symbiont genotypes (Goulet &

Coffroth 2003). Host and symbiont genotypes that are more flexible or able to form new partner combinations during periods of environmental change may have an advantage by forming better-adapted symbioses (Buddemeier & Fautin 1993). Alternatively, if genotypic associations between host and symbiont are fixed following initial infections, then the potential partner recombination may only occur during the early stages of development

(Coffroth et al. 2001, Goulet & Coffroth 2003). While the stability of a symbiont species in an individual colony over time has been shown (Thornhill et al. 2006, Sampayo et al. 2008), little is known concerning the stability a particular clonal genotype (Goulet & Coffroth

2003). Describing stability at this resolution is crucial for understanding the patterns of clonal diversity and predicting the possibility of partner recombination below the species level.

7 To address questions concerning the genetic diversity and connectivity of

Symbiodinium populations, the stability between particular host and symbiont genotypes, and ultimately predict the adaptive potential of particular Symbiodinium spp. to a changing global climate, microsatellite loci were developed for clade D Symbiodinium, for the specific use on the closely related species Symbiodinium glynii (D1) and Symbiodinium trenchi (D1a). S. glynni associates exclusively with the coral genus Pocillopora in the throughout the Pacific

(LaJeunesse et al. 2010b, LaJeunesse unpub.). This symbiont is particularly common in the

Eastern Pacific and can be found in approximately 70 to 100% of the colonies, depending on the geographic location (LaJeunesse et al. 2008, LaJeunesse et al. submitted). Pocillopora is ubiquitous throughout the Pacific (Veron 2000), and is the dominant reef coral in the Eastern

Pacific, covering 100% of the benthos on many reefs (Glynn & Ault 2000). S. trenchi has a global distribution and unlike S. glynni can be found in a wide diversity of invertebrate hosts

(LaJeunesse et al. 2009, 2010a). Both species are abundant in the environment and geographically widespread (LaJeunesse et al. 2009, 2010a, b). Their abundance and distribution make these symbionts ideal for studying both local (within and between adjacent reefs) and regional (between biogeographic regions) genetic connectivity. Lastly, both symbionts have been shown to be stress-tolerant (e.g., LaJeunesse et al. 2009, 2010b), making them ecologically important and particularly relevant for studying the effects of climate change.

References

Baker A (2003) Flexibility and specificity in coral-algal symbiosis: diversity, ecology, and biogeography of Symbiodinium. Annual Review of Ecology and Systematics, 34, 661-689.

8 Baker AC, Jones SH, Lee TS (2008) Symbiont diversity in Arabian corals and its relation to patterns of contemporary and historical environmental stress. Fauna of Saudi Arabia, In press, ???

Brown BE (1997) Coral bleaching: causes and consequences. Coral Reefs, 16, S129-S138.

Buddemeier RW, Fautin DG (1993) Coral Bleaching as an Adaptive Mechanism. BioScience, 43, 320-326.

Coffroth MA, Santos SR (2005) Genetic diversity of symbiotic dinoflagellates in the genus Symbiodinium. Protist, 156, 19-34.

Coffroth MA, Santos SR, Goulet TL (2001) Early ontogenetic expression of specificity in a cnidarian-algal symbiosis. Marine Ecology Progress Series, 222, 85-96.

Drew EA (1972) The Biology and Physiology of Alga-Invertebrate Symbiosis. II. The Density of Symbiotic Algal Cells in a Number of Hermatypic Hard Corals and Alcyonarians from Various Depths. Journal of Experimental Marine Biology and Ecology, 9, 71-75.

Finney JC, Pettay DT, Sampayo EM, et al. (2010) The Relative Significance of Host– Habitat, Depth, and Geography on the Ecology, Endemism, and Speciation of Coral Endosymbionts in the Genus Symbiodinium. Microbial Ecology 60, 250-263.

Fitt WK (1985) Effect of different strains of the zooxanthella Symbiodinium microadriaticum on growth and survival of their coelenterate and molluscan hosts. Proc. 5th Int. Coral Reef Congr., 6, 131-136.

Fitt WK, Brown BE, Warner ME, Dunne RP (2001) Coral bleaching: interpretation of thermal tolerance limits and thermal thresholds in tropical corals. Coral Reefs, 51-65.

Freudenthal H (1962) Symbiodinium gen. nov. and Symbiodinium microadriaticum sp. nov., a zooxanthellae; taxonomy, life cycle, and morphology. Journal of Protozoology, 9, 45-52.

Glynn P, Ault J (2000) A biogeographic analysis of the far eastern Pacific coral reef region. Coral Reefs, 19, 1-23.

Glynn P, Maté J, Baker A, Calderón M (2001) Coral bleaching and mortality in Panama and Ecuador during the 1997-1998 El NinoSouthern Oscillation Event: spatial/temporal patterns and comparisons with the 1982-1983 event. Bulletin of Marine Science, 69, 79-109.

9 Glynn PW, Gassman NJ, Eakin CM, et al. (1991) Reef coral reproduction in the eastern Pacific: Costa Rica, Panama and Galapagos Islands (Ecuador). Part I. Pocilloporidae. Marine Biology, 109, 355–368.

Goreau TJ (1990) Coral bleaching in Jamaica. Nature, 343, 6257.

Goulet T, Coffroth M (2003) Genetic composition of zooxanthellae between and within colonies of the octocoral Plexaura kuna, based on small subunit rDNA and multilocus DNA fingerprinting. Marine Biology, 142, 233-239.

Goulet TL, Coffroth MA (2003) Stability of an octocoral-algal symbiosis over time and space. Marine Ecology-Progress Series, 250, 117-124.

Hartl DL, Clark AG (1997) Principles of Population Genetics, 3 edn. Sinauer Associates, Inc., Sunderland, Massachusetts.

Hoegh-Guldberg O (1999) Climate change, coral bleaching and the future of the world's coral reefs. Marine and Freshwater Research, 50, 839-866.

Hoegh-Guldberg O, Mumby PJ, Hooten AJ, et al. (2007) Coral Reefs Under Rapid Climate Change and Ocean Acidification. Science, 318, 1737-1742.

Howells EJ, van Oppen MJH, Willis BL (2009) High genetic differentiation and cross-shelf patterns of genetic diversity among Great Barrier Reef populations of Symbiodinium. Coral Reefs, 28, 215–225.

Jones AM, R. B, van Oppen MJH, Mieog JC, Sinclair W (2008) A community change in the algal endosymbionts of a scleractinian coral following a natural bleaching event: field evidence of acclimatization. Proceedings of the Royal Society B: Biological Sciences, 275, 1359-1365.

Kawaguti S (1944) On the physiology of reef corals VII. Zooxanthella of the reef coral is Gymnodinium sp., Dinoflagellata; its culture in vitro. Palao Tropical Biological Station Studies, 2, 675—679.

LaJeunesse T (2005) “Species” radiations of symbiotic dinoflagellates in the Atlantic and Indo-Pacific since the Miocene-Pliocene transition. Molecular Biology and Evolution, 22, 570 - 581.

LaJeunesse T, Reyes-Bonilla H, Warner M (2007) Spring “bleaching” among Pocillopora in the Sea of Cortez, Eastern Pacific. Coral Reefs, 26, 265-270.

LaJeunesse T, Reyes-Bonilla H, Warner M, et al. (2008) Specificity and stability in high latitude eastern Pacific coral-algal symbioses. Limnology and Oceanography, 53, 719-727

10 LaJeunesse TC, Pettay DT, Sampayo EM, et al. (2010a) Long-standing environmental conditions, geographic isolation and host-symbiont specificity influence the relative ecological dominance and genetic diversification of coral endosymbionts in the genus Symbiodinium. Journal of Biogeography 37, 785-800.

LaJeunesse TC, Smith R, Walther M, et al. (2010b) Host–symbiont recombination versus natural selection in the response of coral–dinoflagellate symbioses to environmental disturbance. Proceedings of the Royal Society B 277, 2925-2934.

LaJeunesse TC, Smith RT, Finney J, Oxenford H (2009) Outbreak and persistence of opportunistic symbiotic dinoflagellates during the 2005 Caribbean mass coral 'bleaching' event. Proceedings of the Royal Society B: Biological Sciences 276, 4139- 4148.

LaJeunesse TC (2001) Investigating the biodiversity, ecology, and phylogeny of endosymbiotic dinoflagellates in the genus Symbiodinium using the ITS region: in search of a "species" level marker. Journal of Phycology, 37, 866-880.

LaJeunesse TC (2002) Diversity and community structure of symbiotic dinoflagellates from Caribbean coral reefs. Marine Biology, 141, 387-400.

LaJeunesse TC, Loh WKW, van Woesik R, et al. (2003) Low symbiont diversity in southern Great Barrier Reef corals, relative to those of the Caribbean. Limnology and Oceanography, 48, 2046-2054.

LaJeunesse TC, Trench RK (2000) Biogeography of two species of Symbiodinium (Freudenthal) inhabiting the intertidal sea anemone Anthopleura elegantissima (Brandt). Biological Bulletin, 199, 126-134.

Magalon H, Baudry E, Huste´ A, Adjeroud M, Veuille M (2006) High genetic diversity of the symbiotic dinoflagellates in the coral Pocillopora meandrina from the South Pacific. Marine Biology, 148, 913–922.

Mostafavi PG, Fatemi SMR, Shahhosseiny MH, Hoegh-Guldberg O, Loh WKW (2007) Predominance of clade D Symbiodinium in shallow-water reef-building corals off Kish and Larak Islands (Persian Gulf, Iran). Marine Biology, 153, 25–34.

Muscatine L, McCloskey LR, Marian RE (1981) Estimating the daily contribution of carbon from zooxanthellae to coral animal respiration. Limnology and Oceanography, 26, 601-611.

Muscatine L, Porter JW (1977) Reef corals: Mutualistic symbioses adapted to nutrient poor environments. Bioscience, 27, 454-460.

11 Oxenford HA, Roach R, Brathwaite A, et al. (2008) Quantitative observations of a major coral bleaching event in Barbados, Southeastern Caribbean. Climatic Change, 87, 435–449.

Palumbi SR (2001) The ecology of marine protected areas. In: Marine ecology: the new synthesis (eds. Bertness M, Gaines SD, Hay ME), p. 509–530. Sinauer, Sunderland, Massachusetts, USA.

Palumbi SR (2003) Population Genetics, Demographic Connectivity, and the Design of Marine Reserves. Ecological Applications, 13, S146–S158.

Pettay D, LaJeunesse T (2007) Microsatellites from clade B Symbiodinium spp. specialized for Caribbean corals in the Genus Madracis. Molecular Ecology Notes, 7, 1271-1274.

Pettay DT, LaJeunesse TC (2009) Microsatellite loci for assessing genetic diversity, dispersal and clonality of coral symbionts in ‘stress-tolerant’ clade D Symbiodinium. Molecular Ecology Resources, in press.

Reyes-Bonilla H, Carriquiry JD, Leyte-Morales GE, Cupul-Magana AL (2002) Effects of the El Nino-Southern Oscillation and the anti-El Nino event (1997-1999) on coral reefs of the western coast of Mexico. Coral Reefs, 21, 368-372.

Rowan R (2004) Coral Bleaching: Thermal adaptation in reef coral symbionts. Nature, 430, 742-742.

Rowan R, Knowlton N (1995) Intraspecific diversity and ecological zonation in coral-algal symbiosis. Proceedings of the National Academy of Sciences of the United States of America, 92, 2850-2853.

Rowan R, Knowlton N, Baker A, Jara J (1997) Landscape ecology of algal symbionts creates variation in episodes of coral bleaching. Nature, 388, 229-230.

Rowan R, Powers D (1992) Ribosomal RNA sequences and the diversity of symbiotic dinoflagellates (zooxanthellae). Proceedings of the National Academy of Science USA, 89, 3639-3643.

Rowan R, Powers DA (1991) A molecular genetic classification of zooxanthellae and the evolution of animal-algal symbioses. Science, 251, 1348-1351.

Sampayo EM, Dove S, LaJeunesse TC (2009) Cohesive molecular genetic data delineate species diversity in the dinoflagellate genus Symbiodinium. Molecular Ecology, 18, 500–519.

12 Sampayo EM, Ridgway T, P. B, Hoegh-Guldberg O (2008) Bleaching susceptibility and mortality of corals are determined by fine-scale differences in symbiont type. Proceeding of the National Academy of Science, 105, 10444–10449.

Santos S, Taylor D, Kinzie III R, et al. (2002) Molecular phylogeny of symbiotic dinoflagellates inferred from partial chloroplast large subunit (23S)-rDNA sequences. Mol Phylogenet Evol, 23, 97-111.

Santos SR, Coffroth MA (2003) Molecular genetic evidence that dinoflagellates belonging to the genus Symbiodinium Freudenthal are haploid. Biological Bulletin, 204, 10-20.

Santos SR, Kinzie III RA, Sakai K, Coffroth MA (2003) Molecular Characterization of Nuclear Small Subunit (18S)-rDNA Pseudogenes in a Symbiotic Dinoflagellate (Symbiodinium, Dinophyta). Journal of Eukaryotic Microbiology, 50, 417-421.

Santos SR, Shearer TL, Hannes AR, Coffroth MA (2004) Fine-scale diversity and specificity in the most prevalent lineage of symbiotic dinoflagellates (Symbiodinium, Dinophyceae) of the Caribbean. Molecular Ecology, 13, 459-469.

Schoenberg DA, Trench RK (1980) Genetic variation in Symbiodinium (=Gymnodinium) microadriaticum Freudenthal, and specificity in its symbiosis with marine invertebrates. I. Isoenzyme and soluble protein patterns of axenic cultures of S. microadriaticum. Proceedings of the Royal Society of London B, 207, 405-427.

Schoenberg DA, Trench RK (1980) Genetic variation in Symbiodinium (=Gymnodinium) microadriaticum Freudenthal, and specificity in its symbiosis with marine invertebrates. II. Morphological variation in S. microadriaticum. Proceedings of the Royal Society of London B, 207, 429-444.

Schoenberg DA, Trench RK (1980) Genetic variation in Symbiodinium (=Gymnodinium) microadriaticum Freudenthal, and specificity in its symbiosis with marine invertebrates. III. Specificity and Infectivity of S. microadriaticum. Proceedings of the Royal Society of London B, 207, 445-460.

Takabayashi M, Santos SR, Cook CB (2004) Mitochondrial DNA phylogeny of the symbiotic dinoflagellates (Symbiodinium, Dinophyta). Journal of Phycology, 40, 160-164.

Taylor DL (1974) Symbiotic marine algae: taxonomy and biological fitness. In: Symbiosis in the Sea (ed. Vernberg WB), pp. 245-262. University of South Carolina Press, South Carolina.

13 Thornhill D, LaJeunesse TC, Kemp D, Fitt W, Schmidt GW (2006) Multi-year, seasonal genotypic surveys of coral-algal symbioses reveal prevalent stability or post- bleaching reversion. Marine Biology, 148, 711-722.

Toller W, Rowan R, Knowlton N (2001) Zooxanthellae of the Montastraea annularis species complex: Patterns of distribution of four taxa of Symbiodinium on different reefs and across depths. Biological Bulletin, 201, 348-359.

Trench RK (1993) Microalgal-Invertebrate Symbioses: A Review. Endocytobiosis & Cell Research, 9, 135-175.

Veron JEN (2000) Corals of the world Australian Institute of Marine Science, Queensland.

Wilkinson CR (2004) Status of the Coral Reefs of the World: 2004. Australian Institute of Marine Science, Townsville.

14 CHAPTER 1

Genotypic diversity and spatial-temporal distribution of Symbiodinium clone lineages in an abundant reef coral

Abstract

Coral-algal symbioses construct and maintain entire ecosystems, yet despite decades of research the fine-scale dynamics between populations of coral hosts and their dinoflagellate endosymbionts remain unclear. Modern population genetic approaches can elucidate these patterns and substantially advance our understanding of these associations. By applying microsatellite markers developed for both host and symbiont, we investigated the intra- colony diversity, prevalence and stability of Symbiodinium glynni (type D1) multilocus genotypes in association with dense populations of Pocillopora from two sites in the Gulf of California. The genetic diversity and allelic frequencies in reef populations of S. glynni remained stable over time, with common clone lineages persisting and no temporal population subdivision (ΦPT = 0.021 & -0.003) found over three years. Collections from circular plots showed no statistical correlation between related Pocillopora individuals and their associations with particular S. glynni genotypes, with no spatial structuring or clonal aggregation across a reef for the symbiont. From permanent linear transects, samples were analyzed from multiple locations within a colony and some were re-sampled ~1 year later. Many colonies of Pocillopora (~ 53%) were dominated by a single S. glynni genotype and tended to associate with the same symbiont genotype(s) temporally, while colony ramets often possessed unrelated symbiont genotypes. The lack of correlation between host and symbiont lineages and the possibility for temporal turnover of symbiont genotypes suggests that these genotypic combinations are potentially highly flexible. Data such as these on the population structure, genetic diversity and stability of coral/algal symbioses are necessary for inferring the processes by which these symbioses evolve and respond to changes in environmental conditions.

14 Introduction

The mutualistic association between invertebrates and endosymbiotic dinoflagellates, genus Symbiodinium, is arguably the most ecologically important and prevalent symbiosis in the tropical marine environment (Trench 1993). Scleractinian corals, in particular, form the carbonate framework that provides shelter, food and/or nursery grounds for an extensive diversity of tropical marine species. As a result of an ecosystem maintained by these symbioses, Symbiodinium is one of the most abundant microbial eukaryotes on coral reefs.

The use of population genetic markers such as microsatellites provides the precise tools necessary to address questions about the genotypic diversity of Symbiodinium at multiple spatial and temporal scales. Blooms of free-living photosynthetic unicellular eukaryotes like diatoms may contain many thousands of unique genotypes (Evans et al.

2005; Rynearson & Armbrust 2005; Rynearson et al. 2006). Unlike most free-living photosynthetic species, Symbiodinium populations persist year-round, reaching densities in the millions of cells per square centimeter of host tissue (Drew 1972; Stimson et al. 2002) and analogous to a phytoplankton bloom. Population densities of Symbiodinium inside a host colony may fluctuate seasonally responding to changes in temperature, light, and/or nutrients, but their growth appear relatively stable and regulated (Hoegh-Guldberg and Smith 1989;

Jones & Yellowlees 1997; Fitt et al. 2000). Increased rates of cellular division will coincide with the growing regions of the host colony as symbionts populate new and vacant host tissues (Jones & Yellowlees 1997). The mechanisms may be numerous, but host suppression along with symbiont expulsion most likely explains the maintenance of symbiont densities

(Drew 1972; Jones & Yellowlees 1997). While biomass (i.e., cell densities) can change seasonally, virtually nothing is known concerning the stability of genotypic diversity found

15 within individual host colonies.

The genetic diversity of symbiont populations within a coral colony is often studied using markers that approximate a species level of resolution (LaJeunesse 2001; Sampayo et al. 2009). A majority of hosts associate with a particular symbiont, or small subset of genetic types, in a given geographic location or environment (e.g., LaJeunesse 2002; Sampayo et al.

2008; LaJeunesse et al. 2010a). Changes in these associations appear to be influenced by severe thermal stress events that may initiate the opportunistic rise of rare and/or heterologous symbionts whose residence is often short-lived when environmental conditions return to normal (Thornhill et al. 2006; Jones et al. 2008; LaJeunesse et al. 2009; LaJeunesse et al. 2010b). Examinations of Symbiodinium populations at the genotype level (i.e., individual clonal line) within and between colonies are still preliminary. Initial studies using genomic fingerprinting indicated that symbiont populations were homogenous throughout a colony and stable for many years (Goulet & Coffroth 2003a; Goulet & Coffroth 2003b).

Recent studies using microsatellite loci appear to confirm these seminal findings (Santos et al. 2003; Pettay & LaJeunesse 2007, 2009; Thornhill et al. 2009; Kirk et al. 2009; Andras et al. 2011), however most of these investigations based their conclusions on a minimal number of microsatellite loci and/or from single samples taken from individual colonies.

The mode of transmission, abundance and dispersal ability directly influences the evolutionary potential of parasites and/or pathogens (Barrett et al. 2008). Similarly, there is some suggestion that mixed clone infections of the malaria parasite Plasmodium spp., part of a sister phylum to dinoflagellates, leads to intra-host competition for resources and increased virulence as these clones contend for limited resources (Mackinnon & Read 2004; Havryliuk

& Ferreira 2009). Coral-algal symbioses may respond similarly where the interactions

16 between host and symbiont and their respective life histories and spatial structure affect their coevolution. As seen with other microorganisms, distinct genotypes within a species of

Symbiodinium may exhibit variation in physiology giving some lineages an adaptive advantage under the selective pressures created by a warming climate (Bell et al. 2006;

LaJeunesse et al. 2010b). Therefore, the diversity and stability of Symbiodinium clones within a colony and on a reef may dictate the adaptive potential of the symbiosis to environmental changes.

Pocillopora type 1 is a dominant shallow-water coral found throughout much of the tropical and subtropical Eastern Pacific and associates predominantly with the stress-tolerant clade D symbiont, Symbiodinium glynni (type D1; sensu LaJeunesse et al. 2008; LaJeunesse et al. 2010b; Pinzón & LaJeunesse 2011). This coral-algal symbiosis and its ecological response to climate change has emerged as one of the better characterized associations in the world (Glynn et al. 2001; Reyes-Bonilla et al. 2002; Iglesias-Prieto et al. 2004; Baker et al.

2004; LaJeunesse et al. 2007; Chavez-Romo & Reyes-Bonilla 2007; LaJeunesse et al. 2008;

LaJeunesse et al. 2010; Pinzón & LaJeunesse 2011; Pettay & LaJeunesse submitted). Due to the ecological importance of this association, nine microsatellite markers were used to examine the genotypic diversity, degree of clonality, fine-scale spatial distributions and temporal stability of Symbiodinium glynni populations present within and among colonies of

Pocillopora at two reef locations near La Paz in the Gulf of California, Mexico. Although the symbionts in this association are vertically transmitted from parent to offspring, recent data indicate that the host is capable of associating with different regionally adapted S. glynni populations (Pettay & LaJeunesse submitted). It is hypothesized that the flexibility between host and symbiont seen at the population level will be maintained at the individual level. To

17 test this hypothesis of flexibility, host genotypes were also acquired to examine their influence on the distributions of S. glynni clonal lines. These fine-scale analyses will help to further our understanding of the specificity and stability of these associations, and reveal patterns necessary for inferring ecological processes important to their evolution.

Materials and Methods

Sample Collection

Samples were collected from two sites separated by 10 km, Isla Gaviotas (ISLG) and

Punta Galeras (PGAL), at depths ranging from two to ten meters in the Gulf of California

(GoC) near La Paz, Mexico (Figure 1). Branch fragments (~2 cm2) from colonies of

Pocillopora type 1 (sensu Pinzón & LaJeunesse 2011) were collected by SCUBA and preserved in a high salt, 20% DMSO buffer (Seutin et al. 1991) and stored at -20 °C until

DNA extraction. In May 2006, colonies exhibiting different morphologies (representing different host genotypes) were sampled from three 25 meter linear transects (142 colonies total) established at each site. At this time, three colonies at each site were sampled intensively at the tops and bottoms of branches from the north, east, south, west and center of the colony. To examine more broadly the genotypic diversity of S. glynni populations within a single colony, we returned in May of 2007 to sample from 10 colonies being monitored at each site (totaling 20 colonies and 60 samples). Three independent samples from separate regions of each colony were collected. In August 2008 these collections were conducted again but this time attention was paid to the orientation of the colony. Samples were acquired from the north, east, south and west sides of 10 colonies from PGAL. Nine of the ten colonies were then resampled nine months later in May 2009.

18

Figure 1. Description of sampling locations, Pocillopora host and sampling strategies used to investigate the fine-scale clonal diversity, stability and spatial structure of S. glynni in the Gulf of California. (a) The two sites in the GoC, Punta Galeras (PGAL) and Isla Gaviotas (ISLG), and the prevailing currents into and out of the Bay of La Paz. (b) Picture of a colony of Pocillopora at PGAL. (c) The two sampling designs used to collect single samples per colony; (1) 25m transect with haphazard sampling along the transect (2006) and (2) 20 diameter circular plot place on top of transect and samples collected according to random coordinates (2009). Three transects/plots were collected per location (PGAL & ISLG). (d) Pictorial description for colonies sampled multiple times around the colony. Multiple sampling involved either three random locations approximately equidistant from each, four locations corresponding to north, east, south and west, and five locations corresponding to north, east, south, west and center of the colony. (e) Colonies collected at five locations also consisted of sampling each branch at each location from both the top and the bottom of the branch.

19 Lastly, a random sampling strategy was employed to examine the spatial distribution of symbiont clones over a reef and the genotypic relation between host and symbiont.

Twenty colonies were sampled from three twenty-meter diameter circular plots (> 30 meters apart) positioned over the original linear transects established in 2006 (Figure 1). Prior to sampling, random coordinates were generated for each plot with a precision of one degree and fifty centimeters. From the center point of each plot, a compass and measuring tape were used to locate each sampling point. If no colony was found a particular position, the next set of coordinates was used until twenty colonies were sampled from each plot.

DNA Extraction and Symbiont Identification

Nucleic acid extractions were conducted using a modified Promega Wizard genomic

DNA extraction protocol (LaJeunesse et al. 2003). The dominant resident symbiont was identified by denaturing gradient gel electrophoresis (DGGE) fingerprinting of the partial

5.8S and internal transcribed spacer (ITS) region 2 (LaJeunesse 2002). The region was amplified using a touch-down thermal cycle profile with the primers “ITS2clamp” and

“ITSintfor2” (LaJeunesse & Trench 2000), and the PCR products resolved on denaturing gels

(45 – 80% of 7M urea & 40% formamide) using a CBScientific system (Del Mar, CA) for 16 hours at 115 volts. Samples where S. glynni (D1) was detected were utilized for this study.

Microsatellite Analysis

Nine independently sorting polymorphic microsatellite loci originally developed for clade D Symbiodinium were utilized to examine the clonal diversity and spatial structure of S. glynni (D1Sym11, D1Sym14, D1Sym17, D1Sym34, D1Sym67, D1Sym77a, D1Sym77b,

20 D1Sym87, and D1Sym92; Pettay & LaJeunesse 2009; Wham et al. 2011). Each locus was amplified in separate 10 µl reaction volumes containing 2.5 µM dNTP’s, 0.2 U Taq DNA

Polymerase (New England Biolabs or Amplitaq), 1x Mg-free Buffer (2.5 mM MgCl2), 1 µM of the forward and reverse primers and ~ 50 ng of DNA template. PCR amplification conditions consisted of an initial denaturing step of 94°C for 2 min, 32 cycles of 94°C for 15 s, annealing temperature (Ta, see Supplementary Table 1) for 15 s, 72°C for 15 s, and a final extension of 72°C for 5 min. Following amplification, fragment sizes were analyzed on an

ABI 3730 Genetic Analyzer (Applied Biosystems, Foster City, CA) using a 500 bp standard

(LIZ-labeled) at the Pennsylvania State University Genomics Core Facility. Fragments sizes were visually analyzed using GeneMarker v.1.51 (SoftGenetics, State College, PA). The presence of multiple peaks was interpreted to indicate that a sample contained more than one haploid genotype. Multiple alleles at a particular locus were scored when a second clear peak was found in the expected size range and was at least one-third the size of the dominant peak. This method was used to prevent the inclusion of microsatellite stutter peaks, which do not represent true alleles. Based on tests using artificial mixtures of single genotypes of differing concentrations, all loci show the ability to resolve the numerically dominant S. glynni genotype based on fragment peak height (Figure 2). Therefore, there appears to be little to no bias in fragment amplification based on size (i.e., Havryliuk et al. 2008) and the dominant MLG is represented by the dominant fragment peaks, which were used for statistical analyses (see below). The occasional exclusion of small ambiguous peaks may occasionally cause a low background genotype to be missed, however this conservative approach prevents the wrong overestimation of genetic diversity (Anderson et al. 2000). In repeatedly sampled colonies where there were mixtures of two or more symbiont genotypes,

21

Figure 2. An example of artificial mixtures of S. glynni genotypes for one microsatellite locus (D1Sym87) to show their resolution and demonstrate there is little preferential amplification based on fragment size. The ratio of the smaller allele to the larger allele is given in the corner of each electropherogram.

22 the most likely identities of each individual genotype was determined by comparing replicate samples where a single MLG was present and comparing relative peak heights between alleles at loci with multiple peaks. However, the true identity of “background” MLGs may not be completely reliable and must be interpreted cautiously.

Statistical Analysis

Clonal Diversity and Structure

Basic measurements of clonal diversity and spatial organization, along with spatial structure analyses were calculated using GenClone (v. 2.0, Arnaud-Haond & Belkhir 2007).

The descriptive statistics included the number of samples (N), number of genotypes (G), clonal richness (R), clonal heterogeneity as the Simpson complement (D*), clonal evenness as the Simpson evenness (V), and the Pareto distribution of clonal membership. Clonal richness is calculated as R = (G – 1)/(N – 1) and ranges from zero to one, with a value of one indicating all samples are distinct MLGs (i.e., no clones). The Simpson complement, which equals one minus the Simpson index (λ), describes the probability that two random individuals within a sampling are distinct MLGs. Simpson evenness describes the distribution of MLGs with respect to their abundance and ranges from zero to one, with one indicating all MLGs are in equal abundance. For clonal organisms the distribution of ramets into clonal size classes, or the number of samples with identical MLGs, follows the Pareto, or power law, distribution (Arnaud-Haond et al. 2007). The slope of the regression (β) line is a function of both the evenness and richness of a sample set and is a way to graphically represent the heterogeneity of clone abundance. Descriptions of these indices, their derivation and benefits are reviewed in Arnaud-Haond et al. (2007).

23 Spatial analyses of clonal structure were conducted for all circular plots using

GenClone (v. 2.0) to obtain a clonal aggregation index (Ac), an edge effect (Ee), the clonal subrange and a kinship coefficient (F(r)) for each distance class (r) for spatial autocorrelation at both the ramet and genet level. Additionally, a multivariate approach to spatial autocorrelation, whereby multiple genetic loci are assessed simultaneously, was conducted using GenAlEx (version 6.4, Peakall & Smouse 2006) and was run using the multiple pops option. This option not only analyzes each plot individually, but also groups all plots on the same reef together, providing increased sample size and statistical power, and is useful when common processes contributing to spatial structure are believed to be occurring at each location (Peakall & Smouse 2006), like would be expected from multiple circular plots on a single. For these analyses replicate MLGs were included in the dataset making it a spatial analysis at the ramet level. The significance of the autocorrelation analyses was tested using a permutation method, as implement by GenClone (1000 permutations) and GenAlEx (10000 permutations).

Temporal Stability of Reef Populations

To examine whether populations of S. glynni change in genotypic diversity, composition or structure over time, allelic frequencies and the degree of differentiation between populations collected in 2006 (linear transects) with those collected in 2009 (circular plots) were calculated for each site using GenAlEx. For the calculation of allele frequencies and genetic similarity, duplicate MLGs were removed. These allelic frequencies were also used to calculate haploid genetic diversity (h), which gives an indication that two individuals drawn at random will be genetically different, and information index (I), which is a measure

24 of allelic diversity (Peakall and Smouse 2006). Population differentiation between time- points was assessed using an analysis of molecular variance (AMOVA) for both sites. Using these time-point populations for each reef, AMOVA’s along with a permutation procedure were performed to test for significant difference in genetic diversity between populations

(Excoffier et al. 1992). The AMOVA produces variance components along with Φ-statistics

(F-statistic analogs), which partition genetic variation at different hierarchical levels

(Excoffier et al. 1992). The significance of the genetic variance and Φ-statistics were then tested using 10000 permutations.

Correlation between Host and Symbiont Genotypes

Pocillopora transmit their symbionts vertically from mother to offspring (Glynn et al.

1991; Chavez-Romo & Reyes-Bonilla 2007), which may lead to a correlation between host and symbiont genotypes. To test if a statistical relationship exists, Mantel tests were performed on genetic distance based matrices of pairwise comparisons between microsatellite MLGs of both Pocillopora and S. glynni individuals from the circular plots.

The Mantel tests were conducted using GenAlEx and the statistical significance of the correlation coefficient between the two matrices was tested using 9999 random permutations with an α = 0.05 and a null hypothesis that no significant relationship exists between the host and symbiont genetic distances. Two mantel tests were conducted, one per reef, using the combined data from the three circular plots with each combination of host to symbiont genotype represented only once (i.e., holobiont clones were removed). The Pocillopora data was acquired from the analysis of six microsatellite loci from Pinzón et al. (submitted).

25 Results

Symbiont identification

The microsatellite analyses targeted a Symbiodinium whose genome is characterized by a single numerically dominant ITS variant inferred from ITS2-DGGE fingerprinting (type

D1; sensu Thornhill et al. 2007; Sampayo et al. 2009; LaJeunesse et al. 2010a). Genetic comparisons with this and other Clade D Symbiodinium, and its apparent unique association with Pocillopora indicates that this symbiont is a distinct operational taxonomic unit,

Symbiodinium glynni (LaJeunesse et al. 2010a; Wham et al. 2011; unpublished data). While in most of the samples analyzed by ITS2-DGGE detected only S. glynni, C1b-c, a Clade C

Pocillopora specific type was present in a small proportion of samples (< 5%).

Allelic and genotypic diversity, and temporal structure from linear transects and circular plots

The total number of alleles per locus acquired from the linear transects (2006) ranged from one to eight (PGAL) and two to five (ISLG). Similarly, in 2009, one to six (PGAL) and two to six (ISLG) alleles per locus were obtained from the circular plot (Table 1). The effective number of alleles (Ae) per locus ranged from 1.00 to 4.80 (PGAL06), 1.32 to 3.00

(ISLG06), 1.00 to 3.83 (PGAL09) and 1.66 to 4.65 (ISLG09) (Table 1). Allele frequencies at each locus ranged from 0.028 to 0.778 (PGAL06), 0.032 to 1.000 (PGAL09), 0.048 to

0.762 (ISLG06) and 0.045 to 0.727 (ISLG09) (Appendix A). Allele frequencies remained similar and the dominance of a particular allele(s) persisted, in most instances, over the three- year period. Similarly, the temporal population differentiation for each site as calculated by an AMOVA indicated that the population at each reef was temporally stable and

26 Table 1. Description of microsatellite loci used in this study and their diversity according to sampling date and location including allele size range, number of alleles and Ae = effective alleles.

GAV 2006 GAV 2009 PUN 2006 PUN 2009

Size Range Number Ae Size Range Number Ae Size Range Number Ae Size Range Number Ae Locus (bp) of Alleles (bp) of Alleles (bp) of Alleles (bp) of Alleles D1Sym11 151 - 159 3 1.62 153 - 159 4 2.00 151 - 161 6 2.11 149 - 159 6 3.77

D1Sym14 179 - 183 3 2.38 179 - 185 3 2.07 173 - 185 5 3.27 173 - 185 5 2.94

D1Sym17 145 - 149 3 2.41 145 - 149 3 2.78 143 - 149 4 2.56 145 - 149 3 2.18

27 D1Sym34 388 - 404 5 2.92 388 - 400 4 3.06 380 - 404 8 4.80 380 - 400 5 3.60

D1Sym67 134 - 140 3 2.67 134 - 140 3 2.85 134 - 140 3 1.55 134 - 140 2 1.21

D1Sym77a 178 - 181 3 2.46 178 - 181 3 2.40 178 - 184 4 1.33 178 - 181 3 1.22

D1Sym77b 190 - 193 3 1.76 181 - 193 4 2.44 190 - 193 3 2.76 190 - 196 4 2.42

D1Sym87 244 - 256 4 3.00 244 - 264 6 4.65 244 - 264 6 4.29 244 - 264 6 3.83

D1Sym92 124 - 128 2 1.32 124 - 128 2 1.66 128 1 1.00 128 1 1.00

undifferentiated (ΦPT = 0.021 p = 0.150 for ISLG, and ΦPT = -0.003 p = 0.501 for PGAL;

Table 2). However, rare private alleles with a frequency < 0.100 (found once or twice) were found between sampling times for PGAL (10 alleles) and ISLG (9 alleles). The exception was two alleles from ISLG (181 at D1Sym14 & 260 at D1Sym87), which had allele frequencies > 0.200 in one year and was not found in the other. The mean haploid diversity

(H) was 0.56 for ISLG and 0.50 for PGAL, and the information index (I) was 0.98 for ISLG and 0.95 for PGAL (Table 3). The probability of identity (P(ID)) was calculated using allele frequency data at each locus and combining both the linear transects and circular plots for each site. The probability that two samples with the same MLGs may not have originated from the same clone lineage was 1.09 x 10-6 for ISLG to 3.1 x 10-6 for PGAL.

Table 2. Summary of AMOVA examining temporal population differentiation for ISLG and PGAL. Summary AMOVA Table - ISLG

Source df SS MS Est. Var. % Among Pops 1 3.866 3.866 0.056 2% Within Pops 41 109.110 2.661 2.661 98% Total 42 112.977 2.717 100%

Stat Value P(rand >= data) PhiPT 0.021 0.150

Summary AMOVA Table - PGAL

Source df SS MS Est. Var. % Among Pops 1 2.108 2.108 0.000 0% Within Pops 65 150.892 2.321 2.321 100% Total 66 153.000 2.321 100%

Stat Value P(rand >= data) PhiPT -0.003 0.501

28 Table 3. Summary statistics for 2006 linear transects and 2009 circular plots for Isla Gaviotas and Punta Galeras in the GoC. The statistics are as follows: number of samples (N), number of genotypes (G), clonal richness (R), information index (I), haploid diversity (H), clonal heterogeneity as Simpson complement (D*), clonal evenness as Simpson evenness (V), Pareto distribution 2 slope (β) and curve fit (r ) and significance (p), largest size class of clones (MAX), clonal subrange (CS), aggregation index (Ac) and the edge effect (Ee). Numbers in parentheses represent standard error for I & H, while they represent p-values for Ac & Ee. Dark grey boxes represent values not calculated or could not be calculated.

Freq. 2 # of D N G R I H D* V β r P MAX CS Ac Ee 0.017 0.124 1 0.95 18 8 0.412 0.791 0.526 0.42 1.00 0.000 8 13 (0.441) (0.108)

-0.194 0.478 2 0.95 18 6 0.294 0.791 0.767 0.44 0.77 0.023 6 12.7 (0.961) (0.006) 0.203 -0.101 29 Circular Plot 3 1.00 20 12 0.579 1.043 0.593 0.932 0.784 1.03 0.84 0.040 3 8.3 (0.021) (0.639)

All 0.97 56 22 0.382 (0.096) (0.038) 0.869 0.714 0.37 0.98 0.002 18

1 0.67 12 9 0.727 0.939 0.500 1.26 1.00 0.000 3

Isla Gaviotas (ISLG) 2 0.67 13 8 0.583 0.897 0.640 0.69 0.97 0.003 4

Linear 3 0.87 17 11 0.625 0.882 0.306 0.55 0.86 0.015 7

Transect 0.916 0.530 All 0.75 42 21 0.488 (0.091) (0.048) 0.918 0.762 0.58 0.96 0.009 12 0.143 -0.021

1 0.75 15 10 0.643 0.933 0.711 0.84 1.00 0.000 3 11 (0.105) (0.351) 0.036 -0.008 2 1.00 19 16 0.833 0.982 0.711 NA NA NA 2 8 (0.218) (0.269) 0.044 -0.044

Circular Plot 3 1.00 19 12 0.611 0.923 0.482 0.936 0.757 1.08 0.96 0.010 4 12.1 (0.354) (0.489) All 0.92 53 31 0.577 (0.187) (0.095) 0.962 0.858 0.94 0.94 0.027 7

1 0.68 16 15 0.933 0.992 0.000 NA NA NA 2

Punta Galeras (PGAL) 2 0.80 16 11 0.667 0.908 0.352 0.70 0.90 0.013 6

Linear 3 0.67 21 14 0.650 0.948 0.747 1.15 0.97 0.007 4

Transect 0.984 0.514 All 0.71 53 34 0.635 (0.183) (0.088) 0.971 0.856 1.03 0.99 0.004 7

Alleles were amplified for all samples at all nine loci except for D1Sym17,

D1Sym77a and D1Sym77b. The D1Sym17 locus did not amplify in just two instances perhaps due to DNA quality. The frequencies of non-amplifiable alleles for D1Sym77a and

D1Sym77b were high in samples of good DNA quality, and ranged from 0.132 to 0.333

(77a) and 0.245 to 0.714 (77b) and appear to be the result of accumulated sequence differences in the flanking regions of D1Sym77 that inhibit the proper amplification of divergent alleles (i.e., null alleles). For this reason, these null alleles were coded as a distinct allele and included in the data for subsequent analyses.

Table 3 provides a summary of allelic and genotypic diversity, along with measures of clonality, clonal distribution and size, for each transect and circular plot. Excluding the within colony sampling (see below), eighty-five different genotypes were differentiated among two hundred and four MLGs obtained from the ISLG and PGAL transects and circular plots. The number of distinct genotypes per transect/plot ranged from six to sixteen

(Table 3). Many genotypes of S. glynni were found more than once but were often restricted to a particular transect/plot. The maximum clone size found during a sampling time was seven and eighteen for PGAL and ISLG, respectively. The clone recovered in the highest frequency at ISLG in 2006 was also the most abundantly sampled clone on that same site in

2009 (12 times in 2006, 18 in 2009). This pattern was also observed at PGAL where a particular clone was found at high regularity during both sampling efforts (7 times in 2006, 7 in 2009). A total of six (6.1%, ISLG) and thirteen (12.3%, PGAL) clones were shared between the three-year sampling times, while nine distinct clones were shared between sites

(4.4%).

30 Genotypic richness (G) for both sites at each sampling time was 0.49 and 0.38 for

ISLG and 0.63 and 0.58 for PGAL in 2006 and 2009, respectively. Between individual transects and/or plots G was variable and ranged from 0.29 to 0.93 (Table 3). Clonal heterogeneity (D*, described in Materials and Methods) per reef was 0.92 and 0.87 for ISLG and 0.97 and 0.96 for PGAL in 2006 and 2009 respectively. Values of D* per transect and/or plots were variable and ranged from 0.79 to 0.98 (Table 3). Evenness (V) was highly variable per transect and/or plot ranging from 0 (only one clone) to 0.78, with V per reef equal to 0.76 and 0.71 for ISLG and 0.86 and 0.86 for PGAL in 2006 and 2009 (Table 3).

Evenness as measured by the Pareto distribution was also highly variable with β ranging from 0.42 to 1.26. In general, the distribution of clonal membership was more skewed by the dominance of a few clones in ISLG than in PGAL. While the estimates of R, D* and V within a reef in a particular year give an indication of relative clonality and evenness between transects or plots, these measures are particularly sensitive to sampling strategy, which differed between 2006 and 2009 (Arnaud et al. 2007).

Within colony clone diversity & stability

The total number of alleles per colony ranged from nine to twenty, with the number of loci differentiating clones within a colony ranging from one to eight. Many colonies appeared to host homogeneous clonal populations of S. glynni despite being sampled multiple times (Figure 3). The frequency of colonies with two or more MLGs, ranged between 0.07 and 0.70 depending on the site and the within-colony sampling strategy employed (Table 4). Colonies sampled only once tended to have a single S. glynni genotype

(0.87). More genotypes per colony were found when colonies were sampled multiple times

31 as different regions/branches of some colonies possessed different S. glynni genotypes.

Colonies sampled at three separate positions revealed a similar diversity to those sampled at ten places (Figure 3; Table 4; Appendix B). A lower frequency of colonies with mixed symbiont genotypes was observed when genotypes that differed at only a single locus were removed (Table 4).

The estimated number of S. glynni genotypes ranged from one to six in colonies sampled multiple times (average 1.80; Appendix B). Colonies at PGAL consistently possessed a higher incidence of S. glynni populations comprising multiple genotypes whereas colonies at

ISLG tended to be homogenous for one genotype (Table 4). The presence of a certain clone or mixtures clones in a particular area of the colony did not correspond to their orientation in the colony (data not shown). Clonal identity and diversity was identical in almost all cases (n

= 59 out of 60) at the apex and base of a branch indicating that irradiance has little influence on genotype distribution and sampling a single branch multiple times yielded no additional genotypic diversity.

Colonies sampled over nine months on their north, south, east and west sides (Figure

3) suggested that genotype diversity of S. glynni populations exhibited relative temporal stability at the colony level, but may under go dynamic changes across regions of a colony.

Most branches maintained the same symbiont clone(s) over time, with the frequency of a stable genotype(s) equaling 0.67 (north), 0.89 (east), 0.78 (south) and 0.67 (west). For entire colonies, two-thirds experienced a change in clone diversity in at least one location (e.g. north-side, west-side) within the colony, but generally involved the displacement of one clone by another already present in another region of the colony (Figure 3). However, in three of the nine colonies (33%) a new genotype was found that had not been detected in the

32

Table 4. Within colony diversity of S. glynni MLGs in Pocillopora from the GoC. Frequency of branches and colonies where multiple genotypes were detected is based on sampling strategy, or number of samples per colony, for each reef, with an average for both reefs together. The total number of branches (i.e., samples) and colonies sampled are indicated by nsamp and ncol, respectively. Frequencies mixtures were calculated for genotypic differences based on only one locus and for those with alleles differing at two or more loci. *Five branches of each colony were sampled, but each branch was sampled at both the top and bottom effectively making these colonies sampled ten times.

Samples Colonies Samples Colonies Sampling w/ w/ Location w/ w/ Strategy Mixtures Mixtures Mixtures Mixtures (≥2 loci) (≥2 loci) nsamp = 43 0.069 0.047 ISLG ncol = 43 0.069 0.047 nsamp =56 0.179 0.143 Single PGAL ncol = 56 0.179 0.143 nsamp = 99 0.131 0.101 Ave ncol = 99 0.131 0.101 nsamp = 30 0.233 0.167 ISLG ncol = 10 0.400 0.200 nsamp = 30 0.330 0.133 Three PGAL ncol = 10 0.700 0.500 nsamp =60 0.283 0.150 Ave ncol = 20 0.550 0.350 PGAL nsamp = 40 0.125 0.100 2008 ncol = 10 0.400 0.400 Four PGAL nsamp = 36 0.028 0.028 2009 ncol = 9 0.444 0.222 nsamp = 30 0.000 0.000 ISLG ncol = 3 0.000 0.000 nsamp = 30 0.200 0.133 Five* PGAL ncol = 3 0.667 0.667 nsamp =60 0.100 0.067 Ave ncol = 6 0.333 0.333

33

Figure 3. Examples of the distribution of S. glynni clone diversity and stability in Pocillopora type 1. (a) A single clone throughout the colony and both the top and bottom of the branch. (b) Two clones within a colony with zones dominated by a particular clone and an area of overlap in which both clones are found. (c) Colony displaying stability and homogeneity of clones over time. Numbers indicate the frequency stable associations with a symbiont genotype(s) occurred with regard to the location on the colony. (d) Colony showing stable mixtures of two clones over time. (e) Colony showing a mixture of two clones at t = 0 and a subsequent increase of one clone over another (top) and a replacement of two clones by a single “new” clone (right) after nine months.

34 colony nine months earlier (Figure 3).

Spatial Analyses

The multivariate spatial autocorrelation analyses at the ramet level indicated a lack of spatial structure at the scale measured (~ 20 m) for all plots at both ISLG and PGAL (Figure

4 & 5). While an occasional distance class displayed significant correlation (five for ISLG plots & three for PGAL plots), the results were often related to distance classes with low sample sizes, and there appeared no general trends in the correlation coefficients (r) for any plot. Similarly when the data from the three plots were combined per site there was no indication of spatial structure for ISLG or PGAL populations, with no significant distance classes for ISLG and only two significant for PGAL (Figure 5). Spatial autocorrelation analyses at both the ramet and genet level for each plot using GenClone also indicated a lack of spatial structure for all plots (data not shown).

Clonal aggregation (Ac) for each plot ranged from -0.19 to 0.20 and was non- significant except for ISLG #3 (0.20, p = 0.021), suggesting that symbiont clones are not spatially restricted. The maximum distance between clones for each plot ranged from 8.3 to

13 meters and was highest at ISLG #1 and PGAL #3. No significant edge effect (Ee) was found for all plots except ISLG #2 (0.48, p = 0.006) and ranged from -0.01 to 0.48. The general lack of significance for Ee indicates the sampling scheme was adequate and there is a low probability of sampling induced bias on measures such as genotypic richness (Arnaud-

Haond & Belkhir 2007).

35

Figure 4. Representative circular plots from Punta Galeras and Isla Gaviotas depicting the spatial orientation of and association between Pocillopora and S. glynni MLGs, with the shape representing host genotype and the color representing symbiont. MLGs found only once are indicated with a square (host) or solid black (symbiont). Some Pocillopora colonies associate with S. C1b-c and are shown with an open symbol.

36

Figure 5. Correlograms showing the spatial correlation rc at each distance class (1 m) for each circular plot individually and the combined analysis for each reef, 95% confidence interval about the null hypothesis of random distribution of S. glynni clones, and 95% confidence errors bars about rc as determined by bootstrapping.

Relationships between host and symbiont genotypes

While several specific MLGs of host and symbiont were found together multiple times (seven at ISLG & one at PGAL), mantel tests indicated there was no correlation between closely related host associating with closely related symbionts at either ISLG (R =

0.058, p = 0.198) or PGAL (R = -0.045, p = 0.203). Identical Pocillopora genotypes

(ramets) were often found to harbor different genotypes of S. glynni (up to fifteen), and, conversely, host genets (up to seven) frequently harbored the same S. glynni genotype

(Figure 4 & 6). However, host ramets found close to each other, likely the result fragmentation, often associated with the same symbiont genotype. One widespread

Pocillopora genet comprising 43 colonies distributed across 90 m of nearshore habitat at

37 ISLG associated with fifteen different S. glynni genotypes (Figure 6). Pairwise genetic distance between individual symbiont genotypes indicate that most are genotypically divergent and likely originate from sexual recombination (i.e., not apart of the same clonal lineage) (Figure 6 inset).

Figure 6. The distribution of 15 different S. glynni genotypes (indicated by color) across a single Pocillopora clone distributed over approximately 90 m of near shore habitat at depths ranging from 2 to 3 meters. Black shading designates unique multi locus genotypes. Inset (top) shows the distribution of the number of colonies associating with each particular clone of symbiont. Inset (bottom) shows the distribution of genetic distances (number of distinct alleles) between the individual symbiont clones and the frequency of occurrence for each distance class.

38 Discussion

The stress tolerant and ecologically dominant symbiosis between S. glynni and colonies of the Pocillopora type 1 is widespread throughout the eastern Pacific (LaJeunesse et al. 2010b; Pinzón & LaJeunesse 2011). With population genetic markers available for both partners, this symbiosis is uniquely suited for investigating fundamental biological attributes concerning symbiont population dynamics within and between host colonies and populations. The application of population genetic markers may therefore provide novel insight into the ecology and micro-evolutionary processes critical to symbiont speciation and their coevolution with host lineages.

High genotypic diversity and population stability of S. glynni in the GoC

The populations of S. glynni found at PGAL and ISLG are comprised of numerous clonal lineages generated by sexual recombination (Figure 7; LaJeunesse 2001, Santos & Coffroth

2003) with many of these lineages being rare while others are common, locally widespread and persistent in the ecosystem. The ninety-six distinct genotypes characterized from colonies sampled in this study represent a small subset of what exists in the population.

Based on this relatively limited sampling, the number of different clone lines that exist in the region and across the eastern Pacific must be incredibly high.

The genetic compositions of S. glynni populations at each location remained stable over a three-year period (Table 2; LaJeunesse et al. 2007; LaJeunesse et al. 2010b).

Furthermore, genotypes prevalent in the population in 2006 remained common in 2009.

While unique genotypes were found in each sampling time, allele frequencies remained

39 stable and these new genotypes likely represent variation due to random sampling, immigration and/or new genotypes created by sexual recombination. A lack of temporal differentiation also seems to be a common feature of most free-living eukaryotic microbes.

Dinoflagellate and diatom blooms, spanning numerous genera, have shown little to no temporal differentiation for periods ranging from days to several years (Evans et al. 2005;

Rynearson & Armbrust 2005; Rynearson et al. 2006; Casteleyn et al. 2009; Nagai et al.

2009; however see Godhe & Harnstrom 2010). Genetic diversity is high in these blooms, but unlike Symbiodinium few to no clones are shared over time.

Figure 7. The distribution of genetic distances (number of distinct alleles) between individuals and frequency of occurrence for each distance class for both reefs, ISLG & PGAL. The distributions are produced with repeated MLGs removed and are a combination of all samples from plots and transects for each reef.

40 Some clonal lineages of S. glynni appear to persist in the water column for long durations and can disperse 10’s of kilometers, or more, as is the case of “free-living” bloom- forming dinoflagellates and diatoms (Evans et al. 2005; Rynearson & Armbrust 2005;

Rynearson et al. 2006; Nagai et al. 2007, 2009; Casteleyn et al. 2009; Godhe & Harnstrom

2010; Lowe et al. 2010). While the high genotypic diversity found for S. glynni is analogous to that found in populations of a phytoplankton species bloom, the distribution of individual genotypes on a reef is partitioned into separate colonies across the host population. Due to the relatively stable nature of their host habitat, a genotype strain may remain in a colony and continuously generate propagules for many years. The persistence of a symbiont clonal lineage is unlike bloom forming species, and may further enhance their longevity and dispersal. What ultimately limits the longevity of a particular Symbiodinium genotype is unknown, and may ultimately be determined by random events and/or selective pressures.

No correlation between host and symbiont genotypes

The vertical transmission of symbionts in the lifecycle of Pocillopora could lead to the co-speciation of host and symbiont lineages (Glynn et al. 1991; Chavez-Romo & Reyes-

Bonilla 2007; Hirose et al. 2008). Monophyletic ‘species’ lineages of Symbiodinium specific to Pocillopora do exist (LaJeunesse 2005; Pinzón & LaJeunesse 2011), however mantel tests indicate that the relatedness of S. glynni genotypes do not correlate with that of its

Pocillopora host (Figure 4 & 6). The finding that some colonies harbored as many as six unrelated S. glynni genotypes (discussed below), and that host ramets of a genet often associated with distinct symbiont genotypes, indicates that flexibility is high among individual genotypes of host and symbiont. This lineage recombination is notable finding

41 and may allow symbioses to evolve as better-adapted individuals associate with each other

(LaJeunesse et al. 2010b).

Symbiont genotypic diversity and spatial/temporal stability inside a host colony

Are colonies typically dominated by a single Symbiodinium genotype and to what extent are these combinations stable? While multi-locus genotype data are limited, recent findings suggest most colonies associate with a single stable dominant genotype (Goulet and

Coffroth 2003; Santos et al. 2003; Pettay & LaJeunesse 2007, submitted; Kirk et al. 2009;

Thornhill et al. 2009; Pinzón et al. 2010), but others suggests extensive genotypic diversity in a majority of colonies (Magalon et al. 2006; Apprill and Gates 2007; Howells et al. 2009).

Examining multiple samples from single colonies with a high number of optimized loci tested against iso-clonal cell lines (Pettay and LaJeunesse 2009; Wham et al. 2011) suggests that a high proportion of colonies are dominated by a single temporally stable S. glynni genotype, but also indicate that many other colonies may harbor up to several genotypes

(Table 4).

Pocillopora colonies in the GoC are dominated by one or several S. glynni clone(s).

Analyses of samples from different sides/locations on a colony frequently found additional S. glynni genotypic diversity (Figure 3, Table 4). This genotypic diversity is greater than previously thought (Pettay & LaJeunesse submitted), but far less than indicated by previous analyses on Pocillopora (Magalon et al. 2006). Approximately half of the colonies possessed multiple genotypes (Table 4), each typically inhabiting separate regions of the colony unrelated to irradiance levels. In the six colonies where five branches were sub- sampled (Figure 1), the base of a branch, deep inside the colony, had the same genotype

42 found at the apex of the branch exposed to full sunlight in nearly all branches (59 out of 60).

This within branch homogeneity and finding the same symbiont genotype in nearby branches suggests that the symbioses of some colonies are a patch-work of a small number of S. glynni clones with zones dominated by one genotype that overlap in distribution and two or more genotypes occur (Figure 3). While some physiological differences may exist among S. glynni genotypes, there was no indication that a particular genotype favored a particular side/location of the colony (e.g., north vs. south) or zone of the branch canopy (upper vs. lower), suggesting that external environmental factors have a limited influence on S. glynni genotype distribution at the scale of the colony. Similar heterogeneity over a colony has been report for the Montastraea annularis complex in the Caribbean, where the tops and sides of some colonies harbored different genotypes (Thornhill et al. 2009).

Temporal stability of a resident symbiont genotype may be a common feature for coral-Symbiodinium associations (Goulet and Coffroth 2003; LaJeunesse et al. 2005; Kirk et al. 2005; Thornhill et al. 2009). Colonies of the Montastraea annularis species complex sampled repeatedly over time and at different locations within the colony were found to host mostly one genotype of Symbiodinium B1 that were seasonally stable over four years

(Thornhill et al. 2009). A majority (67 to 89%) of S. glynni genotypes sampled from a particular colony orientation (north, south, east or west) remained stable when re-sampled after nine months (Figure 3). Many of the observed temporal “changes” in symbiont identity involved the replacement of one genotype by another already present within the colony.

Finding of a novel genotype nine months later suggest the possibility that the introduction and spread of a distinct S. glynni genotype may occur rapidly. However, it was not possible

43 to determine whether instances of a “new” genotype represented an introduction or one simply missed during the initial sampling effort.

Associations may persist between specific genotypes of host and symbiont lasting many years (Goulet & Coffroth 2003; Thornhill et al. 2009). However, the within colony distribution patterns observed for S. glynni genotypes indicate the possibility that different combinations arise during the life of the host. The examination of multiple ramets of a host genet offers a means to determine the frequency by which a new S. glynni genotype becomes established within a colony. A large host genet comprising forty-three clone mates distributed over ninety meters of near shore habitat at ISLG associated with fifteen genetically distinct S. glynni genotypes (Figure 6; Pinzón et al. submitted). The high S. glynni genotypic diversity within a single host suggests that recombination between lineages of host and symbiont occurs with some regularity over the lifetime of a host genotype (Figure

4 & 6). This confirmed capability for switching among symbiont genotypes suggests the potential for rapid replacement of genotypes through selective sweeps within and across symbiont populations.

Finding numerous genetically diverse S. glynni genotypes within host genets and no correlation between host and symbiont relatedness supports work at the population level showing flexibility between individuals of this symbiosis. Population genetic analyses of both host and symbiont indicate that populations of Pocillopora type I and S. glynni are interconnected throughout the tropical eastern Pacific (Pinzon & LaJeunesse 2011; Pettay &

LaJeunesse submitted). While the host’s population continues into the GoC, there is a definitive genetic break for the symbiont despite the fact that dispersing host larvae carry the symbionts of their mother with them (Pettay & LaJeunesse submitted). Environmental

44 conditions have been implicated to explain this biogeographic break between tropical and sub-tropical S. glynni populations, whereby “less-adapted” genotypes are displaced by

“better-adapted” genotypes (Pettay & LaJeunesse submitted). It is therefore presumed that the co-migrating symbionts of immigrant host larvae are eventually out competed by locally adapted S. glynni genotypes.

Consequences of multi-clone infections

What limits the frequency and abundance of multiple Symbiodinium clones inside a colony? The patterns and consequences of multiple-clone infections of disease causing eukaryotes may provide insight into mechanisms regulating coral-algal mutualisms.

Plasmodium, the causative agent of malaria, is a haploid eukaryote belonging to

Apicomplexa, the sister phylum to Dinoflagellata. Multiple-clone infections of a single

Plasmodium spp. are common, with incidence as high as 80% in some populations (Read &

Taylor 2001; Bell et al. 2006; Havryliuk & Ferreira 2009). Experimental manipulations show that multiple-clone infections lead to competition for resources, resulting in higher virulence (Taylor et al. 1997, de Roode et al. 2005, Bell et al. 2006). Asymptomatic infections may become symptomatic and potentially dangerous to the host following the second infection of a competing parasite genotype (P. malariae; Bruce et al. 2007, however see Vardo-Zalik & Schall 2008). This increased virulence decreases host fitness and increases parasite transmission (Mackinnon & Read 2004). To avoid similar processes from occurring within Pocillopora, mechanisms may exist that limit or suppress multiple clone infections and thereby minimize a competitive arms race for host resources. A rapidly growing Symbiodinium clone that out competes all other genotypes could negatively affect

45 the growth and reproduction of the host. If multiple-clone infections within a colony lead to instability (i.e., increased resource use at the expense of the host) in the symbiosis, then there may selective advantages for mechanisms that minimize the genotypic diversity within individual colonies. Ultimately, there are numerous biotic and abiotic factors controlling the balance, stability and specificity observed between Symbiodinium populations and their host colony (Rowan & Knowlton 1995; Trench 1993; Douglas 1998; Coffroth et al. 2001).

Ongoing research analyzing genome expression libraries may provide insight into matters of cellular recognition and population control (e.g. Volstra et al. 2009).

Conclusions

The application of genetic markers that resolve among individuals ushers a new phase in coral-algal symbioses research and discovery. In summary, (1) populations of S. glynni comprised numerous clonal lineages generated by sexual recombination; (2) the genetic composition of the populations studied were stable for at least three years; (3) many genotypes were found only once while others were prevalent among host colonies at one or both locations; (4) temporally persistent genotypes characterized in at the beginning of the study were subsequently found three years later; (5) most samples contained a single genotype and (6) many colonies sampled at multiple locations also possessed a single detectable genotype (~ 53%), (7) while other colonies contained up to six, but rarely more than three distinguishable genotypes that were spatially partitioned across the colony and apparently uninfluenced by external irradiance; (8) genotypes persisted within a colony for at least nine months; and (9) no correlation was found between host and symbiont genotypes, indicating that lineages of host and symbiont switch through time, providing a mechanism for

46 colonies to acquire physiological variants adapted to local environmental conditions. The numerous observations made from this study alone should initiate a breadth of questions and inquiry. For example, do the patterns observed here differ substantially when symbionts in host communities are examined from warmer more stable environments? Or, how do these observations relate to coral colonies and other cnidarians with horizontal modes of symbiont acquisition? Continued progress in this emerging area of coral-algal symbiosis research will fundamentally advance our ecological and evolutionary understanding of these important partnerships.

References

Anderson T, Haubold B, Williams JT, et al. (2000) Microsatellite markers reveal a spectrum of population structures in the malaria parasite Plasmodium falciparum. Molecular Biology and Evolution 17, 1467-1482.

Anthony TG, Conway DJ, Cox-Singh J, et al. (2005) Fragmented population structure of Plasmodium falciparum in a region of declining endemicity. Journal of Infectious Diseases 191, 1558-1564.

Apprill AM, Gates RD (2006) Recognizing diversity in coral symbiotic dinoflagellate communities. Molecular Ecology 16, 1127-1134.

Arnaud-Haond S, Belkhir K (2007) GENCLONE: a computer program to analyse genotypic data, test for clonality and describe spatial clonal organization. Molecular Ecology Notes 7, 15-17.

Arnaud-Haond S, Duarte CM, Alberto F, Serrão EA (2007) Standardizing methods to address clonality in population studies. Molecular Ecology 16, 5115-5139.

Baker A, Starger C, McClanahan T, Glynn P (2004) Corals' adaptive response to climate change. Nature 430, 741.

Barrett L, Thrall P, Burdon J, Linde C (2008) Life history determines genetic structure and evolutionary potential of host-parasite interactions. Trends in Ecology & Evolution 23, 678-685.

47 Bell A, Roode J, Sim D, Read A (2006) Within-host competition in genetically diverse malaria infections: parasite virulence and competitive success. Evolution 60, 1358- 1371.

Bruce M, Macheso A, Galinski M, Barnwell J (2007) Characterization and application of multiple genetic markers for Plasmodium malariae. Parasitology 134, 637-650.

Buddemeier R, Fautin D (1993) Coral bleaching as an adaptive mechanism. Bioscience 43, 320-326.

Casteleyn G, Evans KM, Backeljau T, et al. (2009) Lack of population genetic structuring in the marine planktonic diatom Pseudo-nitzschia pungens (Bacillariophyceae) in a heterogeneous area in the Southern Bight of the North Sea. Marine Biology 156, 1149-1158.

Chávez-Romo HE, Reyes-Bonilla H (2007) Sexual reproduction of the coral Pocillopora damicornis in the southern Gulf of California, Mexico. Ciencias Marinas 33, 495- 501.

Coffroth M, Santos S, Goulet T (2001) Early ontogenetic expression of specificity in a cnidarian-algal symbiosis. Marine Ecology Progress Series 222, 85-96.

De Roode JC, Pansini R, Cheesman SJ, et al. (2005) Virulence and competitive ability in genetically diverse malaria infections. Proceedings of the National Academy of Sciences 102, 7624-7628.

Douglas A (1998) Host benefit and the evolution of specialization in symbiosis. Heredity 81, 599-603.

Drew E (1972) The biology and physiology of alga-invertebrate symbioses. II. the density of symbiotic algal cells in a number of hermatypic hard corals and alcyonarians from various depths. Journal of Experimental Marine Biology and Ecology 9, 71-75.

Evans K, Kühn S, Hayes P (2005) High levels of genetic diversity and low levels of genetic differentiation in North Sea Pseudonitzschia pungens (Bacillariophyceae) populations. Journal of Phycology 41, 506-514.

Excoffier L, Smouse P, Quattro J (1992) Analysis of molecular variance inferred from metric distances among DNA haplotypes: application to human mitochondrial DNA restriction data. Genetics 131, 479-491.

Fitt W, McFarland F, Warner M, Chilcoat G (2000) Seasonal patterns of tissue biomass and densities of symbiotic dinoflagellates in reef corals and relation to coral bleaching. Limnology and Oceanography 45, 677-685.

48 Fitt W, Trench R (1983) The Relation of Diel Patterns of Cell Division To Diel Patterns of Motility in the Symbiotic Dinoflagellate Symbiodinium Microadriaticum FREUDENTHAL in Culture. New Phytologist 94, 421-432.

Glynn P, Ault J (2000) A biogeographic analysis and review of the far eastern Pacific coral reef region. Coral Reefs 19, 1-23.

Glynn P, Mate J, Baker A, Calderon M (2001) Coral bleaching and mortality in Panama and Ecuador during the 1997-1998 El Niño-Southern Oscillation Event: spatial/temporal patterns and comparisons with the 1982-1983 event. Bulletin of Marine Science 69, 79-109.

Godhe A, Härnström K (2010) Linking the planktonic and benthic habitat: genetic structure of the marine diatom Skeletonema marinoi. Molecular Ecology 19, 4478-4490.

Goulet T, Coffroth M (2003a) Stability of an octocoral-algal symbiosis over time and space. Marine Ecology Progress Series 250, 117-124.

Goulet TL, Coffroth MA (2003b) Genetic composition of zooxanthellae between and within colonies of the octocoral Plexaura kuna, based on small subunit rDNA and multilocus DNA fingerprinting. Marine Biology 142, 233-239.

Havryliuk T, Ferreira M (2009) A closer look at multiple-clone Plasmodium vivax infections: detection methods, prevalence and consequences. Memórias do Instituto Oswaldo Cruz 104, 67-73.

Havryliuk T, Orjuela-Sánchez P, Ferreira MU (2008) Plasmodium vivax: Microsatellite analysis of multiple-clone infections. Experimental Parasitology 120, 330-336.

Hirose M, Reimer JD, Hidaka M, Suda S (2008) Phylogenetic analyses of potentially free- living Symbiodinium spp. isolated from coral reef sand in Okinawa, Japan. Mar Biol 155, 105-112.

Hoegh-Guldberg O, Smith GJ (1989) The effects of sudden changes in light, temperature and salinity on the population density metabolism and export of zooxanthellae from the reef corals Seriatopora hystrix and Stylophora pistillata. Marine Ecology Progress Series 129, 279–303.

Howells E, Van Oppen M, Willis B (2009) High genetic differentiation and cross-shelf patterns of genetic diversity among Great Barrier Reef populations of Symbiodinium. Coral Reefs 28, 215-225.

49 Iglesias-Prieto R, Reyes-Bonilla H, Rodríquez RR (2003) Effects of 1997-1998 ENSO on coral reef communities in the Gulf of California, Mexico. Geofisica Internacional 42, 467-471.

Jones A, Berkelmans R, Van Oppen M, Mieog J, Sinclair W (2008) A community change in the algal endosymbionts of a scleractinian coral following a natural bleaching event: field evidence of acclimatization. Proceedings of the Royal Society B: Biological Sciences 275, 1359-1365.

Jones R, Yellowlees D (1997) Regulation and control of intracellular algae (= zooxanthellae) in hard corals. Philosophical Transactions of the Royal Society B: Biological Sciences 352, 457-468.

Kemp DW, Fitt WK, Schmidt GW (2008) A microsampling method for genotyping coral symbionts. Coral Reefs 27, 289-293.

Kirk N, Andras J, Harvell C, Santos S, Coffroth M (2009) Population structure of Symbiodinium sp. associated with the common sea fan, Gorgonia ventalina, in the Florida Keys across distance, depth, and time. Marine Biology 156, 1609-1623.

LaJeunesse T (2001) Investigating the biodiversity, ecology, and phylogeny of endosymbiotic dinoflagellates in the genus Symbiodinium using the ITS region: in search of a" species" level marker. Journal of Phycology 37, 866-880.

LaJeunesse T (2002) Diversity and community structure of symbiotic dinoflagellates from Caribbean coral reefs. Marine Biology 141, 387-400.

LaJeunesse T, Bonilla H, Warner M, et al. (2008) Specificity and stability in high latitude eastern Pacific coral-algal symbioses. Limnology and Oceanography 53, 719-727.

LaJeunesse T, Loh W, Van Woesik R, et al. (2003) Low symbiont diversity in southern Great Barrier Reef corals, relative to those of the Caribbean. Limnology and Oceanography, 2046-2054.

LaJeunesse T, Reyes-Bonilla H, Warner M (2007) Spring “bleaching” among Pocillopora in the Sea of Cortez, Eastern Pacific. Coral Reefs 26, 265-270.

LaJeunesse TC (2005) ‘‘Species’’ radiations of symbiotic dinoflagellates in the Atlantic and Indo-Pacific since the Miocene-Pliocene transition. Molecular Biology and Evolution 22, 570-581.

LaJeunesse TC, Pettay DT, Sampayo EM, et al. (2010a) Long-standing environmental conditions, geographic isolation and host-symbiont specificity influence the relative ecological dominance and genetic diversification of coral endosymbionts in the genus Symbiodinium. Journal of Biogeography 37, 785-800.

50 LaJeunesse TC, Smith R, Walther M, et al. (2010b) Host–symbiont recombination versus natural selection in the response of coral–dinoflagellate symbioses to environmental disturbance. Proceedings of the Royal Society B 277, 2925-2934.

LaJeunesse TC, Smith RT, Finney J, Oxenford H (2009) Outbreak and persistence of opportunistic symbiotic dinoflagellates during the 2005 Caribbean mass coral 'bleaching' event. Proceedings of the Royal Society B: Biological Sciences 276, 4139- 4148.

LaJeunesse TC, Trench RK (2000) The biogeography of two species of Symbiodinium (Freudenthal) inhabiting the intertidal anemone, Anthopleura elegantissima (Brandt). Biological Bulletin 199, 126-134.

Lowe CD, Montagnes DJS, Martin LE, Watts PC (2010) High genetic diversity and fine- scale spatial structure in the marine flagellate Oxyrrhis marina (Dinophyceae) uncovered by microsatellite loci. PLoS One 5, e15557.

Mackinnon M, Read A (2004) Virulence in malaria: an evolutionary viewpoint. Philosophical Transactions of the Royal Society of London. Series B: Biological Sciences 359, 965-986.

Magalon H, Baudry E, Husté A, Adjeroud M, Veuille M (2006) High genetic diversity of the symbiotic dinoflagellates in the coral Pocillopora meandrina from the South Pacific. Marine Biology 148, 913-922.

Nagai S, Lian C, Yamaguchi S, et al. (2007) Microsatellite markers Reveal Population Genetic Structure of the Toxic Dinoflagellate Alexandrium tamarense (Dinophyceae) in Japanese Coastal Waters. Journal of Phycology 43, 43-54.

Nagai S, Nishitani G, Sakamoto S, et al. (2009) Genetic structuring and transfer of marine dinoflagellate Cochlodinium polykrikoides in Japanese and Korean coastal waters revealed by microsatellites. Molecular Ecology 18, 2337-2352.

Peakall R, Smouse PE (2006) GENALEX 6: genetic analysis in Excel. Population genetic software for teaching and research. Molecular Ecology Notes 6, 288-295.

Pettay D, LaJeunesse T (2007) Microsatellites from clade B Symbiodinium spp. specialized for Caribbean corals in the genus Madracis. Molecular Ecology Notes 7, 1271-1274.

Pettay DT, LaJeunesse TC (2009) Microsatellite loci for assessing genetic diversity, dispersal and clonality of coral symbionts in ‘stress-tolerant’ clade D Symbiodinium. Molecular Ecology Resources 9, 1022-1025.

51 Pettay DT, LaJeunesse TC (in press) Long-range dispersal and high-latitude environments influence the population structure of a “stress-tolerant” endosymbiont harbored by Pocillopora in the Eastern Pacific. Molecular Ecology.

Pinzón JH, Devlin-Durante MK, Weber M, Baums IB, LaJeunesse TC (2010) Microsatellite loci for Symbiodinium A3 (S. fitti) a common algal symbiont among Caribbean Acropora (stony corals) and Indo-Pacific giant clams (Tridacna). Conservation Genetics Resources 3, 45-47.

Pinzón JH, LaJeunesse TC (2011) Species delimitation of common reef corals in the genus Pocillopora using nucleotide sequence phylogenies, population genetics and symbiosis ecology. Molecular Ecology 20, 311-325.

Pinzón JH, Reyes-Bonilla H, Baums IB, LaJeunesse TC (submitted) Local factors influence genotypic diversity among populations of the ecologically dominant Pocillopora (Scleractinia) in the Gulf of California. Coral Reefs

Primmer C, Ellegren H (1998) Patterns of molecular evolution in avian microsatellites. Molecular Biology and Evolution 15, 997-1008.

Read A, Taylor L (2001) The ecology of genetically diverse infections. Science 292, 1099- 1102.

Reyes-Bonilla H, Carriquiry J, Leyte-Morales G, Cupul-Magaña A (2002) Effects of the El Niño-Southern Oscillation and the anti-El Niño event (1997-1999) on coral reefs of the western coast of México. Coral Reefs 21, 368-372.

Richmond R (1987) Energetics, competency, and long-distance dispersal of planula larvae of the coral Pocillopora damicornis. Marine Biology 93, 527-533.

Richmond RH (1985) Variations in the population biology of Pocillopora damincornis across the Pacific. Proceedings of the 5th International Coral Reef Symposium 6, 101-106.

Rowan R, Knowlton N (1995) Intraspecific diversity and ecological zonation in coral-algal symbiosis. Proceedings of the National Academy of Sciences 92, 2850-2853.

Rynearson T, Armbrust E (2005) Maintenance of clonal diversity during a spring bloom of the centric diatom Ditylum brightwellii. Molecular Ecology 14, 1631-1640.

Rynearson T, Newton J, Armbrust E (2006) Spring bloom development, genetic variation, and population succession in the planktonic diatom Ditylum brightwellii. Limnology and Oceanography 51, 1249-1261.

52 Sampayo E, Ridgway T, Bongaerts P, Hoegh-Guldberg O (2008) Bleaching susceptibility and mortality of corals are determined by fine-scale differences in symbiont type. Proceedings of the National Academy of Sciences 105, 10444-10449.

Sampayo EM, Dove S, LaJeunesse TC (2009) Cohesive molecular genetic data delineate species diversity in the dinoflagellate genus Symbiodinium. Molecular Ecology 18, 500-519.

Santos S, Coffroth M (2003) Molecular genetic evidence that dinoflagellates belonging to the genus Symbiodinium Freudenthal are haploid. The Biological Bulletin 204, 10-20.

Santos S, Gutierrez-Rodriguez C, Lasker H, Coffroth M (2003) Symbiodinium sp. associations in the gorgonian Pseudopterogorgia elisabethae in the Bahamas: high levels of genetic variability and population structure in symbiotic dinoflagellates. Marine Biology 143, 111-120.

Santos S, Shearer T, Hannes A, Coffroth M (2004) Fine-scale diversity and specificity in the most prevalent lineage of symbiotic dinoflagellates (Symbiodinium, Dinophyceae) of the Caribbean. Molecular Ecology 13, 459-469.

Seutin G, White BN, Boag PT (1991) Preservation of avian blood and tissue samples for DNA analyses. Canadian Journal of Zoology 69, 82–92.

Stimson J, Sakai K, Sembali H (2002) Interspecific comparison of the symbiotic relationship in corals with high and low rates of bleaching-induced mortality. Coral Reefs 21, 409-421.

Taylor L, Walliker D, Read A (1997) Mixed-genotype infections of the rodent malaria Plasmodium chabaudi are more infectious to mosquitoes than single-genotype infections. Parasitology 115, 121-132.

Thornhill D, Xiang Y, Fitt W, Santos S (2009) Reef endemism, host specificity and temporal stability in populations of symbiotic dinoflagellates from two ecologically dominant Caribbean corals. PLoS One 4, e6262.

Thornhill DJ, LaJeunesse TC, Kemp DW, Fitt WK, Schmidt GW (2006) Multi-year, seasonal genotypic surveys of coral-algal symbioses reveal prevalent stability or post- bleaching reversion. Marine Biology 148, 711-722.

Thornhill DJ, LaJeunesse TC, Santos SR (2007) Measuring rDNA diversity in eukaryotic microbial systems: how intragenomic variation, pseudogenes, and PCR artifacts confound biodiversity estimates. Molecular Ecology 16, 5326-5340.

Trench R (1993) Microalgal-invertebrate symbioses - a review. Endocytoboisis and Cell Research 9, 135-175.

53 Vardo-Zalik A, Schall J (2008) Clonal diversity within infections and the virulence of a malaria parasite, Plasmodium mexicanum. Parasitology 135, 1363-1372.

Voolstra CR, Sunagawa S, Schwarz JA, et al. (2009) Evolutionary analysis of orthologous cDNA sequences from cultured and symbiotic dinoflagellate symbionts of reef- building corals (Dinophyceae: Symbiodinium). Comparative Biochemistry and Physiology - Part D: Genomics and Proteomics 4, 67-74.

Wham DC, Pettay DT, LaJeunesse TC (2011) Microsatellite loci for the host-generalist “zooxanthella” Symbiodinium trenchi and other Clade D Symbiodinium. Conservation Genetic Resources, 1-4.

54 CHAPTER 2

Long-range dispersal and high-latitude environments influence the population structure of a “stress-tolerant” endosymbiont harbored by Pocillopora in the Eastern Pacific

Abstract Symbiotic dinoflagellates (Symbiodinium) are among the most abundant microbial eukaryotes found on coral reef ecosystems. The migration and dispersal of thermally tolerant species may influence how algal-coral symbioses respond to climate warming. Eastern Pacific populations of Symbiodinium glynni (D1) harbored by a regionally dominant coral in the genus Pocillopora were analyzed using eleven microsatellite loci. A total of 402 multilocus haploid genotypes were acquired, of which 215 were distinct. Clone, or strain, diversity was high at most collection sites, with a single genotype detected in most samples (> 80%). Minimal population subdivision was detected throughout the Eastern Tropical Pacific (ETP) encompassing 1000’s of square kilometers from mainland Mexico and Clipperton Atoll to the Gulf of Panama and the Galapagos Islands. The general lack of population differentiation over these distances appears to correspond with recent findings of extensive host genetic connectivity and indicates Pocillopora larvae, which maternally inherit their symbionts, aid in the dispersal of this symbiont. Unlike a confluent host population, high latitude populations in the subtropical Gulf of California (Sea of Cortez) were strongly differentiated from the ETP. While the genetic discontinuity corresponds to a natural barrier to exchange created by prevailing surface currents, selection pressures related to large seasonal changes in temperature and irradiance explains the genetic differentiation of this most northern population. The study of Symbiodinium population genetics may have broader application for evaluating the importance of environmental extremes in driving microbial eukaryote evolution in marine ecosystems.

55 Introduction

Marine microrganisms have the potential to disperse great distances and may therefore show little population differentiation over large geographic regions (Finlay 2002;

Fenchel & Finlay 2004; Foissner 2006). The genetic connectivity of these organisms is likely influenced by numerous intrinsic and extrinsic factors including changes in water quality, nutrients, irradiance and temperature as well as prevailing ocean currents (Foissner

2006). While some species appear to exhibit broad latitudinal and longitudinal distributions, phylogenetic and emerging population genetic data are indicating that (1) many widespread taxa actually comprise numerous cryptic species displaying limited geographic ranges and/or

(2) populations of cosmopolitan species are not panmictic (Rynearson & Armburst 2004;

Darling et al. 2004, 2007; Nagai et al. 2007, 2009; Lilly et al. 2007; Casteleyn et al. 2009,

2010). Despite some progress, the microevolution of eukaryotic microalgae remains poorly understood because culturing is often required in order to obtain enough material upon which to conduct genetic analyses (Rynearson & Armburst 2004; Shankle et al. 2004).

Furthermore, culturing is laborious and highly selective of the natural diversity that exists

(Wintzingerode et al. 1997). In contrast, widespread endosymbiotic dinoflagellates in the genus Symbiodinium are ideal models for population genetic studies since numerous samples may be acquired from various marine invertebrate hosts and habitats over a range of spatial and temporal scales (Santos et al. 2003; Pettay et al. 2007, 2009).

Reef coral communities in the Eastern Pacific prosper under sea surface temperatures

(SSTs), nutrient concentrations and irradiance levels that fluctuate widely between seasons

(Glynn & Ault 2000). These environmental extremes and isolation from the Central Pacific explains the depauperate community assemblage of reef corals that harbor Symbiodinium

56 (Glynn & Ault 2000). The most distinctive feature of these communities it that branching

Pocillopora can represent 90-100% of the live coral cover in near shore hard bottom habitats and encompass a latitudinal range of both tropical (Eastern Tropical Pacific or ETP) and subtropical (Gulf of California or GoC) regions. While approximately eight or nine morphospecies of Pocillopora have been reported from the eastern Pacific (Glynn & Ault

2000), recent genetic and ecological data question the validity of this diversity (Combosch et al. 2008; Pinzón & LaJeunesse 2011). It appears that most of the region’s Pocillopora may actually comprise a single genetically definable species designated as type 1 (Pinzón &

LaJeunesse 2011). Populations of this coral exhibited little discernable differentiation across the entire eastern Pacific despite habitat patchiness, strong environmental gradients and complex surface currents that may interrupt dispersal (Glynn et al. 1996; Glynn & Ault

2000).

The ecological success of type 1 Pocillopora throughout the eastern Pacific may be attributed to their associations with the intracellular symbiont, Symbiodinium glynni (D1)

(sensu LaJeunesse et al. 2010b). As with most other Clade D Symbiodinium, S. glynni appears ecologically specialized to one host genus (LaJeunesse et al. 2008, 2010a). S. glynni is extremely common among type 1 colonies of Pocillopora and dominates a majority of individuals in most near shore locations, especially in central Mexico where it appears to be the only symbiont present in populations of this animal (LaJeunesse et al. 2010b). Like other symbioses involving species in clade D, colonies harboring S. glynni possess a greater thermal tolerance, tend not to bleach (significant reductions in symbiont cell densities) and are more likely to survive when exposed to stressful temperatures (Glynn et al. 2001;

LaJeunesse et al. 2010b). As a consequence of its abundance and thermal tolerance the

57 symbiosis between Pocillopora type 1 and S. glynni is important to the present and future stability of shallow water communities of the region.

Given the ecological importance that thermally tolerant Symbiodinium may have as sea surface temperatures rise (Baker et al. 2004; Berkelmans & van Oppen 2006; LaJeunesse et al. 2009, 2010a), genetic analyses of their populations may reveal patterns of migration and dispersal important for understanding how environmental conditions and a warming climate influence the ecological and geographic spread of these symbionts. The recent application of microsatellite markers supported phylogenetic and ecological data indicating that members of clade D comprise numerous and mostly host specific operational taxonomic units, or species (Pettay & LaJeunesse 2009; LaJeunesse et al. 2010a; Wham et al. 2011).

Microsatellite analyses examining genetic connectivity of particular Symbiodinium in clades

B and C indicate that populations separated by small distances (e.g., 5-10 km) exhibit significant differences in allelic composition and suggests that dispersal is extremely limited for these symbionts (Santos et al. 2003; Howells et al. 2009; Kirk et al. 2009; Thornhill et al.

2009; Andras et al. 2011; but see Magalon et al. 2006). However, population genetic studies of Symbiodinium remain few and substantially more data are needed to evaluate the generality of genetic structuring of metapopulations.

S. glynni’s association with Pocillopora throughout the eastern Pacific offers an opportunity to study the genetic diversity and connectivity among populations of a “stress- tolerant” dinoflagellate. The maternal inheritance, or vertical transmission, of S. glynni by

Pocillopora larvae may significantly enhance the long distance dispersal of this symbiont

(Glynn et al. 1991; Chavez-Romo & Reyes-Bonilla 2007). It is hypothesized that the general lack of genetic differentiation of the host across both tropical and subtropical regions of the

58 eastern Pacific (Pinzón & LaJeunesse 2011) will impart similar patterns on populations of its co-migrating symbiont. Samples from populations of Pocillopora with S. glynni were collected from locations encompassing several biogeographic provinces spanning over 3,400 km from the Galapagos Islands at the equator to the southern Gulf of California (Sea of

Cortez) at approximately 24˚N. In the first population genetic examination of a Clade D

Symbiodinium, eleven microsatellite loci were analyzed and the population structure compared with that of the host.

Materials and Methods

Sample Collections

Samples analyzed in this study were part of a larger biogeographical analysis of

Pocillopora spp. symbionts across the eastern Pacific (LaJeunuesse et al. 2008, 2010b).

Fragments (~2 cm2) of Pocillopora colonies were collected by SCUBA from six sampling locations (Figure 8) and preserved in either a high salt, 20% DMSO buffer (Seutin et al.

1991) or 95% ethanol and stored at -20 °C until DNA extraction. At each regional location, samples were collected along linear transects at depths ranging from two to ten meters at each of one to four sites separated by as much as 50 km. As Pocillopora in the ETP are known to reproduce asexually by fragmentation (Richmond 1985; unpubl. data), morphologically distinct colonies were sampled at least 3 meters apart to avoid sampling clonemates resulting from fragmentation. Sample collection dates for each location are as follows: Gulf of California (GoC) in May 2006 and June 2008, Banderas Bay (BB) and

Clipperton Atoll (CLP) in April 2007, Gulf of Tehuantepec (OAX) in May 2008, Gulf of

Panama (PAN) in January 2009, and the Galapagos Islands (GAL) in March 2007. A total of

59

Figure 8. Sampling locations in the Tropical and Subtropical Eastern Pacific. Collections at multiple sites separated by 2-50 km were made during trips to the following locations: Gulf of California (GoC), Banderas Bay (BB), Gulf of Tehuantepec (OAX), Clipperton Atoll (CLP), Gulf of Panama (PAN) and Galapagos Islands (GAL).

402 samples representing five different Pocillopora morphospecies were analyzed to determine the population structure of S. glynni across the Eastern Pacific, with the sampling locations and sample sizes given in Figure 8 and Table 6, respectively. While samples included different morphospecies of host, genetic analysis shows them all to belong to

Pocillopora type 1 and is the only host in the eastern Pacific to associate with S. glynni.

Molecular-Genetic Identification

Nucleic acid extractions were conducted using a modified Promega Wizard genomic

DNA extraction protocol (LaJeunesse et al. 2003). The dominant resident symbiont was

60 identified by denaturing gradient gel electrophoresis (DGGE) fingerprinting of the partial

5.8S and internal transcribed spacer (ITS) region 2 (LaJeunesse 2002). The region was amplified using a touch-down thermal cycle profile with the primers “ITS2clamp” and

“ITSintfor2” (LaJeunesse & Trench 2000), and the PCR products resolved on denaturing gels

(45 – 80% of 7M urea & 40% formamide) using a CBScientific system (Del Mar, CA) for 16 hours at 115 volts. Samples where S. glynni (D1) was detected were utilized for this study.

Additional colonies from the Clipperton Atoll and the Galapagos Islands that appeared to harbor only the clade C species (i.e. C1b-c) contained low abundance background populations of S. glynni and were utilized for microsatellite analysis.

Microsatellite Analysis

A subset of samples was initially screened with fifteen microsatellite loci developed for Clade D Symbiodinium to determine the degree of allelic polymorphism for these populations (Pettay & LaJeunesse 2009; Wham et al. 2011). One primer set, D1Sym77, amplified two distinct loci and each was scored separately (see Results). Eleven out of fifteen loci were sufficiently polymorphic and used to determine the population structure of

Symbiodinium glynni in the eastern Pacific (Table 5). The remaining loci were either monomorphic or provided minimal genetic resolution and were therefore not assessed (data not shown). GenePop (version 4.0.11, Rousett 2008) was used to examine the degree of linkage among these loci in pairwise comparisons.

The forward primers of all loci were fluorescently labeled and the microsatellite fragments amplified, and fragments analyzed at using the Pennsylvania State University

Genomics Core Facility according the methods of Pettay & LaJeunesse (2007 & 2009) and

61 Wham et al. (2011). Since microsatellites are known to produce stutter due to polyermase slippage during PCR, only the dominant peaks were scored. Multiple alleles at a particular locus were scored when a second clear peak was found in the expected size range and was at least one-third the size of the dominant peak. The presence of multiple peaks was interpreted to indicate that a sample contained more than one haploid genotype. The occasional exclusion of small ambiguous peaks may occasionally cause a low background genotype to be missed, however this conservative approach prevents the overestimation of genetic diversity (Anderson et al. 2000). Additionally, while this approach may slightly skew the results of diversity on a per colony basis, it should have little impact on the population structure analyses since these mixed genotype samples did not result in additional allelic diversity.

Multilocus genotypes (MLGs) were constructed from fragment size data gathered for each sample. Samples were occasionally reanalyzed to confirm the existence of mixed genotypes and unusual or rare fragment sizes. Recovering multiple alleles at a locus (or loci) indicates the presence of two or more MLGs of symbiont within a single sample. In cases where mixtures were found, only the dominant fragment at each locus (i.e., highest peak) was used to construct a single MLG for that sample. While it has been shown that preferential amplification of certain fragments, usually the smallest, can occur with microsatellites, experimental mixtures using the loci in this study have been shown to correctly amplify the fragment in highest concentration (data not shown). Therefore, it is believed that constructing MLGs using the dominant fragments adds little bias, yet allows a maximum number of MLGs to be recovered.

62 Data Analysis

Populations of organisms that commonly reproduce asexually will tend to have a high frequency of clones. Therefore statistical calculations based on allele frequencies can be negatively biased (Arnaud-Haond et al. 2007). For example, linkage analysis on a population that is highly clonal may indicate that loci are in linkage disequilibrium, when they may not. Similarly, AMOVA’s based on genetic distance measurements between pairs of samples are potentially wrongly influenced by high numbers of duplicated MLGs in a data set. Therefore, a population with high clonality often appears genetically different from other populations not dominated by a particular clone. For these reasons, all analyses (unless stated otherwise) were conducted with duplicated MLGs removed at each sampling location.

2 2 4 The probability of identity (PI = (Σpi ) - Σpi , where pi is the frequency of the i-th allele at a locus) was calculated to determine the power to resolve genetically distinct individuals. The PI is an estimate of the probability that two unrelated individuals drawn at random will by chance have the same MLG (Peakall & Smouse 2005). Individual PI’s are calculated for each locus and an overall PI for all loci is the product of each individual locus

PI. PI values between 0.01 – 0.0001 are believed to be reasonably low enough for population studies (Waits et al. 2001), with values lower than 0.01 adequate for mark-recapture studies on population size estimation (Mills et al. 2000). Since these values may be affected by population substructure (Waits et al. 2001), PI was calculated for each location and for the entire dataset using GenAlEx (version 6.4, Peakall & Smouse 2006).

The software Structure (Version 2.3.2) was used to overcome biases of assigning populations by location by using the microsatellite data to cluster MLG based on their genetic similarities irrespective of sample origin. Briefly, this software uses a Bayesian

63 clustering approach to probabilistically assign individuals to populations (Pritchard et al.

2000). The model assumes there are K populations characterized by a set of allele frequencies, with the assumptions of unlinked loci that are in linkage and Hardy-Weinberg equilibrium within populations (Pritchard et al. 2000). While the assumption of HWE does not directly apply to haploid organisms, as long as the organism recombines, which

Symbiodinium does, and the there is not a large signal of LD between loci the ploidy is irrelevant. Symbiodinium does reproduce asexually and removing repetitive MLGs prevents the biasing of Structure analyses due to clonality. Concerning HWE specifically, this theory poses that in an idealized theoretical population allele frequencies remain constant from generation to generation. In the case of haploid organisms, the maintenance of constant allele frequencies (e.g., HWE) can still be assumed for neutral loci with a balance between factors such as mutation, selection, migration and drift, and that violation of this assumption

(e.g., nonrandom mating, strictly asexual reproduction) will yield populations that are in linkage disequilibrium.

Structure analyses to investigate differentiation across the eastern Pacific were conducted using an admixture model with correlated allele frequencies and were run from K

= 1 to 10 with five runs per K and a burnin of 100,000 and 1,000,000 reps after the burnin. A plot of the log probability of the data for a given K (ln P(D)) versus K was derived from the structure results, along with an analysis of the second order rate of change of K following the method by Evanno et al. (2005) and implemented using Structure Harvester (v0.56.4) by Earl

(2009) to determine the appropriate clustering of individuals. Five runs per K were utilized to verifying consistency between runs, with the run having the highest ln P(D) for the appropriate K used to construct Structure plots and inform clustering. Graphic displays of

64 Structure plots were manipulated (i.e., color and sample order) using DISTRUCT (Rosenberg

2004). Additional analyses were performed on the ETP region only with the additional use of the location prior feature, which performs better in the presence of weak population structuring (Hubisz et al. 2009; See results). To graphically compare population clustering for host and symbiont Structure analyses were also conducted on previously published population data for the host, Pocillopora type 1 (Pinzón & LaJeunesse 2011), using a correlated allele model with admixture and location prior and were run for K = 2 with five runs per K and a burnin of 100,000 and 1,000,000 reps after the burnin.

To investigate the possibility of population differentiation according to Pocillopora morphospecies, Structure analyses were conducted on data from the three locations where sampling was adequate to conduct such a comparison; GoC, BB and OAX. For these analyses, data was grouped according host morphology with identical MLGs removed within each morphological grouping. Analyses were identical to those for geography, with the exception that the location prior setting was used to better assist with weak structuring

(Hubisz et al. 2009) and the maximum K was set to eight.

To confirm the clustering determined by Structure, similar Bayesian analyses were implemented using the program BAPS (Version 5.3; Corander et al. 2003, 2004). BAPS utilizes a different algorithm that relies mostly on a greedy stochastic optimization procedure, only using a Markov chain Monte Carlo (MCMC) procedure for complex data sets (Corander et al. 2006; Latch et al. 2006). This procedure allows for faster analyses, yet typically yields results similar to Structure for known population structuring (Latch et al. 2006). Mixture analyses were run in BAPS for both the geographic and morphospecies clustering, using the

“cluster by groups” option, K ranging from one to ten (geographic) or one to eight

65 (morphospecies) and five replicates for each K. The K with the highest log marginal likelihood value was considered the appropriate number of clusters.

Analyses of molecular variance (AMOVA) where then conducted on geographic populations with regional structuring as defined by Structure and by morphospecies to determine the degree of differentiation. AMOVA’s along with a permutation procedure were performed in GenAlEx to test for significant difference in genetic diversity between populations (Excoffier et al. 1992). The AMOVA produces variance components along with

Φ-statistics (F-statistic analogs), which partition genetic variation at different hierarchical levels (Excoffier et al. 1992). The significance of the variance components and Φ-statistics were then tested using 10000 permutations and a Bonferroni corrected α = 0.05. In addition to quantifying genetic differentiation, GenAlEx was used to calculate several summary statistics for each Structure population and sampling location, including h and I. Haploid

2 genetic diversity (h = 1 - Σpi ) gives an indication that two individuals drawn at random will be genetically different, while information index (I) is a measure of allelic diversity (Peakall and Smouse 2006). Lastly, clonal richness (R) which is equal to (G – 1)/(N – 1), where G =

# of unique MLG’s and N = total sample size, was also calculated to give the frequency of unique genotypes and an indication of the contribution of asexual reproduction.

Environmental Data

Monthly averages for sea surface temperatures (SSTs; °C) and photosynthetically active radiation (PAR; Einsteins m-2 day-1) were calculated for each sampling location using data for years 2000 to 2009 obtained from the Giovanni online data system, developed and maintained by NASA Goddard Earth Sciences (GES) Data and Information System (DISC).

66 Approximately 24 km2 around each sampling location was selected and data retrieved from the SeaWiFS (PAR) and MODIS-Aqua (SST) databases. Scatter plots were drawn using monthly averages and standard deviations. To compare environmental variation among locations, box plots were drawn using annual means, first and third quartiles, and extreme maximum and minimum annual values.

Results

ITS2-DGGE fingerprinting and sequencing

Each of the MLGs characterized in this study possessed a single ITS2 sequence (D1) that dominated their ribosomal array (Thornhill et al. 2007; Sampayo et al. 2009; LaJeunesse et al. 2010a). This sequence contrasts with most other Clade D types characterized by this method, indicating S. glynni represents a distinct operational taxonomic unit (LaJeunesse et al. 2010a)

Microsatellite Data

The number of alleles per locus ranged from 3 to 29, however, the effective number of alleles (Ae) per locus ranged from 1.02 to 6.22 (Table 5). Allele frequencies for each location ranged from 0.011 to 1.00 (Appendix C). The frequencies of putative null alleles were rare among loci ranging from 0 for loci D1Sym88 and D1Sym92 to a value of 0.079 for

D1Sym11. Primers for locus D1Sym77 consistently amplified two independent alleles. The presence of two alleles at this locus appears to be from a partial chromosome duplication event early in the evolution of this symbiont lineage (Wham et al. 2011). Detailed analysis of allelic data throughout the sampling range has revealed two distributions

67 Table 5. Description of microsatellite loci used in this study. Ta = annealing temperature, Ae = effective alleles with standard error in parenthesis. Subscript numbers following the repeat motif indicate the number of repeats in the initial cloned sequence that was used to develop locus primers.

Size Range Number of Ae Locus Primer Sequence (5' - 3') Repeat Motif Ta (°C) (bp) Alleles 1.14 (0.09) D1Sym9 F - CAGAAGCCCAATTATATGCGGCA (FAM) (GTT)6 57 106 - 115 4 R - AGGATGATGAGCATGCCGACG 2.42 (030) D1Sym11 F - TGAAATCTCACTCAGAGTCGGAC (FAM) (AC)13 57 151 - 161 6 R - GCAGACAGTGATTTCAGTTCCGA 1.57 (0.33) D1Sym14 F - TCTCAGTGGAAAGCATTGTGG (FAM) (CT)11 AT (CT)4 55 173 - 185 7 R - TCGTCTGAATCAGGATCTGACG 4.36 (0.78) D1Sym17 F - TGTGAATGCTTCTTGGGGTG (HEX) (CA)8 57 143 - 167 13

68 R - TCATGCTTGTCCGTGAGCAG

D1Sym34 F - ACCTGAGACCTGAGTGTTGC (FAM) 55 332 - 428 29 6.22 (1.15) (CAAA)9 CACA (CAAA)4

R - ATCATGGGCAGAGCTCCTGG (GAAACAAA)2 (CAAA)13 2.40 (0.37) D1Sym67 F - GAATCCAGATGGTGCCTGC (VIC) (ATC)8 57 131 - 149 7 R - CAAAGGTAGCCGATTGTCTC 2.03 (0.35) D1Sym77a F - CCACTGAGATTGGTAGGTGAA (PET) (TTC)5 CT (CTTCCT)2 C (TTC)4 55 169 - 184 6 R - ACCGATGGTGTTTGTGACTCG 1.32 (0.14) D1Sym77b F - CCACTGAGATTGGTAGGTGAA (PET) (TTC)5 CT (CTTCCT)2 C (TTC)4 55 184 - 193 4 R - ACCGATGGTGTTTGTGACTCG 2.91 (0.48) D1Sym87 F - CCTATGACTCCAAGGGTGACG (FAM) (GAAG)7 57 244 - 268 7 R - AGACATACCTCGGTCTTGTC D1Sym88 F - TTGTCAGACTGAATGCTCCA (NED) (CTTT)3 G (TTTC)7 55 227 - 235 3 1.02 (0.02) (TCTCTTTC)2 TCTTTTT (CT)3

R - GTGTTCAAGCGACATCCCA (TCTT)2 (TTTC)2 T (CTT)4 1.07 (0.03) D1Sym92 F - GCGTTTGACACAAGGATCCCT (FAM) (CCTA)6 (CCTG)3 57 124 - 132 3 R - TTGGGATGCTCTTGGCGAC

of alleles that are distinct and non-overlapping, and easily binned into low (77a) and high- sized (77b) fragments with two alleles from a single size distribution never co-occurring within an individual in ninety-nine percent of the samples. Because alleles were relatively independent of each other, each was scored as a separate locus with the smaller allele sized designated as one locus (77a) and the larger alleles as a second locus (77b).

Numerous alleles were private to a particular sampling location or region (Appendix

C). The GoC possessed 12 alleles involving five loci (D1Sym14, 17, 34, 77b & 88) that were unique to the region. Banderas Bay possessed 13 alleles from three loci (D1Sym9, 34

& 77b), GAL possessed 2 alleles from one locus (D1Sym34), and CLP and OAX each possessed 1 allele at one locus (D1Sym34 & D1Sym14) found only in these locations. All alleles in PAN were found in at least one other location.

Two hundred and fifteen different genotypes were scored among four hundred and two MLGs obtained. Seventy-five, or 18.7 percent, of samples possessed multiple MLGs

(i.e. more than one allele at one or more loci). The majority of these samples (n = 51) possessed two alleles at just one locus. Another 24 samples contained two alleles at multiple loci (11, 5, 2, and 6 samples possessed allele variation at 2, 3, 4, 5 loci, respectively).

Multiple alleles were never observed for six or more loci or for the conserved loci D1Sym88 and D1Sym92.

The probability of identity (P(ID)) that two samples with the same MLGs may not have originated from the same clone lineage was exceedingly low for most locations and ranged from 7.55 x 10-4 to 7.76 x 10-7 (Table 6). All of these values, except one, were below the

0.0001 criteria recommended by Waits et al. (2001). The overall P(ID) based on data from all locations was 4.09 x 10-9. Therefore, there was a 1 in two hundred and fifty million

69 Table 6. Summary statistics by location including number of colonies sampled, number of unique MLGs of S. glynni, clonal richness (R), haploid diversity (h), information index (I), and probability of identity (PI). Numbers in parenthesis represent the standard error for a particular statistic.

Probability Number of Unique Clonal Haploid Information Private of Identity Location Samples MLG's Richness (R) Diversity (h) Index (I) Alleles (PI) Gulf of California (GoC) 137 64 0.46 0.43 (0.09) 0.85 (0.20) 12 1.87E-06 Banderas Bay (BB) 155 87 0.56 0.40 (0.10) 0.92 (0.26) 13 7.76E-07 Gulf of Tehuantepec (OAX) 71 40 0.56 0.44 (0.08) 0.87 (0.18) 1 1.97E-06 Clipperton Atoll (CLIP) 7 6 0.83 0.33 (0.10) 0.58 (0.19) 1 6.49E-05 Bay of Panama (PAN) 21 7 0.30 0.30 (0.09) 0.46 (0.14) 0 7.55E-04

70 Galapagos Islands (GAL) 11 11 1.00 0.44 (0.11) 0.88 (0.23) 2 3.44E-07

Total 402 215 0.53 0.39 (0.04) 0.76 (0.08) 4.09E-09

probability that identical genotypes from separate locations are of independent origins.

Clonal richness (R) or genotypic richness ranged from 0.30 to 1.00 for each location

(Table 6). Values approaching 0 indicate high clonality whereas values of 1.00 are indicative of populations where no repeated genotypes were sampled. Haploid diversity indices ranged from 0.30 to 0.44 with an overall average of 0.39 for all locations, while the information index ranged from 0.92 to 0.46 with an average of 0.76 (Table 6). The mean genotypic richness for all locations was 0.51 indicating that a large proportion (~ 49%) of MLGs were detected two or more times. Some “locally prevalent” clones were unusually common within and among sites in all sampling locations except GAL and CLP. In one case a MLG was recovered from nineteen different colonies.

A pairwise analysis of linkage among loci revealed that ~93% of the pairwise comparisons were unlinked (Bonferroni corrected; α = 0.05). In general, linkage between paired loci changed based on population under analysis, with no pair of loci linked in all locations. Loci linked at one location were often in linkage equilibrium at another location.

The linkage for some loci is believed to be a consequence of the clonal evolution of these dinoflagellates or, in some cases, evidence of weak differentiation over small spatial scales within and between sites from a particular location.

Population Differentiation

Two well supported populations were detected using the Evanno et al. method to process the Structure results (k = 2; ln P(D) = -2351.4; Figure 9, Appendix D), and corresponded to the subtropical GoC region and the ETP (Figure 10). The posterior

71

Figure 9. Structure plot of S. glynni populations from the tropical and subtropical Eastern Pacific based on allelic frequencies at 11 microsatellite loci. Each bar in the graph represents the probability that a sample belongs to a particular color-coded population. Each graph represents analyses for a particular number of populations (K) run under an admixture with a correlated allele frequency model. (a) Major differentiation occurred between the GoC population and populations in the ETP (k = 2). Additional clustering occurred in the ETP, yet did not correspond to location, depth or host morphospecies (k = 3).

probabilities for increasing values of K continued to rise until k = 6, at which they stabilize and/or decline (Appendix D). Further examination of Structure runs at k = 3 show consistency between runs and additional structuring within the ETP region and may represent two cryptic populations within the region (Figure 9; ln P(D) = -2230.4 & Δruns = 1.0;

Appendix D & E), however this additional subdivision appears haphazard and does not related to geographic location or depth. Subsequent analyses on the ETP region only, and using the location prior feature, reveal similar patterns as above with no further subdivision in relation to geographic location in the ETP (data not shown). Additional analyses using

BAPS support the two regional populations (k = 2; log(ml) = -2489.7) with no further subdivision of the ETP, even when analyzed in isolation (k = 1; log(ml) = -1780.3).

72

Figure 10. Biogeographic map of the eastern Pacific showing differentiated populations of S. glynni as determined by statistical analyses of microsatellite data. The GoC (shaded in purple) represents a northern subtropical population, while locations in the ETP comprise a single undifferentiated population (orange). Structure plot to the left shows how Pocillopora type 1 populations are relatively homogeneous throughout the GoC and ETP (sensu Pinzón & LaJeunesse 2011). The inset in the upper right identifies cases were a S. glynni genotype was found at two different locations (proportions were calculated by individuals/total # unique MLGs between two locations).

None of the Structure analyses indicated meaningful population subdivision according to host morphospecies at any location (Figure 11). Analysis of the second order rate of change of K suggests two populations in the GoC and OAX, neither of which corresponds to morphospecies (Figure 11). In one case, four populations were suggested for

BB, but these did not align with morphospecies. Examination of the posterior probabilities for each K show similar increases between different K’s up to four and then large drop in probability between k = 4 and 5. Since the method by Evanno et al. considers change in

73

Figure 11. Structure plots of S. glynni populations showing a lack of population subdivision according to Pocillopora morphospecies in the GoC (a), BB (b) and OAX (c). Structure plots are arranged in order of increasing K and the data is ordered according morphology of the colony from which it was collected.

probabilities both before and after a specific K, the analysis in this case is artificially influence by the large decrease the posterior probability after k = 4 and a single population among morphospecies is the most logical conclusion. Examination of Structure plots for each K support the above conclusions (Figure 11). Subsequent analyses of these data using

BAPS also support a single population among morphospecies (log(ml) for k = 1: GoC, -

656.3; BB, -806.1; OAX, -596.0). Lastly, AMOVA’s for each location were not significant at p = 0.0001 (ΦPT = -0.015, GoC; -0.004, BB; -0.007, OAX), as were all pairwise comparisons between morphospecies at each location (P < 0.001, Table 7).

The pairwise comparisons of Phi (Φ) values between the GoC population and populations within the ETP showed that differentiation was always greatest between the GoC and all other locations (Table 9). Regional variance analysis based on clusters corresponding to the GoC and ETP yielded highly significant values (p = 0.0001, Table 8)

74 Table 7. A pairwise matrix comparing the genetic relationship (ΦPT) between S. glynni populations based on Pocillopora morphospecies in the eastern Pacific. Analyses were conducted in three different locations: (a) Gulf of California, (b) Banderas Bay and (c) Gulf of Tehuantepec. Significant values (sequential Bonferroni corrected P < 0.05) represented in bold.

(a)

Morphospecies P. capitata P. damincornis P. meandrina P. verrucosa

P. capitata ---

P. damincornis 0.000 ---

P. meandrina 0.000 0.006 ---

P. verrucosa 0.000 0.000 0.000 ---

(b)

Morphospecies P. capitata P. damincornis P. meandrina P. verrucosa

P. capitata ---

P. damincornis 0.000 ---

P. meandrina 0.000 0.000 ---

P. verrucosa 0.026 0.000 0.046 ---

(c)

Morphospecies P. capitata P. damincornis P. eydouxi P. meandrina P. verrucosa

P. capitata ---

P. damincornis 0.000 ---

P. eydouxi 0.000 0.000 ---

P. meandrina 0.000 0.000 0.038 ---

P. verrucosa 0.027 0.000 0.050 0.000 ---

ΦRT = 0.373, ΦPR = 0.040 and ΦPT = 0.398, further validating a tropical-subtropical subdivision. Some significance was observed between populations within the ETP (ΦPR) and

75 indicates there may be weak geographical differentiation within the ETP between PAN and

BB and OAX, respectively, not detected by Structure or BAPS (Table 9).

Table 8. Analysis of Molecular Variance (AMOVA) for S. glynni in the eastern Pacific from each of six collection locations with regional tropical and subtropical populations (GoC and ETP).

Estimated Percent of Fixation Source df SS MS Variance Variation Index ΦRT = Among Regions 1 145.39 145.39 1.52 37% 0.373* ΦPR = Among Pops w/in Regions 4 18.80 4.70 0.10 2% 0.040* ΦPT = Within Pops 209 513.08 2.45 2.45 60% 0.398*

Total 214 677.27 4.07 100%

* P < 0.0001

Table 9. A pairwise matrix comparing the genetic relationship (ΦPT) between all six locations of S. glynni in the eastern Pacific. Significant values (sequential Bonferroni corrected P < 0.05) represented in bold.

Eastern Tropical Pacific

GoC BB OAX CLP PAN GAL

Gulf of --- California

Banderas 0.411 --- Bay Gulf of Pacific 0.374 0.017 --- Tehuantepec Clipperton 0.388 0.018 0.022 --- Atoll Bay of 0.379 0.146 0.124 0.079 --- Panama Galapagos 0.337 0.045 0.036 0.016 0.129 --- Eastern Tropical Islands

76 While the majority of genotypes were restricted to a single sample location, nine

MLGs were detected in more than one location (Figure 10, inset). The most geographically widespread of these clones was one found in both PAN and GAL, approximately 1,600 km apart. Seven cloned genotypes were shared between BB and OAX (~ 1,200 km apart) consistent with the predominant surface currents in the region (Figure 10). All 64 MLGs found in the GoC were unique to this location.

Figure 12. Ten year averages of monthly means (2000 – 2009) of sea surface temperatures (SSTs) and photosynthetically available irradiance (PAR) for the six sampling locations. (a) and (c) depict monthly averages and standard deviations at all locations, while (b) and (d) are box plots showing annual mean, first and third quartiles, and maximum and minimum seasonal ranges by location. Monthly averages were acquired from NASA’s Giovanni website.

77 Environmental Data

SSTs and PAR from each location were compared on a monthly and yearly basis from 2000 to 2009 to investigate the relation of environmental conditions with population structuring of S. glynni (Figure 12a-d). Sites in the GoC endured 1) the largest yearly fluctuation in monthly averages in both environmental parameters (~10 °C and 28 Einsteins m-2 day-1, Figure 12c and 12d, respectively); 2) had the coldest monthly average temperature

(~20.0 °C SD ± 1.1 in February), at least 4.3 ˚C colder than any other region (Figure 4a); 3) were exposed to the lowest and highest monthly PAR averages (~31.0 Einsteins m-2 day-1 SD

± 1.3 in December and ~59.5 Einsteins m-2 day-1 SD ± 1.4 in June Figure 12b); and 4) maximal average monthly temperatures (~30.1 °C SD ± 0.5) were only 0.8 ˚C lower than the highest average among all locations.

Discussion

Population genetic approaches offer the resolution necessary to infer key processes in the ecology and evolution coral-algal symbioses. Symbiodinium glynni in association with

Pocillopora type 1 is nearly ubiquitous throughout the eastern Pacific, especially in coastal areas where annual temperatures and water turbidity fluctuate considerably (LaJeunesse et al.

2010b). Stress-tolerant symbioses involving symbionts like S. glynni may become increasingly important in the response of coral communities to a warming climate (Baker et al. 2004); but too many uncertainties remain with regard to their potential for dispersal. The patterns observed here and discussed below should redirect speculation on the nature of these associations and generate further interest in the population genetics and microevolution of these symbionts.

78

Subdivision and connectivity among tropical and subtropical S. glynni populations

A distinct genetic break occurred between subtropical and tropical populations of

Symbiodinium glynni. In contrast, no apparent genetic partitioning was found among populations distributed across the tropical region encompassing 1000’s of square kilometers from mainland Mexico and Clipperton Atoll to the Gulf of Panama and the Galapagos

Islands (Figures 9 & 10). Identical MLGs were recovered from locations separated by as much as 1,600 km (Figure 10 inset), and indicates that certain clones can successfully disperse over long distances, an observation not previously documented in population genetic studies of other marine eukaryotic microbes. Cloned genotypes were routinely observed within and between sites involving spatial scales from 2-100 km indicating S. glynni (and

Pocillopora larvae) frequently disperse over these distances. The extensive regional connectivity as well as strong population differentiation appears similar to recent studies of free-living planktonic dinoflagellates (Nagai et al. 2007, 2009) and other marine microalgae

(Rynearson & Armburst 2004, 2006).

The maintenance of gene flow over large distances by S. glynni contrasts with assertions that Clade B Symbiodinium in the Caribbean, exhibiting limited dispersal and/or geneflow, are highly structured across populations of hosts with horizontal modes of symbiont acquisition (Santos et al. 2003; Kirk et al. 2009; Andras et al. 2011). Competitive interactions between symbionts for host habitat are favored when hosts’ acquire symbionts from the environment at the beginning of each generation. Host colonies expel millions of viable symbiont cells daily as an apparent mechanism for regulating symbiont densities

(Hoegh-Guldberg et al. 1987; Jones & Yellowlees 1997), a process that generates numerous

79 symbiont propagules. Therefore, larvae that arrive and settle on a reef are likely exposed to high environmental concentrations of symbiont genotypes (i.e. clones) that dominate the surrounding host population. In order for a particular symbiont clone, or individual, to establish itself in a new region it must disperse passively via water currents and compete with the locally dominant clone(s) by successfully infecting and proliferating within the cells of a compatible host (Fitt & Trench 1983; Santos et al. 2003). This density dependency model assumes that all genotypes infect and compete equally well (Thornhill et al. 2009). Over successive generations the proliferation of a particularly common clonal line among coral recruits will lead to a few MLGs dominating a given reef, significantly effecting genetic structuring over relatively short distances (Santos et al. 2003; Howells et al. 2009; Kirk et al.

2009; Thornhill et al. 2009; Andras et al. 2011).

The dispersal and recruitment of S. glynni genotypes may be significantly influenced by the mode of symbiont transmission exhibited by its Pocilloporid host. Recent genetic analysis (Pinzón & LaJeunesse 2011) found little evidence of population differentiation among Pocillopora type 1 across the eastern Pacific (Figure 10), and is consistent with several other studies reporting weak population structure among populations separated by

500 km or more (Ayre et al. 1997; Ridgway et al. 2008; Magalon et al. 2005). The apparent lack of geographic structuring throughout much of the ETP suggests that the broadcast spawned larvae of Pocillopora persist for many days in the water column, similar to laboratory experiments with brooded larvae (e.g. Richmond 1987). Aided by the homogenizing action of complex surface currents (Fielder et al. 1992; Glynn et al. 1996;

Glynn & Ault 2000), their larvae are probably disbursed throughout the ETP region (Victor et al. 2001). Pocillopora transmit their symbionts horizontally during the final stages of

80 oogenesis (Glynn et al. 1991; Hirose et al. 2000; Chavez-Romo & Reyes-Bonilla 2007), with fertilized eggs containing over one hundred symbiont cells that divide as larvae begin to develop (Hirose et al. 2000). Assuming a larva settles in an adequate location, survives and grows; the co-migrating symbiont, already well established in the animal, proliferates as the colony matures. Over time these migrant symbionts can contribute significantly to the local

S. glynni gene pool. While nothing is known about the longevity of “free-living” S. glynni cells, the entrainment of symbionts in the life history of Pocillopora explains in large part the high genetic connectivity observed for symbiont populations in the ETP (Figure 10).

The significant differentiation between S. glynni populations in the GoC and ETP is inconsistent with the connectivity observed among host populations between these regions

(Figure 10). Opposing currents, and differences in temperature and salinity, can restrict larval dispersal and migration between the Gulf and the rest of the ETP (Figure 10; Castro et al. 2000, Hurtado et al. 2007). However, seasonal changes in surface currents periodically allow flow into and out of the GoC (Castro et al. 2000) and may explain why Pocillopora type 1 shows no indication of population subdivision between mainland Mexico and the GoC

(Pinzón & LaJeunesse 2011; but see Chavez-Romo et al. 2008). The discordant patterns of population differentiation between the host and symbiont suggest that selective pressures within the region and unrelated to dispersal may influence the unique genetic structure of S. glynni populations.

Local adaptation of S. glynni genotypes in the Gulf of California

The contrast between host and symbiont population structure highlights the importance of analyzing both partners. It appears that Pocillopora individuals recruit

81 successfully into or out of the GoC, while the symbionts do not (Figure 10). The dichotomy between host and symbiont populations in the GoC suggests that symbiont populations are responding to selection pressures that do not affect the host. Environmental selection based on thermal tolerance and/or irradiance may offer a possible explanation. The GoC experiences the largest annual extremes and fluctuation in temperature (~10°C) and irradiance (~27 Einsteins m-2 day-1) of all the locations sampled in the eastern Pacific (Figure

12a-d). These large fluctuations and/or extremes may exert selective pressures on the symbiont populations harboured by Pocillopora. The differential survival among S. glynni genotypes in the GoC may have occurred following a range expansion from the south following the last glaciation (~ 10-18 KYA). A second possibility is that the genetic differentiation of the GoC population developed under more extreme conditions during the last glacial period. Fossil records indicate that Pocillopora were extensive throughout the

Gulf during the late Pleistocene (~125 KYA reviewed in López-Peréz 2005), but whether a refuge population endured the last ice age is unknown.

What explains the apparent lack of migrants between regions? ETP populations of S. glynni may indeed disperse into the GoC carried by their larval hosts. Upon settlement, however, successful “migrant” symbionts are presumably out competed by the locally adapted genotypes of the Gulf (Pettay et al. submitted). Although Pocillopora vertically transmit their symbionts, the juvenile stages may be capable of acquiring different S. glynni strains from the environment. The apparent ability of animals that normally inherit their symbionts maternally to form new associations with “better-adapted” symbiont genotypes indicates that, to a limited extent, these vertical symbioses are flexible. While external environments influence the distribution and prevalence of particular symbionts (Rowan et al.

82 1997; Rodriquez-Lanettey et al. 2001; Ulstrup & van Oppen 2003; LaJeunesse et al. 2004,

2010a), these would be among the first observations of local adaptation to environmental conditions that result in population subdivision and ultimately leading to speciation.

No subdivision between host morphospecies

The lack of population differentiation for S. glynni among the Pocillopora morphospecies sampled further supports a body of evidence indicating that Pocillopora type

1 is a cohesive species. Other genetically differentiated Pocillopora in the ETP (i.e., types 2

& 3) do not appear to associate with S. glynni (Pinzón & LaJeunesse 2011). In Hawaii were morphospecies correspond to genetically defined groupings each Pocillopora associates with a different Clade C Symbiodinium (LaJeunesse et al. 2004; Pinzón 2011). Reproductive isolation due to host specialization is believed to be a critical process in generating and maintaining Symbiodinium diversity (LaJeunesse 2005). The specificity for a particular host species, genus, or family is a common ecological trait among Symbiodinium spp. (LaJeunesse

2002; LaJeunesse et al. 2004, 2010a, LaJeunesse 2005, Frade et al. 2007; Sampayo et al.

2009; Bongaerts et al. 2010; Pinzón & LaJeunesse 2011). A symbiont’s fitness is dependent on the establishment, persistence, reproduction and successful dispersal from the cellular and biochemical surroundings influenced by the host genetic makeup. The lack of subdivision among S. glynni populations associated with different Pocillopora morphospecies suggests that there are no intracellular differences imparting selective pressure to influence the evolution of reproductive isolation (Schluter 2009; Figure 11, Table 7). Therefore, the symbiont associates with Pocillopora morphospecies of type 1 as if they are a single homogeneous habitat (i.e., species).

83

A single S. glynni genotype detected in most samples

The routine detection of only a single genotype of Symbiodinium per sample makes population genetic studies like this one possible. Of the four hundred and two samples analyzed, less than twenty percent contained detectable mixtures of multiple genotypes and this proportion lowers to less than seven percent when only samples with multiple alleles at more than one locus are counted. The possibility that Symbiodinium populations tend to be homogeneous over a portion and/or the entire colony is supported by recent analyses of scleractinians and gorgonians in the Caribbean where only a single genotype of a particular

Symbiodinium spp. is recovered from > 70 to 95% of samples (Pseudopterogorgia elisabethae, Santos et al. 2003; Madracis spp., Pettay & LaJeunesse 2007; Gorgonia ventalina, Kirk et al. 2009; Montastrea annularis and M. faveolata, Thornhill et al. 2009;

Acropora palmata, Pinzón et al. 2011; Gorgonia ventalina, Andras et al. 2011). These microsatellite data are consistent with the pioneering work of Goulet and Coffroth (2003a, b) who, utilizing DNA fingerprinting, detected a single dominant genotype from the basal, middle and apical regions of gorgonian colonies monitored for as long as 10 years.

Samples from Pocillopora meandrina and the soft coral Sinularia flexibilis from

French Polynesia and the GBR, respectively, found more than one and often several Clade C genotypes in 85 and 94% of cases (Magalon et al. 2005; Howells et al. 2009). Processes regulating clone diversity within a colony may differ for Clade C Symbiodinium, allowing more genotypes to persist. However, given that Symbiodinum genomes contain regions of highly repetitive DNA that are readily apparent during microsatellite development (Pettay &

LaJeunesse 2007, 2009; Pinzón et al. 2011), a more likely explanation for the excessive

84 frequency of samples with multiple genotypes may be that microsatellite primers developed in these studies amplify multiple products within the genome due to non-specific binding and/or loci that have undergone recent duplication events (Pettay & LaJeunesse 2009; Wham et al. 2011). Testing microsatellite loci on isoclonal cultures and sequencing different fragment products (when they occur) would resolve the possibility that a high number of alleles reported per sample resulted from the PCR amplification of duplicated genomic regions and/or instances of non-specific primer binding (Santos & Coffroth 2003, Pettay &

LaJeunesse 2007, 2009; Pinzón et al. 2011).

Implications for conservation in the ETP

The ability of Pocillopora type 1 and S. glynni to disperse throughout much of the eastern Pacific suggests that the shallow water ecosystems they create may be resilient to local disturbances and capable of recovery by migration of individuals from surrounding sources. However, the lack of gene flow between S. glynni populations in the GoC and the

ETP indicates that these high latitude populations may be adapted to the unique environmental conditions that are characteristic of the Gulf region. The GoC population is small relative to the ETP and may be particularly vulnerable to extinction if its host population were lost. Conversely, as the climate continues to warm symbiont populations adapted to more marginal conditions like the GoC may become a source of physiological diversity. While more data are needed to evaluate the relative exchange of the eastern Pacific with the rest of the Pacific, the thermal tolerance of S. glynni and its ability to disperse large distances with its host suggest that this region may provide a source of Pocillopora whose symbionts are adapted to a warmer more environmentally unstable planet.

85 References

Anderson T, Haubold B, Williams JT, et al. (2000) Microsatellite markers reveal a spectrum of population structures in the malaria parasite Plasmodium falciparum. Molecular Biology and Evolution 17, 1467-1482.

Andras JP, Kirk NL, Harvell CD (2011) Range-wide population genetic structure of Symbiodinium associated with the Caribbean Sea fan coral, Gorgonia ventalina. Molecular Ecology, online first.

Arnaud-Haond S, Duarte CM, Alberto F, Serrão EA (2007) Standardizing methods to address clonality in population studies. Molecular Ecology 16, 5115-5139.

Ayre DJ, Hughes TP, Standish RJ (1997) Genetic differentiation, reproductive mode, and gene flow in the brooding coral Pocillopora damicornis along the Great Barrier Reef, Australia. Marine Ecology Progress Series 159, 175-187.

Ayre DJ, Minchinton TE, Perrin C (2009) Does life history predict past and current connectivity for rocky intertidal invertebrates across a marine biogeographic barrier? Molecular Ecology 18, 1887-1903.

Baker A, Starger C, McClanahan T, Glynn P (2004) Corals' adaptive response to climate change. Nature 430, 741.

Bohonak A (1999) Dispersal, gene flow, and population structure. Quarterly review of biology 74, 21-45.

Bongaerts P, Riginos C, Ridgway T, Sampayo E (2010) Genetic divergence across habitats in the widespread coral Seriatopora hystrix and its associated Symbiodinium. PLoS ONE 5, e10871.

Casteleyn G, Evans KM, Backeljau T, et al. (2009) Lack of population genetic structuring in the marine planktonic diatom Pseudo-nitzschia pungens (Bacillariophyceae) in a heterogeneous area in the Southern Bight of the North Sea. Marine Biology 156, 1149-1158.

Casteleyn G, Leliaert F, Backeljau T, et al. (2010) Limits to gene flow in a cosmopolitan marine planktonic diatom. Proceedings of the National Academy of Sciences 107, 12952-12957.

Castro R, Mascarenhas AS, Durazo R, Collins CA (2000) Seasonal variation of the temperature and salinity at the entrance to the Gulf of California, Mexico. Ciencias Marinas 26, 561-583.

86 Chávez-Romo H, Correa-Sandoval F, Paz-García D, et al. (2008) Genetic structure of the scleractinian coral, Pocillopora damicornis, from the Mexican Pacific. Proceedings of the 11th International Coral Reef Symposium, 429-433.

Chávez-Romo HE, Reyes-Bonilla H (2007) Sexual reproduction of the coral Pocillopora damicornis in the southern Gulf of California, Mexico. Ciencias Marinas 33, 495- 501.

Coffroth M, Santos S (2005) Genetic Diversity of Symbiotic Dinoflagellates in the Genus Symbiodinium. Protist 156, 19 - 34.

Combosch D, Guzman H, Schuhmacher H, Vollmer S (2008) Interspecific hybridization and restricted trans-Pacific gene flow in the Tropical Eastern Pacific Pocillopora. Molecular Ecology 17, 1304-1312.

Corander J, Marttinen P, Mäntyniemi S (2006) A Bayesian method for identification of stock mixtures from molecular marker data. Fishery Bulletin 104, 550-558.

Corander J, Waldmann P, Marttinen P, Sillanpaa M (2004) BAPS 2: enhanced possibilities for the analysis of genetic population structure. Bioinformatics 20, 2363-2369.

Corander J, Waldmann P, Sillanpaa MJ (2003) Bayesian analysis of genetic differentiation between populations. Genetics 163, 367-274.

Cowen R, Lwiza K, Sponaugle S, Paris C, Olson D (2000) Connectivity of marine populations: open or closed? Science 287, 857.

Darling K, Kucera M, Pudsey C, Wade C (2004) Molecular evidence links cryptic diversification in polar planktonic protests to Quarternary climate dynamics. Proceedings of the National Academy of Science 101, 7657-7662.

Darling KF, Kucera M, Wade CM (2007) Global molecular phylogeography reveals persistent Arctic circumpolar isolation in a marine planktonic protist. Proceedings of the National Academy of Sciences 104, 5002-5007.

Earl DA (2009) Structure Harvester, from website: v0.3 http://users.soe.ucsc.edu/~dearl/software/struct_harvest/.

Evanno G, Regnaut S, Goudet J (2005) Detecting the number of clusters of individuals using the software STRUCTURE: a simulation study. Molecular Ecology 14, 2611-2620.

Excoffier L, Smouse P, Quattro J (1992) Analysis of molecular variance inferred from metric distances among DNA haplotypes: application to human mitochondrial DNA restriction data. Genetics 131, 479-491.

87 Falush D, Stephens M, Pritchard J (2003) Inference of population structure using multilocus genotype data: Linked loci and correlated allele frequencies. Genetics 164, 1567- 1587.

Fenchel T, Finlay B (2004) The ubiquity of small species: patterns of local and global diversity. Bioscience 54.

Fiedler P, Chavez F, Behringer D, Reilly S (1992) Physical and biological effects of Los Niños in the eastern tropical Pacific, 1986-1989. Deep Sea Research Part A. Oceanographic Research Papers 39, 199-219.

Finlay B (2002) Global dispersal of free-living microbial eukaryote species. Science 296, 1061-1063.

Fitt W, Trench R (1983) The Relation of Diel Patterns of Cell Division To Diel Patterns of Motility in the Symbiotic Dinoflagellate Symbiodinium Microadriaticum FREUDENTHAL in Culture. New Phytologist 94, 421-432.

Foissner W (2006) Biogeography and dispersal of micro-organisms: a review emphasizing protists. Acta Protozoologica 45, 111-136.

Frade PR, De Jongh F, Vermeulen F, Van Bleijswijk J, Bak RPM (2007) Variation in symbiont distribution between closely related coral species over large depth ranges. Molecular Ecology 17, 691-703.

Glynn P, Ault J (2000) A biogeographic analysis and review of the far eastern Pacific coral reef region. Coral Reefs 19, 1-23.

Glynn P, Gassman N, Eakin C, et al. (1991) Reef coral reproduction in the eastern Pacific: Costa Rica, Panama, and Galapagos Islands (Ecuador). I. Pocilloporidae. Reproducción de los arrecifes de coral en el Pacífico oriental: Costa Rica, Panamá y las Islas Galápagos (Ecuador). I. Pocilloporidae. Marine Biology. 109, 355-368.

Glynn P, Mate J, Baker A, Calderon M (2001) Coral bleaching and mortality in Panama and Ecuador during the 1997-1998 El Niño-Southern Oscillation Event: spatial/temporal patterns and comparisons with the 1982-1983 event. Bulletin of Marine Science 69, 79-109.

Glynn P, Veron J, Wellington G (1996) Clipperton Atoll (eastern Pacific): oceanography, geomorphology, reef-building coral ecology and biogeography. Coral Reefs 15, 71- 99.

Goulet T, Coffroth M (2003a) Stability of an octocoral-algal symbiosis over time and space. Marine Ecology Progress Series 250, 117-124.

88 Goulet TL, Coffroth MA (2003b) Genetic composition of zooxanthellae between and within colonies of the octocoral Plexaura kuna, based on small subunit rDNA and multilocus DNA fingerprinting. Marine Biology 142, 233-239.

Hirose M, Kinzie R, Hidaka M (2000) Early development of zooxanthella-containing eggs of the corals Pocillopora verrucosa and P. eydouxi with special reference to the distribution of zooxanthellae. Biological Bulletin 199, 68-75.

Hoegh-Guldberg O, McCloskey LR, Muscatine L (1987) Expulsion of zooxanthellae by cnidarians from the Red Sea. Coral Reefs 5, 201-204.

Howells E, Van Oppen M, Willis B (2009) High genetic differentiation and cross-shelf patterns of genetic diversity among Great Barrier Reef populations of Symbiodinium. Coral Reefs 28, 215-225.

Hurtado L, Frey M, Gaube P, Pfeiler E, Markow T (2007) Geographical subdivision, demographic history and gene flow in two sympatric species of intertidal snails, Nerita scabricosta and Nerita funiculata, from the tropical eastern Pacific. Marine Biology 151, 1863-1873.

Jones R, Yellowlees D (1997) Regulation and control of intracellular algae (= zooxanthellae) in hard corals. Philosophical Transactions of the Royal Society B: Biological Sciences 352, 457-468.

Kirk N, Andras J, Harvell C, Santos S, Coffroth M (2009) Population structure of Symbiodinium sp. associated with the common sea fan, Gorgonia ventalina, in the Florida Keys across distance, depth, and time. Marine Biology 156, 1609-1623.

LaJeunesse T (2001) Investigating the biodiversity, ecology, and phylogeny of endosymbiotic dinoflagellates in the genus Symbiodinium using the ITS region: in search of a" species" level marker. Journal of Phycology 37, 866-880.

LaJeunesse T (2002) Diversity and community structure of symbiotic dinoflagellates from Caribbean coral reefs. Marine Biology 141, 387-400.

LaJeunesse T, Bonilla H, Warner M, et al. (2008) Specificity and stability in high latitude eastern Pacific coral-algal symbioses. Limnology and Oceanography 53, 719-727.

LaJeunesse T, Loh W, Van Woesik R, et al. (2003) Low symbiont diversity in southern Great Barrier Reef corals, relative to those of the Caribbean. Limnology and Oceanography, 2046-2054.

LaJeunesse TC (2005) ‘‘Species’’ radiations of symbiotic dinoflagellates in the Atlantic and Indo-Pacific since the Miocene-Pliocene transition. Molecular Biology and Evolution 22, 570-581.

89 LaJeunesse TC, Bhagooli R, Hidaka M, et al. (2004) Closely related Symbiodinium spp. differ in relative dominance in coral reef host communities across environmental, latitudinal and biogeographic gradients. Marine Ecology Progress Series 284, 147- 161.

LaJeunesse TC, Pettay DT, Sampayo EM, et al. (2010a) Long-standing environmental conditions, geographic isolation and host-symbiont specificity influence the relative ecological dominance and genetic diversification of coral endosymbionts in the genus Symbiodinium. Journal of Biogeography 37, 785-800.

LaJeunesse TC, Smith R, Walther M, et al. (2010b) Host–symbiont recombination versus natural selection in the response of coral–dinoflagellate symbioses to environmental disturbance. Proceedings of the Royal Society B 277, 2925-2934.

LaJeunesse TC, Smith RT, Finney J, Oxenford H (2009) Outbreak and persistence of opportunistic symbiotic dinoflagellates during the 2005 Caribbean mass coral 'bleaching' event. Proceedings of the Royal Society B: Biological Sciences 276, 4139- 4148.

LaJeunesse TC, Trench RK (2000) The biogeography of two species of Symbiodinium (Freudenthal) inhabiting the intertidal anemone, Anthopleura elegantissima (Brandt). Biological Bulletin 199, 126-134.

Latch EK, Dharmarajan G, Glaubitz JC, Rhodes OE (2006) Relative performance of Bayesian clustering software for inferring population substructure and individual assignment at low levels of population differentiation. Conservation Genetics 7, 295- 302.

Lilly EL, Halanych KM, Anderson DM (2007) Species boundaries and global biogeography of the Alexandrium tamarense complex (Dinophyceae). Journal of Phycology 43, 1329-1338.

López-Pérez RA (2005) The Cenozoic hermatypic corals in the eastern Pacific: History of research. Earth-Science Reviews 72, 67-87.

Magalon H, Adjeroud M, Veuille M (2005) Patterns of genetic variation do not correlate with geographical distance in the reef-building coral Pocillopora meandrina in the South Pacific. Molecular Ecology 14, 1861-1868.

Magalon H, Baudry E, Husté A, Adjeroud M, Veuille M (2006) High genetic diversity of the symbiotic dinoflagellates in the coral Pocillopora meandrina from the South Pacific. Marine Biology 148, 913-922.

90 Miller K, Ayre D (2008) Population structure is not a simple function of reproductive mode and larval type: insights from tropical corals. Journal of Animal Ecology 77, 713-724.

Mills L, Citta J, Lair K, Schwartz M, Tallmon D (2000) Estimating animal abundance using noninvasive DNA sampling: promise and pitfalls. Ecological Applications 10, 283- 294.

Nagai S, Lian C, Yamaguchi S, et al. (2007) Microsatellite markers Reveal Population Genetic Structure of the Toxic Dinoflagellate Alexandrium tamarense (Dinophyceae) in Japanese Coastal Waters. Journal of Phycology 43, 43-54.

Nagai S, Nishitani G, Sakamoto S, et al. (2009) Genetic structuring and transfer of marine dinoflagellate Cochlodinium polykrikoides in Japanese and Korean coastal waters revealed by microsatellites. Molecular Ecology 18, 2337-2352.

Peakall R, Smouse PE (2006) GENALEX 6: genetic analysis in Excel. Population genetic software for teaching and research. Molecular Ecology Notes 6, 288-295.

Pettay D, LaJeunesse T (2007) Microsatellites from clade B Symbiodinium spp. specialized for Caribbean corals in the genus Madracis. Molecular Ecology Notes 7, 1271-1274.

Pettay DT, LaJeunesse TC (2009) Microsatellite loci for assessing genetic diversity, dispersal and clonality of coral symbionts in ‘stress-tolerant’ clade D Symbiodinium. Molecular Ecology Resources 9, 1022-1025.

Pettay DT, LaJeunesse TC (submitted) The genotype diversity, distribution and spatial- temporal dominance of Symbiodinium clone lineages among colonies of Pocillopora from the Gulf of California. Molecular Ecology.

Pinzón JH (2011) Phylogenetics, population genetics, and ecology to understand the evolution of coral-algal mutualisms. PhD dissertation. The Pennsylvania State University pp. 154

Pinzón JH, Devlin-Durante MK, Weber M, Baums IB, LaJeunesse TC (2011) Microsatellite loci for Symbiodinium A3 (S. fitti) a common algal symbiont among Caribbean Acropora (stony corals) and Indo-Pacific giant clams (Tridacna). Conservation Genetics Resources 3, 45-47.

Pinzón JH, LaJeunesse TC (2011) Species delimitation of common reef corals in the genus Pocillopora using nucleotide sequence phylogenies, population genetics and symbiosis ecology. Molecular Ecology 20, 311-325.

Pritchard J, Stephens M, Donnelly P (2000) Inference of population structure using multilocus genotype data. Genetics 155, 945.

91 Richmond R (1987) Energetics, competency, and long-distance dispersal of planula larvae of the coral Pocillopora damicornis. Marine Biology 93, 527-533.

Richmond RH (1985) Variations in the population biology of Pocillopora damincornis across the Pacific. Proceedings of the 5th International Coral Reef Symposium 6, 101-106.

Ridgway T, Riginos C, Davis J, Hoegh-Guldberg O (2008) Genetic connectivity patterns of Pocillopora verrucosa in southern African Marine Protected Areas. Marine Ecology Progress Series 354, 161-168.

Rodriguez-Lanetty M, Loh W, Carter D, Hoegh-Guldberg O (2001) Latitudinal variability in symbiont specificity within the widespread scleractinian coral Plesiastrea versipora. Marine Biology 138, 1175-1181.

Rosenberg N (2004) DISTRUCT: a program for the graphical display of population structure. Molecular Ecology Notes 4, 137-138.

Rousset F (2008) GENEPOP'007: a complete re-implementation of the GENEPOP software for Windows and Linux. Molecular Ecology Resources 8, 103-106.

Rowan R, Knowlton N, Baker A, Jara J (1997) Landscape ecology of algal symbionts creates variation in episodes of coral bleaching. Nature 388, 265-269.

Rynearson TA, Armbrust EV (2004) Genetic differentiation among populations of the planktonic marine diatom Ditylum bright- wellii (Bacillariophyceae). Journal of Phycology 40, 34-43.

Sampayo EM, Dove S, LaJeunesse TC (2009) Cohesive molecular genetic data delineate species diversity in the dinoflagellate genus Symbiodinium. Molecular Ecology 18, 500-519.

Santos S, Coffroth M (2003) Molecular genetic evidence that dinoflagellates belonging to the genus Symbiodinium Freudenthal are haploid. The Biological Bulletin 204, 10-20.

Santos S, Gutierrez-Rodriguez C, Lasker H, Coffroth M (2003) Symbiodinium sp. associations in the gorgonian Pseudopterogorgia elisabethae in the Bahamas: high levels of genetic variability and population structure in symbiotic dinoflagellates. Marine Biology 143, 111-120.

Santos S, Shearer T, Hannes A, Coffroth M (2004) Fine-scale diversity and specificity in the most prevalent lineage of symbiotic dinoflagellates (Symbiodinium, Dinophyceae) of the Caribbean. Molecular Ecology 13, 459-469.

92 Seutin G, White BN, Boag PT (1991) Preservation of avian blood and tissue samples for DNA analyses. Canadian Journal of Zoology 69, 82–92.

Shankle AM, Mayali X, Franks PJS (2004) Temporal patterns in population genetic diversity of Prorocentrum micans (Dinophyceae). Journal of Phycology 40, 239-247.

Thornhill D, Xiang Y, Fitt W, Santos S (2009) Reef endemism, host specificity and temporal stability in populations of symbiotic dinoflagellates from two ecologically dominant Caribbean corals. PLoS One 4, e6262.

Thornhill DJ, LaJeunesse TC, Santos SR (2007) Measuring rDNA diversity in eukaryotic microbial systems: how intragenomic variation, pseudogenes, and PCR artifacts confound biodiversity estimates. Molecular Ecology 16, 5326-5340.

Trench RK (1987) Dinoflagellates in non-parasitic symbioses. In: The biology of Dinoflagellates, botantical monographs (ed. Taylor FJR), pp. 531-570. Blackwell Scientific Publications, Oxford.

Ulstrup KE, van Oppen MJH (2003) Geographic and habitat partitioning of genetically distinct zooxanthellae (Symbiodinium) in Acropora corals on the Great Barrier Reef. Molecular Ecology 12, 3477-3484.

Victor BC, Wellington GM, Robertson DR, Ruttenberg BI (2001) The effect of the El Nino- Southern Oscillation event on the distribution of reef-associated labrid fishes in the eastern Pacific Ocean. Bulletin of Marine Science 69, 279-288.

Waits LP, Luikart G, Taberlet P (2001) Estimating the prob- ability of identity among genotypes in natural populations: cautions and guidelines. Molecular Ecology 10, 249-256.

Wham DC, Pettay DT, LaJeunesse TC (2011) Microsatellite loci for the host-generalist “zooxanthella” Symbiodinium trenchi and other Clade D Symbiodinium. Conservation Genetic Resources, 1-4.

Wintzingerode F, Göbel U, Stackebrandt E (1997) Determination of microbial diversity in environmental samples: pitfalls of PCR-based rRNA analysis. FEMS Microbiology Reviews 21, 213-229.

93 CHAPTER 3

Population structure of the pan-Pacific Pocillopora symbiont, Symbiodinium glynni

Abstract

Relatively little is known about the dispersal and gene flow among populations of microbial eukaryotes across the extensive longitudinal range of the Pacific Ocean. The high abundance of symbiotic dinoflagellates (Symbiodinium) in reef corals allows the investigation of their population differentiation across the tropical Pacific through the sampling of independent colonies from various locations. Symbiont populations acquired from colonies of Pocillopora spp. from the tropical (ETP) and subtropical (Gulf California, GoC) eastern Pacific, Palau in the west Pacific, and the Gulf of Thailand were analyzed with eleven microsatellite loci. A fifth population of Pocillopora spp. that associates with a phylogenetically different type of Clade D (D5) from the Andaman Sea in the northeastern Indian Ocean was also examined. Bayesian clustering resolved several distinct populations

(ΦRT = 0.280, p = 0.0001) including the previously described latitudinal separation between the GoC and ETP, which were well differentiated from populations in the west Indo-Pacific region. Additional structuring between Palau, the Gulf of Thailand and the Andaman Sea was statistically significant (ΦPR = 0.179, p = 0.0001). Latitudinal environments and large oceanic expanses appear to have structured S. glynni into genetically differentiated populations. However, several genotypes from Palau were similar in allelic composition to genotypes found in the GoC, suggesting that occasional dispersal may occur between east and west Pacific regions. While S. glynni associates with Pocillopora type 1 across the GoC and ETP, it associates with a different species of Pocillopora (genetically-defined as type 5) in western regions and this may in part explain the longitudinal differentiation. These findings highlight the complex nature of population structuring of coral-algal symbioses.

94 Introduction

Isolation and marginal environmental conditions characterize the Eastern Tropical

Pacific (ETP) and have been implicated in the depauperate marine communities of the region

(Glynn 1997). The large expanse of open-ocean between the central Pacific and ETP (~5000 km or more) is the most likely dispersal route for marine organism between the two regions and has been largely regarded as impassible and consequently labeled the Eastern Pacific

Barrier (EPB; Ekman 1953). This expanse of ocean has existed since the Cenozoic (Grigg &

Hey 1992) and has led to distinct biota on either side of the barrier, yet some trans-Pacific species do exist (Briggs 1961, Vermeij 1987, Glynn et al. 1996, Robertson & Allen 1996,

Lessios et al. 1998, Lessios et al. 2003, Lessios & Robertson 2006). It is widely believed that these trans-Pacific species represent relatively recent long-distance dispersal events

(Dana 1975, Grigg & Hey 1992, Glynn et al. 1996, Robertson & Allen 1996, Glynn & Ault

2000) and research suggests that the EPB may be more permeable than once believed

(Lessios et al. 1998, Lessios & Robertson 2006, Duda & Lessios 2009).

Macro-invertebrates and fishes have been the primary focus of research on gene flow across the Pacific (e.g., McMillan & Palumbi 1995, Palumbi 1996, Lessios et al. 1998,

Lessios et al. 1999, Lessios et al. 2001, Lessios et al. 2003, Williams & Reid 2004, Lessios

& Robertson 2006, Volger et al. 2008, Duda & Lessios 2009), yet the population patterns of microorganisms across such vast regions remain poorly understood (reviewed in Green &

Bohannan 2006, Foissner 2006). Microorganisms comprise a large portion of the planet’s biodiversity and are necessary for biogeochemical cycling and ecosystem function; so it is important to understand the processes that influence their distribution and diversity (Morin &

McGrady-Steed 2004, Green & Bohannan 2006). In marine environments microbial

95 eukaryotes (e.g., diatoms, dinoflagellates, foraminiferans) are one group of microorganisms whose population processes are beginning to receive attention. In general, populations of these organisms appear well differentiated over large geographic areas (e.g., between ocean basins). Of the few studies examining biogeographic patterns at this scale many are based on data from conserved genetic markers and/or haphazardly cultured isolates (Iglesias-

Rodríguez et al. 2006, Darling et al. 2007, Lilly et al. 2007, Masseret et al. 2009, McCauley et al. 2009, Casteleyn et al. 2010, Penna et al. 2010).

The symbioses between reef corals and their dinoflagellate endosymbionts, genus

Symbiodinium, are widespread in tropical marine waters and construct the calcium carbonate framework on which the coral reef ecosystem depends (Trench 1993). While their ability to calcify makes these symbioses ecologically important, their abundance makes Symbiodinium one of the most prevalent microbial eukaryotes in the tropics. Many Symbiodinium species form highly specific associations with their hosts (reviewed in Coffroth & Santos 2005) and this specificity, along with their abundance and sessile nature, makes finding and collecting particular symbionts relatively easily. Therefore, Symbiodinium is an ideal model organism to study population diversity and dispersal of dinoflagellates. While several studies have documented that various Symbiodinium spp. can occur over large distances and across ocean basins (LaJeunesse 2002, LaJeunesse et al. 2004a, Finney et al. 2010, LaJeunesse et al.

2010), nothing is known of the population structuring and potential for gene flow over large distances.

Pocilloporid corals are one of the most ubiquitous corals in the Indo/Pacific (Veron

2000) and associate with specific types of Symbiodinium (LaJeunesse et al. 2004b, Sampayo et al. 2007, LaJeunesse et al. 2010, Pinzón & LaJeunesse 2011). In the eastern Pacific the

96 symbiosis between the stress-tolerant clade D symbiont Symbiodinium glynni and

Pocillopora type I appears to dominate mainland locations and also extends offshore into the

Galapagos Islands and Clipperton Atoll (LaJeunesse et al. 2010, Pinzón & LaJeunesse 2011).

Recent investigations into the population structuring of this symbiosis has revealed that both the host and symbiont show little differentiation across the ETP, suggesting the propensity for long-distance dispersal at regional scales (Pinzón & LaJeunesse 2011, Pettay &

LaJeunesse in press). To build upon these regional analyses this study reports on the pan-

Pacific distribution S. glynni populations from the Gulf of Thailand, Palau and the eastern

Pacific through the analyses of eleven microsatellite loci to assess the population structuring, gene flow, and the possible dispersal of this symbiont across the Pacific Ocean.

Materials and Methods

Sample Collections

Samples analyzed in this study were part of a larger biogeographical analysis of

Pocillopora spp. symbionts across the Indo/Pacific and include previously described samples acquired from the ETP (LaJeunuesse et al. 2008, 2010b). Fragments (~2 cm2) of Pocillopora colonies were collected by SCUBA from ten sampling locations (Figure 13) and preserved in either a high salt, 20% DMSO buffer (Seutin et al. 1991) or 95% ethanol and stored at -20 °C until DNA extraction. At each location, samples were collected along linear transects at depths ranging from two to twenty meters at each of one to four sites separated by as much as

50 km. As Pocillopora are known to fragment (Richmond 1985, unpubl. data), morphologically distinct colonies were sampled at least 3 meters apart to avoid sampling clonemates. Samples were collected from Adang-Rawi Island (Satun Province) in February

97

Figure 13. Sampling locations in the West Indo-Pacific, including the Andaman Sea (AND, 1), the Gulf of Thailand (GHT, 2), Palau (PAL, 3) and from the Tropical and Subtropical Eastern Pacific including the Gulf of California (GoC, 4), Banderas Bay (BB, 5), Gulf of Tehuantepec (OAX, 6), Clipperton Atoll (CLP, 7), Gulf of Panama (PAN, 8) and Galapagos Islands (GAL, 9). At each location, collections of samples were usually made at several sites separated by 2-150 km.

of 2009, KangKao Island (Chonburi Province) in February of 2009, Man Islands (Rayong

Province) Thailand in February of 2007, and Palau (PAL) in May of 2009. Both KangKao and Man Islands are in the northern Gulf of Thailand (GTH), while Adang-Rawi is on the other side of the Malaysian peninsula in the Andaman Sea (AND). The six sample locations in the eastern Pacific were previously described by Pettay & LaJeunesse (in prep) and are as follows: Gulf of California (GoC), Banderas Bay (BB), Clipperton Atoll (CLP), Gulf of

Tehuantepec (OAX), Gulf of Panama (PAN), and the Galapagos Islands (GAL). Samples collected from the west Pacific, except four from Palau and described below, all belonged to the Pocillopora damicornis morpho-species and represented type 5 as designated by Pinzón

& LaJeunesse (unpub.), while the samples from the eastern Pacific spanned several morpho- species, yet were all Pocillopora type 1 (Pinzón & LaJeunesse 2011). A total of 482 samples were analyzed to determine the population structure of S. glynni across the Pacific, with the sampling locations and sample sizes given in Figure 13 and Table 11.

98 Molecular-Genetic Identification

Nucleic acid extractions were conducted using a modified Promega Wizard genomic

DNA extraction protocol (LaJeunesse et al. 2003). The dominant resident symbiont was identified by denaturing gradient gel electrophoresis (DGGE) fingerprinting of the partial

5.8S and internal transcribed spacer (ITS) region 2 (LaJeunesse 2002). The region was amplified using a touch-down thermal cycle profile with the primers “ITS2clamp” and

“ITSintfor2” (LaJeunesse & Trench 2000), and the PCR products resolved on denaturing gels

(45 – 80% of 7M urea & 40% formamide) using a CBScientific system (Del Mar, CA) for 16 hours at 115 volts. Samples where S. glynni (D1) was detected were utilized for this study, as were samples containing Symbiodinium D5 (See Results).

Microsatellite Analysis

Eleven variable microsatellite loci developed for Clade D Symbiodinium were used to determine the degree of population structuring of S. glynni across Pacific Ocean (Pettay &

LaJeunesse 2009; Wham et al. 2011). These eleven loci proved adequate to resolve populations of S. glynni in the eastern Pacific and were thoroughly described therein (Pettay

& LaJeunesse in prep). Therefore, samples from the western Pacific are newly analyzed as part of this study, while all data from the eastern Pacific were previously generated for analyses on that region.

The forward primers of all loci were fluorescently labeled and the microsatellite fragments amplified using the following PCR mix; 10 µl reaction volumes containing 2.5 µM dNTP’s, 0.2 U Taq DNA Polymerase (New England Biolabs or Amplitaq), 1x Mg-free

Buffer, 2.5 mM MgCl2, 1 µM of each forward and reverse primer and ~ 50 ng of DNA

99 template. Reaction conditions consisted of: initial denaturing step of 94°C for 2 min, 32 cycles of 94°C for 15 s, annealing temperature (Ta, see Table10) for 15 s, 72°C for 15 s, and a final extension of 72°C for 5 min. Following amplification, fragment sizes were analyzed on an ABI 3730 Genetic Analyzer (Applied Biosystems, Foster City, CA) using a 500 bp standard (LIZ-labeled) at the Pennsylvania State University Genomics Core Facility.

Fragments sizes were visually analyzed using GeneMarker v.1.51 (SoftGenetics, State

College, PA). Since microsatellites are known to produce stutter due to polyermase slippage during PCR, only the dominant peaks were scored. Multiple alleles at a particular locus were scored when a second clear peak was found in the expected size range and was at least one- third the size of the dominant peak. The presence of multiple peaks was interpreted to indicate that a sample contained more than one haploid genotype. The occasional exclusion of small ambiguous peaks may occasionally cause a low background genotype to be missed, however this conservative approach prevents the overestimation of genetic diversity

(Anderson et al. 2000). Additionally, while this approach may slightly skew the results of diversity on a per colony basis, it should have little impact on the population structure analyses since these mixed genotype samples did not result in additional allelic diversity.

Multilocus genotypes (MLGs) were constructed from fragment size data gathered for each sample. A subset of samples was reanalyzed to confirm the existence of mixed genotypes and unusual or rare fragment (e.g. allele) sizes. When a sample possessed two or more alleles at one or more loci, only the dominant (i.e., highest peak) fragment size was used to construct the MLG. The amplification of loci from experimental mixtures containing different ratios of two genotypes correctly amplified the fragment highest in concentration and did not show preferential amplification for the smaller fragment (Pettay & LaJeunesse in

100 prep). Therefore, constructing MLGs using the dominant fragments adds little bias, yet allows a maximum number of MLGs to be recovered.

Data Analysis

Populations of S. glynni are composed of both sexually and asexually (i.e., clones) derived individuals (Pettay & LaJeunesse in prep, Pettay et al. in prep.). The inclusion of repetitive MLGs due to asexual division can greatly skew statistical analyses based on allele frequencies (discussed in Pettay & LaJeunesse in prep). Therefore, all analyses (unless stated otherwise) were conducted with duplicated MLGs removed at each sampling location

(e.g., GoC or PAL).

Linkage among pairs of loci was measured using GenePop (version 4.0.11, Rousett

2 2 2008) and tests were conducted for each location. The probability of identity (PI = (Σpi ) -

4 Σpi , where pi is the frequency of the i-th allele at a locus) was calculated to determine the power to resolve genetically distinct individuals. The PI is an estimate of the probability that two unrelated individuals drawn at random will by chance have the same MLG (Peakall &

Smouse 2005). Individual PI’s are calculated for each locus and an overall PI for all loci is the product of each individual locus PI. PI values between 0.01 – 0.0001 are believed to be reasonably low enough for population studies (Waits et al. 2001), with values lower than

0.01 adequate for mark-recapture studies on population size estimation (Mills et al. 2000).

Since these values may be affected by population substructure (Waits et al. 2001), PI was calculated for each location and for the entire dataset using GenAlEx (version 6.4, Peakall &

Smouse 2006).

101 The software Structure (Version 2.3.2) was used to overcome biases of assigning populations by location by using the microsatellite data to cluster MLGs based on their genetic similarities irrespective of sample origin. Briefly, this software uses a Bayesian clustering approach to probabilistically assign individuals to populations (Pritchard et al.

2000). The model assumes there are K populations characterized by a set of allele frequencies, with the assumptions of unlinked loci that are in linkage and Hardy-Weinberg equilibrium within populations (Pritchard et al. 2000).

Structure analyses were conducted on all data using a correlated allele model with admixture and were run from K = 1 to 15 with five runs per K and a burnin of 100,000 and

1,000,000 reps after the burnin. A plot of the log probability of the data for a give K (ln

P(D)) versus K was derived from the structure results, along with an analysis of the second order rate of change of K following the method by Evanno et al. (2005) and implemented using Structure Harvester (v0.56.4) by Earl (2009) to determine the appropriate clustering of individuals. Five runs per K were utilized to verifying consistency between runs, with the run having the highest ln P(D) for the appropriate K used to construct Structure plots and inform clustering. Graphic displays of Structure plots were manipulated (i.e., color and sample order) using DISTRUCT (Rosenberg 2004). Following the analyses on the entire

Pacific, separate Structure analyses were also conducted on the western Pacific data only.

These analyses were conducted to investigate clustering within the western Pacific without the influence of the eastern Pacific data and were also conducting using a correlated allele model with admixture and were run from K = 1 to 10, with five runs per K and a burnin of

100,000 and 1,000,000 reps after the burnin.

102 An AMOVA was performed to determine the degree of differentiation between the three regions (GoC, ETP and western Pacific), between locations within regions and within locations. Additionally, pairwise comparisons between all locations were also performed to determine pairwise Φ-statistics. The AMOVA and pairwise comparisons, along with a permutation procedure were performed in GenAlEx to test for significant difference in genetic diversity between populations (Excoffier et al. 1992). The AMOVA produces variance components along with Φ-statistics (F-statistic analogs), which partition genetic variation at different hierarchical levels (Excoffier et al. 1992). The significance of the variance components and Φ-statistics were then tested using 10000 permutations and a

Bonferroni corrected α = 0.05. In addition to quantifying genetic differentiation, GenAlEx was used to calculate several summary statistics for each Structure population and sampling

2 location, including h and I. Haploid genetic diversity (h = 1 - Σpi ) gives an indication that two individuals drawn at random will be genetically different, while information index (I) is a measure of allelic diversity (Peakall and Smouse 2006). Lastly, clonal richness (R) which is equal to (G – 1)/(N – 1), where G = # of unique MLG’s and N = total sample size, was also calculated to give the frequency of unique genotypes and an indication of the contribution of asexual reproduction.

Results

ITS2-DGGE fingerprinting and sequencing

Each of the MLGs characterized in this study possessed a single ITS2 sequence, either D1 or D5, which dominated their ribosomal array (Thornhill et al. 2007; Sampayo et al. 2009, LaJeunesse et al. 2010a). These sequences contrasts with each other and most other

103 Clade D types characterized by this method, indicating they each represents a distinct operational taxonomic unit (LaJeunesse et al. 2010a). The P. damicornis in Palau and the

Gulf of Thailand all possessed S. glynni (D1), similar to the Pocillopora in the eastern

Pacific. In contrast, the P. damicornis in the Andaman Sea associated with S. D5. Four additional Pocillopora colonies representing the P. verrucosa and meandrina morpho-species from Palau were found to associate with S. glynni and were also included in the population analyses.

Microsatellite Data

The number of alleles per locus ranged from 3 to 33, however, the effective number of alleles (Ae) per locus ranged from 1.03 to 5.18 and averaged 2.12 (Table 10). Allele frequencies for each location ranged from 0.011 to 1.00 (Appendix F). The frequencies of putative null alleles were rare among loci ranging from 0 for loci D1Sym67, D1Sym88 and

D1Sym92 to a value of 0.045 for D1Sym77a, except for D1Sym77b that had a frequency of

0.168. The high frequency of non-amplifying alleles for D1Sym77b is attributed to the high frequency of individuals from the GoC (0.406) and Palau (0.727) missing alleles at this locus. The frequency of non-amplifying alleles in the rest of the ETP (0.046) and the western

Pacific (0.000) for D1Sym77b are similar to other loci. The high frequency in the GoC was noted previously by Pettay & LaJeunesse (in prep) and is believed to be due to sequence divergence in the flanking regions in populations from the GoC.

Numerous alleles were private to a particular sampling location or region (Appendix

F). In the western Pacific, Palau possessed five alleles involving five loci (D1Sym 11, 14, 87

& 77b), while GTH possessed three alleles at two loci (D1Sym 17 & 34), and AND (S. D5)

104 Table 10. Description of microsatellite loci used in this study. Ta = annealing temperature, Ae = effective alleles with standard error in parenthesis. Subscript numbers following the repeat motif indicate the number of repeats in the initial cloned sequence that was used to develop locus primers. Size Number Range of Ae Locus Primer Sequence (5' - 3') Repeat Motif Ta (°C) (bp) Alleles

D1Sym9 F - CAGAAGCCCAATTATATGCGGCA (FAM) (GTT)6 57 106 - 115 4 1.26 (0.18) R - AGGATGATGAGCATGCCGACG

D1Sym11 F - TGAAATCTCACTCAGAGTCGGAC (FAM) (AC)13 57 151 - 167 8 2.24 (0.26) R - GCAGACAGTGATTTCAGTTCCGA

D1Sym14 F - TCTCAGTGGAAAGCATTGTGG (FAM) (CT)11 AT (CT)4 55 173 - 189 8 1.57 (0.28) R - TCGTCTGAATCAGGATCTGACG

D1Sym17 F - TGTGAATGCTTCTTGGGGTG (HEX) (CA)8 57 141 - 167 14 3.21 (0.67) 94 R - TCATGCTTGTCCGTGAGCAG

D1Sym34 F - ACCTGAGACCTGAGTGTTGC (FAM) (CAAA)9 CACA (CAAA)4 55 332 - 428 33 5.18 (GAAACAAA) (CAAA) (0.83) R - ATCATGGGCAGAGCTCCTGG 2 13

D1Sym67 F - GAATCCAGATGGTGCCTGC (VIC) (ATC)8 57 131 - 149 7 2.22 (0.28) R - CAAAGGTAGCCGATTGTCTC

D1Sym77a F - CCACTGAGATTGGTAGGTGAA (PET) (TTC)5 CT (CTTCCT)2 C 55 169 - 187 7 1.75 (TTC) (0.26) R - ACCGATGGTGTTTGTGACTCG 4

D1Sym77b F - CCACTGAGATTGGTAGGTGAA (PET) (TTC)5 CT (CTTCCT)2 C 55 184 - 196 5 1.27 (TTC) (0.11) R - ACCGATGGTGTTTGTGACTCG 4

D1Sym87 F - CCTATGACTCCAAGGGTGACG (FAM) (GAAG)7 57 236 - 268 9 2.58 (0.50) R - AGACATACCTCGGTCTTGTC

D1Sym88 F - TTGTCAGACTGAATGCTCCA (NED) (CTTT)3 G (TTTC) 55 227 - 235 3 1.03 R - GTGTTCAAGCGACATCCCA (0.02)

D1Sym92 F - GCGTTTGACACAAGGATCCCT (FAM) (CCTA)6 (CCTG)3 57 120 - 132 4 1.04 (0.02) R - TTGGGATGCTCTTGGCGAC

possessed only one allele at D1Sym34 that was unique to the location. In the ETP, BB possessed 10 alleles involving two loci (D1Sym34 & 77b), while the GoC possessed three alleles involving two loci (D1Sym14 & 88), and CLP and GAL each possessed 1 allele at one locus (D1Sym34) found only in these locations. All alleles in the OAX and PAN were found in at least one other site. Comparing the western to the eastern Pacific, a total of thirteen alleles from eight loci (D1Sym11, 14, 17, 34, 92, 87, 77a & 77b) were found only in the west, while forty-five alleles from ten loci (all but D1Sym9) were found in eastern populations.

Two hundred and forty-four different genotypes were scored among the 482 MLGs obtained. One hundred and twenty-three, or 25.5 percent, of samples possessed multiple

MLGs (i.e. more than one allele at one or more loci). A total of 24 samples (~5%) contained several loci with two alleles (11, 3, 4, 4, and 2 samples possessed allele variation at 2, 3, 4, 5,

6 loci, respectively). Multiple alleles at seven or more loci were never observed or at

D1Sym88 and D1Sym92.

The probability of identity (P(ID)) was calculated for samples from each location as well as for all locations together using allele frequency data at each locus (Table 11). The probability that two samples with the same MLG’s may not have originated from the same clone lineage was low for most locations and ranged from 1.46 x 10-2 to 7.76 x 10-7. Since a majority of these values, except those from PAN and AND, were at or below the 0.0001 criteria recommended by Waits et al. (2001), it is believed that these loci possess adequate resolving power for population studies and that the higher values of Thailand and Panama are a consequence of historical processes for these locations (Arnaud-Haond et al. 2007) and the fact that only one or two reefs were sampled for each of these locations. The overall P(ID)

106 Table 11. Summary statistics by location including number of colonies sampled, number of unique MLGs, clonal richness (R), haploid diversity (h), information index (I), and probability of identity (PI). Numbers in parenthesis represent the standard error for a particular statistic.

Clonal Probability Number of Unique Richness Haploid Information of Identity Location Samples MLG's (R) Diversity (h) Index (I) (PI) Palau (PAL) 18 11 0.59 0.54 (0.08) 1.03 (0.18) 6.02E-08 Gulf of Thailand (GTH) 39 11 0.26 0.32 (0.09) 0.55 (0.18) 1.35E-04 Andaman Sea (AND; S. D5) 23 7 0.27 0.18 (0.08) 0.27 (0.12) 1.46E-02 Gulf of California (GoC) 137 64 0.46 0.43 (0.09) 0.85 (0.20) 1.87E-06 Banderas Bay (BB) 155 87 0.56 0.40 (0.10) 0.92 (0.26) 7.76E-07 Gulf of Tehuantepec (OAX) 71 40 0.56 0.44 (0.08) 0.87 (0.18) 1.97E-06 Clipperton Atoll (CLIP) 7 6 0.83 0.33 (0.10) 0.58 (0.19) 6.49E-05 Bay of Panama (PAN) 21 7 0.30 0.30 (0.09) 0.46 (0.14) 7.55E-04 Galapagos Islands (GAL) 11 11 1.00 0.44 (0.11) 0.88 (0.23) 3.44E-07 Total 482 244 0.51 0.38 (0.03) 0.71 (0.07) 3.92E-10

based on data from all locations was 3.92 x 10-10. Therefore, there was a 1 in two and one half billion probability of encountering the same genotype, but from independent origins, in separate locations.

Clonal richness (R) or genotypic richness ranged from 0.26 to 1.00 for each location

(Table 11). Values approaching 0 indicate high clonality whereas values of 1.00 are indicative of populations where no repeated genotypes were sampled. Haploid diversity indices ranged from 0.18 to 0.54 with an overall average of 0.38 for all locations, while the information index ranged from 0.27 to 1.03 with an average of 0.71 (Table 11). The mean genotypic richness for all locations was 0.51 indicating that a large proportion (~ 49%) of

MLGs were detected two or more times. Several “locally prevalent” clones were unusually common within and among sites in a sampling location. In the western Pacific the largest clone was found in seventeen colonies in the Gulf of Thailand, while in the eastern Pacific

107 the largest clone was recovered from nineteen different colonies. Lastly, one S. D5 MLG was detected in nine different Pocillopora colonies.

A pairwise analysis of linkage among loci based on each sample location revealed that ~95% of all pairwise comparisons were unlinked (Bonferroni corrected; α = 0.05). In general, linkage between pairs of loci changed based on population, with no two loci linked in more than two locations. The linkage for some loci is believed to be a consequence of the clonal evolution of these dinoflagellates, population history or, in some cases, evidence of weak differentiation over smaller spatial scales.

Population Differentiation

Two well supported populations were detected using Structure (k = 2; ln P(D) = -

3067.1; Figure 14). The first population corresponded to the GoC and the western Pacific, while the second included the rest of the ETP locations: BB, OAX, CLP, PAN and GAL

(Figure 14). Structure may have difficulty with hierarchical differentiation depending on the sampling and strength of signal (Pritchard et al. 2010, Hubisz et al. 2009) and for this reason the method of Evanno et al. is meant to be guide for discerning population differentiation based on Structure analyses (Evanno et al. 2005, Pritchard et al. 2010). While the Evanno et al. method showed the largest change in the between k = 1 and 2 (Δk = 721.2), further examination of the analyses revealed three regional populations (k = 3; ln P(D) = -2766.3;

Figure 14) with a relatively large change in average ln P(D) (Δk = 300.7) and corresponding to GoC, ETP and the western Pacific. Interestingly, six samples from Palau showed a high probability (> 0.6) of clustering with the GoC population, suggesting there may be occasionally gene flow between the two populations. Runs at higher values of k fail to form

108

Figure 14. Structure plot of S. glynni populations from the Western Pacific and the tropical (GoC) and subtropical (ETP) Eastern Pacific based on allelic frequencies at 11 microsatellite loci. Each bar in the graph represents the probability that a sample belongs to a particular color-coded population. Each graph represents analyses for a particular number of populations (K) run under an admixture with a correlated allele frequency model. (a) Major differentiation occurred between the GoC population and populations in the ETP (k = 2) as noted previously by Pettay & LaJeunesse (in prep), with the western Pacific populations clustering with the GoC. Additional clustering occurred which differentiation the western Pacific from the GoC (k = 3), with not further subdivision by geographic location for higher values of K (k = 4). (b) Population subdivision for western Pacific only, with additional with increasing values of K. Populations from Palau, the two reef locations in the Gulf of Thailand and the Andaman Sea (S. D5) are all highly subdivided (k = 4).

109 any biogeographically meaningful clusters (data not shown).

Since Structure will detect the strongest signal of differentiation and may therefore miss population structuring at smaller scales (Pritchard et al. 2010, Evanno et al. 2005), the locations in the western Pacific were also analyzed separate from those of the GoC and ETP.

These analyses in conjunction with the method by Evanno et al. show the largest change in average ln P(D) (Δk = 74.2; ln P(D) = -334.3) at k = 2 and separates PAL and RAY from

CHO and AND. The clustering of CHO with SAT is surprising since the SAT samples are characterized as D5 because they possesses a different ITS2 sequence from S. glynni

(LaJeunesse et al. 2010). Runs at higher values of k continue to resolve additional clusters by location but with lower values of Δk, with k = 3 (Δk = 56.8; ln P(D) = -278.2) separating

CHO and SAT into two different clusters and k = 4 (Δk = 22.8; ln P(D) = -253.3) distinguishing all four locations into distinct clusters (Figure 14). The ln P(D) values for each k continue to increase up to k = 4, after which they begin to decrease, suggesting that this is the highest level of structuring (Pritchard et al. 2010).

Population genetic data were divided into the three regional groupings resolved by

Structure (i.e., GoC, ETP & W. Pacific) and AMOVA analysis yielded ΦRT = 0.292, ΦPR =

0.151 and ΦPT = 0.398. All of these values were significant (p = 0.0001, Table 12) and supported the Structure defined clusters. Pairwise comparisons between individual locations yielded Φ-values ranging from 0.016 to 0.648, with a majority (78%) of pairwise comparison significant after sequential Bonferroni correction (Table 13). When considering only the pairwise comparisons between locations from different regions, all of the comparisons are significant and supports the three clusters define by Structure.

110 Table 12. Analysis of Molecular Variance (AMOVA) for S. glynni across the Pacific Ocean with populations grouped by regions: Western Pacific, tropical Eastern Pacific and subtropical Eastern Pacific.

AMOVA for 3 Regions (Western Pacific, Gulf of California and Eastern Tropical Pacific) Estimated Percent of Fixation Source df SS MS Variance Variation Index

Among Regions 2 193.38 96.69 1.18 29% ΦRT = 0.292*

Among Pops w/in Regions 6 61.25 10.21 0.43 11% ΦPR = 0.151*

Within Pops 235 573.08 2.44 2.44 60% ΦPT = 0.398* Total 243 827.70 4.05 100%

* P < 0.0001

Table 13. A pairwise matrix comparing the genetic relationship (ΦPT) between all eight Pacific locations of S. glynni and one location of S. D5 in the eastern Indian Ocean. Significant values (sequential Bonferroni corrected P < 0.05) represented in bold.

Pairwise Matrix Western Pacific Eastern Tropical Pacific

AND GTH PAL GoC BB GoT CLP PAN GAL

Andaman Sea (S. D5) ---

Gulf of Thailand 0.521 --- Pacific Western Palau 0.498 0.403 ---

Gulf of California 0.558 0.441 0.237 ---

Banderas Bay 0.541 0.442 0.410 0.411 --- Gulf of Tehuantepec --- 0.519 0.419 0.368 0.374 0.017

Clipperton Atoll 0.604 0.433 0.337 0.388 0.018 0.022 --- Pacific Bay of Panama 0.648 0.519 0.407 0.379 0.146 0.124 0.079 --- Eastern Tropical Galapagos Islands 0.539 0.375 0.315 0.337 0.045 0.036 0.016 0.129 ---

While the genotypic richness of S. glynni was low for most of the locations sampled

(Table 11), with many MGLs found numerous times, the majority of genotypes were restricted to a single sample location. None of the MLGs from the west Pacific locations or

111 the GoC were found in a second sampling location. In contrast, nine different MLGs in the

ETP were regionally widespread and detected in two sampling locations (see Figure 3 in

Pettay & LaJeunesse in prep). The most geographically widespread of these clones occurred in both PAN and the GAL, approximately 1,600 km apart.

Discussion

Comparisons of morphological traits have indicated that microbial eukaryotes comprise geographically widespread panmictic populations. Molecular genetic data is overturning this notion by showing that unicellular eukaryotes often display population structure over relatively small distances and can possess complex population patterns. S. glynni is a dinoflagellate that forms endosymbiotic associations with coral in the genus

Pocillopora in both the eastern and western Pacific. The widespread distribution of this dinoflagellate offers the opportunity to identify whether population structuring occurs over an entire ocean basin and including a region that is classically known as a dispersal barrier for marine organisms, the Eastern Pacific Barrier.

Population differentiation on each side of the Pacific

Populations in the western Pacific are clearly differentiated from those in the eastern

Pacific as shown by both cluster analysis and AMOVA (Figure 14 & Table 12). The genetic break between the GoC and the ETP is resolved by Structure prior to the break between east and west, with the west clustering with the GoC. However, this clustering may be an artifact of sample size and when a subset of samples from the east are drawn at random to mirror the

112 sample sizes in the west, the east/west break is resolved first (data not shown). Interestingly, several samples from Palau cluster more closely with the GoC, regardless of sample size, and may represent occasional gene flow over evolutionary time scales (Figure 14).

The significant differentiation between the eastern and western Pacific for S. glynni corresponds with historical interpretations of species and community assemblages that place them into two distinct biogeographic regions with little to no connectivity and has been attributed to the dispersal barrier in the eastern Pacific (EPB), a vast stretch of open ocean that extends for greater than 5,000 km (Ekman 1953, Briggs 1974). While recent genetic data on individual marine species have demonstrated the permeability of this dispersal barrier, the presence and magnitude of the barrier appears species specific and often related to larval biology (Lessios et al. 1998, Lessios & Robertson 2006, Duda & Lessios 2009). S. glynni and its Pocillopora host appear to disperse over hundreds to thousands of kilometers in the eastern Pacific (Pinzón & LaJeunesse 2011, Pettay & LaJeunesse in prep). Identical

MLGs of S. glynni were found as far apart as 1,200 km and environmental conditions in this region have been suggested as more effective barriers to dispersal than distance (Pettay &

LaJeunesse in prep). In contrast, the significant differentiation between the east and west populations suggests that dispersal across the Pacific Ocean by S. glynni is inhibited by distance. Similar processes appear to occur for other dinoflagellates where a combination of environmental conditions and distance explain genetic differentiation over relatively small geographic areas (Santos et al. 2003, Kirk et al. 2009, Nagai et al. 2009). However, over larger regions like the Pacific Ocean strong genetic structuring by distance emerges as with the diatom Pseudo-nitzchia pugens (clade I), where distinct population differentiation occurred between Japan, New Zealand and the western United States (Casteleyn et al. 2010).

113 Swift moving equatorial currents provide the potential dispersal routes for long-lived planktonic larvae across the Pacific. It was hypothesized that the long-lived and highly dispersive larvae of Pocillopora may aid in maintaining gene flow of S. glynni across the

Pacific (Richmond 1987, Pinzón & LaJeunesse 2011, Pettay & LaJeunesse in prep), but significant ΦRT and ΦPT values suggest otherwise. The clustering of several Palau individuals with the GoC does suggest, however, that gene flow may occur occasionally over evolutionary timescales. Changes in current conditions during and as a result of El Niño events may provide the opportunity for individuals to cross the Pacific as current velocity increases and transit time between suitable habitat decreases (Glynn et al. 1996). The fact that these individuals cluster with the GoC and not the ETP is interesting and suggests that proportionally more gene flow may proceed from east to west, as opposed to west to east

(Lessios & Robertson 2006). Gene flow in this direction would suggest that the equatorial counter current (i.e., NEC) beginning near the mouth of the GoC and ending at the islands of the Philippines, Mariana Islands and Palau in the west (Wyrtki 1967, Morey et al. 1999), may act to connect populations between regions. Additional samples from the western locations and others in the region (e.g., Philippines & Guam) may resolve the western cluster more accurately and negate any influence due to sample sizes. Alternatively, if occasional gene flow does occur, then the possibility of isolation by distance may exist between the two regions that further sampling of the remote islands in the central Pacific would resolve.

Western Pacific populations of S. glynni

Vicariance due to sea-level fluctuations (~130 m) during the Pleistocene along with complex oceanographic processes are largely implicated in promoting genetic diversification

114 in the Indo-West Pacific at both the species and intraspecific level (reviewed in Carpenter et al. 2011). However, additional hypotheses to explain the extreme diversity of this region include the accumulation of species dispersing from peripheral regions, the overlap of Indian and Pacific Ocean marine fauna, and refugia from extinction provided by the diverse marine habitats of the region (Carpenter et al. 2011). Phylogeographic analyses of various invertebrate and fish species indicate that there are numerous biogeographic breaks within the region and that many of these breaks are species specific, suggesting that a combination of the above hypotheses are or have been potentially contributing to diversification (Barber

2009). While many of these breaks show little concordance, several major zoogeographic breaks are beginning to emerge (Carpenter et al. 2011); including the Sunda Shelf along the

Malaysian Peninsula, Sumatra and Java, the Mindanao and Halmahera Eddies in east

Indonesia, and north/south and east/west divisions of the Philippines.

Populations of S. glynni in Palau, the Gulf of Thailand and the Andaman Sea appear well differentiated from each other (Figure 14 & Table 13). The differentiation between

Palau and the Gulf of Thailand can be explained by a combination of distance and oceanographic features where landmasses along with complex ocean currents and the shallow waters of the Sudan Shelf greatly limit and prevent gene flow, forming the biogeographic breaks as described above (Wyrtki 1961; Morey et al. 1999; Carpenter et al.

2010). Likewise, the differentiation between the Andaman Sea and both the Gulf and Palau is not only related to distance and oceanographic features, but also the fact that the

Pollicopora symbionts in the Andaman Sea represent a different taxonomic group within clade D (S. D5) (LaJeunesse et al. 2010a). Pairwise comparisons between S. D5 and S.

115 glynni were consistently greater than those within S. glynni indicative species-level differentiation (Table 13).

The population subdivision, as determine by Structure, between reefs within Thai waters corresponds to three of the four distinct marine regions of Thailand that are characterized by unique oceanographic conditions (Sudara & Yeemin 1997, Yeemin et al.

2006). The strong population differentiation seen for S. glynni within these waters can be partially explained by oceanographic conditions and distance, but may additionally relate to the brooding mode of reproduction by Pocillopora damicornis the Gulf of Thailand (Kuanui et al. 2008). Many brooding Pocilloporid species produce larvae that settle soon after release, leading to local recruitment and strong population structuring over short distances

(Miller & Ayre 2004, Maier et al. 2005, Underwood et al. 2006, van Oppen et al. 2008,

Bongaerts et al. 2010, Starger et al. 2010), such as that seen for S. glynni within the GTH.

Similarly, no clones were shared between the two locations sampled in the GTH further suggesting limited gene flow and contrasting the lack of differentiation in eastern Pacific over similar distances by broadcast spawning Pocillopora spp. (Pettay & LaJeunesse in prep).

Interestingly, Bayesian clustering divides the Gulf of Thailand population by the two reefs within this location prior to dividing Palau from the Gulf of Thailand or S. D5 from the

Gulf (Figure 14). In addition to being located on the opposite of side of the Malaysian

Peninsula, S. D5 is believed to be its own taxonomic unit (i.e., species) and was therefore expected to cluster in isolation (LaJeunesse et al. 2010a). These two symbionts differ by only two base pair differences in their dominant ITS2 sequence and given that they both associate with the same Pocillopora species (type 5) in each location, these symbionts may

116 represent sister species. Microsatellite analysis of sister species in plants have shown higher divergence within a species than between, primarily due to shared ancestral alleles

(Drummond & Hamilton 2007). Similar processes may explain the case for S. glynni and S.

D5, which may be in the earlier stages of the continuum of speciation. The clustering of a S. glynni population with a S. D5 population may also reflect the sampling strategy, where additional samples from these and other locations for both symbionts may be needed to accurately resolve species boundaries between evolutionarily similar lineages (Funk &

Omland 2003). Lastly, the ITS region has been shown to be conservative and may not delimit species lineages among certain Symbiodinium and it is possible that the S. glynni populations sampled here belong to different evolutionary lineages, representing two different taxonomic units (Santos et al. 2004, Finney et al. 2010). Continued research with larger sample sizes are needed to accurately address these questions concerning.

Host shifts versus geographic isolation in S. glynni

It appears that the ability of S. glynni to disperse great distances in the ETP may not translate into long distance dispersal across the Pacific Ocean, and that the expanse of the

Pacific Ocean is an effective barrier to dispersal for this symbiont. However, distance may not be the only barrier at work. The symbiont populations sampled in the western Pacific reside within a different species of Pocillopora than found in the eastern Pacific (Pinzón

2011). In the eastern Pacific, S. glynni associates with type 1 Pocillopora; while in the western Pacific it associates predominantly with type 5. Therefore, the differentiation between the two regions may represent structuring according to host association, either alone or in addition to distance-based differentiation.

117 Population differentiation and subsequent speciation for Symbiodinium as a result of host would not be surprising. It is believed that host specificity is one of the primary forces driving symbiont diversification and leading to reproductive isolation (Sanots et al. 2004,

LaJeunesse 2005, Finney et al. 2010). Symbiont speciation according Pocillopora species has been shown for Clade C in the eastern Pacific (Pinzón & LaJeunesse 2011), and similar symbiont diversification has been documented in the Caribbean octocoral Pseudopterogorgia for clade B (Sanots et al. 2004). Symbiodinium B1 in the Florida Keys shows population- level differentiated according to the species of Montastraea with which it associates, however this sub-structuring may be geographically influences since it does not occur in the

Bahamas (Thornhill et al. 2009). Teasing apart the hierarchical influences of host versus distance in limiting gene flow is complicated and will require further research.

While the shift in host association for S. glynni prevents implicating distance as the sole factor driving population differentiation, it highlights the complexity of coral/symbiont associations over large regions. Pocillopora type 1 does in fact exist in Palau, however S. glynni did not occur in any of the samples collected from two different reefs (LaJeunesse unpub.). Instead, all of the samples possessed a clade C type related to the S. C1b-c found in the eastern Pacific. Unfortunately this shift prevented the population-level analysis for the holobiont to investigate dispersal between the two regions as was done in the eastern Pacific

(Pinzón & LaJeunesse 2011, Pettay & LaJeunesse in prep). These geographical patterns in host-symbiont associations present new questions concerning how and why shifts occur.

Understanding these processes may enlighten us on the evolution of host/symbiont associations and how they will be affected during major changes in climate.

118

References

Anderson T, Haubold B, Williams JT, et al. (2000) Microsatellite markers reveal a spectrum of population structures in the malaria parasite Plasmodium falciparum. Molecular Biology and Evolution 17, 1467-1482.

Barber PH (2009) The challenge of understanding the Coral Triangle biodiversity hotspot. Journal of Biogeography 36, 1845-1846.

Bongaerts P, Riginos C, Ridgway T, Sampayo E (2010) Genetic divergence across habitats in the widespread coral Seriatopora hystrix and its associated Symbiodinium. PLoS ONE 5, e10871.

Briggs JC (1961) The east Pacific barrier and the distribution of marine shore fishes. Evolution 15, 545-554.

Briggs JC (1974) Marine Zoogeography McGraw-Hill, New York.

Carpenter KE, Barber PH, Crandall ED, et al. (2011) Comparative Phylogeography of the Coral Triangle and Implications for Marine Management. Journal of Marine Biology 2011, 1-14.

Casteleyn G, Leliaert F, Backeljau T, et al. (2010) Limits to gene flow in a cosmopolitan marine planktonic diatom. Proceedings of the National Academy of Sciences 107, 12952-12957.

Coffroth M, Santos S (2005) Genetic Diversity of Symbiotic Dinoflagellates in the Genus Symbiodinium. Protist 156, 19 - 34.

Dana T (1975) Development of contemporary eastern Pacific coral reefs. Marine biology 33, 355-374.

Darling KF, Kucera M, Wade CM (2007) Global molecular phylogeography reveals persistent Arctic circumpolar isolation in a marine planktonic protist. Proceedings of the National Academy of Sciences 104, 5002-5007.

Drummond CS, Hamilton MB (2006) Hierarchical components of genetic variation at a species boundary: population structure in two sympatric varieties of Lupinus microcarpus (Leguminosae). Molecular Ecology 16, 753-769.

Duda TF, Lessios HA (2009) Connectivity of populations within and between major biogeographic regions of the tropical Pacific in Conus ebraeus, a widespread marine gastropod. Coral Reefs 28, 651-659.

119 Earl DA (2009) Structure Harvester, from website: v0.3 http://users.soe.ucsc.edu/~dearl/software/struct_harvest/.

Ekman S (1953) Zoogeography of the sea, London.

Evanno G, Regnaut S, Goudet J (2005) Detecting the number of clusters of individuals using the software STRUCTURE: a simulation study. Molecular Ecology 14, 2611-2620.

Excoffier L, Smouse P, Quattro J (1992) Analysis of molecular variance inferred from metric distances among DNA haplotypes: application to human mitochondrial DNA restriction data. Genetics 131, 479-491.

Finney JC, Pettay DT, Sampayo EM, et al. (2010) The Relative Significance of Host– Habitat, Depth, and Geography on the Ecology, Endemism, and Speciation of Coral Endosymbionts in the Genus Symbiodinium. Microbial Ecology 60, 250-263.

Foissner W (2006) Biogeography and dispersal of micro-organisms: a review emphasizing protists. Acta Protozoologica 45, 111-136.

Funk DJ, Omland KE (2003) Species-level paraphyly and polyphyly: frequency, causes, and consequences, with insights from animal mitochondrial DNA. Annual Review of Ecology, Evolution, and Systematics 34, 397-423.

Glynn P, Ault J (2000) A biogeographic analysis and review of the far eastern Pacific coral reef region. Coral Reefs 19, 1-23.

Glynn P, Veron J, Wellington G (1996) Clipperton Atoll (eastern Pacific): oceanography, geomorphology, reef-building coral ecology and biogeography. Coral Reefs 15, 71- 99.

Glynn PW (1997) Eastern Pacific reef coral biogeography and faunal flux: Durham's dilemma revisited. Proceedings of the 8th International Coral Reef Symposium 1, 371-378.

Green J, Bohannan B (2006) Spatial scaling of microbial biodiversity. Trends in Ecology & Evolution 21, 501-507.

Grigg RW, Hey R (1992) Paleoceanography of the tropical eastern Pacific Ocean. Science 255, 172-178.

Hubisz MJ, Falush D, Stephens M, Pritchard JK (2009) Inferring weak population structure with the assistance of sample group information. Molecular Ecology Resources 9, 1322-1332.

120 Iglesias-Rodríguez DM, Schofield OM, Batley J, Medlin LK, Hayes PK (2006) Intraspecific genetic diversity in the marine coccolithophore Emiliania huxleyi (Prymnesiophyceae): the use of microsatellite analysis in marine phytoplankton population studies. Journal of Phycology 42, 526-536.

Kirk N, Andras J, Harvell C, Santos S, Coffroth M (2009) Population structure of Symbiodinium sp. associated with the common sea fan, Gorgonia ventalina, in the Florida Keys across distance, depth, and time. Marine Biology 156, 1609-1623.

Kuanui P, Chavanich S, Raksasab C, Viyakarn V (2008) Lunar periodicity of larval release and larval development of Pocillopora damicornis in Thailand. Proceedings of the 11th International Coral Reef Symposium, 375-377.

LaJeunesse T (2002) Diversity and community structure of symbiotic dinoflagellates from Caribbean coral reefs. Marine Biology 141, 387-400.

LaJeunesse T, Bonilla H, Warner M, et al. (2008) Specificity and stability in high latitude eastern Pacific coral-algal symbioses. Limnology and Oceanography 53, 719-727.

LaJeunesse T, Loh W, Van Woesik R, et al. (2003) Low symbiont diversity in southern Great Barrier Reef corals, relative to those of the Caribbean. Limnology and Oceanography, 2046-2054.

LaJeunesse TC (2005) ‘‘Species’’ radiations of symbiotic dinoflagellates in the Atlantic and Indo-Pacific since the Miocene-Pliocene transition. Molecular Biology and Evolution 22, 570-581.

LaJeunesse TC, Pettay DT, Sampayo EM, et al. (2010a) Long-standing environmental conditions, geographic isolation and host-symbiont specificity influence the relative ecological dominance and genetic diversification of coral endosymbionts in the genus Symbiodinium. Journal of Biogeography 37, 785-800.

LaJeunesse TC, Smith R, Walther M, et al. (2010b) Host–symbiont recombination versus natural selection in the response of coral–dinoflagellate symbioses to environmental disturbance. Proceedings of the Royal Society B 277, 2925-2934.

LaJeunesse TC, Thornhill DJ, Cox EF, et al. (2004) High diversity and host specificity observed among symbiotic dinoflagellates in reef coral communities from Hawaii. Coral Reefs 23.

LaJeunesse TC, Trench RK (2000) The biogeography of two species of Symbiodinium (Freudenthal) inhabiting the intertidal anemone, Anthopleura elegantissima (Brandt). Biological Bulletin 199, 126-134.

121 Lessios H, Kane J, Robertson D (2003) Phylogeography of the pantropical sea urchin Tripneustes: Contrasting patterns of population structure between oceans. Evolution 57, 2026-2036.

Lessios H, Kessing B, Pearse J (2001) Population structure and speciation in tropical seas: Global phylogeography of the sea urchin Diadema. Evolution 55, 955-975.

Lessios H, Kessing B, Robertson D, Paulay G (1999) Phylogeography of the pantropical sea urchin Eucidaris in relation to land barriers and ocean currents. Evolution 53, 806- 817.

Lessios HA, Kessing BD, Robertson DR (1998) Massive gene flow across the world's most potent marine biogeographic barrier. Proceedings of the Royal Society B: Biological Sciences 265, 583-588.

Lessios HA, Robertson DR (2006) Crossing the impassable: genetic connections in 20 reef fishes across the eastern Pacific barrier. Proceedings of the Royal Society B 273, 2201-2208.

Lilly EL, Halanych KM, Anderson DM (2007) Species boundaries and global biogeography of the Alexandrium tamarense complex (Dinophyceae). Journal of Phycology 43, 1329-1338.

Maier E, Tollrian R, Rinkevich B, Nürnberger B (2005) Isolation by distance in the scleractinian coral Seriatopora hystrix from the Red Sea. Marine biology 147, 1109- 1120.

Masseret E, Grzebyk D, Nagai S, et al. (2009) Unexpected Genetic Diversity among and within Populations of the Toxic Dinoflagellate Alexandrium catenella as Revealed by Nuclear Microsatellite Markers. Applied and Environmental Microbiology 75, 2037- 2045.

McCauley LAR, Erdner DL, Nagai S, Richlen ML, Anderson DM (2009) Biogeographic analysis of the globally distributed harmful algal bloom species Alexandrium minutum (Dinophyceae) based on rRNA gene sequences and microsatellite markers. Journal of Phycology 45, 454-463.

McMillan W, Palumbi S (1995) Concordant evolutionary patterns among Indo-West Pacific butterflyfishes. Proceedings of the Royal Society B 260, 229-236.

Miller KJ, Ayre DJ (2004) The role of sexual and asexual reproduction in structuring high latitude populations of the reef coral Pocillopora damicornis. Heredity 92, 557-568.

122 Mills L, Citta J, Lair K, Schwartz M, Tallmon D (2000) Estimating animal abundance using noninvasive DNA sampling: promise and pitfalls. Ecological Applications 10, 283- 294.

Morey S, Shriver J, O'Brien J (1999) The effects of Halmahera on the Indonesian throughflow. Journal of Geophysical Research 104, 23281-23296.

Morin P, McGrady-Steed J (2004) Biodiversity and ecosystem functioning in aquatic microbial systems: a new analysis of temporal variation and species richness- predictability relations. Oikos 104, 458-466.

Nagai S, Nishitani G, Sakamoto S, et al. (2009) Genetic structuring and transfer of marine dinoflagellate Cochlodinium polykrikoides in Japanese and Korean coastal waters revealed by microsatellites. Molecular Ecology 18, 2337-2352.

Nystrom M, Moberg F, Tedengren M (1997) Natural and anthropogenic distrubance on reef corals in the inner Gulf of Thailand: physiological effects of reduced salinity, copper, and siltation. Proceedings of the 8th International Coral Reef Symposium 2, 1893- 1898.

Palumbi SR (1996) What can molecular genetics contribute to marine biogeography? An urchin's tale. Journal of Experimental Marine Biology and Ecology 203, 75-92.

Peakall R, Smouse PE (2006) GENALEX 6: genetic analysis in Excel. Population genetic software for teaching and research. Molecular Ecology Notes 6, 288-295.

Penna A, Fraga S, Battocchi C, et al. (2010) A phylogeographical study of the toxic benthic dinoflagellate genus Ostreopsis Schmidt. Journal of Biogeography 37, 830-841.

Pettay DT, LaJeunesse TC (2009) Microsatellite loci for assessing genetic diversity, dispersal and clonality of coral symbionts in ‘stress-tolerant’ clade D Symbiodinium. Molecular Ecology Resources 9, 1022-1025.

Pettay DT, LaJeunesse TC (in press) Long-range dispersal and high-latitude environments influence the population structure of a “stress-tolerant” endosymbiont harbored by Pocillopora in the Eastern Tropical Pacific. Molecular Ecology.

Pinzón JH, LaJeunesse TC (2011) Species delimitation of common reef corals in the genus Pocillopora using nucleotide sequence phylogenies, population genetics and symbiosis ecology. Molecular Ecology 20, 311-325.

Pritchard J, Stephens M, Donnelly P (2000) Inference of population structure using multilocus genotype data. Genetics 155, 945.

Pritchard J, Wen X, Falush D (2010) Documentation for structure software: Version 2.3.

123 Richmond R (1987) Energetics, competency, and long-distance dispersal of planula larvae of the coral Pocillopora damicornis. Marine Biology 93, 527-533.

Richmond RH (1985) Variations in the population biology of Pocillopora damincornis across the Pacific. Proceedings of the 5th International Coral Reef Symposium 6, 101-106.

Robertson D, Allen G (1996) Zoogeography of the shorefish fauna of Clipperton Atoll. Coral Reefs 15, 121-131.

Rosenberg N (2004) DISTRUCT: a program for the graphical display of population structure. Molecular Ecology Notes 4, 137-138.

Rousset F (2008) GENEPOP'007: a complete re-implementation of the GENEPOP software for Windows and Linux. Molecular Ecology Resources 8, 103-106.

Sampayo EM, Dove S, LaJeunesse TC (2009) Cohesive molecular genetic data delineate species diversity in the dinoflagellate genus Symbiodinium. Molecular Ecology 18, 500-519.

Sampayo EM, Franceschinis L, Hoegh-Guldberg O, Dove S (2007) Niche partitioning of closely related symbiotic dinoflagellates. Molecular Ecology 16, 3721-3733.

Santos S, Gutierrez-Rodriguez C, Lasker H, Coffroth M (2003) Symbiodinium sp. associations in the gorgonian Pseudopterogorgia elisabethae in the Bahamas: high levels of genetic variability and population structure in symbiotic dinoflagellates. Marine Biology 143, 111-120.

Santos S, Shearer T, Hannes A, Coffroth M (2004) Fine-scale diversity and specificity in the most prevalent lineage of symbiotic dinoflagellates (Symbiodinium, Dinophyceae) of the Caribbean. Molecular Ecology 13, 459-469.

Seutin G, White BN, Boag PT (1991) Preservation of avian blood and tissue samples for DNA analyses. Canadian Journal of Zoology 69, 82–92.

Starger CJ, Barber PH, Ambariyanto, Baker AC (2010) The recovery of coral genetic diversity in the Sunda Strait following the 1883 eruption of Krakatau. Coral Reefs 29, 547-565.

Sudara S, Yeemin T (1997) Status of Coral Reefs in Thailand. In: Status of Coral Reefs in the Pacific, pp. 133-144.

Suvapepun S (1991) Long-term ecological changes in the Gulf of Thailand. Marine Pollution Bulletin 23, 213-217.

124 Thornhill D, Xiang Y, Fitt W, Santos S (2009) Reef endemism, host specificity and temporal stability in populations of symbiotic dinoflagellates from two ecologically dominant Caribbean corals. PLoS One 4, e6262.

Thornhill DJ, LaJeunesse TC, Santos SR (2007) Measuring rDNA diversity in eukaryotic microbial systems: how intragenomic variation, pseudogenes, and PCR artifacts confound biodiversity estimates. Molecular Ecology 16, 5326-5340.

Trench R (1993) Microalgal-invertebrate symbioses - a review. Endocytoboisis and Cell Research 9, 135-175.

Underwood JN, Smith LD, Van Oppen MJH, Gilmour JP (2006) Multiple scales of genetic connectivity in a brooding coral on isolated reefs following catastrophic bleaching. Molecular Ecology 16, 771-784.

Van Oppen MJH, Lutz A, De'ath G, Peplow L, Kininmonth S (2008) Genetic traces of recent long-distance dispersal in a predominantly self-recruiting coral. PLoS ONE 3, e3401.

Vermeij GJ (1987) The dispersal barrier in the tropical Pacific: implications for molluscan speciation and extinction. Evolution 41, 1046-1058.

Veron JEN (2000) Corals of the world Australian Institute of Marine Science, Queensland.

Vogler C, Benzie J, Lessios H, Barber P, rheide GW (2008) A threat to coral reefs multiplied? Four species of crown-of-thorns starfish. Biological Letters online, 1-4.

Waits L, Luikart G, Taberlet P (2001) Estimating the probability of identity among genotypes in natural populations: cautions and guidelines. Molecular Ecology 10, 249-256.

Wham DC, Pettay DT, LaJeunesse TC (2011) Microsatellite loci for the host-generalist “zooxanthella” Symbiodinium trenchi and other Clade D Symbiodinium. Conservation Genetic Resources, 1-4.

Williams S, Reid D (2004) Speciation and diversity on tropical rocky shores: a global phylogeny of snails of the genus Echinolittorina. Evolution 58, 2227-2251.

Wyrtki K (1961) Physical oceanography of the southeast Asian waters. Naga Report 2.

Wyrtki K (1967) Circulation and water masses in the eastern equatorial Pacific Ocean. International Journal of Oceanology and Limnology 1, 117-147.

Yeemin T, Sutthacheep M, Somkleeb N (2006) Status of Coral Reefs in Thailand. Proceedings of the 10th International Coral Reef Symposium, 1062-1067.

125 CHAPTER 4

High clonality and low genetic diversity characterize populations of Symbiodinium trenchi in the Caribbean relative to the Indo- Pacific

Abstract

The loss of endosymbiotic dinoflagellates (i.e., coral bleaching) due to thermal anomalies can lead to disease outbreaks and mass mortality and has cause widespread degradation of many reef coral communities. However, some host-symbiont combinations show higher resistance to thermal perturbations and their proliferation has the potential to influence the response of reef communities to climate change. Colonies associating with symbionts of the phylogenetic grouping Clade D exhibit tolerance to physiological stress, most notable of which is the recently characterized Symbiodinium trenchi (D1a). This symbiont differs ecologically from other genetically defined types within Clade D in that it is the most geographically widespread, associates with a diversity of host taxa and in the Caribbean S. trenchi has increased in prevalence and abundance in coral communities under prolonged exposure to raised sea surface temperatures. The unique ecology and distribution of this symbiont prompted an investigation into its genotypic diversity and population structuring in representative reefs from the Indo-Pacific and throughout the Atlantic using twelve diallelic microsatellite loci. Populations of S. trenchi from the Indo-Pacific were genetically diverse (I > 1.14) with most genotypes the products of sexual recombination (R > 0.80). In contrast, populations throughout the greater Caribbean were limited in genetic diversity (I = 0.67) and excessively clonal (R = 0.28). While no multilocus genotypes (MLGs) were shared among reefs in the Indo-Pacific, one unusual MLG was found forty-four times in the Caribbean and from every sample location except the Flower Garden Banks. Genotypic diversity is typically high among other Caribbean Symbiodinium spp. suggesting something is unusual about the population dynamics and dispersal of S. trenchi in this region. Severe thermal anomalies including recent events in 2005 and 2009 possibly facilitated expansion of a particularly aggressive clone lineage and/or populations of this symbiont, which may stem from a small number of opportunistic introductions originating from the Indo-Pacific.

126 Introduction

Many coral reef invertebrates form symbiotic associations with endosymbiotic dinoflagellates in the genus Symbiodinium and are critical to ecosystem functioning (Trench

1993). The association between reef-building coral and Symbiodinium is specifically responsible for the construction and maintenance of the entire ecosystem. Unfortunately these associations are sensitive to slight changes in water temperatures of only a few degrees

Celsius, which can lead to a disassociation of the symbiosis called coral bleaching (Hoegh-

Guldberg 1999). Coral bleaching involves the expulsion of their dinoflagellate symbionts and if prolonged can result in coral mortality and ecosystem degradation (Hoegh-Guldberg

1999). The extent and frequency of bleaching episodes is increasing due to anthropogenic related atmospheric warming and has been particularly evident in the Caribbean basin (Porter et al. 1989; Fitt et al. 2000; McWilliams et al. 2005; Oxenford et al. 2008). Coral reefs in this region have experience several mass bleaching events over the last several decades, decreasing the live coral cover by as much as 50% in some locations (Wilkinson & Souter

2008).

Symbiodinium diversity is divided into eight major genetically defined clades (A –

H), with additional diversity in the form of numerous “types” (i.e., species) within in each clade (reviewed in Coffroth & Santos 2005). Some host-symbiont combinations show higher resistance to environmental perturbations, and this “stress-tolerance” appears particularly true for associations involving members of clade D (Glynn et al. 2001; Toller et al. 2001; Rowan

2004; Berkelmans et al. 2006; Jones et al. 2008; LaJeunesse et al. 2009). In addition to acute stress-tolerance, Clade D symbionts have been shown to have a global distribution and various species are now known to be ecologically important in marginal reef environments

127 such as those exposed to large temperature fluctuations, air-exposed reef flats and high turbidity (Ulstrup et al. 2003; Fabricius et al. 2004; Garren et al. 2006; LaJeunesse et al.

2007; Mostafavi et al. 2007; LaJeunesse et al. 2008; LaJeunesse et al. 2010a). The recently characterized symbiont Symbiodinium trenchi (D1a) represents one of these clade D members. The worldwide distribution of S. trenchi and its ability to associate with a diversity of host taxa makes it unique among other genetically defined clade D types

(LaJeunesse et al. 2009; LaJeunesse et al. 2010a). With a warming climate significantly impacting reefs around the planet, S. trenchi is of particular interest given its thermal tolerance, widespread distribution and host generalist nature.

Thermal tolerance is not the only unique characteristic of S. trenchi. Recent findings indicate that S. trenchi can exist as low density background populations and subsequently increase in prevalence and abundance during prolonged exposure to raised SST’s (Smith

2008; LaJeunesse et al. 2009). This increase was observed in Barbados during the 2005 mass coral bleaching event in the Caribbean and provided enhanced thermal tolerance for many of the corals in the community (Smith 2008; LaJeunesse et al. 2009). This method of increased prevalence and abundance is contrary to previous research that suggests S. trenchi enters “new” symbioses in response to bleaching (i.e., Adaptive Bleaching; Buddemeier &

Fautin 1993). Surprisingly, many of the host colonies that associate with S. trenchi during thermal anomalies revert back to their homologous symbiont following a return to stable environmental conditions (Thornhill et al. 2006; LaJeunesse et al. 2009). While some have touted this symbiont as a savior (Baker 2001; Baker et al. 2004), the semi-permanent nature of some S. trenchi associations has led others to question the ecological benefit of this

128 symbiont and its role in the long-term sustainability of coral reefs during climate change

(Thornhill et al. 2006; LaJeunesse et al. 2009).

Genetic diversity and gene flow are important factors in the evolutionary potential of a species (Franklin & Frankham 1998; Stockwell et al. 2003). Understanding this diversity and these processes for coral reef organisms will be critical as inferences and management plans are made to confront their degradation. Stress-tolerant symbionts such as S. trenchi will be of particular interest given its resilience to environmental perturbations. Questions have already been raised on the ability of this symbiont to shape the contemporary evolution of coral reefs in any meaningful way (Baker 2001; Thornhill et al 2006; LaJeunesse et al.

2009; Correa & Baker 2010). To address these questions, a population genetics approach must be taken to investigate the species diversity throughout its range. Additionally, the population in Barbados can be compared to others around the Caribbean to ascertain its

“uniqueness” and whether similar results can be expected around the region in the future.

Lastly, the widespread distribution of S. trenchi allows the comparison of genetic diversity and gene flow over large geographic regions. To better understand the evolutionary potential of S. trenchi in the Caribbean twelve diallelic microsatellite loci were used to analyze populations from around the Caribbean, and three locations from its Indo-Pacific range;

Palau, Thailand and Zanzibar. It is predicted that S. trenchi will show high levels of genetic diversity, with population subdivision according geography, similar to other symbionts studied to date.

129 Materials and Methods

Sample Collections

Samples analyzed in this study were part of larger regional Symbiodinium diversity studies in the Caribbean Sea (LaJeunesse 2002; Finney et al. 2010), the eastern and western

Indian Ocean (LaJeunesse et al. 2010a) and the western Pacific Ocean. Cnidarians were collected by SCUBA from seven locations around the Caribbean (Barbados, Belize, Florida

Keys, Flower Garden Banks, Mexico, Panama & St. Croix), three locations in the western

Indian Ocean (Zanzibar, Tanzania), two locations in the eastern Indian Ocean (Andaman Sea,

Thailand) and two locations in the western Pacific Ocean (Palau) (Figure 15). Sample fragments were preserved in either a high salt, 20% DMSO buffer (Seutin et al. 1991) or

95% ethanol and stored at -20 °C until DNA extraction. Sample collection dates and totals for each regional location are as follows: Caribbean (CAR; n = 88) in 1999 to 2009, Zanzibar

(ZAN; n = 13) in 2007, Thailand (THA; n = 58) in 2007, and Palau (PAL; n = 31) in 2009.

The diversity of hosts sampled in each region varied (Table 15), with the total number of genera in each location equal to eight (CAR), twenty-one (PAL), twenty-two (THA) and nine

(ZAN).

Molecular-Genetic Identification

Nucleic acid extractions were conducted using a modified Promega Wizard genomic

DNA extraction protocol (LaJeunesse et al. 2003). The dominant resident symbiont was identified by denaturing gradient gel electrophoresis (DGGE) fingerprinting of the partial

5.8S and internal transcribed spacer (ITS) region 2 (LaJeunesse 2002). The region was amplified using a touch-down thermal cycle profile with the primers “ITS2clamp” and

130

Figure 15. Sampling locations for S. trenchi in the Caribbean, Western Pacific Ocean, Eastern Indian Ocean and Western Indian Ocean. Collections were made at multiple sites within each location and were separated by <100 km. The specific locations in the Indo- Pacific (left) are as follows: Zanzibar, Tanzania (1); Andaman Sea, Thailand (2); and Palau (3). The specific locations in the Caribbean (right) are as follows: the Flower Garden Banks (4); Key West (5); St. Croix (6); Barbados (7); Panama (8); Belize (9); and Mexico (10).

“ITSintfor2” (LaJeunesse & Trench 2000), and the PCR products resolved on denaturing gels

(45 – 80% of 7M urea & 40% formamide) using a CBScientific system (Del Mar, CA) for 16 hours at 115 volts. All samples used in this study were verified to contain S. trenchi (D1a).

Microsatellite Analysis

Twelve microsatellite loci developed for Clade D Symbiodinium were used to determine the genetic diversity of S. trenchi populations (Table 14; Pettay & LaJeunesse

2009; Wham et al. 2011). These loci are the same loci that have been used previously on other species of clade D Symbiodinium (i.e., S. glynni), however S. trenchi has experience a partial or whole genome duplication making these loci diallelic in S. trenchi (Pettay &

LaJeunesse 2009; Wham et al. 2011). Although it is still believed that S. trenchi is a haploid organism, like other Symbiodinium spp., the diallelic nature of these microsatellite loci cause the data to be treated as diploid and Hardy-Weinberg equilibrium was measured for each

131 Table 14. Description of microsatellite loci used in this study. Ae = effective alleles with standard error in parenthesis. Subscript numbers following the repeat motif indicate the number of repeats in the initial cloned sequence that was used to develop locus primers.

Locus Primer Sequence (5' - 3') Repeat Motif Size Range (bp) Number of Alleles Ae

D1Sym9 F - CAGAAGCCCAATTATATGCGGCA (FAM) (GTT)6 106 - 112 3 2.03 (0.05) R - AGGATGATGAGCATGCCGACG

D1Sym11 F - TGAAATCTCACTCAGAGTCGGAC (FAM) (AC)13 149 - 161 7 3.52 (0.59) R - GCAGACAGTGATTTCAGTTCCGA

D1Sym14 F - TCTCAGTGGAAAGCATTGTGG (FAM) (CT)11 AT (CT)4 167 - 191 7 2.37 (0.13) R - TCGTCTGAATCAGGATCTGACG

D1Sym17 F - TGTGAATGCTTCTTGGGGTG (HEX) (CA)8 137 - 147 6 2.91 (0.34) R - TCATGCTTGTCCGTGAGCAG D1Sym34 F - ACCTGAGACCTGAGTGTTGC (FAM) (CAAA)9 CACA (CAAA)4 350 - 429 30 9.92 (3.00)

126 (GAAACAAA)2

R - ATCATGGGCAGAGCTCCTGG (CAAA)13

D1Sym66 F - CTCTGGGATGGACCCCGAA (PET) (TCT)8 288 - 294 5 2.08 (0.39) R - GACAGTGTTTCACTGGTCGCA

D1Sym67 F - GAATCCAGATGGTGCCTGC (VIC) (ATC)8 125 - 140 6 2.45 (0.54) R - CAAAGGTAGCCGATTGTCTC

D1Sym82 F - AAGACTTCCGCAAGCACAC (HEX) (ACT)12 225 - 285 21 8.25 (2.44) R - CGCATCTACTGGTGGTCAGAC

D1Sym87 F - CCTATGACTCCAAGGGTGACG (FAM) (GAAG)7 236 - 256 6 2.42 (0.12) R - AGACATACCTCGGTCTTGTC

D1Sym88 F - TTGTCAGACTGAATGCTCCA (NED) (CTTT)3 G (TTTC)7 232 - 250 5 1.63 (0.25) R - GTGTTCAAGCGACATCCCA

D1Sym92 F - GCGTTTGACACAAGGATCCCT (FAM) (CCTA)6 (CCTG)3 124 - 136 4 1.30 (0.16) R - TTGGGATGCTCTTGGCGAC

D1Sym93 F - GCTCAAAGGAGCTCTAGGGGT (NED) (AATC)6 141 - 173 8 2.72 (0.23) R - TGTCAAGGTAGAGAGCCTGGT

regional location in GenAlEx (version 6.4, Peakall & Smouse 2006). GenePop (version

4.0.11, Rousett 2008) was used to examine the degree of linkage among these loci in pairwise comparisons.

The forward primers of all loci were fluorescently labeled and the microsatellite fragments amplified, and fragments analyzed at using the Pennsylvania State University

Genomics Core Facility according the methods of Pettay & LaJeunesse (2007 & 2009) and

Wham et al. (2011). Since microsatellites are known to produce stutter due to polyermase slippage during PCR, only the dominant peaks were scored. Multiple alleles at a particular locus were scored when a second clear peak was found in the expected size range and was at least one-third the size of the dominant peak. The presence of multiple peaks was interpreted to indicate that a sample contained more than one haploid genotype. The occasional exclusion of small ambiguous peaks may occasionally cause a low background genotype to be missed, however this conservative approach prevents the overestimation of genetic diversity (Anderson et al. 2000). Additionally, while this approach may slightly skew the results of diversity on a per colony basis, it should have little impact on the population structure analyses since these mixed genotype samples did not result in additional allelic diversity. Multilocus genotypes (MLGs) were constructed from fragment size data gathered for each sample according the methods of Pettay & LaJeunesse (in prep.).

Data Analysis

Populations of organisms that commonly reproduce asexually will tend to have a high frequency of clones. Therefore statistical calculations based on allele frequencies can be negatively biased (Arnaund-Haond et al. 2007; Pettay & LaJeunesse in prep.). For this

133 reason, all analyses (unless stated otherwise) were conducted with duplicated MLGs removed at each sampling location.

2 2 4 The probability of identity (PI = (Σpi ) - Σpi , where pi is the frequency of the i-th allele at a locus) was calculated to determine the power to resolve genetically distinct individuals. The PI is an estimate of the probability that two unrelated individuals drawn at random will by chance have the same MLG (Peakall & Smouse 2005). Individual PI’s are calculated for each locus and an overall PI for all loci is the product of each individual locus

PI. PI values between 0.01 – 0.0001 are believed to be reasonably low enough for population studies (Waits et al. 2001), with values lower than 0.01 adequate for mark-recapture studies on population size estimation (Mills et al. 2000). Since these values may be affected by population substructure (Waits et al. 2001), PI was calculated for each regional location

GenAlEx (Peakall & Smouse 2006).

Clonal richness (R) which is equal to (G – 1)/(N – 1), where G = # of unique MLGs and N = total sample size, was calculated to give the frequency of unique genotypes and an indication of the contribution of asexual reproduction. To examine the genetic diversity between individuals within regions, genetic distance (GD) between unique MLGs was calculated and their frequencies plotted. Due to the lower sample size of the Zanzibar population, GD plots were not produced for this population. Clonal richness, GD plots and other measures of clonality were conducted using GenClone (v. 2.0, Arnaud-Haond &

Belkhir 2007). A principal coordinate analysis (PCoA) based on GD was conducted using

GenAlEx to graphically depict the relatedness between individuals within and between regional populations. Lastly, the phylogenetic relationship between unique MLGs was analyzed using an Unweighted Pair Group Method with Arithmetic Mean (UPGMA) using

134 Neighbor from the PHYLIP package (Felsenstein 2004). The GD was calculated calculated in GenAlEx.

The software Structure (Version 2.3.2) was used to overcome biases of assigning populations by location by using the microsatellite data to cluster MLG based on their genetic similarities irrespective of sample origin. Briefly, this software uses a Bayesian clustering approach to probabilistically assign individuals to populations (Pritchard et al.

2000). The model assumes there are K populations characterized by a set of allele frequencies, with the assumptions of unlinked loci that are in linkage and Hardy-Weinberg equilibrium within populations (Pritchard et al. 2000).

Structure analyses to investigate differentiation across the eastern Pacific were conducted using an admixture model with correlated allele frequencies and were run from K

= 1 to 10 with five runs per K and a burnin of 100,000 and 1,000,000 reps after the burnin. A plot of the log probability of the data for a given K (ln P(D)) versus K was derived from the structure results, along with an analysis of the second order rate of change of K following the method by Evanno et al. (2005) and implemented using Structure Harvester (v0.56.4) by Earl

(2009) to determine the appropriate clustering of individuals. Five runs per K were utilized to verifying consistency between runs, with the run having the highest ln P(D) for the appropriate K used to construct Structure plots and inform clustering. Graphic displays of

Structure plots were manipulated (i.e., color and sample order) using DISTRUCT (Rosenberg

2004).

An analysis of molecular variance (AMOVA) was conducted to determine the degree of differentiation between regional locations. The AMOVA along with a permutation procedure were performed in GenAlEx to test for significant difference in genetic diversity

135 between populations (Excoffier et al. 1992). The AMOVA produces variance components along with Φ-statistics (F-statistic analogs), which partition genetic variation at different hierarchical levels (Excoffier et al. 1992). The calculation of Φ-statistics as opposed to F- statistics was considered more appropriate since this method suppresses the within individual variation usually considered with true diploid organisms (Peakall & Smouse 2006). The significance of the variance components and Φ-statistics were then tested using 10000 permutations and a Bonferroni corrected α = 0.05. In addition to quantifying genetic differentiation, GenAlEx was used to calculate several summary statistics for each regional location, including observed and expected heterozygosity (Ho & He) and I.

Results

ITS2-DGGE fingerprinting and sequencing

Each of the MLGs characterized in this study possessed two co-dominant ITS2 sequences (D1 & a) that dominated their ribosomal array (Thornhill et al. 2007; Sampayo et al. 2009, LaJeunesse et al. 2010a). These co-dominant sequences along with their unique

ITS2-DGGE fingerprint contrasts with most other Clade D types characterized by this method, indicating S. trenchi represents a distinct operational taxonomic unit (LaJeunesse et al. 2010a).

Microsatellite Data

The number of alleles per locus ranged from 3 to 30, however, the effective number of alleles (Ae) per locus ranged from 1.30 to 9.92 (Table 14). Allele frequencies for each location ranged from 0.010 to 1.00 (Appendix G). The frequencies of putative null alleles were rare among loci ranging from 0 for loci D1Sym9 to a value of 0.069 for D1Sym82.

136 While all of the alleles in the Caribbean were also found in other sampling regions, numerous alleles were private to each of the other three regions (Table 15). PAL possessed 7 alleles involving six loci (D1Sym34, 66, 67, 82, 88 & 93) that were unique to the region. THA possessed 19 alleles from seven loci (D1Sym14, 34, 67, 82, 87, 92 & 93) that were unique.

ZAN possessed 3 alleles from two loci (D1Sym14 & 34) that were unique. A total of one hundred and nineteen different genotypes were scored among one hundred and eighty-nine

MLGs obtained from the entire sample set. Twenty-five of these MLGs were from CAR, thirty from PAL, fifty-three from THA and eleven from ZAN (Table 15).

Table 15. Summary statistics by location including number of colonies sampled (NS), number of host taxa (genera) sampled (NH), number of unique MLGs of S. trenchi (G), clonal richness (R), information index (I), observed (Ho) and expected (He) heterozygosity, number of private alleles (PA) and probability of identity (PI). Numbers in parenthesis represent the standard error for a particular statistic.

Location NS NH G R I Ho He PA PI 0.67 0.64 0.40 Caribbean (CAR) 88 8 25 0.28 (0.15) (013) (0.08) 0 1.47E-05 1.27 0.75 0.61 Palau (PAL) 31 21 30 0.97 (0.21) (0.07) (0.06) 7 1.92E-10 1.43 0.76 0.66 Thailand (THA) 58 22 53 0.91 (0.21) (0.06) (0.05) 19 9.30E-12 1.14 0.73 0.59 Zanzibar (ZAN) 13 9 11 0.83 (0.15) (0.07) (0.06) 3 2.71E-09

The probability of identity (P(ID)) was calculated for samples from each location using allele frequency data at each locus. The probability that two samples with the same MLGs may not have originated from the same clone lineage was exceedingly low for most locations and ranged from 1.47 x 10-5 to 9.30 x 10-12 (Table 15). All of these values were below the

0.0001 criteria recommended by Waits et al. (2001) and are therefore believed to be adequate for population level investigations.

137 Clonal richness (R) or genotypic richness was high for all Indo-Pacific locations ranging from 0.83 to 0.97 (Table 15). The most often a clone was detected was four times

(THA) and no clones were shared among regions or reefs within a region (Figure 16). In contrast, R was low in the CAR (0.28) and clones were frequently found multiple times both within sample sites and between locations as far as 3,000 km apart (Table 15, Figure 17). In the most extreme case, one MLG was found in forty-four out of the eighty-eight samples from around the CAR and at every location except the Flower Garden Banks (Figure 16 &

17).

Figure 16. The occurrence of increasing clone sizes (number of individuals with identical MLGs) for S. trenchi based on four regional samplings. Colors correspond to sampling location. A majority of clones were only found only once or twice, with the exception of one clone (arrow) found forty-four times in the Caribbean.

138

Figure 17. The distribution and abundance of particular clones of S. trenchi at each location in the Caribbean, with the locations corresponding to Figure 15. The size of the pie charts reflects the number of samples at each location, with black indicating MLGs found only once in the Caribbean and grey representing MLGs unique to Barbados. Colors shared between locations indicate identical MLGs found at multiple sampling sites. Barbados, Panama and the Flower Gardens each had at least one MLG private to that location, and the red MLG was found at and prevalent at all locations except the Flower Gardens.

A pairwise analysis of linkage among loci revealed that ~95% of the pairwise comparisons were unlinked (Bonferroni corrected; α = 0.05). In general, linkage between pairs of loci changed based on population, with no two loci linked in all locations. The linkage for some loci is believed to be a consequence of the clonal evolution of these dinoflagellates or possible evidence of weak differentiation over smaller spatial scales or according to host taxa.

139 Population Differentiation

Observed heterozygosity ranged from 0.64 to 0.76 and was lowest in the CAR (Table

15). Observed was higher than expected heterozygosity in every region. The number of loci in Hardy-Weinberg equilibrium varied for each region, ranging from ten in ZAN to three in

PAL. Similar to heterozygosity, the information index (I) was higher in the Indo-Pacific

(1.14 to 1.43) than in the Caribbean (0.67).

Figure 18. The distribution of genetic distances (number of distinct alleles) between individuals and their frequency of occurrence for samples within the Caribbean, Andaman Sea and Palau.

140 The GD between individuals within regions was only determined for CAR, PAL and

THA. The average GD between individuals was high in the Indo-Pacific regions (15.68 &

13.23), but was much lower in the CAR (5.00; Figure 18). Additionally, both THA and PAL showed a bell-shaped distribution of distances around the average for each population. In contrast, the CAR population showed a bimodal distribution skewed towards a distance of two. The skewed distribution of the CAR is attributed to the high similarity between individuals within this region, many of which (~75%) differ by only four alleles or less. The

PCoA (axis 1 = 29% & axis 2 = 20% of variation) shows a similar lack of diversity in the

CAR, where all but two individuals (one in the Flower Garden Banks and one in Barbados) cluster tightly together and in contrast to the larger cluster from the Indo-Pacific regions

(Figure 19). Similarly, the UPGMA tree clusters all but two samples from the Caribbean together and with short branch lengths indicative of closely related individuals (Figure 20).

In contrast, most of the individuals from the Indo-Pacific fall at the end of long braches.

Lastly, the Caribbean samples are embedded within the larger Indo-Pacific tree as opposed to representing their own distinct branch like would be expected of subdivide populations.

Figure 19. Principal coordinate analysis of S. trenchi MLGs from the Caribbean, Palau, Andaman Sea and Zanzibar. The two axes describe 29% (axis 1) and 20% (axis 2) of the total variation. Colors and shapes correspond to sampling locations.

141

Figure 20. UPGMA tree produced by PHYLYP showing the phylogenetic relationship between unique MLGs from each regional sampling site. Branch lengths correlate to genetic distance and the colors of branches indicate location of origin for a MLG. Colors correspond to those used in Figure 19.

142 Two regional populations were detected using the Evanno et al. method to process the

Structure results (k = 2; ln P(D) = -3553.8), and corresponded to the Caribbean and the Indo-

Pacific. Examination of Structure runs of k > 2 showed greater variation indicative poor fit for the data. Subsequent analyses using the location prior option or without the Caribbean showed a similar lack of population subdivision between PAL, THA and ZAN.

Table 16. Analysis of Molecular Variance (AMOVA) for S. trenchi in all four sampling locations. Estimated Percent of Fixation Source df SS MS Variance Variation Index

Among Populations 3 88.99 29.66 0.86 12% ΦPT = 0.119*

Within Populations 115 728.70 6.34 6.34 88%

Total 118 817.69 7.19 100% * P < 0.0001

Table 17. A pairwise matrix comparing the genetic relationship (ΦPT) between all four locations of S. trenchi in the Caribbean and Pacific. Significant values (Bonferroni corrected P < 0.05) represented in bold.

Indo-Pacific Andaman Caribbean Palau Zanzibar Sea

Caribbean ---

Palau --- 0.209

Pacific Andaman Sea --- - 0.213 0.032

Indo Zanzibar --- 0.268 0.025 0.041

*Bold indicate significance at P < 0.05 after sequential Bonferroni correction

The population genetic data were divided into the four populations (CAR, PAL, THA

& ZAN) and AMOVA analysis yielded ΦPT = 0.119, which was highly significant (p <

0.0001, Table 16). Φ-values of pairwise comparisons were significant for all comparisons

143 except for two involving ZAN with PAL and THA (Table 17). While the lack of significance for ZAN may indicate gene flow between these regions, it is most likely a function of the lower sample size in ZAN since the comparison between PAL and THA is highly significant.

Discussion

The worldwide decline of coral reefs due to climate related bleaching events has been widely documented and the frequency of such events appears to be increasing. The capacity of some algal symbionts (Symbiodinium spp.) of coral to better withstand thermal anomalies has allowed individuals harboring these symbionts to survive bleaching events at a higher rate. Symbiodinium trenchi (D1a) is one of these symbionts and has demonstrated its stress tolerance during thermal bleaching events and by its ability to thrive in marginal habitats.

Behavior of this symbiont during the 2005 mass bleaching event in Barbados indicates that while most Symbiodinium spp. were negatively impacted, S. trenchi has the ability to spread to coral taxa where it was not found prior to bleaching and subsequently increase its ecological footprint during this environmental perturbation. A population genetic approach was used to better understand this Caribbean population of S. trenchi and its evolutionary potential in relation to populations in the Indian and Pacific Oceans.

Population differentiation and dispersal

Regional analyses of S. trenchi populations indicate there is significant differentiation between the four regions, with significant pairwise comparisons between most pairs of populations (Table 17) and numerous private alleles found in each region (except the

144 Caribbean; Table 15). While the differing of allele frequencies shows subdivision within the

Indo-Pacific, PCoA only differentiates the Caribbean population with the rest of Indo-Pacific forming one large cluster (Figure 19). The lack of regional clusters in the Indo-Pacific is supported by Structure analyses, which only differentiate the Caribbean and the Indo-Pacific.

Additionally, pairwise ΦPT-values are much lower between Indo-Pacific locations and indicate that gene flow may occasionally occur over this vast region. Similarly, the domination of a single clone along with the sharing of numerous less abundant clones between locations in the Caribbean basin suggests a high dispersal potential for this unusual symbiont (Figure 17).

The apparent large dispersal potential of S. trenchi is in stark contrast to observations for most other Symbiodinium spp. and other marine microbial eukaryotes (Santos et al. 2003,

Rynearson et al. 2006, Howells et al. 2009, Kirk et al. 2009, Nagai et al. 2009, Thornhill et al. 2009, Casteleyn et al. 2010). Symbiodinium spp. typically show highly structured populations over distances as small as 5-10 km with clones rarely found outside of a single reef or adjacent reefs. S. trenchi populations contrast these patterns with only slight population subdivision over the Indo-Pacific and a single clone that dominates most of the sampled locations in the Caribbean (Figure 16 & 17, Table 17). A similar lack of population subdivision has been observed for another clade D symbiont, S. glynni, in the Eastern Pacific

(Pettay & LaJeunesse in prep). This symbiont is vertically transmitted to the offspring however, and the ability of this symbiont to co-migrate with its host is believed to greatly enhance the dispersal of S. glynni (Pettay & LaJeunesse in prep.). S. trenchi predominantly associates with horizontally transmitting hosts and must therefore rely on its own dispersal

145 mechanisms and ocean currents. Given a passive form of dispersal, it is remarkable that S. trenchi can disperse such great distances.

Clonality in the Caribbean

Symbiodinium populations are maintained by a combination of clonal and sexual reproduction (Fitt & Trench 1983; LaJeunesse 2001; Santos & Coffroth 2003). Many factors can influence the contribution of sexual versus asexual reproduction including environment, disturbance, population density and the intrinsic biology of the organism (Bell et al. 1992,

Otto 2009, Vallejo-Marín et al. 2010). For Symbiodinium, the degree of clonality can be wide-ranging depending on the species. Symbiodinium B1 associating with Gorgonia ventalina in the Florida Keys shows little clonality (R = 0.84; Kirk et al. 2009), while S. glynni in the Eastern Pacific shows high clonality through much of its range (R = 0.53, Pettay

& LaJeunesse in prep). Further examination of individual populations of S. glynni shows that environment may drive the contribution of asexual reproduction with offshore island populations being more genotypically diverse than inshore locations, however this pattern needs to be examined in more depth (Pettay & LaJeunesse in prep).

At first glance the large increase in clonality within the Caribbean (R = 0.28) compared to the Indo-Pacific (R > 0.83) could be explained by differing environmental conditions or disturbance histories (i.e., bleaching). The majority of the reefs sampled in the

Indo-Pacific were near-shore lagoon habitats with high turbidity and/or average temperature

(LaJeunesse et al. 2010a), while many of the reefs sampled in the Caribbean are considered more pristine environments (e.g., Barbados, Belize, St. Croix). However, until the 2005 bleaching event in Barbados the distribution of S. trenchi in the Caribbean was typically

146 patchy and relegated to marginal habitats such as near-shore lagoons or the extreme limits of its host’s depth range (Toller et al. 2001; Garren et al. 2006; LaJeunesse et al. 2010b).

Similarly, while the Caribbean has received much attention for thermal bleaching events, similar patterns of bleaching and the decline of coral reefs worldwide is now recognized

(Williams & Bunkley-Williams 1990; Bruno & Selig 2007). Therefore, the environmental factors that could be contributing to differences in clonality between the regions do not appear to be driving the patterns we see here and suggests some other explanation must exist to account for the drastic reduction of genotypic diversity within the Caribbean (Figure 16 &

18).

Lack of allelic and genotypic diversity in the Caribbean

High clonality is not the only unusual characteristic of S. trenchi in the Caribbean.

This region also possesses lower genetic diversity between individuals (Figure 18). The high similarity in the Caribbean is in stark contrast to the Indo-Pacific where less than five percent of individuals at a particular location show a similar relatedness. Additionally, these related individuals are spread over the entire Caribbean and encompass thousands of square kilometers, while the Indo-Pacific locations show higher diversity and more divergence over reefs separated by less than 100 km. Lastly, the distribution of genetic diversity in Indo-

Pacific populations of S. trenchi is typical of other Symbiodinium, both within clade D (S

.glynni) and within the Caribbean (S. B7), and further suggests that unusual historical population processes are affecting S. trenchi in the Caribbean (Pettay & LaJeunesse 2007;

Pettay & LaJeunesse in prep). Whatever has caused this lack of diversity, it is most likely that the dominant clone of S. trenchi originated in a single location and has subsequently

147 spread throughout the region and began to diversify (e.g., mutation or sexual recombination).

Dispersing from a point of origin and subsequent diversification would explain the high similarity of individuals around the Caribbean.

S. trenchi has been introduced to the Caribbean

The lack of diversity, along with high clonality, suggests that there has been a significant reduction in population size for S. trenchi in the Caribbean and a subsequent spread of a particularly infectious MLG (or MLL) throughout the region. Two divergent processes could be expected to yield the population characteristics seen in the Caribbean: a population bottleneck or a founder event (Hartl & Clark 1997). In the case of a Caribbean- wide bottleneck, some event (e.g., thermal anomaly) must have drastically reduced the number of individuals around the Caribbean and subsequently caused a reduction in genetic diversity. However, the expectation under a Caribbean-wide bottleneck where some individuals survived and others did not, would yield populations around the region composed of random MLGs and retain a relatively high amount of diversity, a pattern not seen in the

Caribbean. Perhaps a bottleneck occurred that was so severe that only one or a few isolated populations survived. A bottleneck such as this and subsequent dispersal could result in the patterns seen in the Caribbean, but the catastrophic event needed has not occurred in this region. Similarly, you would expect other Symbiodinium spp. to experience similar bottlenecks and show analogous patterns of low diversity, but those studied to date show no evidence of such an event and make this scenario highly improbable (Pettay & LaJeunesse

2007; Kirk et al. 2009). Additionally, the Caribbean has a high diversity of Symbiodinium spp. and if an event occurred that was so devastating to a stress-tolerant species like S.

148 trenchi, then species-level diversity would also suffer (LaJeunesse 2002; LaJeunesse et al.

2003; Finney et al. 2010).

The outcome of a founder event seems a better fit for the data gathered in the

Caribbean. In this scenario, one or a few individuals recruit into a new location and are genetically limited by the genomes of the new individual(s) (Hartl & Clark 1997). The founding population can then disperse to additional locations and/or diversify through sexual recombination and/or mutation. The results of this scenario would mirror that found for S. trenchi in the Caribbean, where individuals within and between locations are genetically similar and overall genetic diversity is low (Figure 17 & 18). In the case of clonal organisms like Symbiodinium, it would even be possible to find clonemates between locations. While founding events do occur through natural processes, it is most likely that S. trenchi’s founding population represents the introduction of an Indo-Pacific symbiont species into the

Caribbean. Based the distribution of genotypic diversity and clonality the most probable site of the introduction occurred in Panama or Barbados (Figure 17), however sampling more locations and at a higher rate would be needed to determine the site of introduction. From here, the founding population has spread throughout the Caribbean, with one clonal lineage dominating a majority of the associations. Whether the diversity of the current population is due to sexual recombination between founding individuals or is the result of the accumulation of mutations in single founding clone is unknown at this time. There appears to have been multiple introductions though, with one sample in the Flower Garden Banks and one in Barbados representing more divergent lineages (Figure 18), and may indicate that the diversity is a combination of the two processes.

149 The introduction of nonindigenous dinoflagellates to new regions around the world has been occurring for some time, primarily through ballast water (Williams et al. 1988;

Hallegraeff & Bolch 1992; Carlton & Geller 1993; Carlton 1996). Ballast water in cargo ships entering the Caribbean through the Panama Canal is one possibility for the introduction of S. trenchi. Regardless of the route the fact that a single clonal lineage could disperse and dominate the entire Caribbean is remarkable, but may not be uncommon for clonal species introduced into new ecosystems. Asexually reproducing microbial eukaryotes, invertebrates and plants have all shown the ability to maintain populations and further spread through clonal reproduction when introduced into a new region (Provan et al. 2005; Liu et al. 2006;

Mergeay et al. 2006; Lilly et al. 2007; Milgroom et al. 2008; Okada et al. 2009; Lambertini et al. 2010). An extreme example is the water flea, Daphnia hybrid, where a single clone has spread and displaced the native species throughout its entire range in Africa within sixty years of the introduction (Mergeay et al. 2006). This Daphnia clone is a particularly good disperser and highly competitive, qualities that must also apply to the S. trenchi clone in the

Caribbean. If the S. trenchi introduction did coincide with the opening of the Panama Canal, the ability for a single clonal lineage to disperse throughout the Caribbean over a >90 year period seems highly probable in light of the Daphnia introduction. This possibility seems even more plausible considering that disease agents appear to have spread throughout the

Caribbean in less than one year (Lessios et al. 1984).

Conclusions

Understanding the diversity and population structuring of the stress-tolerant S. trenchi will begin to answer questions concerning the ability of this symbiont positively affect coral

150 reefs as the climate continues to warm. Genetic diversity and gene flow are important factors in the evolutionary potential of a species, and it appears that most S. trenchi populations are highly diverse and have the ability to disperse great distances. However, the low diversity of this symbiont in the Caribbean and possibility that this population is the result of an introduction raises even more questions concerning the ability of this symbiont to shape the contemporary evolution of coral reefs in this region. While this symbiont can enhance the thermal tolerance of its host, these associations are typically short-lived in the Caribbean.

The temporary nature of these associations may indicate they are more beneficial to this introduced symbiont and provide very little long-term gain for its host. Continued research into the physiological interplay between S. trenchi and Caribbean coral may truly determine the evolutionary potential of these symbioses.

References

Anderson T, Haubold B, Williams JT, et al. (2000) Microsatellite markers reveal a spectrum of population structures in the malaria parasite Plasmodium falciparum. Molecular Biology and Evolution 17, 1467-1482.

Arnaud-Haond S, Belkhir K (2007) GENCLONE: a computer program to analyse genotypic data, test for clonality and describe spatial clonal organization. Molecular Ecology Notes 7, 15-17.

Arnaud-Haond S, Duarte CM, Alberto F, Serrão EA (2007) Standardizing methods to address clonality in population studies. Molecular Ecology 16, 5115-5139.

Baker A (2001) Reef corals bleach to survive change. Nature 411, 765-766.

Baker A, Starger C, McClanahan T, Glynn P (2004) Corals' adaptive response to climate change. Nature 430, 741.

Bell G (1982) The masterpiece of nature: the evolution and genetics of sexuality Croom Helm, London.

151 Berkelmans R, Van Oppen M (2006) The role of zooxanthellae in the thermal tolerance of corals: a ‘nugget of hope’for coral reefs in an era of climate change. Proceedings of the Royal Society B: Biological Sciences 273, 2305-2312.

Bruno JF, Selig ER (2007) Regional Decline of Coral Cover in the Indo-Pacific: Timing, Extent, and Subregional Comparisons. PLoS One 2, e711.

Buddemeier R, Fautin D (1993) Coral bleaching as an adaptive mechanism. Bioscience 43, 320-326.

Carlton J (1996) Marine bioinvasions: the alteration of marine ecosystems by nonindigenous species. Oceanography 9, 36-43.

Carlton J, Geller JB (1993) Ecological roulette: the global transport of nonindigenous marine organisms. Science 261, 78-82.

Casteleyn G, Leliaert F, Backeljau T, et al. (2010) Limits to gene flow in a cosmopolitan marine planktonic diatom. Proceedings of the National Academy of Sciences 107, 12952-12957.

Coffroth M, Santos S (2005) Genetic Diversity of Symbiotic Dinoflagellates in the Genus Symbiodinium. Protist 156, 19 - 34.

Correa AMS, Baker AC (2010) Disaster taxa in microbially mediated metazoans: how endosymbionts and environmental catastrophes influence the adaptive capacity of reef corals. Global Change Biology 17, 68-75.

Earl DA (2009) Structure Harvester, from website: v0.3 http://users.soe.ucsc.edu/~dearl/software/struct_harvest/.

Evanno G, Regnaut S, Goudet J (2005) Detecting the number of clusters of individuals using the software STRUCTURE: a simulation study. Molecular Ecology 14, 2611-2620.

Excoffier L, Smouse P, Quattro J (1992) Analysis of molecular variance inferred from metric distances among DNA haplotypes: application to human mitochondrial DNA restriction data. Genetics 131, 479-491.

Fabricius KE, Mieog JC, Colin PL, Idip D, H. Van Oppen MJ (2004) Identity and diversity of coral endosymbionts (zooxanthellae) from three Palauan reefs with contrasting bleaching, temperature and shading histories. Molecular Ecology 13, 2445-2458.

Felsenstein J (2004) PHYLIP (Phylogeny Inference Package) version 3.6. Distributed by the author. Department of Genome Sciences, University of Washington, Seattle.

152 Finney JC, Pettay DT, Sampayo EM, et al. (2010) The Relative Significance of Host– Habitat, Depth, and Geography on the Ecology, Endemism, and Speciation of Coral Endosymbionts in the Genus Symbiodinium. Microbial Ecology 60, 250-263.

Fitt W, McFarland F, Warner M, Chilcoat G (2000) Seasonal patterns of tissue biomass and densities of symbiotic dinoflagellates in reef corals and relation to coral bleaching. Limnology and Oceanography 45, 677-685.

Fitt W, Trench R (1983) The Relation of Diel Patterns of Cell Division To Diel Patterns of Motility in the Symbiotic Dinoflagellate Symbiodinium Microadriaticum FREUDENTHAL in Culture. New Phytologist 94, 421-432.

Franklin I, Frankham R (1998) How large must populations be to retain evolutionary potential? Animal Conservation 1, 69-70.

Garren M, Walsh S, Caccone A, Knowlton N (2006) Patterns of association between Symbiodinium and members of the Montastraea annularis species complex on spatial scales ranging from within colonies to between geographic regions. Coral Reefs 25, 503-512.

Glynn P, Mate J, Baker A, Calderon M (2001) Coral bleaching and mortality in Panama and Ecuador during the 1997-1998 El Niño-Southern Oscillation Event: spatial/temporal patterns and comparisons with the 1982-1983 event. Bulletin of Marine Science 69, 79-109.

Hallegraeff G, Bolch C (1992) Transport of diatom and dino- flagellate resting spores in ships’ ballast water: implications for plankton biogeography and aquaculture. Journal of Plankton Research 14, 1067-1084.

Hartl D, Clark A (1997) Principles of Population Genetics, 3rd edn. Sinauer Associates, Sunderland, Massachusetts.

Hoegh-Guldberg O (1999) Climate change, coral bleaching and the future of the world's coral reefs. Marine & Freshwater Research 50, 839-866.

Howells E, Van Oppen M, Willis B (2009) High genetic differentiation and cross-shelf patterns of genetic diversity among Great Barrier Reef populations of Symbiodinium. Coral Reefs 28, 215-225.

Jones A, Berkelmans R, Van Oppen M, Mieog J, Sinclair W (2008) A community change in the algal endosymbionts of a scleractinian coral following a natural bleaching event: field evidence of acclimatization. Proceedings of the Royal Society B: Biological Sciences 275, 1359-1365.

153 Kirk N, Andras J, Harvell C, Santos S, Coffroth M (2009) Population structure of Symbiodinium sp. associated with the common sea fan, Gorgonia ventalina, in the Florida Keys across distance, depth, and time. Marine Biology 156, 1609-1623.

LaJeunesse T (2001) Investigating the biodiversity, ecology, and phylogeny of endosymbiotic dinoflagellates in the genus Symbiodinium using the ITS region: in search of a" species" level marker. Journal of Phycology 37, 866-880.

LaJeunesse T (2002) Diversity and community structure of symbiotic dinoflagellates from Caribbean coral reefs. Marine Biology 141, 387-400.

LaJeunesse T, Bonilla H, Warner M, et al. (2008) Specificity and stability in high latitude eastern Pacific coral-algal symbioses. Limnology and Oceanography 53, 719-727.

LaJeunesse T, Loh W, Van Woesik R, et al. (2003) Low symbiont diversity in southern Great Barrier Reef corals, relative to those of the Caribbean. Limnology and Oceanography, 2046-2054.

LaJeunesse T, Reyes-Bonilla H, Warner M (2007) Spring “bleaching” among Pocillopora in the Sea of Cortez, Eastern Pacific. Coral Reefs 26, 265-270.

LaJeunesse TC, Pettay DT, Sampayo EM, et al. (2010a) Long-standing environmental conditions, geographic isolation and host-symbiont specificity influence the relative ecological dominance and genetic diversification of coral endosymbionts in the genus Symbiodinium. Journal of Biogeography 37, 785-800.

LaJeunesse TC, Smith R, Walther M, et al. (2010b) Host–symbiont recombination versus natural selection in the response of coral–dinoflagellate symbioses to environmental disturbance. Proceedings of the Royal Society B 277, 2925-2934.

LaJeunesse TC, Smith RT, Finney J, Oxenford H (2009) Outbreak and persistence of opportunistic symbiotic dinoflagellates during the 2005 Caribbean mass coral 'bleaching' event. Proceedings of the Royal Society B: Biological Sciences 276, 4139- 4148.

Lambertini C, Riis T, Olesen B, Clayton J, Sorrell B (2010) Genetic diversity in three invasive clonal aquatic species in New Zealand. BMC Genetics 11, 1-18.

Lessios H, Robertson D, Cubit J (1984) Spread of Diadema mass mortality through the Caribbean. Science 226, 335-337.

Lilly EL, Halanych KM, Anderson DM (2007) Species boundaries and global biogeography of the Alexandrium tamarense complex (Dinophyceae). Journal of Phycology 43, 1329-1338.

154 Liu J, Dong M, Miao SL, et al. (2006) Invasive alien plants in China: role of clonality and geographical origin. Biological Invasions 8, 1461-1470.

McWilliams JP, Cô té IM, Gill JA, Sutherland WJ, Watkinson AR (2005) Accelerating impacts of temperature-induced coral bleaching in the Caribbean. Ecology 86, 2055- 2060.

Mergeay J, Verschuren D, Meester L (2006) Invasion of an asexual American water flea clone throughout Africa and rapid displacement of a native sibling species. Proceedings of the Royal Society B: Biological Sciences 273, 2839-2844.

Milgroom MG, Sotirovski K, Spica D, et al. (2008) Clonal population structure of the chestnut blight fungus in expanding ranges in southeastern Europe. Molecular Ecology 17, 4446-4458.

Mills L, Citta J, Lair K, Schwartz M, Tallmon D (2000) Estimating animal abundance using noninvasive DNA sampling: promise and pitfalls. Ecological Applications 10, 283- 294.

Mostafavi PG, Fatemi S, Shahhosseiny M, Hoegh-Guldberg O, Loh W (2007) Predominance of clade D Symbiodinium in shallow-water reef-building corals off Kish and Larak Islands (Persian Gulf, Iran). Marine Biology 153, 25-34.

Nagai S, Nishitani G, Sakamoto S, et al. (2009) Genetic structuring and transfer of marine dinoflagellate Cochlodinium polykrikoides in Japanese and Korean coastal waters revealed by microsatellites. Molecular Ecology 18, 2337-2352.

Okada M, Lyle M, Jasieniuk M (2009) Inferring the introduction history of the invasive apomictic grass Cortaderia jubata using microsatellite markers. Diversity and Distributions 15, 148-157.

Otto Sarah P (2009) The Evolutionary Enigma of Sex. The American Naturalist 174, S1-S14.

Oxenford HA, Roach R, Brathwaite A, et al. (2008) Quantitative observations of a major coral bleaching event in Barbados, Southeastern Caribbean. Climatic Change 87, 435-449.

Peakall R, Smouse PE (2006) GENALEX 6: genetic analysis in Excel. Population genetic software for teaching and research. Molecular Ecology Notes 6, 288-295.

Pettay D, LaJeunesse T (2007) Microsatellites from clade B Symbiodinium spp. specialized for Caribbean corals in the genus Madracis. Molecular Ecology Notes 7, 1271-1274.

155 Pettay DT, LaJeunesse TC (2009) Microsatellite loci for assessing genetic diversity, dispersal and clonality of coral symbionts in ‘stress-tolerant’ clade D Symbiodinium. Molecular Ecology Resources 9, 1022-1025.

Pettay DT, LaJeunesse TC (in press) Long-range dispersal and high-latitude environments influence the population structure of a “stress-tolerant” endosymbiont harbored by Pocillopora in the Eastern Pacific. Molecular Ecology.

Porter JW, Fitt WK, Spero HJ, Rogers CS, White MW (1989) Bleaching in reef corals: physiological and stable isotopic responses. Proceedings of the National Academy of Sciences 86, 9342-9346.

Pritchard J, Stephens M, Donnelly P (2000) Inference of population structure using multilocus genotype data. Genetics 155, 945.

Provan J, Murphy S, Maggs CA (2004) Tracking the invasive history of the green alga Codium fragile ssp. tomentosoides. Molecular Ecology 14, 189-194.

Rosenberg N (2004) DISTRUCT: a program for the graphical display of population structure. Molecular Ecology Notes 4, 137-138.

Rousset F (2008) GENEPOP'007: a complete re-implementation of the GENEPOP software for Windows and Linux. Molecular Ecology Resources 8, 103-106.

Rowan R (2004) Coral bleaching: thermal adaptation in reef coral symbionts. Nature 430, 742.

Rowan R, Powers D (1991) A molecular genetic classification of zooxanthellae and the evolution of animal-algal symbioses. Science 251, 1348-1351.

Rynearson T, Newton J, Armbrust E (2006) Spring bloom development, genetic variation, and population succession in the planktonic diatom Ditylum brightwellii. Limnology and Oceanography 51, 1249-1261.

Sampayo EM, Dove S, LaJeunesse TC (2009) Cohesive molecular genetic data delineate species diversity in the dinoflagellate genus Symbiodinium. Molecular Ecology 18, 500-519.

Santos S, Coffroth M (2003) Molecular genetic evidence that dinoflagellates belonging to the genus Symbiodinium Freudenthal are haploid. The Biological Bulletin 204, 10-20.

Santos S, Gutierrez-Rodriguez C, Lasker H, Coffroth M (2003) Symbiodinium sp. associations in the gorgonian Pseudopterogorgia elisabethae in the Bahamas: high levels of genetic variability and population structure in symbiotic dinoflagellates. Marine Biology 143, 111-120.

156 Seutin G, White BN, Boag PT (1991) Preservation of avian blood and tissue samples for DNA analyses. Canadian Journal of Zoology 69, 82–92.

Smith R (2008) Specificity, stability and comparative physiology of the coral-Symbiodinium mutualisms: evaluating the potential for acclimation and/or adaptation in reef corals, Florida International University.

Stockwell CA, Hendry AP, Kinnison MT (2003) Contemporary evolution meets conservation biology. Trends in Ecology & Evolution 18, 94-101.

Thornhill D, Xiang Y, Fitt W, Santos S (2009) Reef endemism, host specificity and temporal stability in populations of symbiotic dinoflagellates from two ecologically dominant Caribbean corals. PLoS One 4, e6262.

Thornhill DJ, LaJeunesse TC, Kemp DW, Fitt WK, Schmidt GW (2006) Multi-year, seasonal genotypic surveys of coral-algal symbioses reveal prevalent stability or post- bleaching reversion. Marine Biology 148, 711-722.

Thornhill DJ, LaJeunesse TC, Santos SR (2007) Measuring rDNA diversity in eukaryotic microbial systems: how intragenomic variation, pseudogenes, and PCR artifacts confound biodiversity estimates. Molecular Ecology 16, 5326-5340.

Toller W, Rowan R, Knowlton N (2001) Repopulation of zooxanthellae in the Caribbean corals Montastraea annularis and M. faveolata following experimental and disease- associated bleaching. The Biological Bulletin 201, 360.

Trench R (1993) Microalgal-invertebrate symbioses - a review. Endocytoboisis and Cell Research 9, 135-175.

Ulstrup KE, van Oppen MJH (2003) Geographic and habitat partitioning of genetically distinct zooxanthellae (Symbiodinium) in Acropora corals on the Great Barrier Reef. Molecular Ecology 12, 3477-3484.

Vallejo-Marín M, Dorken ME, Barrett SCH (2010) The Ecological and Evolutionary Consequences of Clonality for Plant Mating. Annual Review of Ecology, Evolution, and Systematics 41, 193-213.

Waits L, Luikart G, Taberlet P (2001) Estimating the probability of identity among genotypes in natural populations: cautions and guidelines. Molecular Ecology 10, 249-256.

Wham DC, Pettay DT, LaJeunesse TC (2011) Microsatellite loci for the host-generalist “zooxanthella” Symbiodinium trenchi and other Clade D Symbiodinium. Conservation Genetic Resources, 1-4.

157 Wilkinson C, Souter D (2008) Status of Caribbean Coral Reefs after Bleaching and Hurrincanes in 2005 Global Coral Reef Monitoring Network, and Reef and Rainforest Research Centre, Townsville.

Williams EH, Bunkley-Williams L (1990) The world-wide coral reef bleaching cycle and related sources of coral mortality. Atoll Res. Bull 335, 1-71.

Williams R, Griffiths F, Van der Wal E, Kelly J (1988) Cargo vessel ballast water as a vector for the transort of non- indigenous marine species. Estuarine, Coastal and Shelf Science 26, 409-420.

158 APPENDICES

Appendix A. Allele frequencies for each reef according to sample date. All frequencies were calculated with repeated MLGs removed within each time point and reef. Locus Allele ISLG_06 ISLG_09 Locus Allele PGAL_06 PGAL_09 Sym11 N 21 22 Sym11 N 36 31 151 0.048 0.000 149 0.000 0.032 153 0.762 0.682 151 0.028 0.129 155 0.190 0.136 153 0.639 0.419 157 0.000 0.091 155 0.028 0.065 159 0.000 0.091 157 0.250 0.226 Sym14 N 21 22 159 0.028 0.129 179 0.571 0.591 161 0.028 0.000 181 0.238 0.000 Sym14 N 36 31 183 0.190 0.364 173 0.028 0.032 185 0.000 0.045 179 0.333 0.258 Sym17 N 20 22 181 0.250 0.194 145 0.100 0.318 183 0.361 0.484 147 0.450 0.227 185 0.028 0.032 149 0.450 0.455 Sym17 N 35 31 Sym34 N 21 22 143 0.029 0.000 388 0.143 0.364 145 0.543 0.613 392 0.048 0.136 147 0.257 0.258 396 0.524 0.409 149 0.171 0.129 400 0.190 0.091 Sym34 N 36 31 404 0.095 0.000 380 0.028 0.032 Sym67 N 21 22 384 0.056 0.000 134 0.476 0.409 386 0.028 0.000 137 0.190 0.364 388 0.056 0.065 140 0.333 0.227 392 0.306 0.323 Sym92 N 21 22 396 0.194 0.290 124 0.143 0.273 400 0.250 0.290 128 0.857 0.727 404 0.083 0.000 Sym87 N 21 22 Sym67 N 36 31 244 0.429 0.136 134 0.778 0.903 248 0.190 0.182 137 0.028 0.000 252 0.333 0.318 140 0.194 0.097 256 0.048 0.045 Sym92 N 36 31 260 0.000 0.227 128 1.000 1.000 264 0.000 0.091 Sym87 N 36 31 Sym77 N 21 22 244 0.333 0.355 100 0.333 0.136 248 0.139 0.129 178 0.143 0.318 252 0.222 0.097

159 181 0.524 0.545 256 0.028 0.032 Sym77b N 21 22 260 0.222 0.323 100 0.714 0.545 264 0.056 0.065 181 0.000 0.045 Sym77 N 36 31 190 0.048 0.318 100 0.111 0.065 193 0.238 0.091 178 0.000 0.032 181 0.861 0.903 184 0.028 0.000 Sym77b N 36 31 100 0.389 0.194 190 0.417 0.581 193 0.194 0.194 196 0.000 0.032

160 Appendix B. Summary plots of S. glynni genotypic diversity within individual Pocillopora colonies based on sampling strategy. (a) Percent of colonies possessing multiple genotypes of S. glynni and (b) estimate number of genotypes per colony based on the number times a colony was sampled. *Five branches of each colony were sampled, but each branch was sampled at both the top and bottom effectively making these colonies sampled ten times.

161 Appendix C. Haploid allele frequencies and sample size by location for S. glynni in the Eastern Pacific.

Locus Allele GoC BB OAX CLP PAN GAL D1Sym9 N 73 98 48 8 8 18 106 0.000 0.020 0.000 0.000 0.000 0.000 109 1.000 0.847 0.792 1.000 1.000 1.000 112 0.000 0.102 0.208 0.000 0.000 0.000 115 0.000 0.031 0.000 0.000 0.000 0.000 D1Sym11 N 73 92 45 8 8 16 151 0.219 0.065 0.000 0.375 0.375 0.125 153 0.438 0.739 0.644 0.375 0.625 0.500 155 0.055 0.141 0.133 0.000 0.000 0.188 157 0.233 0.043 0.111 0.125 0.000 0.188 159 0.041 0.011 0.089 0.125 0.000 0.000 161 0.014 0.000 0.022 0.000 0.000 0.000 D1Sym14 N 73 98 48 8 8 16 173 0.014 0.000 0.000 0.000 0.000 0.000 175 0.000 0.010 0.021 0.000 0.000 0.000 177 0.000 0.949 0.813 0.875 0.375 1.000 179 0.301 0.041 0.167 0.000 0.625 0.000 181 0.274 0.000 0.000 0.125 0.000 0.000 183 0.397 0.000 0.000 0.000 0.000 0.000 185 0.014 0.000 0.000 0.000 0.000 0.000 D1Sym17 N 73 93 48 8 8 18 143 0.014 0.000 0.000 0.000 0.000 0.000 145 0.479 0.011 0.000 0.000 0.000 0.000 147 0.247 0.011 0.063 0.000 0.000 0.000 149 0.260 0.140 0.083 0.000 0.125 0.167 151 0.000 0.086 0.333 0.000 0.000 0.000 153 0.000 0.194 0.333 0.250 0.375 0.389 155 0.000 0.172 0.125 0.375 0.375 0.056 157 0.000 0.108 0.021 0.125 0.125 0.000 159 0.000 0.054 0.000 0.000 0.000 0.111 161 0.000 0.097 0.021 0.125 0.000 0.167 163 0.000 0.086 0.000 0.125 0.000 0.056 165 0.000 0.022 0.021 0.000 0.000 0.000 167 0.000 0.022 0.000 0.000 0.000 0.056 D1Sym34 N 73 98 47 8 8 18 332 0.000 0.010 0.000 0.000 0.000 0.000 336 0.000 0.010 0.000 0.000 0.000 0.000 340 0.000 0.051 0.000 0.000 0.000 0.000 344 0.000 0.173 0.021 0.000 0.000 0.000

162 348 0.000 0.020 0.000 0.000 0.000 0.000 356 0.000 0.010 0.000 0.000 0.000 0.000 370 0.000 0.031 0.000 0.000 0.000 0.000 372 0.014 0.000 0.000 0.000 0.000 0.000 374 0.000 0.010 0.000 0.000 0.000 0.056 376 0.027 0.000 0.000 0.000 0.000 0.000 378 0.014 0.010 0.000 0.000 0.000 0.000 380 0.055 0.020 0.000 0.000 0.000 0.000 384 0.082 0.000 0.043 0.000 0.000 0.000 388 0.055 0.010 0.021 0.000 0.000 0.000 390 0.000 0.041 0.021 0.000 0.000 0.000 392 0.137 0.000 0.000 0.000 0.000 0.000 394 0.000 0.061 0.000 0.000 0.000 0.000 396 0.301 0.082 0.043 0.000 0.500 0.167 400 0.247 0.020 0.170 0.000 0.125 0.056 402 0.000 0.000 0.000 0.000 0.000 0.167 404 0.068 0.122 0.191 0.375 0.000 0.056 408 0.000 0.153 0.170 0.250 0.375 0.222 410 0.000 0.010 0.000 0.000 0.000 0.000 412 0.000 0.051 0.213 0.125 0.000 0.167 414 0.000 0.010 0.000 0.000 0.000 0.000 416 0.000 0.010 0.085 0.125 0.000 0.056 418 0.000 0.000 0.000 0.125 0.000 0.000 420 0.000 0.041 0.021 0.000 0.000 0.056 424 0.000 0.031 0.000 0.000 0.000 0.000 428 0.000 0.010 0.000 0.000 0.000 0.000 D1Sym67 N 73 98 48 8 8 18 131 0.014 0.010 0.000 0.250 0.000 0.000 134 0.740 0.000 0.021 0.000 0.000 0.000 137 0.082 0.082 0.000 0.000 0.000 0.000 140 0.164 0.327 0.354 0.500 1.000 0.278 143 0.000 0.449 0.542 0.250 0.000 0.333 146 0.000 0.112 0.063 0.000 0.000 0.167 149 0.000 0.020 0.021 0.000 0.000 0.222 D1Sym77 N 62 98 48 7 8 16 169 0.000 0.041 0.188 0.000 0.000 0.063 172 0.000 0.010 0.000 0.143 0.000 0.000 175 0.000 0.082 0.083 0.000 0.000 0.313 178 0.032 0.765 0.667 0.857 0.625 0.375 181 0.952 0.092 0.063 0.000 0.250 0.250 184 0.016 0.010 0.000 0.000 0.125 0.000 D1Sym77b N 62 98 48 7 8 16 178 0.032 0.061 0.021 0.000 0.000 0.000

163 181 0.290 0.000 0.000 0.000 0.000 0.000 184 0.000 0.020 0.000 0.000 0.000 0.000 187 0.016 0.857 0.729 1.000 1.000 0.938 190 0.403 0.061 0.250 0.000 0.000 0.063 193 0.258 0.000 0.000 0.000 0.000 0.000 D1Sym87 N 73 98 45 8 8 18 244 0.356 0.010 0.133 0.000 0.000 0.000 248 0.137 0.184 0.156 0.000 0.000 0.278 252 0.164 0.531 0.533 0.625 0.375 0.333 256 0.041 0.204 0.156 0.375 0.625 0.056 260 0.096 0.020 0.000 0.000 0.000 0.278 264 0.178 0.051 0.000 0.000 0.000 0.000 268 0.027 0.000 0.022 0.000 0.000 0.056 D1Sym88 N 73 98 48 8 8 18 227 0.055 0.000 0.000 0.000 0.000 0.000 231 0.000 1.000 1.000 1.000 1.000 1.000 235 0.945 0.000 0.000 0.000 0.000 0.000 D1Sym92 N 73 98 48 8 8 18 124 0.027 0.000 0.000 0.000 0.000 0.056 128 0.973 0.969 0.979 1.000 1.000 0.944 132 0.000 0.031 0.021 0.000 0.000 0.000

164 Appendix D. Plots derived from the method by Evanno et al. 2005 using the second order rate of change of the Ln P(D) to determine the appropriate value of K based on runs in Structure. a) Plot of the average Ln P(D) (±SD) over five runs for each value of K. b) Result plot of the second order rate of change showing a clear peak at K = 2, representing the genetic break between the GoC and the ETP. c) Raw data from the ten consecutive Structure runs with five replicates at each K and used for the above analyses.

a) Ln P(D) = L(K)

b) ∆K

165 K lnP(D) = L(K) 1 -3019.1 1 -3019.5 1 -3019.4 1 -3019.5 1 -3019.2 2 -2351.6 2 -2351.4 2 -2351.5 2 -2351.4 2 -2351.3 3 -2230.7 3 -2231.4 3 -2230.9 3 -2230.4 3 -2230.5 4 -2201.2 4 -2200.8 4 -2201.4 4 -2092.1 4 -2201.3 5 -2175.5 5 -2169.5 5 -2174.5 5 -2176.8 5 -2171.4 6 -2305 6 -2246 6 -2217.6 6 -2286.1 6 -2244.7 7 -2233.5 7 -2222.8 7 -2052.8 7 -2395.2 7 -2090.1 8 -2071.3 8 -2066 8 -2149.1 8 -3888.8 8 -2081.1 9 -2200.1 9 -3642.8 9 -2087.5 9 -2276.1 9 -2074.2 10 -2163.5 10 -2417.5 10 -2124.7 10 -2095.8 10 -2138.5

166 Appendix E. Principal coordinate analysis of S. glynni populations from the ETP. Colors correspond to the two potential cryptic populations in this region as revealed by Structure, while shapes represent a sample’s location of origin. While two clearly defined clusters fail to be resolved, the two cryptic populations do show distributions shifted to upper left or lower right.

167 Appendix F. Haploid allele frequencies and sample size by location for S. glynni across the Pacific. Locus Allele PAL GoT THA LaP BB OAX CLIP PA GAL Sym9 N 11 11 7 64 87 40 6 7 11 106 0.364 0.000 1.000 0.000 0.023 0.000 0.000 0.000 0.000 109 0.091 1.000 0.000 1.000 0.839 0.825 1.000 1.000 1.000 112 0.455 0.000 0.000 0.000 0.103 0.175 0.000 0.000 0.000 115 0.091 0.000 0.000 0.000 0.034 0.000 0.000 0.000 0.000 Sym11 N 11 10 7 64 82 38 6 7 10 151 0.000 0.000 0.000 0.031 0.061 0.000 0.333 0.429 0.100 153 0.000 0.100 0.000 0.563 0.768 0.711 0.333 0.571 0.500 155 0.091 0.900 0.429 0.078 0.122 0.105 0.000 0.000 0.200 157 0.091 0.000 0.429 0.281 0.037 0.105 0.167 0.000 0.200 159 0.545 0.000 0.143 0.031 0.012 0.053 0.167 0.000 0.000 161 0.091 0.000 0.000 0.016 0.000 0.026 0.000 0.000 0.000 163 0.091 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 167 0.091 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 Sym14 N 11 11 7 64 87 40 6 7 10 173 0.000 0.000 0.000 0.016 0.000 0.000 0.000 0.000 0.000 175 0.000 0.000 1.000 0.000 0.000 0.025 0.000 0.000 0.000 177 0.455 1.000 0.000 0.000 0.954 0.850 1.000 0.429 1.000 179 0.273 0.000 0.000 0.359 0.046 0.125 0.000 0.571 0.000 181 0.091 0.000 0.000 0.219 0.000 0.000 0.000 0.000 0.000 183 0.091 0.000 0.000 0.391 0.000 0.000 0.000 0.000 0.000 185 0.000 0.000 0.000 0.016 0.000 0.000 0.000 0.000 0.000 189 0.091 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 Sym17 N 11 11 7 64 82 40 6 7 11 141 0.000 0.636 0.000 0.000 0.000 0.000 0.000 0.000 0.000 143 0.455 0.364 0.571 0.016 0.000 0.000 0.000 0.000 0.000 145 0.545 0.000 0.429 0.469 0.000 0.000 0.000 0.000 0.000 147 0.000 0.000 0.000 0.266 0.000 0.025 0.000 0.000 0.000 149 0.000 0.000 0.000 0.250 0.146 0.100 0.000 0.143 0.091 151 0.000 0.000 0.000 0.000 0.098 0.325 0.000 0.000 0.000 153 0.000 0.000 0.000 0.000 0.195 0.350 0.333 0.429 0.273 155 0.000 0.000 0.000 0.000 0.183 0.125 0.333 0.429 0.091 157 0.000 0.000 0.000 0.000 0.122 0.025 0.167 0.000 0.000 159 0.000 0.000 0.000 0.000 0.061 0.000 0.000 0.000 0.182 161 0.000 0.000 0.000 0.000 0.085 0.025 0.000 0.000 0.182 163 0.000 0.000 0.000 0.000 0.061 0.000 0.167 0.000 0.091 165 0.000 0.000 0.000 0.000 0.024 0.025 0.000 0.000 0.000 167 0.000 0.000 0.000 0.000 0.024 0.000 0.000 0.000 0.091 Sym34 N 11 11 7 64 87 39 6 7 11 332 0.000 0.000 0.000 0.000 0.011 0.000 0.000 0.000 0.000

168 340 0.000 0.000 0.000 0.000 0.046 0.000 0.000 0.000 0.000 344 0.000 0.000 0.000 0.000 0.161 0.026 0.000 0.000 0.000 348 0.000 0.000 0.000 0.000 0.023 0.000 0.000 0.000 0.000 356 0.000 0.000 0.000 0.000 0.011 0.000 0.000 0.000 0.000 366 0.000 0.091 0.000 0.000 0.000 0.000 0.000 0.000 0.000 368 0.000 0.091 0.571 0.000 0.000 0.000 0.000 0.000 0.000 370 0.091 0.091 0.000 0.000 0.034 0.000 0.000 0.000 0.000 372 0.000 0.000 0.429 0.000 0.000 0.000 0.000 0.000 0.000 374 0.000 0.091 0.000 0.000 0.000 0.000 0.000 0.000 0.091 376 0.091 0.000 0.000 0.031 0.000 0.000 0.000 0.000 0.000 378 0.091 0.091 0.000 0.016 0.011 0.000 0.000 0.000 0.000 380 0.000 0.000 0.000 0.063 0.011 0.000 0.000 0.000 0.000 382 0.000 0.091 0.000 0.000 0.000 0.000 0.000 0.000 0.000 384 0.091 0.000 0.000 0.094 0.000 0.026 0.000 0.000 0.000 386 0.000 0.182 0.000 0.016 0.000 0.000 0.000 0.000 0.000 388 0.000 0.000 0.000 0.047 0.011 0.026 0.000 0.000 0.000 390 0.091 0.273 0.000 0.000 0.034 0.026 0.000 0.000 0.000 392 0.364 0.000 0.000 0.188 0.000 0.000 0.000 0.000 0.000 394 0.000 0.000 0.000 0.000 0.080 0.000 0.000 0.000 0.000 396 0.091 0.000 0.000 0.281 0.057 0.051 0.000 0.429 0.182 400 0.000 0.000 0.000 0.203 0.023 0.179 0.000 0.143 0.091 402 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.182 404 0.000 0.000 0.000 0.063 0.138 0.179 0.333 0.000 0.000 408 0.000 0.000 0.000 0.000 0.149 0.231 0.167 0.429 0.182 410 0.000 0.000 0.000 0.000 0.011 0.000 0.000 0.000 0.000 412 0.091 0.000 0.000 0.000 0.069 0.154 0.167 0.000 0.091 414 0.000 0.000 0.000 0.000 0.011 0.000 0.000 0.000 0.000 416 0.000 0.000 0.000 0.000 0.023 0.077 0.167 0.000 0.091 418 0.000 0.000 0.000 0.000 0.000 0.000 0.167 0.000 0.000 420 0.000 0.000 0.000 0.000 0.046 0.026 0.000 0.000 0.091 424 0.000 0.000 0.000 0.000 0.023 0.000 0.000 0.000 0.000 428 0.000 0.000 0.000 0.000 0.011 0.000 0.000 0.000 0.000 Sym67 N 11 11 7 64 87 40 6 7 11 131 0.000 0.364 0.000 0.016 0.011 0.000 0.333 0.000 0.000 134 0.182 0.364 0.000 0.703 0.000 0.025 0.000 0.000 0.000 137 0.455 0.000 0.286 0.078 0.092 0.000 0.000 0.000 0.000 140 0.000 0.273 0.714 0.203 0.322 0.400 0.500 1.000 0.273 143 0.273 0.000 0.000 0.000 0.471 0.500 0.167 0.000 0.364 146 0.000 0.000 0.000 0.000 0.080 0.075 0.000 0.000 0.182 149 0.091 0.000 0.000 0.000 0.023 0.000 0.000 0.000 0.182 Sym92 N 11 11 7 64 87 40 6 7 11 120 0.000 0.636 1.000 0.000 0.000 0.000 0.000 0.000 0.000 124 0.000 0.000 0.000 0.031 0.000 0.000 0.000 0.000 0.091

169 128 1.000 0.364 0.000 0.969 0.977 0.950 1.000 1.000 0.909 132 0.000 0.000 0.000 0.000 0.023 0.050 0.000 0.000 0.000 Sym87 N 11 11 7 64 87 38 6 7 11 236 0.273 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 240 0.182 1.000 1.000 0.000 0.000 0.000 0.000 0.000 0.000 244 0.091 0.000 0.000 0.297 0.011 0.105 0.000 0.000 0.000 248 0.000 0.000 0.000 0.125 0.184 0.158 0.000 0.000 0.273 252 0.182 0.000 0.000 0.234 0.563 0.553 0.667 0.429 0.364 256 0.182 0.000 0.000 0.031 0.195 0.158 0.333 0.571 0.000 260 0.091 0.000 0.000 0.125 0.023 0.000 0.000 0.000 0.273 264 0.000 0.000 0.000 0.156 0.023 0.000 0.000 0.000 0.000 268 0.000 0.000 0.000 0.031 0.000 0.026 0.000 0.000 0.091 Sym77a N 11 11 7 55 87 40 5 7 10 169 0.000 0.000 0.000 0.000 0.011 0.150 0.000 0.000 0.100 172 0.000 0.000 0.000 0.000 0.011 0.000 0.200 0.000 0.000 175 0.000 0.000 0.000 0.000 0.092 0.100 0.000 0.000 0.300 178 0.091 0.000 1.000 0.055 0.782 0.675 0.800 0.571 0.300 181 0.455 0.000 0.000 0.927 0.092 0.075 0.000 0.286 0.300 184 0.000 0.364 0.000 0.018 0.011 0.000 0.000 0.143 0.000 187 0.455 0.636 0.000 0.000 0.000 0.000 0.000 0.000 0.000 Sym77 b N 3 11 7 38 83 39 5 7 10 184 0.000 0.000 0.000 0.000 0.024 0.000 0.000 0.000 0.000 187 0.000 0.000 0.000 0.026 0.904 0.744 1.000 1.000 0.900 190 0.667 0.000 0.000 0.658 0.072 0.256 0.000 0.000 0.100 193 0.000 1.000 1.000 0.316 0.000 0.000 0.000 0.000 0.000 196 0.333 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 Sym88 N 11 11 7 64 87 40 6 7 11 227 0.000 0.000 0.000 0.047 0.000 0.000 0.000 0.000 0.000 231 0.091 0.636 1.000 0.000 1.000 1.000 1.000 1.000 1.000 235 0.909 0.364 0.000 0.953 0.000 0.000 0.000 0.000 0.000

170

Appendix G. Allele frequencies and sample size by location for S. trenchi Indian, Pacific and Atlantic Oceans. Locus Allele/n CAR PHU PAL ZAN Sym9 N 25 53 30 11 106 0.520 0.585 0.417 0.591 109 0.480 0.330 0.567 0.409 112 0.000 0.085 0.017 0.000 Sym11 N 25 51 29 10 149 0.000 0.049 0.052 0.000 151 0.500 0.343 0.345 0.500 153 0.440 0.216 0.207 0.200 155 0.060 0.127 0.207 0.250 157 0.000 0.118 0.155 0.050 159 0.000 0.118 0.017 0.000 161 0.000 0.029 0.017 0.000 Sym14 N 25 52 29 11 167 0.000 0.000 0.000 0.045 173 0.000 0.019 0.000 0.045 175 0.000 0.038 0.000 0.000 177 0.500 0.462 0.448 0.364 179 0.500 0.462 0.414 0.500 181 0.000 0.010 0.138 0.045 191 0.000 0.010 0.000 0.000 Sym17 N 24 48 29 10 137 0.000 0.021 0.000 0.050 139 0.000 0.010 0.000 0.100 141 0.479 0.458 0.517 0.350 143 0.479 0.198 0.276 0.300 145 0.021 0.271 0.190 0.200 147 0.021 0.042 0.017 0.000 Sym34 N 25 52 29 10 350 0.000 0.010 0.000 0.000 353 0.000 0.010 0.017 0.000 361 0.000 0.019 0.000 0.000 363 0.000 0.010 0.000 0.000 365 0.000 0.029 0.034 0.000 367 0.000 0.058 0.017 0.000 369 0.000 0.038 0.017 0.000 370 0.000 0.000 0.000 0.050 371 0.000 0.077 0.000 0.050 372 0.000 0.000 0.000 0.100 373 0.000 0.115 0.138 0.000 375 0.000 0.048 0.000 0.000 377 0.020 0.048 0.172 0.100 378 0.000 0.029 0.000 0.000 379 0.040 0.019 0.017 0.000 381 0.040 0.087 0.069 0.150 383 0.140 0.048 0.069 0.050 385 0.360 0.048 0.121 0.200 387 0.360 0.038 0.052 0.200 389 0.040 0.058 0.052 0.050 391 0.000 0.019 0.052 0.000

171 393 0.000 0.058 0.017 0.050 395 0.000 0.029 0.017 0.000 397 0.000 0.058 0.069 0.000 399 0.000 0.010 0.017 0.000 401 0.000 0.019 0.000 0.000 403 0.000 0.010 0.017 0.000 405 0.000 0.010 0.000 0.000 409 0.000 0.000 0.017 0.000 429 0.000 0.000 0.017 0.000 Sym66 N 24 49 29 11 288 0.000 0.031 0.017 0.091 291 0.000 0.224 0.172 0.136 294 1.000 0.633 0.603 0.500 297 0.000 0.112 0.190 0.273 300 0.000 0.000 0.017 0.000 Sym93 N 25 51 29 11 141 0.000 0.000 0.017 0.000 145 0.000 0.098 0.017 0.000 149 0.400 0.402 0.500 0.591 151 0.200 0.010 0.000 0.000 153 0.000 0.127 0.138 0.091 157 0.400 0.343 0.328 0.318 161 0.000 0.010 0.000 0.000 173 0.000 0.010 0.000 0.000 Sym92 N 25 53 29 11 124 0.000 0.255 0.121 0.045 128 1.000 0.717 0.845 0.955 132 0.000 0.019 0.034 0.000 136 0.000 0.009 0.000 0.000 Sym67 N 24 52 30 10 125 0.000 0.000 0.033 0.000 128 0.000 0.106 0.000 0.000 131 0.979 0.423 0.483 0.550 134 0.000 0.221 0.367 0.300 137 0.021 0.173 0.050 0.100 140 0.000 0.077 0.067 0.050 Sym82 N 24 47 28 11 225 0.000 0.011 0.000 0.000 228 0.000 0.011 0.000 0.000 231 0.000 0.043 0.018 0.000 234 0.458 0.074 0.071 0.000 237 0.104 0.085 0.071 0.000 240 0.042 0.170 0.089 0.045 243 0.292 0.043 0.107 0.045 246 0.042 0.032 0.071 0.182 249 0.021 0.064 0.018 0.364 252 0.000 0.085 0.089 0.136 255 0.000 0.106 0.143 0.091 258 0.000 0.074 0.036 0.000 261 0.021 0.032 0.036 0.045 264 0.000 0.043 0.089 0.000 267 0.000 0.021 0.036 0.000 270 0.021 0.032 0.054 0.000 273 0.000 0.000 0.018 0.045

172 276 0.000 0.000 0.018 0.000 279 0.000 0.011 0.018 0.045 282 0.000 0.032 0.000 0.000 285 0.000 0.032 0.018 0.000 Sym87 N 24 51 29 10 236 0.000 0.029 0.000 0.000 240 0.500 0.471 0.379 0.450 244 0.479 0.422 0.483 0.400 248 0.021 0.059 0.138 0.150 252 0.000 0.010 0.000 0.000 256 0.000 0.010 0.000 0.000 Sym88 N 25 51 29 11 232 0.000 0.049 0.000 0.227 234 0.000 0.343 0.121 0.045 240 0.980 0.588 0.828 0.682 244 0.020 0.020 0.017 0.045 250 0.000 0.000 0.034 0.000

173 VITA Daniel T. Pettay [email protected] EDUCATION The Pennsylvania State University, State College, PA Ph.D. in Biology 2011 Dissertation: “Diversity, Connectivity and Stability of Coral/Algal Symbioses over Various Spatial Scales” Dissertation Advisor: Dr. Todd C. LaJeunesse College of Charleston, Charleston, SC M.S. in Marine Biology 2006 Thesis: “Effects of the Antifouling Algaecide, Irgarol 1051, on Cultured Zooxanthellae (Genus Symbiodinium)” Thesis Advisor: Dr. Cheryl M. Woodley Clemson University, Clemson, SC B.S. in Biological Sciences 2000

TEACHING EXPERIENCE The Pennsylvania State University, State College, PA Laboratory Instructor – Invertebrate Zoology 2009

Florida International University, Miami, FL Laboratory Instructor – Invertebrate Zoology 2006 - 2007

Laboratory Instructor – General Biology 2005 - 2007

College of Charleston, Charleston, SC Laboratory Instructor – General Biology 2001 - 2002

RELATED EXPERIENCE The Pennsylvania State University, State College, PA Research Assistant/Student – Dinoflagellate Symbiont Lab 2008 - Present

Florida International University, Miami, FL Research Assistant/Student – Dinoflagellate Symbiont Lab 2005 - 2007

Research Assistant – Forensic DNA Profiling Facility

Hollings Marine Lab, NOAA/NOS, Charleston, SC 2005 Research Assistant/Student – Coral Health and Disease Lab

Research Assistant/Student – Coral Health and Disease Lab 2002 - 2004

AWARDS & FUNDING

• C.V. Starr Scholarship Endowment – Bermuda Biological Station for Research, Marine Ecotoxicology Summer Course (2003). • Braddock Award ($2,000) – The Pennsylvania State University, Department of Biology (2008). • Braddock Research Award ($1,500) – The Pennsylvania State University, Department of Biology (2010). • PADI Foundation Grant ($2,075) – “Dispersal of thermally tolerant Clade D Symbiodinium in the Caribbean” (2010).