Thermal Tolerance of Coral Photosymbionts: Genetic Factors and Strategies to Pursue Genetic Enhancement

Thermal tolerance of coral photosymbionts: genetic factors and strategies to pursue genetic enhancement

Rachel A. Levin

A thesis in fulfilment of the requirements for the degree of Doctor of Philosophy

School of Biological, Earth, and Environmental Sciences Faculty of Science

July 2017

THE UNIVERSITY OF NEW SOUTH WALES Thesis/Dissertation Sheet

Surname or Family name: Levin

First name: Rachel Other name/s: Ashley

Abbreviation for degree as given in the University calendar: PhD

School: Biological, Earth, and Environmental Sciences Faculty: Science

Title: Thermal tolerance of coral photosymbionts: genetic factors and strategies to pursue genetic enhancement

Abstract 350 words maximum: Dinoflagellates of the genus Symbiodinium form essential symbioses with reef building corals, underpinning the entire ecological foundation of coral reefs. Corals rely on photosynthate produced by Symbiodinium for their growth and calcification, which in turn forms the reef framework. Increased sea surface temperature due to climate change triggers the loss of Symbiodinium from corals (coral bleaching), which can result in coral death. Different genetic variants of Symbiodinium exhibit diverse thermal tolerances that influence the thermal bleaching thresholds of their coral host. However, despite decades of research into Symbiodinium biology, determinants of Symbiodinium thermal tolerance are still largely unresolved. Therefore, I aimed to unlock the basis of Symbiodinium thermal tolerance using a comparative physiology-genomics approach, and subsequently using this new knowledge, aimed to develop novel strategies that promote genetic manipulation of Symbiodinium. In this thesis, I discovered that thermal tolerance of type C1 Symbiodinium is driven by up-regulation of genes and functional gene groups responsible for sexual reproduction, scavenging of reactive oxygen species, and protein folding that maintain photosynthetic ability and limit reactive oxygen species production under heat stress. I also uncovered the first entire genome of a Symbiodinium virus along with hundreds of transcripts from viruses that infect Symbiodinium, whose transcriptional regulation under heat stress may contribute to Symbiodinium thermal sensitivity. Next, I successfully removed cell walls from live Symbiodinium to create the first Symbiodinium protoplasts and achieved Symbiodinium protoplast fusion, a key step in creating hybrid Symbiodinium cells with novel genetic combinations for ideal traits. Finally, using these discoveries, I developed a theoretical framework for Symbiodinium genetic engineering that incorporates Symbiodinium genetic elements, viral genetic elements, and Symbiodinium protoplasts, along with original genomic analyses of the potential for CRISPR/Cas9 gene editing in Symbiodinium. Together, the studies presented in this thesis unveil factors that govern, as well as strategies that may genetically enhance, Symbiodinium thermal tolerance.

Declaration relating to disposition of project thesis/dissertation

I hereby grant to the University of New South Wales or its agents the right to archive and to make available my thesis or dissertation in whole or in part in the University libraries in all forms of media, now or here after known, subject to the provisions of the Copyright Act 1968. I retain all property rights, such as patent rights. I also retain the right to use in future works (such as articles or books) all or part of this thesis or dissertation.

I also authorise University Microfilms to use the 350 word abstract of my thesis in Dissertation Abstracts International (this is applicable to doctoral theses only).

10/07/2017 ……………………………………………………………………… ……………………………………………………………………… ……….………….. Signature Witness Signature Date The University recognises that there may be exceptional circumstances requiring restrictions on copying or conditions on use. Requests for restriction for a period of up to 2 years must be made in writing. Requests for a longer period of restriction may be considered in exceptional circumstances and require the approval of the Dean of Graduate Research.

FOR OFFICE USE ONLY Date of completion of requirements for Award:

i COPYRIGHT STATEMENT

'I hereby grant the University of New South Wales or its agents the right to archive and to make available my thesis or dissertation in whole or part in the University libraries in all forms of media, now or here after known, subject to the provisions of the Copyright Act 1968. I retain all proprietary rights, such as patent rights. I also retain the right to use in future works (such as articles or books) all or part of this thesis or dissertation. I also authorise University Microfilms to use the 350 word abstract of my thesis in Dissertation Abstract International (this is applicableto doctoral theses only). I have either used no substantial portions of copyright material in my thesis or I have obtained permission to use copyright material; where permission has not been granted I have applied/will apply fora partial restriction of the digital copy of my thesis or dissertation.'

Signed

Date ......

AUTHENTICITY STATEMENT

'I certify that the Library deposit digital copy is a direct equivalent of the final officially approved version of my thesis. No emendation of content has occurred and if there are any minor variations in formatting, they are the result of the conversion to dig�al format.'

/ Date ...... /P J/t.T...... Originality statement

I hereby declare that this submission is my own work, and to the best of my knowledge, it contains no materials previously published or written by another person or substantial proportions of material which have been accepted for the award of any other degree or diploma at UNSW or any other educational institution, except where due acknowledgement is made in the thesis. Any contribution made to the research by others, with whom I have worked at UNSW or elsewhere, is explicitly acknowledged in the thesis. I also declare that the intellectual content of this thesis is the product of my own work, except to the extent that assistance from others in the project’s design and conception or in style, presentation and linguistic expression is acknowledged.

Signed

Date 10/07/2017

ii Publications during candidature

Journal articles

Levin RA, Beltran VH, Hill R, Kjelleberg S, McDougald D, Steinberg PD, van Oppen MJH (2016). Sex, scavengers, and chaperones: transcriptome secrets of divergent Symbiodinium thermal tolerances. Molecular Biology and Evolution 33: 2201-2215.

Levin RA, Voolstra CR, Weynberg KD, van Oppen MJH (2017a). Evidence for a role of viruses in the thermal sensitivity of coral photosymbionts. The ISME Journal 11: 808-812.

Levin RA, Suggett DJ, Nitschke MR, van Oppen MJH, Steinberg PD (2017b). Expanding the Symbiodinium (Dinophyceae, Suessiales) toolkit through protoplast technology. Journal of Eukaryotic Microbiology doi: 10.1111/jeu.12393.

Levin RA, Suggett DJ, Voolstra, CR, Agrawal S, Steinberg PD, Suggett DJ, van Oppen MJH (2017c). Engineering strategies to decode and enhance the genomes of coral symbionts biology. Frontiers in Microbiology 8: 1220.

Conference abstracts

Levin RA, Hill R, van Oppen MJH, Steinberg PD, McDougald D (2014). From Symbiodinium heat stress to coral bleaching: a physiological timeline. Poster Presentation at the 15th International Symposium on Microbial Ecology. Gangnam-gu, Seoul, South Korea.

Levin RA, Hill R, van Oppen MJH, Steinberg PD, McDougald D (2015). Comparative transcriptomics of Symbiodinium to identify genes that reduce coral bleaching. Poster Presentation at the 115th American Society for Microbiology General Meeting. New Orleans, Louisiana, United States of America.

Levin RA, Beltran VH, Hill R, Steinberg PD, van Oppen MJH (2016). Symbiodinium exposed: transcriptomic basis of thermal tolerance. Oral Presentation at the 13th International Coral Reef Symposium. Honolulu, Hawaii, United States of America.

iii List of main figures

Chapter 2 Figure 1. Physiological detection of Symbiodinium heat stress ...... 13 Figure 2. Hierarchical clustering of DEGs ...... 16 Figure 3. Unsuccessful vs. successful acclimation to elevated temperature ...... 21 Figure 4. Regulation of meiosis, ROS scavenging, and molecular chaperone genes ...... 26 Figure 5. Model of the molecular basis of Symbiodinium thermal tolerance and its impacts on Symbiodinium-coral symbiosis ...... 30

Chapter 3 Figure 1. Genome of the novel +ssRNAV and its expression in Symbiodinium transcriptomes ...... 56 Figure 2. Viral infections and antiviral responses of Symbiodinium under heat stress ...... 59

Chapter 4 Figure 1. Cell wall digestion confirmed by cellulose staining and by swelling of protoplasts in hypotonic culture medium ...... 91 Figure 2. Somatic hybridization (fusion) of Symbiodinium protoplasts ...... 93 Figure 3. Single-cell photobiology throughout protoplast generation and cell wall regeneration ...... 95 Figure 4. Physiological status of cultures throughout protoplast generation and cell wall regeneration ...... 96 Figure 5. Bright-field imaging of cell morphology throughout protoplast generation and cell wall regeneration ...... 98

Chapter 5 Figure 1. Breakthroughs in NGS of Symbiodinium ...... 106 Figure 2. Design of a tailored expression construct for Symbiodinium ...... 113

iv Acknowledgements

Moving to Australia and completing my PhD degree has truly been a life-defining adventure. I am immensely appreciative of all the amazing people who enabled and shaped this invaluable experience.

To my PhD supervisors, Peter Steinberg and Madeleine van Oppen, I cannot thank you enough for your support, guidance, and input. Peter, you have made my PhD at UNSW possible and had my back through tough (dramatic!) times. Madeleine, you have been a constant source of inspiration and someone I could always turn to for scientific and personal advice. I would also like to acknowledge my PhD panel members: Iain Suthers and Adriana Verges for their feedback during annual review meetings, Ross Hill for teaching me PAM fluorometry, and Chris Marquis for sharing his technical expertise and for introducing me to industry contacts.

I am extremely grateful to Dave Suggett and Peter Ralph for giving me a lab-home at UTS when I had none. Dave, you have been an incredibly generous mentor and so welcoming of me into your research group. Sam Goyen, Lisa Fujise, Steph Gardner, Matt Nitsche, and Emma Camp, I will never forget our trip to Hawaii for ICRS, except for the parts we can’t remember thanks to Fireball. Caitlin Lawson, my birthday twin, thank goodness we had each other to tackle our totally real 18th and 21st birthdays. Audrey Commault, Liz Deschaseaux, Michele Fabris, Mathieu Pernice, Bojan Tamburic, Marco Giardina, Peter Davey, Jack Adriaans, and An Tran, you have filled my PhD with hysterical laughter, endless drinks, and NBA viewing parties – I couldn’t possibly have had more fun.

To Rhys Graham, thank you ×1,000,000 for listening to countless hours of me over- explaining my research, patiently waiting for me when my experiments ran late, proof reading my manuscript drafts, and being an unwavering source of love and care

v throughout my PhD. With us both finishing our graduate degrees this year, I can’t wait to see what’s in store for us and Carlton next!

Finally, thank you to my mom. You encouraged me to move to Australia, were always just a phone call away, and have been the sturdiest support system imaginable. I absolutely would not be where I am today without you.

vi Table of contents

Abstract ...... i Originality statement ...... ii Publications during candidature ...... iii List of main figures ...... iv Acknowledgements ...... v

Chapter 1 General introduction ...... 1 1.1 Thesis structure ...... 1 1.2 Symbiodinium and its ecological importance ...... 1 1.3 Anthropogenic climate change and its effects on the Symbiodinium-coral symbiosis ...... 3 1.4 Thesis aims ...... 4

Chapter 2 Sex, scavengers, and chaperones: transcriptome secrets of divergent Symbiodinium thermal tolerances ...... 7 2.1 Abstract ...... 8 2.2 Introduction ...... 9 2.3 Results and discussion ...... 12 2.3.1 Physiological responses of Symbiodinium to heat stress ...... 12 2.3.2 Plasticity of Symbiodinium transcriptomes under heat stress ...... 14 2.3.3 Gene ontology (GO) analysis of DEGs to identify functional gene groups involved in thermal tolerance ...... 17 2.3.4 Regulation of hallmark genes involved in adaptation and thermal tolerance .... 22 2.3.5 Linking Symbiodinium transcriptional heat stress responses to thermal history, physiological heat stress responses, and coral bleaching susceptibility ...... 28 2.4 Materials and methods ...... 31 2.4.1 Culture maintenance and genotyping ...... 31 2.4.2 Experimental setup ...... 32 2.4.3 Photosynthesis measurements ...... 33 2.4.4 ROS measurements ...... 33 2.4.5 Culture viability measurements ...... 34 2.4.6 Statistical analysis of physiological measurements ...... 34 2.4.7 Preparation and sequencing of RNA samples ...... 35 2.4.8 Transcriptome assembly and differential gene expression analysis ...... 36

vii 2.4.9 Annotation and GO analysis ...... 37 2.4.10 Isolation of a Fe-Sod gene ...... 37 2.5 Acknowledgements ...... 38 2.6 Supplementary tables ...... 39 2.7 Supplementary figures ...... 45 2.8 Supplementary dataset ...... 50

Chapter 3 Evidence for a role of viruses in the thermal sensitivity of coral photosymbionts ...... 51 3.1 Abstract ...... 52 3.2 Main text ...... 53 3.3 Acknowledgements ...... 60 3.4 Supplementary materials and methods ...... 60 3.4.1 Generation of transcriptomes from type C1 Symbiodinium populations ...... 60 3.4.2 Identification of viral and antiviral transcripts ...... 61 3.4.3 Differential expression analysis ...... 63 3.4.4 PCR amplification of the +ssRNAV MCP gene ...... 63 3.5 Supplementary discussion ...... 64 3.6 Supplementary tables ...... 67 3.7 Supplementary figures ...... 74 3.8 Supplementary datasets ...... 79

Chapter 4 Expanding the Symbiodinium (Dinophyceae, Suessiales) toolkit through protoplast technology ...... 81 4.1 Abstract ...... 82 4.2 Introduction ...... 83 4.3 Materials and methods ...... 85 4.3.1 Culture maintenance and genotyping ...... 85 4.3.2 Enzymatic digestion and regeneration of the cell wall ...... 86 4.3.3 Protoplast physiology ...... 88 4.4 Results and discussion ...... 90 4.5 Acknowledgements ...... 99 4.6 Supplementary video ...... 99

Chapter 5 Engineering strategies to decode and enhance the genomes of coral symbionts ...... 101 5.1 Abstract ...... 102 5.2 Introduction ...... 103 5.3 Tailoring a genetic engineering framework for Symbiodinium ...... 106

viii 5.3.1 Transcriptional promoters and terminators ...... 107 5.3.2 Genes of interest ...... 109 5.3.3 Selectable marker genes ...... 110 5.3.4 Viral elements ...... 111 5.4 CRISPR/Cas9 genome editing and Symbiodinium ...... 113 5.5 Intracellular delivery of constructs and complexes ...... 117 5.6 Can we reduce coral bleaching with genetically enhanced Symbiodinium? ...... 117 5.7 Acknowledgements ...... 120 5.8 Supplementary materials and methods ...... 120 5.8.1 sgRNA design for functional genomics studies ...... 120 5.8.2 sgRNA design for promoter studies ...... 122 5.9 Supplementary tables ...... 123 5.10 Supplementary figure ...... 129 5.11 Supplementary dataset ...... 130

Chapter 6 General discussion ...... 131

References ...... 135

Appendix A Sex, scavengers, and chaperones: transcriptome secrets of divergent Symbiodinium thermal tolerances ...... 165

Appendix B Evidence for a role of viruses in the thermal sensitivity of coral photosymbionts ...... 181

Appendix C Expanding the Symbiodinium (Dinophyceae, Suessiales) toolkit through protoplast technology ...... 187

Appendix D Engineering strategies to decode and enhance the genomes of coral symbionts ...... 199

ix x Chapter 1 General introduction

1.1 Thesis structure The thesis is structured as a “thesis by publication” (a format for PhD submission recognized by the Faculty of Science, UNSW). It contains a brief general introduction chapter, four published data chapters, and a concise general discussion chapter followed by references and appendices. The general introduction and discussion chapters are condensed to limit redundancy with the data chapters, which are each a stand-alone manuscript accompanied by supplementary information. For each data chapter, I designed and conducted the experiments, analyzed the results, and wrote the manuscript; my co-authors contributed to study design, analysis, or manuscript preparation. Manuscripts in published format are provided as appendices.

1.2 Symbiodinium and its ecological importance Symbiodinium is the largest known genus of endosymbiotic dinoflagellates and forms relationships with many marine invertebrates including jellyfish, giant clams, sponges, anemones, and – most notably – reef building corals (LaJeunesse 2002). Symbiodinium cells photosynthesize and provide corals with a large portion of their nutritional requirements for survival and growth; in turn corals provide Symbiodinium cells with inorganic nutrients for photosynthesis (Gordon and Leggat 2010, Muscatine 1990). In addition to the contributions from the coral host, essential nutrients are also made available to Symbiodinium cells by associated bacteria through processes such as nitrogen fixation and cycling of sulfur and phosphorus (Bourne et al. 2016, Croft et al. 2005, Pernice et al. 2012). This highly efficient nutrient recycling forms the foundation of coral reef ecosystems, allowing reefs to flourish in oligotrophic waters (Muscatine and Porter 1977). While covering less than 1% of the sea floor, coral reefs support approximately one third

1 of all described marine species (Reaka-Kudla 2001, Reaka-Kudla et al. 1996) and ~500 million people (Burke et al. 2011, Wilkinson 1996).

Currently, nine genetically distinct clades of Symbiodinium (A-I) have been discovered (Coffroth and Santos 2005, Pochon and Gates 2010), which comprise hundreds of identified sub-clades or “types” (Tonk et al. 2013) (http://www.symbiogbr.org/). Extensive diversity of phenotypic traits – such as photosynthetic ability, growth rate, and stress tolerance – is exhibited across members of the genus Symbiodinium (Suggett et al. 2015, Suggett et al. 2008). Furthermore, some Symbiodinium types only associate with certain coral species (e.g., type B13) (Diekmann et al. 2003), while others are host generalists (e.g., type C1) (LaJeunesse et al. 2004b).

Symbiodinium can be vertically transmitted from parent to eggs or larvae in most brooding corals (Baird et al. 2009) in order to equip coral offspring with Symbiodinium that are well suited for the local environment (Byler et al. 2013). However, closed-system symbioses can be vulnerable to changing conditions like temperature, irradiance, and disease if the inherited Symbiodinium are unable to adequately acclimate or adapt (Byler et al. 2013). Alternatively, many Symbiodinium types also persist as free-living (ex hospite) cells that can be acquired by the majority of broadcast spawning corals and some brooding corals (Baird et al. 2009, Boulotte et al. 2016, Hirose et al. 2001). Horizontal transmission of Symbiodinium from the environment to corals allows Symbiodinium-coral associations to be updated based on present conditions (Byler et al. 2013, Cumbo et al. 2013, Lewis and Coffroth 2004) – assuming that the free-living Symbiodinium have evolved fast enough to match present conditions and that the coral is able to form a symbiosis with the available free-living Symbiodinium types.

Another coral strategy to cope with a changing environment is Symbiodinium shuffling. Corals often harbor a dominant Symbiodinium type that serves as the ideal partner under basal conditions along with low levels of other Symbiodinium types that

2 may be more suited to other conditions like increased sea surface temperature (Jones and Berkelmans 2011, Little et al. 2004). When exposed to altered conditions, background Symbiodinium types can become dominant to continue the supply of photosynthate to the coral host and prevent coral bleaching (Bay et al. 2016, Berkelmans and van Oppen 2006).

1.3 Anthropogenic climate change and its effects on the Symbiodinium-coral symbiosis Elevated production of greenhouse gases from human activities involving the burning of fossil fuels, industrial processes, and livestock farming have been rapidly altering Earth’s climate (Hoegh-Guldberg et al. 2007, Le Quéré et al. 2015, Oreskes 2004). Consequently, organisms are struggling to adapt fast enough, putting them at risk of extinction (Carpenter et al. 2008, Donner et al. 2005, Hoegh-Guldberg et al. 2007, Thomas et al. 2004). Approximately 30% of the world’s coral reefs are already degraded (Burke et al. 2011, Cesar et al. 2003, Wilkinson 2002), and impact models project that irreversible destruction of coral reefs will occur worldwide in a matter of decades unless emissions of greenhouse gases are minimized and reef protection efforts are maximized (Carpenter et al. 2008, Donner 2009, Hoegh-Guldberg 1999, Hoegh-Guldberg et al. 2007, Hughes et al. 2017, Mora et al. 2016, Pandolfi et al. 2011, Veron et al. 2009).

Globally increasing sea surface temperature due to climate change is considered a major threat to coral reefs as it breaks down the Symbiodinium-coral symbiosis, resulting in loss of Symbiodinium from corals (i.e., coral bleaching) that may lead to coral starvation and mortality from the lack of Symbiodinium photosynthate (Hoegh-Guldberg 1999). A temperature increase of 2-3 °C above the mean summer maximum for less than a week can trigger significant coral bleaching and mortality (Berkelmans and Willis 1999). Prolonged periods of heating as little as 1 °C above the mean summer maximum across broad reef areas and regions can also transition to mass bleaching and mortality (Donner et al. 2005, Hoegh-Guldberg 1999).

3 The increasing frequency and severity of coral bleaching events with climate change are of great concern (Ainsworth et al. 2016, Donner et al. 2007, Glynn 1993, Hughes et al. 2017, Veron et al. 2009). For instance, the recent El Niño event from 2014- 2016 was intensified by climate change (Black and Karoly 2016, Cai et al. 2014, Pala 2016) and caused global increases in sea surface temperature that spurred the longest coral- bleaching event on record (Hughes et al. 2017, Normile 2016, Wijffels et al. 2016). On the Great Barrier Reef alone, 92% of 873 surveyed sites experienced coral bleaching, which resulted in death for 22% of corals (Great Barrier Reef Marine Park Authority 2016).

Symbiodinium are typically more susceptible to thermally induced dysfunction than corals, making Symbiodinium the determinant partner of thermal thresholds for coral bleaching (Fitt et al. 2001, Strychar and Sammarco 2009). When exposed to heat stress, Symbiodinium cells suffer reduced photosynthetic ability (Warner et al. 1999), unstable thylakoid membrane structure (Tchernov et al. 2004), and oxidative stress (Rehman et al. 2016, Suggett et al. 2008). Furthermore, excess reactive oxygen species generated under heat stress may leak out of Symbiodinium cells and oxidatively damage coral macromolecules and tissues (Downs et al. 2002, Weis 2008). Additionally, virus-like particles appear to increase under heat stress in Symbiodinium, thus increased sea surface temperature has been hypothesized to promote virus infections in Symbiodinium that may disrupt symbiosis and contribute to coral bleaching (Davy et al. 2006, Wilson et al. 2005, Wilson et al. 2001). However, genetic understanding of these factors in Symbiodinium that can define coral bleaching susceptibility is scant, and filling this knowledge gap will be critical for innovating progressive strategies to protect coral reefs.

1.4 Thesis aims The broad aims of the thesis were to unveil the basis of Symbiodinium thermal tolerance and to design approaches for advancing future investigations of Symbiodinium that may ultimately produce novel coral reef protection strategies. Specifically:

4 In chapter 2, I aimed to explore genetic architectures of different Symbiodinium populations ex hospite in order to pinpoint key genes that establish Symbiodinium thermal tolerance and may be responsible for previously determined in hospite bleaching thresholds (Levin et al. 2016).

In chapter 3, I aimed to uncover a link between viral infections, Symbiodinium antiviral responses, Symbiodinium thermal tolerance, and bleaching thresholds through novel ‘omics analyses (Levin et al. 2017a).

In chapter 4, I aimed to generate the first Symbiodinium protoplasts (viable cells with their cell walls removed) and achieve Symbiodinium protoplast fusion as a tool for genetic modification of Symbiodnium (Levin et al. 2017b).

In chapter 5, I aimed to integrate Symbiodinium genetic elements, virus genetic elements, and protoplast technology into a proposed framework that may facilitate genetic engineering of Symbiodinium to advance functional genomic studies and potentially increase Symbiodinium thermal tolerance (Levin et al. 2017c).

6 Chapter 2 Sex, scavengers, and chaperones: transcriptome secrets of divergent Symbiodinium thermal tolerances

Publication I Levin RA, Beltran VH, Hill R, Kjelleberg S, McDougald D, Steinberg PD, van Oppen MJH (2016). Sex, scavengers, and chaperones: transcriptome secrets of divergent Symbiodinium thermal tolerances. Molecular Biology and Evolution 33: 2201-2215.

Declaration I certify that this publication was a direct result of my research towards this PhD and that reproduction in this thesis does not breach copyright regulations.

...... Rachel A. Levin

7 2.1 Abstract Corals rely on photosynthesis by their endosymbiotic dinoflagellates (Symbiodinium spp.) to form the basis of tropical coral reefs. High sea surface temperatures driven by climate change can trigger the loss of Symbiodinium from corals (coral bleaching), leading to declines in coral health. Different putative species (genetically distinct types) as well as conspecific populations of Symbiodinium can confer differing levels of thermal tolerance to their coral host, but the genes that govern dinoflagellate thermal tolerance are unknown. Here we show physiological and transcriptional responses to heat stress by a thermo-sensitive (physiologically susceptible at 32°C) type C1 Symbiodinium population and a thermo-tolerant (physiologically healthy at 32°C) type C1 Symbiodinium population. After nine days at 32°C, neither population exhibited physiological stress, but both displayed up-regulation of meiosis genes by ≥ 4-fold and enrichment of meiosis functional gene groups, which promote adaptation. After 13 days at 32°C, the thermo-sensitive population suffered a significant decrease in photosynthetic efficiency and increase in reactive oxygen species (ROS) leakage from its cells while the thermo-tolerant population showed no signs of physiological stress. Correspondingly, only the thermo-tolerant population demonstrated up-regulation of a range of ROS scavenging and molecular chaperone genes by ≥ 4-fold and enrichment of ROS scavenging and protein folding functional gene groups. The physiological and transcriptional responses of the Symbiodinium populations to heat stress directly correlate with the bleaching susceptibilities of corals that harbored these same Symbiodinium populations. Thus, our study provides novel, foundational insights into the molecular basis of dinoflagellate thermal tolerance and coral bleaching.

8 2.2 Introduction Corals and their dinoflagellate endophotosymbionts of the genus Symbiodinium create the foundation of tropical coral reefs, which support hundreds of thousands of plant and animal species (Reaka-Kudla et al. 1996). Tropical reef-building corals require metabolites provided by Symbiodinium for their nutrition and high rates of calcification (Barnes and Chalker 1990, Gordon and Leggat 2010, Muscatine and Porter 1977). Efficient recycling of nutrients between Symbiodinium and corals allows entire ecosystems to flourish in low nutrient waters (Roth 2014). Rising sea surface temperatures due to climate change cause the breakdown of the Symbiodinium-coral symbiosis resulting in the loss of Symbiodinium from the coral host (i.e., coral bleaching) and, consequently, drastic declines in coral health and cover worldwide (Hoegh-Guldberg 1999, Hoegh-Guldberg et al. 2007). Climate change impact models predict that many coral reefs will be irreversibly damaged in a matter of decades (Carpenter et al. 2008, Pandolfi et al. 2011). While the exact mechanistic role that Symbiodinium plays in coral bleaching has yet to be uncovered, increased production of ROS, such as superoxide and hydrogen peroxide, by Symbiodinium cells in response to heat stress is considered to be a key factor (McGinty et al. 2012, Suggett et al. 2008). Leakage of excess ROS from Symbiodinium cells when inside the coral tissues (in hospite) may exacerbate stress-induced oxidative damage of coral tissues and lead to Symbiodinium expulsion (Downs et al. 2002, Krueger et al. 2015).

The genus Symbiodinium is highly diverse, and substantial physiological differences exist among and even within ‘types’, i.e., genetic variants typically designated by the nuclear ribosomal DNA internal transcribed spacer 2 (ITS2) to notionally represent species (Arif et al. 2014). Different Symbiodinium can strongly influence coral gene expression and bleaching susceptibility (DeSalvo et al. 2010, Howells et al. 2012, Oliver and Palumbi 2011, Yuyama et al. 2012), and it is generally thought that Symbiodinium are more vulnerable to heat stress than their coral host (Fitt et al. 2001). Unravelling the molecular basis of variation in Symbiodinium thermal tolerance is thus an essential step required to understand variation in coral bleaching susceptibility.

While Symbiodinium physiological responses to heat stress are well studied (Howells et al. 2012, McGinty et al. 2012, Suggett et al. 2008, Tchernov et al. 2004, Warner et al. 1999), the underlying gene regulation is still unresolved. Much of the evidence to date suggests that Symbiodinium lack a transcriptional response to heat stress (Barshis et al. 2014, Krueger et al. 2015, Leggat et al. 2011b, Putnam et al. 2013), which contradicts the strong evidence in other organisms that physiological changes are largely driven by regulation of mRNA synthesis and degradation (Arbeitman et al. 2002, Harb et al. 2010, Rossouw et al. 2009, Wilusz and Wilusz 2004). In Symbiodinium, translational regulation and post-translational modifications have been hypothesized to primarily drive changes in the proteome under heat stress (Barshis et al. 2014), as only a small collection of transcription factors have been identified in the transcriptome and genome of Symbiodinium (Bayer et al. 2012, Shoguchi et al. 2013). Symbiodinium transcriptomes have also been found to contain microRNAs (Baumgarten et al. 2013), molecules that repress translation of mRNA into proteins as well as direct and accelerate mRNA degradation (Valencia-Sanchez et al. 2006, Wu et al. 2006). Regulation of mRNA abundance may therefore be an important contributor to physiological responses by Symbiodinium.

Several previous gene expression studies in Symbiodinium have applied acute heat stress on the scale of hours to a few days (Barshis et al. 2014, Baumgarten et al. 2013, Krueger et al. 2015, Rosic et al. 2014), while a study on mRNA stability in the dinoflagellate Karenia brevis found dinoflagellate mRNA half-lives to be considerably longer than in other organisms (Morey and Van Dolah 2013). The majority of transcripts involved in the stress response, metabolism, and transcriptional regulation had half-lives over 24 hours, and in some cases over four days (e.g., catalase/peroxidase, thioredoxin, chaperone protein DnaJ) (Morey and Van Dolah 2013). Thus, some dinoflagellate genes may simply require longer periods of time to develop significant, detectable mRNA expression changes. However, Morey and Van Dolah (2013) did not measure mRNA half-

10 lives under temperature stress, which can significantly alter mRNA stability (Castells-Roca et al. 2011, Chiba et al. 2013).

In this study, we used two heterogeneous populations of type C1 Symbiodinium, an ecologically important and globally distributed type associated with a diverse range of coral species (LaJeunesse 2005, LaJeunesse et al. 2004b, Tonk et al. 2013). Despite having identical ITS1 and ITS2 sequences, the populations exhibit different thermal tolerances. Physiological and transcriptional analyses were conducted for each population at ambient (27°C) and elevated (32°C) temperatures in culture in order to investigate the molecular basis of Symbiodinium thermal tolerance. The populations were originally isolated from the coral Acropora tenuis at two separate sites on the Great Barrier Reef: South Molle Island (SM; 20°16’33” S, 148°49’36” E) and Magnetic Island (MI; 19°9’6” S, 146°51’56” E) that have average summer daily maximums of 28.2°C and 30.1°C, respectively. Corals harboring the thermo-sensitive SM population were previously shown to bleach after 11 days at 32°C, while corals harboring the thermo-tolerant MI population remained unaffected (Howells et al. 2012). A significant reduction in photosynthetic capacity due to heat stress, a diagnostic trait of Symbiodinium thermal sensitivity and coral bleaching (Warner et al. 1999), accompanied loss of the SM population from its coral host at 32°C (Howells et al. 2012). The susceptibility of each population to elevated temperature in hospite correlated with thermal tolerance in culture (Howells et al. 2012).

Here we report on thousands of differentially expressed genes (DEGs) in both populations exposed to elevated temperature (32°C) that align with physiological responses. Our findings demonstrate how distinct transcriptomic plasticity and regulation of hallmark thermal tolerance genes and functional gene groups (i.e., gene ontology categories) can allow allopatric, conspecific Symbiodinium populations to exhibit contrasting thermal tolerances.

11 2.3 Results and discussion

2.3.1 Physiological responses of Symbiodinium to heat stress Each population was cultured at 27°C and 32°C in two replicate incubators (Table S1) to avoid potential artifacts from individual incubators in our results. Physiological measurements for detection of cellular heat stress were used to determine sampling time points for transcriptomics that were anticipated to identify DEGs between temperature treatments (Figure 1A-C, Figure S1A-D). On day 13, both the maximum relative electron transport rate for photosynthesis (rETRm) and initial photosynthetic rate (α) significantly decreased (p < 0.05) at 32°C compared to 27°C in the SM population only (Figure 1A-B). Decreased photosynthetic ability of Symbiodinium has been strongly connected to Symbiodinium thermal sensitivity and coral bleaching susceptibility (Howells et al. 2012, Ragni et al. 2010, Takahashi et al. 2009, Warner et al. 1999). Additionally, a significant increase (p < 0.05) in general ROS leakage out of Symbiodinium cells was detected in the SM population at 32°C beginning on day 13 (Figure 1C), an observation that is consistent with evidence that coral bleaching is largely driven by increased ROS inside coral tissues (Downs et al. 2002, Suggett et al. 2008). Therefore, day 13 was chosen as a sampling time point for transcriptomics, along with day -1 to account for any pre-experimental DEGs between groups. Day 9, the potential start of the declining trend in rETRm in the SM population, was also selected as a sampling time point for transcriptomics to determine if the transcriptional response to heat stress precedes significant physiological damage. The overall lower photosynthetic efficiency of the SM population may be due to the lower amounts of photosynthetic pigments (chlorophyll a and β-carotene) in cells from the SM population compared to those from the MI population (Howells et al. 2012).

Figure 1. Physiological detection of Symbiodinium heat stress. Intact lines represent the 27°C temperature treatment, and dashed lines represent the 32°C temperature treatment. Before heating, all samples were kept at 27°C (values in the grey regions). (A) rETRm (mean ± s.e.m., n = 4). (B) α (mean ± s.e.m., n = 4). (C) Leakage of ROS out of cells (mean ± s.e.m., n = 4); unitless fluorescence intensities of CellROX® Orange reagent for oxidative stress detection were normalized across days by setting the fluorescence intensities of the 27°C samples to 100%. Asterisks indicate statistically significant (PERMANOVA) differences between temperature treatments at p < 0.05. Sampling time points for transcriptomics are boxed. Additional physiological measurements are shown in Figure S1. 13 2.3.2 Plasticity of Symbiodinium transcriptomes under heat stress The de novo assembled transcriptomes from the SM and MI populations were comprised of 106,097 and 93,377 putative genes, respectively. However, the number of genes in each transcriptome likely overestimates the number of genes expressed by a single genotype because our study used heterogeneous populations rather than clonal cultures. Each population consisted of an unknown diversity of individuals within type C1 and therefore an unknown diversity of transcript variants and alleles. The SM and MI populations, rather than clonal cultures, were chosen in our study as their bleaching responses at 32°C have been characterized in hospite (Howells et al. 2012) and as heterogeneous populations are more representative of symbiont communities inhabiting Great Barrier Reef corals. Average transcript lengths (SM: 858.1 bp and MI: 911.4 bp; Table S2) for the SM and MI transcriptomes were in range with those for previously published Symbiodinium transcriptomes (Bayer et al. 2012, Parkinson et al. 2016). Quantitative assessment of conserved eukaryotic orthologs (Simão et al. 2015) revealed that the SM and MI transcriptomes are the most complete Symbiodinium transcriptomes of the publicly accessible, published Symbiodinium transcriptomes to date (Baumgarten et al. 2013, Bayer et al. 2012, Ladner et al. 2012, Parkinson et al. 2016, Rosic et al. 2015, Xiang et al. 2015) (Table S3). The biological coefficient of variance (BCV) for gene expression across replicates in each population was found to be < 0.2 on all time points, well below the commonly accepted variance threshold of 0.4 (Chen et al. 2014, McCarthy et al. 2012).

For differential gene expression analysis, we defined significant biological relevance as ≥ 4-fold differential expression between temperature treatments combined with a conservative false discovery rate (FDR) ≤ 0.001. On day -1 prior to heat treatment, only one DEG (TR83958|c0_g1, a putative 10 kDa chaperonin) in the SM population and no DEGs in the MI population were found between the experimental groups of each population that had been pre-assigned to the different temperature treatments. TR83958|c0_g1 from the SM population was not differentially expressed on either of the

14 later time points. The lack of DEGs between experimental groups in both populations before heating corroborates that DEGs detected on days 9 and 13 were in response to the temperature treatment (Figure 2A-D) and that differential expression cutoffs (fold ≥ 4 and FDR ≤ 0.001 between temperature treatments) and replication (n = 4) were adequate to achieve a high signal to noise ratio.

Figure 2. Hierarchical clustering of DEGs. Heat maps show genes (rows) with differential expression (Trinity/edgeR: fold ≥ 4, FDR ≤ 0.001) between 27°C and 32°C samples (columns) for each population on (A) day 9 and (B) day 13. Expression values (FPKM) are log2-transformed and then median-centered by gene. Heat map values were calculated by subtracting each gene’s median log2(FPKM) value from its log2(FPKM) value in each sample. The proportions (%) of DEGs that were up- or down-regulated due to heat stress are noted to the right of the two main gene clusters of each heat map. Genes are independently clustered for each population at each time point. Samples from replicate cultures at each temperature treatment are presented in the same order for each time point. The experimental incubator (A, B, C, or D) that housed each sample is noted below the temperature treatment. 16 On day 9, a total of 4,608 and 2,379 DEGs were identified between the temperature treatments in the SM and MI populations, respectively. The vast majority of DEGs in the SM population (4,199 or 91%) and MI population (2,179 or 92%) were down- regulated at 32°C relative to expression levels at 27°C (Figure 2A). Down-regulation of the majority of DEGs in response to elevated temperature has been previously observed in marine organisms including Symbiodinium and corals (Baumgarten et al. 2013, Bay and Palumbi 2015, Yampolsky et al. 2014) and may be a strategy to conserve energy when confronted with environmental stress (Yampolsky et al. 2014).

On day 13, a total of 4,272 and 3,513 DEGs were identified between the temperature treatments in the SM and MI populations, respectively. The SM population responded similarly to 32°C on day 13 as on day 9 by down-regulating the majority of DEGs (3,341 or 78%). Conversely, the MI population up-regulated the majority of DEGs (2,201 or 63%) at 32°C, suggesting acclimation to 32°C (Figure 2B). Our results demonstrate that some Symbiodinium do exhibit transcriptomic plasticity and are capable of up-regulating a large number of genes in response to elevated temperature.

In the SM population at 32°C, 239 and 1,925 genes remained up- and down- regulated, respectively, on both days 9 and 13. In the MI population at 32°C, 113 and 585 genes remained up- and down-regulated, respectively, on both days 9 and 13. Interestingly, 353 genes in the MI population at 32°C that were down-regulated on day 9 became up-regulated on day 13, while no genes switched from down- to up-regulation in the SM population at 32°C. No up-regulated genes on day 9 became down-regulated on day 13 in either population at 32°C.

2.3.3 Gene ontology (GO) analysis of DEGs to identify functional gene groups involved in thermal tolerance GO analysis (FDR < 0.05) of genes at 32°C further supported that only the MI population acclimated to elevated temperature (Figure 3, Dataset S1A-M), consistent with

17 only the SM population suffering physiological damage after 13 days of heat stress. Acclimation to stressful conditions through transcriptional changes has been observed in other marine organisms including corals (Bay and Palumbi 2015, Hennon et al. 2015, Nymark et al. 2009, Yampolsky et al. 2014), but never before in Symbiodinium, nor to our knowledge, in any dinoflagellate species.

On day 9, the down-regulated genes in the SM population at 32°C were enriched for 133 GO categories consisting of 26 metabolic and biosynthetic categories, while the down-regulated genes in the MI population at 32°C were enriched for 311 GO categories that included 45 metabolic and biosynthetic categories (Dataset S1B, D). Reduced metabolic and biosynthetic activity has been shown to correlate with increased survival time of organisms under stress, as it allows for substantial energetic savings (Hand and Hardewig 1996). Specifically in the case of heat stress, such metabolic compensation is considered an acclimatory mechanism to elevated temperature in the zooplankton Daphnia pulex (Yampolsky et al. 2014).

The small number of significantly up-regulated genes in the SM and MI populations at 32°C on day 9 were enriched for six and seven GO categories, respectively (Figure 3; Dataset S1A, C). The majority of enriched GO categories in both populations were specific to meiosis, suggesting that Symbiodinium cells were participating in sexual rather than strictly asexual reproduction under heat stress. Potential sexual reproduction by Symbiodinium is particularly noteworthy since meiosis creates genetic diversity through chromosomal modifications and recombination, therefore promoting adaptation (Becks and Agrawal 2012, D'Souza and Michiels 2010, Tamburini and Tyler 2005). Meiosis-specific genes have been previously identified in Symbiodinium (Chi et al. 2014, Rosic et al. 2015), but so far, sexual reproduction has not been directly observed. However, recent studies have hypothesized that recombination during meiosis may be a mechanism of adaptation in Symbiodinium (Chi et al. 2014, Wilkinson et al. 2015). Other dinoflagellate species can rapidly increase genetic diversity by switching from mitosis to meiosis and can enter a

18 sexual cyst life cycle stage when exposed to stressful conditions in order to survive and adapt (Bravo and Figueroa 2014, Figueroa et al. 2010), though no visually apparent Symbiodinium cysts were observed in our study.

On day 13, the dramatic increase in up-regulated genes in the MI population at 32°C was characterized by enrichment of 60 GO categories (Figure 3, Dataset S1G), which included nine metabolic categories along with several of the corresponding catabolic categories for maintaining cellular homeostasis. Importantly, GO categories for unfolded protein binding, protein folding, glutamate dehydrogenase (NAD+) activity, and the oxidoreducatase complex – all of which are involved in stress tolerance (Bita and Gerats 2013, Singh and Grover 2008, Srivastava and Singh 1987, Tercé-Laforgue et al. 2015) – also became enriched in the up-regulated genes in the MI population at 32°C. The subset of 353 up-regulated genes at 32°C, which had been down-regulated at 32°C on day 9, was enriched for 29 GO categories including seven for metabolism and biosynthesis, one for oxidoreductase activity, and one for motile cilium (Dataset S1M). The down-regulated genes in the MI population at 32°C on day 13 were enriched for 160 GO categories covering a broad array of processes, though only 13 metabolic and biosynthetic GO categories were present (Dataset S1H). The largely reduced metabolic compensation at 32°C on day 13 (down-regulated genes enriched for 13 metabolic and biosynthetic GO categories) relative to day 9 (down-regulated genes enriched for 45 metabolic and biosynthetic GO categories), along with up-regulation of genes enriched for metabolic and stress tolerance GO categories on day 13, suggests that the MI population had acclimated to 32°C.

In contrast, GO enrichment analysis indicated that the SM population was unable to acclimate to 32°C by day 13. The down-regulated genes in the SM population at 32°C on day 13 were enriched for 135 GO categories, including 19 for metabolism and biosynthesis and one for the oxidoreductase complex (Dataset S1F). Only three GO categories were enriched in the up-regulated genes in the SM population at 32°C, one of

19 which was the meiosis category, chiasma assembly, potentially signifying a continued attempt to adapt by producing genetic diversity through sexual recombination (Figure 3, Dataset S1E). No metabolic, biosynthetic, or stress tolerance GO categories became enriched in the up-regulated genes at 32°C. The extended duration of metabolic compensation experienced by the SM population compared to the MI population under heat stress could potentially cause starvation of its coral host, which may contribute to the higher bleaching susceptibility of corals harboring the SM population (Howells et al. 2012).

Figure 3. Unsuccessful vs. successful acclimation to elevated temperature. GO relationship graphs for enriched biological process GO categories (Goseq: FDR < 0.05) at 32°C were generated using REVIGO (Supek et al. 2011). Bubble size indicates the frequency of the GO category in the UniProt database relative to the other GO categories that are in the same section. Lines link similar GO categories, and the line width indicates the degree of similarity between the GO categories relative to others in the same section. Redundant GO categories (similarity > 0.9) were collapsed into the category that is most frequent in the UniProt database. Graphs for enriched molecular function GO categories (e.g., unfolded protein binding and glutamate dehydrogenase activity) and enriched cellular component GO categories (e.g., oxidoreductase complex and motile cilium) at 32°C are not included in this figure. Full GO analysis results are listed in Dataset S1A-M.

21 2.3.4 Regulation of hallmark genes involved in adaptation and thermal tolerance While the responses of the two Symbiodinium populations to heat stress differed, both transcriptomes contained comparably extensive suites of meiosis-specific and thermal tolerance genes (Figure 4A, Figure S2A) that are consistent with gene content found in other Symbiodinium (Bayer et al. 2012, Chi et al. 2014, Krueger et al. 2015, Rosic et al. 2015). However, a striking difference in gene content between the transcriptomes of the SM and MI populations was the expression of eight iron superoxide dismutase (Fe- Sod) genes in the MI transcriptome while no Fe-Sod genes were expressed in the SM transcriptome, implying that these genes are either absent from the SM population or their expression is suppressed through epigenetic regulation. Detectable Fe-Sod gene expression is inconsistent among other Symbiodinium (Krueger et al. 2015), and phylogenetic evidence that the acquisition of several ROS scavenging genes by Symbiodinium has resulted from horizontal gene transfer (Krueger et al. 2015) indicates that some Symbiodinium genomes may lack Fe-Sod genes entirely. Successful PCR amplification of the most highly expressed Fe-Sod gene (TR20255|c0_g1, open reading frame: 674-78[-]) from the genomic DNA of the MI population but not the SM population highlights the robustness of our transcriptome assemblies and supports that some gene content varies between the populations (Figure S3A). However, our PCR results cannot confirm that no Fe-Sod genes are present in the SM population, as the primers were specific to the open reading frame of TR20255|c0_g1. Differences in nucleotide sequence between the open reading frame of TR20255|c0_g1 and the other seven Fe-Sod genes in the MI population, as well as the Fe-Sod genes identified by Krueger et al. (2015) in types B1, E, and F1 Symbiodinium (Figure S3B), suggest that alternative Fe-Sod genes could be in the SM population but be silenced or expressed below the detectable level. Though, it should also be noted that Krueger et al. (2015) failed to find any Fe-Sod genes expressed in types C1, C3, C15, and D Symbiodinium.

On days 9 and 13, both populations maintained up-regulation of meiosis-specific genes at 32°C (Figure 4B-C), though statistically significant up-regulation (fold ≥ 4, FDR ≤

22 0.001) of mutS homolog 5 (Msh5) was limited to day 9 in the MI population. The heterodimer partners mutS homolog 4 (Msh4) and Msh5 are members of the Msh gene family. Unlike the other Msh genes that are involved in mismatch repair, DNA damage repair, and mitotic recombination (Modrich and Lahue 1996, Stojic et al. 2004, Wang and Qin 2003), studies in a wide range of organisms (including humans, mice, yeast, Caenorhabditis elegans, Arabidopsis thaliana, and Tetrahymena thermophile) show that Msh4 and Msh5 genes are essential and specific to meiosis (Argueso et al. 2004, Bocker et al. 1999, Higgins et al. 2004, Hollingsworth et al. 1995, Kelly et al. 2000, Kneitz et al. 2000, Novak et al. 2001, Shodhan et al. 2014). MSH4 and MSH5 proteins form a meiosis-specific sliding clamp that holds and pairs homologous chromosomes during meiosis (Kneitz et al. 2000, Snowden et al. 2004). Mutations to Msh4 and Msh5 genes have both been shown to affect crossing over of homologous chromosomes but not to affect mismatch repair (Hollingsworth et al. 1995, Ross-Macdonald and Roeder 1994). However, some studies indicate that MSH4 and/or MSH5 proteins may have additional functions outside of meiosis processes such as DNA damage response (Her et al. 2003, Sekine et al. 2007, Tompkins et al. 2009). To consider whether Symbiodinium Msh4 and Msh5 genes may atypically function in non-meiotic pathways with other Msh genes, we investigated the gene expression of Msh1, 2, 3, and 6 in the Symbiodinium transcriptomes. Interestingly, none of the Msh genes besides Msh4 and Msh5 were up-regulated at 32°C, supporting that meiosis-specific processes were induced rather than mismatch repair, DNA damage repair, or mitotic recombination.

The meiotic recombination protein Spo11-2 (Spo11-2) gene was also up-regulated in both Symbiodinium transcriptomes on days 9 and 13 at 32°C. Spo11-2 and its paralog, the meiotic recombination protein Spo11-1 (Spo11-1) gene, are the meiosis-specific members of the Spo11 gene family. SPO11-1 and SPO11-2 proteins create meiosis-specific double-strand breaks in DNA and form the synaptonemal complex to initiate meiosis (Cao et al. 1990, Keeney 2007, Keeney et al. 1997, Tsubouchi and Roeder 2005). Mutations to Spo11-2 in Arabidopsis thaliana cause sterility and aneuploidy (Hartung et al. 2007, Stacey

23 et al. 2006). While we are not aware of any examples in which the Spo11-2 gene acts outside of meiosis, future detailed studies will be important to confirm that the meiosis- specific functions of Spo11-2, as well as Msh4 and Msh5, are conserved in Symbiodinium.

Superoxide dismutases are key scavengers of superoxide, peroxidases are engaged in the removal of hydrogen peroxide, and molecular chaperones are essential for refolding damaged proteins (Gill and Tuteja 2010, Vierling 1991) – making them key contributors to thermal tolerance. Despite the many examples that up-regulation of these genes confers thermal tolerance in numerous photosynthetic species (Bita and Gerats 2013, Singh and Grover 2008, Tang et al. 2006, Van Breusegem et al. 1999), many studies report no notable differential expression of these genes in Symbiodinium at elevated temperature (Barshis et al. 2014, Krueger et al. 2015, Leggat et al. 2011b, Putnam et al. 2013). Yet, limited evidence suggests that the transcriptional heat stress response of Symbiodinium may involve up-regulation of some genes classically associated with thermal tolerance. The first study demonstrated through qPCR that cytochrome P450 (Cyp450) gene expression by type C3 Symbiodinium increased at 26°C and 29°C compared to 23-24°C, while exposure to 32°C resulted in decreased Cyp450 expression (Rosic et al. 2010). The next study also used qPCR and showed that heat shock protein 70 (Hsp70) expression in type C1 Symbiodinium was slightly increased at approximately 30°C, but down-regulation of Hsp70 occurred at 32°C (Rosic et al. 2011). An RNA-Seq study of type A1 Symbiodinium found up-regulation of one peroxiredoxin (Prdx) gene, one Hsp gene, and one chaperone protein DnaJ (DnaJ) gene from exposure to 34°C for 12 hours (Baumgarten et al. 2013). However, the importance of the differential gene expression at this extreme temperature was not substantiated by sample replication, correspondence to a physiological heat stress response, or relation to a coral bleaching response (Baumgarten et al. 2013). Finally, a recent RNA-Seq study detected minor up-regulation of Hsp90 by in hospite Symbiodinium after 24 hours of exposure to 30°C relative to 23-24°C, but not after 72 hours of exposure to 30°C (Rosic et al. 2014).

24 In our study, general down-regulation of thermal tolerance genes was observed on day 9 in both populations at 32°C (Figure 4B, Figure S2B, Tables S4-5). One Hsp90 gene, one Cyp450 gene and two DnaJ genes were up-regulated by the SM population at 32°C compared to just one DnaJ gene up-regulated in the MI population at 32°C. Hsp genes were uniquely found to be down-regulated in the MI population at 32°C, though the MI population also showed no signs of physiological heat stress throughout the study. Elevated temperature has previously been shown to reduce the expression of Hsp genes and Cyp450 genes in Symbiodinium (Rosic et al. 2011, Rosic et al. 2010) as well as the expression of Hsp genes and ROS scavenging genes in corals (Bay and Palumbi 2015, Rosic et al. 2014). Down-regulation of some thermal tolerance genes may be attributed to the general down-regulation of > 90% of all DEGs in both populations at 32°C on day 9. The down-regulated genes in each population were not enriched for GO categories related to thermal tolerance (e.g., unfolded protein binding, the oxidoreductase complex), supporting the notion that down-regulation of thermal tolerance genes may simply reflect a non-targeted, global reduction in transcription to conserve energy at 32°C.

On day 13, only one glutaredoxin (Glrx) gene was up-regulated in the SM population at 32°C. In contrast, three Fe-Sod genes, one cytochrome c peroxidase (Ccpr) gene, one glutathione peroxidase (Gpx) gene, three Prdx genes, two thioredoxin (Txn) genes, and one Cyp450 gene were significantly up-regulated in the MI population at 32°C, highlighting the importance of ROS scavenging genes in type C1 Symbiodinium thermal tolerance. Additionally, 11 Hsp70 genes, four Hsp90 genes, and eight DnaJ genes were up- regulated by the MI population at 32°C compared to only one Hsp70 gene, three Hsp90 genes, and four DnaJ genes up-regulated in the SM population at 32°C (Figure 4C, Figure S2C, Tables S6-7). Up-regulation of Hsp90 genes by both populations under heat stress is consistent with findings that Symbiodinium HSP90 protein abundance increases under heat stress (Ross 2014).

Figure 4. Regulation of meiosis, ROS scavenging, and molecular chaperone genes. (A) The number of genes for gene types involved in sexual reproduction or thermal tolerance in the SM and MI transcriptomes. The number of DEGs at 32°C from each gene type (Trinity/edgeR: fold ≥ 4 and FDR ≤ 0.001 relative to 27°C) are shown for each population on (B) day 9 and (C) day 13. Gene types that had no DEGs in either population are excluded from (B) and (C). Gene abbreviations are as follows: mutS protein homolog 4 (Msh4), mutS protein homolog 5 (Msh5), meiotic recombination protein Spo11-2 (Spo11-2), copper/zinc superoxide dismutase (Cu/Zn-Sod), iron superoxide dismutase (Fe-Sod), manganese superoxide dismutase (Mn-Sod), nickel superoxide dismutase (Ni-Sod), catalase- peroxidase (KatG), ascorbate peroxidase (Apx), cytochrome c peroxidase (Ccpr), glutathione peroxidase (Gpx), peroxiredoxin (Prdx), heat shock protein 70 (Hsp70), heat shock protein 90 (Hsp90). Additional genes are shown in Figure S2. DEG annotation and differential expression details are provided in Tables S4-7.

2.3.5 Linking Symbiodinium transcriptional heat stress responses to thermal history, physiological heat stress responses, and coral bleaching susceptibility The SM and MI Symbiodinium populations have been kept in culture for more than four years at approximately 27°C, and their relative thermal tolerances, reported in 2012 (Howells et al. 2012), were confirmed in our current study conducted in 2015. The marked difference in their transcriptional responses to elevated temperature may therefore be driven by stable, heritable stress memory due to the different thermal regimes of the SM and MI reefs. Stress memory (or “priming”) is the process in which previous exposure to a particular stress causes epigenetic and/or chromosomal modifications that allow for a faster and stronger acclimation response to subsequent exposures, which can be stably passed on to future generations (Bruce et al. 2007). The warmer MI reef reaches ≥ 32°C on approximately 12% of summer days, unlike the cooler SM reef where no summer days reach ≥ 32°C (Howells et al. 2012), suggesting that only the MI population has been primed and/or genetically adapted for efficient acclimation to 32°C. Successful PCR amplification of a Fe-Sod gene from only the genomic DNA of the MI population (Figure S3A) indicates that genetic adaptation is involved in acclimation to 32°C, but epigenomic and genomic analysis will be necessary to determine whether stress memory also contributes to the transcriptional acclimation response.

Acclimation to elevated temperature by the MI population highlights the importance of up-regulating hallmark thermal tolerance genes. Particularly, significant up- regulation of genes for unfolded protein binding, protein folding, and the oxidoreductase complex likely minimizes damage to photosynthetic apparatuses and ROS leakage from cells – both of which were observed in the heat stressed SM population. We hypothesize that the observed transcriptional response by the MI population in culture may also allow the MI population to maintain symbiosis with its coral host at elevated temperature (Howells et al. 2012). Conversely, the observed leakage of ROS out of cells in the SM population due to unsuccessful acclimation to elevated temperature may cause oxidative damage to the coral host, resulting in bleaching as previously seen with corals harboring

28 the SM population when exposed to heat stress (Howells et al. 2012) (Figure 5). While Hsp gene expression has been found to be indistinguishable between Symbiodinium in culture and in hospite (Rosic et al. 2011), more extensive temporal studies of in hospite Symbiodinium gene expression will be necessary to determine the effect of symbiosis on the comprehensive collection of DEGs identified here. Metabolomics should also be utilized to determine if metabolic compensation of in hospite Symbiodinium over extended periods of heat stress factors into the breakdown of Symbiodinium-coral symbiosis.

Figure 5. Model of the molecular basis of Symbiodinium thermal tolerance and its impacts on Symbiodinium-coral symbiosis. Schematics of Symbiodinium cells from the SM population and MI population after 13 days at 32°C hypothesize the impacts of their respective up-regulated thermal tolerance genes (Trinity/edgeR: fold ≥ 4 and FDR ≤ 0.001 relative to 27°C) and enriched thermal tolerance GO categories (Goseq: FDR < 0.05) on their coral hosts. The main organelles contributing to ROS production are depicted. The shape containing ‘ROS’ represents oxidative damage to the Symbiodinium cell or the coral host. Gene abbreviations are as follows: iron superoxide dismutase (Fe-Sod), cytochrome c peroxidase (Ccpr), glutathione peroxidase (Gpx), peroxiredoxin (Prdx), heat shock protein 70 (Hsp70), heat shock protein 90 (Hsp90), glutaredoxin (Glrx), thioredoxin (Txn), cytochrome P450 (Cyp450), chaperone protein DnaJ (DnaJ).

In this study, we have detailed gene regulation by a thermo-sensitive type C1 Symbiodinium population and a thermo-tolerant type C1 Symbiodinium population in response to heat stress that parallels their respective physiological responses to heat stress and previously described bleaching responses in hospite (Howells et al. 2012). Furthermore, our study is the first to identify individual genes as well as overarching functional gene groups that influence dinoflagellate thermal tolerance. Our results provide critical insights into the impacts of Symbiodinium gene regulation on coral bleaching and present genes (e.g., Msh4, Msh5, Spo11-2) that could be used to detect heat stress in Symbiodinium before potential physiological damage occurs.

2.4 Materials and methods

2.4.1 Culture maintenance and genotyping The SM and MI heterogeneous Symbiodinium populations (aims-aten-C1-WSY and aims-aten-C1-MI, respectively) were provided by the Symbiont Culture Facility at the Australian Institute of Marine Science and are the same as reported in Howells et al. (2012). Following isolation from Acropora tenuis, the Symbiodinium populations were initially cultured in filtered seawater supplemented with Daigo IMK (Wako Pure Chemical Industries, Ltd.) and bacterial antibiotics for one month, which minimized the bacterial community to prevent bacterial overgrowth. Cultures were then routinely subcultured in media without antibiotics and monitored regularly by microscopy to ensure no increase in the remaining bacterial community was observed. Complete removal of all bacteria originating from the coral holobiont was not desirable, as optimal growth of dinoflagellate cultures has been shown to require associated bacteria (Alavi et al. 2001, Croft et al. 2005). Dinoflagellate cultures may depend on bacteria to provide necessary components, including but not limited to vitamin B12, in order to thrive (Croft et al. 2005, Ritchie 2012). Unlike the free-living life cycle stage in which Symbiodinium naturally live without a coral host (Granados-Cifuentes et al. 2015, Yamashita and Koike 2013), we are not aware of any stage in which Symbiodinium naturally live without associated bacteria. Therefore,

31 complete removal of Symbiodinium-associated bacteria may have unnatural effects on Symbiodinium transcriptomes.

For genotyping of the Symbiodinium populations, DNA from cultured cells in exponential growth phase was extracted using the DNeasy Plant Mini Kit (Qiagen). The ITS1 region was amplified with the PCR primers and conditions from van Oppen et al. (2001). The partial 5.8S rDNA, ITS2, and partial 28S rDNA region was amplified with the PCR primers and conditions from Stat et al. (2009). Purified PCR products were sequenced by the Australian Genome Research Facility. The ITS1 and ITS2 sequences of each population were reconfirmed to be type C1 and to be identical between the SM and MI populations, as previously reported by Howells et al. (2012). In future studies, alternative molecular markers such as the non-coding region from the psbA minicircle could be assessed through next generation sequencing to investigate finer-scale evolutionarily divergence that may exist between the two C1 populations as well as within each C1 population (LaJeunesse and Thornhill 2011). Increased genetic resolution may provide valuable insight into the different transcriptional responses to heat stress observed in our study.

2.4.2 Experimental setup Each population (~1 x 106 cells/ml, 50 ml total volume) was added to eight replicate culture flasks (n = 4 for each temperature treatment, Table S1). Two flasks per population were randomly assigned to each of four experimental incubators and acclimated at 27°C. Light was provided to cells at an intensity of 30 µmol quanta m-2 s-1 (Crompton 36W cool white fluorescent tubes, 4000 K) with a 12:12 hour light:dark cycle. After 10 days of acclimation, fresh media was supplied to the cultures. Following an additional four days of acclimation (two weeks of acclimation total), two incubators were ramped on day 0 at 0.5°C/hour to 32°C for the heat stress temperature treatment, while two incubators remained at 27°C for the control temperature treatment. Temperature and light intensity in the incubators were monitored with HOBO data loggers (Onset

32 Computer Corporation). Cultures remained in exponential growth phase, determined by the average of three replicate hemocytometer counts for each sample recorded throughout the experiment (Figure S1D).

2.4.3 Photosynthesis measurements A Mini-PAM fluorometer (Walz, Germany) was used to measure effective quantum yield (Figure S1A) and rapid light curves (RLCs) (Figure 1A-B, Figure S1B). RLCs are ideal for providing quick snapshots of Symbiodinium responses to a range of irradiances with results that are reasonably comparable to steady-state light curves (Suggett et al. 2015). With RLCs, < 90 seconds of exposure to high irradiances is applied per sample, which was short enough to avoid significant long-term damage that could greatly affect the physiology and gene expression of the Symbiodinium and allowed for the concurrent analysis of all 16 samples within an equivalent period of the light cycle on each measurement day.

A RLC protocol adapted from Ralph et al. (2002) was used in our study. After seven hours of light exposure, the fiber-optic cable of the Mini-PAM fluorometer was held against the bottom of each culture flask where the Symbiodinium cells had settled. Symbiodinium were exposed to nine steps of increasing actinic light (0 - 1,775 μmol m-2 s-1 PAR) for 10 seconds each, separated by a saturating pulse (0.8 seconds, > 4,000 μmol m-2 s-1 PAR). The light responses of each population at each temperature were determined by fitting the RLCs to the model by Platt et al. (1980). The variables rETRm, α, and Ek (Figure 1A-B, Figure S1B) were calculated using SigmaPlot as per Hill et al. (2004).

2.4.4 ROS measurements Cultures were gently agitated to evenly distribute cells in the media, and aliquots (300 µl per sample) were centrifuged at 3000 g x 5 minutes. Media (for measuring ROS leakage) was collected without disturbing the cell pellet and incubated with CellROX®

33 Orange reagent (5 µM, Thermo Fisher Scientific) for oxidative stress detection in a 96-well black clear bottom plate (Costar) for two hours at 27°C in the dark. CellROX® reagent is irreversibly converted to a fluorescent state in the presence of ROS without requiring the activity of intracellular esterases, making it an appropriate dye for measuring general ROS content in media. Fluorescence intensity of the CellROX® reagent was measured at excitation 540 nm and emission 565 nm with an EnSpire® Multimode Plate Reader (PerkinElmer). The use of CellROX® reagent with Symbiodinium culture media was validated through CellROX® reagent signal quenching from addition of antioxidant chemicals (Figure S4).

2.4.5 Culture viability measurements Culture viability was measured with SYTOX® Green nucleic acid stain (Life Technologies), which is unable to penetrate live Symbiodinium cells. Cultures were gently agitated to evenly distribute cells in the media, and aliquots (50 µl per sample) were incubated with SYTOX® Green nucleic acid stain (1 μM) in the dark for 15 minutes. An Olympus fv1000 confocal microscope with a 488 nm argon-ion laser was used to quantify the proportion of live cells in each sample based on counts of stained and unstained cells averaged across three separate fields of view (Figure S1C).

2.4.6 Statistical analysis of physiological measurements The PRIMER software with the PERMANOVA+ package was used to determine significant differences (p < 0.05) between temperature treatments for each physiological measurement using PERMANOVA with two replicate incubators as a nested factor within each level of the factor temperature (27°C and 32°C) and two flasks of each population in each incubator for each temperature. Where the effect of incubators was not significant (p > 0.2), the incubator factor was pooled, and each temperature treatment within each population (n = 4) was compared using a one-way PERMANOVA.

34 2.4.7 Preparation and sequencing of RNA samples Precisely after six hours of light exposure, cultures were gently agitated to evenly distribute cells in the media. Aliquots containing 2 - 4 x 106 cells per sample were immediately snap frozen in liquid nitrogen within 10 seconds of removal from the experimental incubators. Instant snap freezing of Symbiodinium cells while still in media (rather than the standard method of pelleting by centrifugation for 5 - 10 minutes, removing media, and then snap freezing (Baumgarten et al. 2013, Krueger et al. 2015, Rosic and Hoegh-Guldberg 2010)) caused no sign of cell lysis or loss of RNA integrity (Figure S5). We developed this method to ensure that the effects of experimental temperature treatments on gene expression remained unaltered during sample preservation because gene expression can be affected by centrifugation and extended handling (Baldi and Hatfield 2002). Our method is the only one of which we are aware to immediately preserve Symbiodinium RNA since the compatibility of RNAlater (Thermo Fisher Scientific) with Symbiodinium has not yet been validated. Samples were stored at - 80°C until completion of the heat stress experiment and were processed together on the same day to prevent batch effect.

Snap frozen cells were thawed at room temperature and pelleted at 4°C (3,000 g x 5 minutes). Media was removed, and pellets were lysed in buffer RLT (RNeasy Plant Mini Kit, Qiagen) containing β-mercaptoethanol by bead beating with 0.3 g of 710 - 1,180 µm acid-washed glass beads (Sigma) using a TissueLyser II (Qiagen) for 90 seconds at 30 Hz. RNA was then extracted and purified using the RNeasy Plant Mini Kit (Qiagen) with an added on-column DNase I treatment (Qiagen). Total RNA (150-500 ng) of each sample was sent to the Australian Genome Research Facility for confirmation of high quality RNA using an Agilent 2100 bioanalyzer, poly(A)-purification, Illumina TruSeq stranded library preparation, and sequencing with an Illumina HiSeq2500 (single end 100 bp, ~107 reads per sample, Table S2).

35 2.4.8 Transcriptome assembly and differential gene expression analysis Illumina Truseq (TruSeq3-SE) adapters were removed from RNA sequence reads using Trimmomatic (Bolger et al. 2014). Prinseq (Schmieder and Edwards 2011) was then used to remove poly(A) tails (min tail: 6-A) and to filter out short (min length: 60 bp), low quality (min mean quality score: 20, base window: 1, base step: 1), and low complexity sequences (dust method threshold: 7). The sequence reads for the 24 samples per population (four replicates, two temperature treatments, three time points) that remained after quality filtering were combined for de novo assembly of the SM population transcriptome and MI population transcriptome using Trinity (Grabherr et al. 2011a, Haas et al. 2013) (version: 2.0.6). Minimum transcript length for de novo assembly was set to 150 bp. To focus on transcripts with higher coverage, only transcripts ≥ 250 bp were retained for analysis, as in Baumgarten et al. (2013). Redundant transcripts (99% sequence similarity over 99% of the shorter transcript) in each de novo assembly were collapsed into the longest representative transcript using cd-hit-est (Huang et al. 2010) (Table S2). Completeness of the SM, MI, and other publicly accessible, published Symbiodnium transcriptomes (Baumgarten et al. 2013, Bayer et al. 2012, Ladner et al. 2012, Parkinson et al. 2016, Rosic et al. 2015, Xiang et al. 2015) was assessed using BUSCO with the set of 429 conserved eukaryotic orthologs that have been found to be present in > 90% of surveyed eukaryotic species (though the surveyed species currently lack protist representatives leading BUSCO to be biased towards lower metrics for protists than would otherwise be expected) (Simão et al. 2015) (Table S3). Non-redundant (nr) genes (transcript clusters determined by Trinity based on shared sequence content) were then analyzed for differential expression (fold ≥ 4 and FDR ≤ 0.001 between temperature treatments) according to the standard Trinity pipeline (Haas et al. 2013) (http://trinityrnaseq.github.io/) using RSEM (Li and Dewey 2011) and edgeR (Robinson et al. 2010). Additionally, the BCV of expression counts for all genes across replicates at each time point was separately calculated in edgeR according to Chen et al. (2014).

36 2.4.9 Annotation and GO analysis Transcriptomes were functionally annotated with Trinotate (http://trinotate.github.io/), using the SwissProt and UniRef90/TrEMBL databases (NCBI BLAST+, e-value ≤ 10-5) and the Pfam-A database (HMMR, domain noise cutoff). Top hits from SwissProt were used to annotate transcripts. If a hit was not generated against SwissProt, then the top hit from UniRef90/TrEMBL determined annotation. In the absence of a UniRef90/TrEMBL hit, Pfam-A annotation was used. GOseq (Young et al. 2010), which corrects for transcript length bias, was used as detailed with Trinity (http://trinityrnaseq.sourceforge.net/analysis/run_GOseq.html) for GO analysis (FDR < 0.05, ancestral terms included) of DEGs (fold ≥ 4 and FDR ≤ 0.001 between temperature treatments). SwissProt was used to assign GO categories to transcripts. In the absence of a SwissProt assignment, GO categories provided by Pfam-A were used.

In the SM and MI populations, 33% and 34% of genes received a hit from SwissProt, 46% and 49% of genes received a hit from UniRef90/TrEMBL, and 34% and 36% of genes received a hit from Pfam-A; respectively. In total, 50% of genes in the SM population and 52% of genes in the MI population received annotation from at least one database, and 35% of genes in the SM population and 36% of genes in the MI population were annotated with GO categories – similar to what has been previously reported for annotation of other Symbiodinium transcriptomes (Baumgarten et al. 2013, Parkinson et al. 2016, Rosic et al. 2015). Raw sequence reads, assembled transcriptomes, gene read count matrices, and transcript annotation results are available on NCBI GEO (GEO: GSE72763).

2.4.10 Isolation of a Fe-Sod gene The SM and MI populations were cultured for one week in sterile media with 300 μg/ml of ampicillin, followed by one week in sterile media with 300 μg/ml kanamycin, and finally for one week in sterile media with 100 μg/ml spectinomycin. Afterwards, Symbiodinium cells were pelleted (3,000 g x 5 minutes) and washed three times in sterile

37 media. Genomic DNA was extracted with a PureLink® Genomic DNA Mini Kit (Thermo Fisher Scientific). To confirm that the genomic DNA was of high quality and amplifiable, ITS2 PCR primers (Stat et al. 2011) were successfully used to amplify the ITS2 region from 25 ng of SM or MI genomic DNA. Primers for amplification of a full-length Symbiodinium Fe-Sod gene were based on TR20255|c0_g1 (open reading frame: 674-78[-]) from the MI population (forward: 5’ ATG GCC TTC TCC ATC CCA CCG 3’; reverse: 5’ TCA CAG GTT GGA CTC GGC GAA C 3’) and used for PCR reactions containing 125 ng of SM or MI genomic DNA (Figure S3A). The purified Fe-Sod PCR product that was amplified from the MI genomic DNA was sequenced by the Australian Genome Research Facility and confirmed to match TR20255|c0_g1 (open reading frame: 674-78[-]). The sequence of TR20255|c0_g1 (open reading frame: 674-78[-]) was aligned to the sequences of Symbiodinium Fe-Sod genes identified by Krueger et al. (2015) using ClustalW (Thompson et al. 2002). Alignments were visualized with UCSF Chimera (Pettersen et al. 2004) (Figure S3B).

2.5 Acknowledgements The Biomedical Imaging Facility at The University of New South Wales covered expenses for the use of confocal microscopes and plate readers. Iveta Slapetova provided technical support for imaging. Alexandra Campbell and Ezequiel Marzinelli advised on statistical analysis with PRIMER/PERMANOVA+. Duncan Smith provided technical support and training for the Katana computational cluster at The University of New South Wales. Brian Haas, Cristina Diez-Vives, Maxine Lim, Peter Davey, Cheong Xin Chan, and Zhiliang Chen provided technical support for RNA-Seq analysis. David Suggett provided valuable feedback on the writing of this manuscript. The Centre for Marine Bio-Innovation at The University of New South Wales, the Australian Institute of Marine Science, and the Linnean Society of New South Wales contributed financial support for this study. Raw and processed transcriptomics data are available through NCBI GEO (GSE72763).

38 2.6 Supplementary tables Table S1. Experimental design and sampling time points for transcriptomics.

Population Temperature Incubator Day -1 Day 9 Day 13 SM 27°C A 2 samples 2 samples 2 samples SM 27°C B 2 samples 2 samples 2 samples SM 32°C C 2 samples 2 samples 2 samples SM 32°C D 2 samples 2 samples 2 samples MI 27°C A 2 samples 2 samples 2 samples MI 27°C B 2 samples 2 samples 2 samples MI 32°C C 2 samples 2 samples 2 samples MI 32°C D 2 samples 2 samples 2 samples

Table S2. De novo assemblies of Symbiodinium population transcriptomes.

Nr assembled Average nr Filtered c d Read Assembled b Nr genes N50 transcript Population Raw reads a transcripts e reads transcripts (≥ 250 bp) (≥ 250 bp) length alignment (≥ 250 bp) (≥ 250 bp)

SM 238,256,872 236,965,564 205,494 131,066 106,097 1,253 858.1 88.0%

MI 235,479,270 234,027,212 183,989 116,479 93,377 1,353 911.4 89.6% a Number of reads retained after quality filtering of raw sequence reads b Number of non-redundant (nr) transcripts after removing transcripts < 250 bp and collapsing highly similar transcripts c Transcript (putative isoform) clusters based on shared sequence content d Weighted median statistic of transcript length (50% of the nr transcripts ≥ the N50 value) e Proportion of filtered reads that mapped back to the nr transcripts

Table S3. Completeness of publicly accessible, published Symbiodinium transcriptomes based on the presence of 429 eukaryotic benchmarking universal single-copy orthologs (BUSCOs) from OrthoDB (C: complete [D: duplicated], F: fragmented, M: missing). Asterisks mark transcriptomes that were generated in this study.

Symbiodinium type Reference C[D] F M A1 (S. microadriaticum) Baumgarten et al. (2013) 55%[38%] 13% 32% A2 Rosic et al. (2015) 16%[3%] 22% 62% A Bayer et al. (2012) 47%[8%] 14% 39% B1 (S. minutum) Bayer et al. (2012) 30%[8%] 19% 51% B1 (S. minutum) Parkinson et al. (2016) 65%[13%] 6% 29% B1 (S. pseudominutum) Parkinson et al. (2016) 59%[12%] 8% 33% B19 (S. psygmophilum) Parkinson et al. (2016) 66%[13%] 5% 29% B19 (S. aenigmaticum) Parkinson et al. (2016) 53%[10%] 11% 36% B2 Rosic et al. (2015) 14%[6%] 25% 61% B Xiang et al. (2015) 68%[16%] 4% 28% 39 C1 Rosic et al. (2015) 26%[5%] 24% 50% C1 (SM population)* Levin et al. (2016) 71%[32%] 7% 22% C1 (MI population)* Levin et al. (2016) 72%[30%] 5% 23% C3k Ladner et al. (2012) 16%[3%] 23% 61% D1 Rosic et al. (2015) 26%[5%] 29% 45% D2 Ladner et al. (2012) 26%[5%] 22% 52%

Table S4. Up-regulated (+) and down-regulated (-) meiosis, ROS scavenging, and molecular chaperone genes in the SM population after 9 days at 32°C (fold ≥ 4 and FDR ≤ 0.001 relative to 27°C).

Counts per Gene ID Gene annotation log (fold) FDR 2 million reads TR26080|c0_g1 MutS protein homolog 4 +2.52 6.17 1.95E-08 TR15578|c0_g1 MutS protein homolog 5 +2.24 14.79 8.26E-14 TR41849|c0_g2 Meiotic recombination protein Spo11-2 +2.32 1.77 2.06E-05 TR72293|c1_g1 Heat shock protein 83 +2.75 2.36 6.13E-09 TR47492|c0_g1 Cytochrome P450 +4.06 0.67 5.65E-05 TR63594|c0_g1 DnaJ homolog subfamily B member 6-B +2.73 1.44 4.21E-06 TR73294|c0_g1 DnaJ homolog subfamily C member 7 +3.08 2.51 2.88E-10 TR90535|c0_g1 Superoxide dismutase [Mn] -2.28 9.51 7.25E-05 TR107728|c0_g1 Catalase-peroxidase -4.09 0.58 3.33E-04 TR58725|c0_g2 Probable L-ascorbate peroxidase 4 -2.15 11.12 9.54E-04 TR67510|c0_g1 Glutathione peroxidase 2 -2.42 1.11 1.52E-04 TR48775|c0_g1 Peroxiredoxin-6 -3.55 3.23 2.22E-10 TR49236|c0_g1 Peroxiredoxin-2B -2.86 1.05 9.99E-05 TR65128|c0_g1 Peroxiredoxin-2D -2.56 2.63 1.59E-04 TR92045|c0_g1 Peroxiredoxin TSA1 -2.69 0.87 2.15E-04 TR77174|c0_g1 Alternative oxidase -2.26 4.41 4.24E-04 TR60729|c0_g2 Glutaredoxin -2.45 1.29 6.99E-05 TR106479|c0_g1 Monothiol glutaredoxin-S15 -4.12 1.07 8.36E-08 TR22361|c0_g1 Thioredoxin -2.50 1.52 2.62E-05 TR13040|c0_g1 Thioredoxin -2.06 1.79 2.44E-05 TR2965|c0_g1 Thioredoxin -5.06 0.46 4.61E-04 TR127388|c0_g1 Thioredoxin-1 -2.63 0.75 4.04E-04 TR48398|c0_g1 Thioredoxin domain-containing protein -3.12 1.18 3.17E-06 TR46153|c0_g1 Thioredoxin domain-containing protein -2.52 1.99 1.05E-05 TR49313|c0_g2 Cytochrome P450 -3.01 4.00 1.11E-06 TR9469|c0_g1 DnaJ homolog subfamily A member 4 -2.47 2.73 2.78E-07

40 Table S5. Up-regulated (+) and down-regulated (-) meiosis, ROS scavenging, and molecular chaperone genes in the MI population after 9 days at 32°C (fold ≥ 4 and FDR ≤ 0.001 relative to 27°C).

Counts per Gene ID Gene annotation log (fold) FDR 2 million reads TR3359|c0_g1 MutS protein homolog 4 +2.65 4.20 2.50E-14 TR34615|c0_g1 MutS protein homolog 5 +2.12 10.83 2.23E-19 TR47116|c0_g1 Meiotic recombination protein Spo11-2 +2.21 4.51 6.19E-11 TR57555|c0_g1 Molecular chaperone DnaJ +2.12 1.90 2.20E-07 TR49599|c0_g2 Superoxide dismutase [Cu-Zn] -2.40 2.13 5.69E-11 TR35071|c0_g1 Superoxide dismutase [Fe] -2.74 0.70 7.62E-04 Cytochrome c peroxidase, TR47713|c0_g1 -2.47 2.11 1.36E-06 mitochondrial TR18639|c0_g1 Peroxiredoxin-2 -2.25 1.48 2.39E-07 TR91822|c0_g1 Peroxiredoxin-2C -3.89 0.78 9.12E-06 TR99969|c0_g1 Peroxiredoxin-2F, mitochondrial -2.47 0.82 1.54E-04 TR61543|c0_g1 Thioredoxin peroxidase -2.55 1.24 1.26E-06 TR121144|c0_g1 Heat shock 70 kDa protein -2.08 4.56 8.52E-07 TR41580|c0_g1 Heat shock 70-related protein 1 -2.04 1.05 4.39E-04 TR44032|c0_g1 Heat shock protein SSA3 -2.48 1.38 3.79E-05 Heat shock 70 kDa protein, TR50028|c0_g1 -2.19 1.28 4.48E-05 mitochondrial TR58449|c0_g1 Heat shock 70 kDa protein 4 -2.53 1.32 2.58E-05 TR62486|c1_g1 Heat shock 70 kDa protein 4 -2.51 13.04 1.71E-09 TR71496|c0_g1 Heat shock cognate 70 kDa protein -2.49 1.11 1.49E-05 TR63145|c0_g1 Heat shock protein hsp88 -2.25 1.15 1.67E-04 TR56344|c0_g1 Heat shock-like 85 kDa protein -2.14 3.08 1.88E-04 TR59291|c0_g1 Heat shock-like 85 kDa protein -2.17 9.03 2.42E-05 TR64058|c0_g1 Heat shock protein 81-3 -2.17 3.61 6.56E-06 TR11461|c0_g1 Thioredoxin-1 -3.04 1.19 3.06E-07 Thioredoxin domain-containing protein TR9775|c0_g1 -3.33 0.76 1.28E-05 12 Thioredoxin domain-containing protein TR112961|c0_g1 -2.07 5.77 2.74E-06 5 TR25578|c0_g2 Thioredoxin-like protein 1 -2.64 0.68 6.37E-04 TR82304|c0_g1 Cytochrome P450 4V2 -2.99 0.66 5.16E-04 TR3328|c0_g2 Cytochrome P450 4V2 -3.85 1.64 1.27E-09 TR63302|c0_g1 Sterol 14-alpha demethylase -2.25 0.99 8.92E-05 TR16011|c0_g2 DnaJ homolog subfamily B member 6-A -4.05 0.59 9.15E-05 TR58329|c0_g1 DnaJ homolog subfamily A member 1 -2.40 1.51 6.79E-06 TR52403|c0_g1 DnaJ homolog subfamily A member 2 -2.58 2.14 1.31E-09 TR60716|c0_g1 DnaJ homolog subfamily A member 4 -2.92 0.76 9.61E-05

Table S6. Up-regulated (+) and down-regulated (-) meiosis, ROS scavenging, and molecular chaperone genes in the SM population after 13 days at 32°C (fold ≥ 4 and FDR ≤ 0.001 relative to 27°C).

Counts per Gene ID Gene annotation log (fold) FDR 2 million reads TR26080|c0_g1 MutS protein homolog 4 +3.69 8.22 6.95E-44 TR15578|c0_g1 MutS protein homolog 5 +2.18 14.06 1.96E-28 TR41849|c0_g1 Meiotic recombination protein Spo11-2 +4.27 0.73 8.85E-06 TR41849|c0_g2 Meiotic recombination protein Spo11-2 +3.19 2.69 3.47E-16 TR72002|c0_g4 Heat shock protein 105 kDa +2.04 5.61 9.77E-11 TR65031|c0_g1 Endoplasmin homolog +2.00 2.37 2.60E-07 TR58135|c0_g1 Heat shock protein 90-1 +4.93 0.47 4.82E-04 TR72293|c1_g1 Heat shock protein 83 +2.67 2.29 1.44E-08 TR56879|c0_g1 Glutaredoxin-C2 +2.21 1.03 3.30E-04 TR14920|c0_g1 Chaperone protein DnaJ +2.60 0.88 9.47E-05 TR63594|c0_g1 DnaJ homolog subfamily B member 6-B +4.05 1.27 4.09E-08 TR73294|c0_g1 DnaJ homolog subfamily C member 7 +2.96 2.22 5.56E-09 TR55386|c0_g2 DnaJ protein homolog +2.42 1.03 5.95E-05 TR90535|c0_g1 Superoxide dismutase [Mn] -2.24 3.18 5.61E-04 Probable phospholipid hydroperoxide TR61288|c0_g1 -2.71 1.03 4.24E-04 glutathione peroxidase 6 TR94914|c0_g1 Glutaredoxin-C4 -4.20 0.59 2.90E-04 TR8663|c0_g1 Peroxiredoxin Q, chloroplastic -2.48 0.77 3.63E-04 TR73087|c0_g1 Heat shock 70 kDa protein 14 -2.05 17.71 1.30E-04 TR1229|c0_g1 Heat shock 70 kDa protein -2.27 31.00 1.66E-05 TR92001|c0_g1 Heat shock 70 kDa protein -2.73 20.20 6.80E-06 TR126276|c0_g1 Heat shock protein SSA3 -3.20 24.45 9.35E-10 TR60729|c0_g2 Glutaredoxin -2.40 1.00 3.55E-05 TR33105|c0_g3 Glutaredoxin-1 -2.08 12.46 5.78E-27 TR25363|c0_g1 Glutaredoxin-1 -5.44 0.52 3.73E-04 Monothiol glutaredoxin-S15, TR106479|c0_g1 -3.99 0.55 1.73E-04 mitochondrial TR22361|c0_g1 Thioredoxin -3.11 0.65 1.06E-04 TR27587|c0_g2 Thioredoxin -3.32 1.69 7.15E-11 TR13040|c0_g1 Thioredoxin -3.41 1.03 2.76E-05 TR49951|c0_g1 Cytochrome P450 704C1 -3.33 0.71 8.99E-05 TR73168|c0_g1 Probable cytochrome P450 49a1 -2.92 0.94 1.57E-06 TR70469|c0_g1 Cytochrome P450 3A13 -2.08 1.13 1.72E-04 TR73331|c0_g1 Cytochrome P450 3A19 -2.82 2.25 5.98E-12

42 TR49313|c0_g2 Cytochrome P450 -2.34 1.18 5.08E-04 TR74009|c0_g1 DnaJ homolog subfamily B member 14 -3.15 2.06 7.91E-14 TR96409|c0_g1 Chaperone protein DnaJ -2.42 1.00 3.85E-04 TR56702|c1_g1 DnaJ homolog subfamily A member 1 -2.72 1.52 5.30E-05 TR56318|c0_g2 DnaJ homolog subfamily B member 5 -2.91 6.68 7.58E-11

Table S7. Up-regulated (+) and down-regulated (-) meiosis, ROS scavenging, and molecular chaperone genes in the MI population after 13 days at 32°C (fold ≥ 4 and FDR ≤ 0.001 relative to 27°C).

Counts per Gene ID Gene annotation log (fold) FDR 2 million reads TR3359|c0_g1 MutS protein homolog 4 +2.44 4.26 4.43E-14 TR47116|c0_g1 Meiotic recombination protein Spo11-2 +2.07 5.02 1.95E-09 TR60336|c0_g1 Superoxide dismutase [Fe] +2.03 1.37 2.70E-05 TR20255|c0_g1 Superoxide dismutase [Fe] +2.35 3.02 1.14E-08 TR53519|c0_g1 Superoxide dismutase [Fe] +3.15 0.76 7.18E-05 Cytochrome c peroxidase, TR46907|c0_g1 +2.24 1.60 3.15E-06 mitochondrial Probable phospholipid hydroperoxide TR55391|c0_g1 +4.01 0.63 6.15E-05 glutathione peroxidase TR31467|c0_g1 Peroxiredoxin-2 +3.12 1.05 2.08E-06 TR48822|c0_g1 Peroxiredoxin +2.20 1.40 4.56E-04 TR61543|c0_g1 Thioredoxin peroxidase +2.44 6.55 8.22E-15 TR40730|c0_g1 Heat shock cognate 70 kDa protein 1 +5.47 0.58 5.81E-05 TR41580|c0_g1 Heat shock 70-related protein 1 +2.59 4.33 3.24E-08 TR43532|c0_g1 Heat shock cognate 70 kDa protein 3 +3.75 0.56 7.31E-04 TR44032|c0_g1 Heat shock protein SSA3 +2.33 4.74 1.33E-12 Heat shock 70 kDa protein, TR50028|c0_g1 +2.55 6.64 3.41E-20 mitochondrial TR59506|c0_g1 Hsp70 protein +2.69 1.68 1.09E-06 TR61972|c0_g1 Heat shock cognate 71 kDa protein +2.52 15.11 1.32E-22 Heat shock 70 kDa protein, TR62486|c0_g3 +2.04 7.48 2.49E-12 mitochondrial TR62486|c1_g1 Heat shock 70 kDa protein 4 +2.64 71.80 1.39E-16 TR63145|c0_g1 Heat shock protein hsp88 +2.39 4.43 1.38E-14 TR89622|c0_g1 Heat shock cognate 70 kDa protein 3 +3.58 0.73 3.98E-04 TR56344|c0_g1 Heat shock-like 85 kDa protein +2.32 11.17 3.36E-15 TR59291|c0_g1 Heat shock-like 85 kDa protein +2.24 32.54 6.03E-16 TR64058|c0_g1 Heat shock protein 81-3 +2.51 15.46 1.11E-25 TNF receptor-associated protein 1 TR110286|c0_g1 +2.70 2.38 1.57E-09 homolog, mitochondrial TR59710|c0_g1 Thioredoxin-1 +2.29 0.92 3.76E-04

43 TR51677|c0_g1 Thioredoxin-1 +2.19 1.22 9.42E-06 Probable cytochrome P450 12c1, TR62574|c0_g1 +2.18 1.56 3.90E-06 mitochondrial TR50785|c0_g1 Chaperone protein DnaJ +2.73 1.24 1.24E-06 TR16011|c0_g2 DnaJ homolog subfamily B member 6-A +2.02 1.21 2.38E-04 TR80199|c0_g1 DnaJ homolog subfamily A member 2 +2.57 1.68 1.11E-06 TR61668|c1_g2 DnaJ homolog subfamily A member 2 +2.64 1.97 5.65E-07 TR51427|c0_g1 DnaJ homolog subfamily A member 4 +2.66 1.19 9.90E-05 TR60716|c0_g1 DnaJ homolog subfamily A member 4 +2.25 1.79 3.88E-06 TR40249|c0_g1 DnaJ homolog subfamily C member 3 +2.92 1.89 6.65E-06 DnaJ homolog subfamily C member 7 TR10977|c0_g1 +2.85 0.93 1.17E-05 homolog TR47513|c0_g1 DnaJ protein homolog 2 +2.89 0.95 7.80E-05 TR49599|c0_g2 Superoxide dismutase [Cu-Zn] -4.08 0.90 4.39E-04 Cytochrome c peroxidase, TR47713|c0_g1 -2.89 1.05 1.01E-05 mitochondrial Cytochrome c peroxidase, TR899|c0_g1 -3.28 0.79 2.64E-04 mitochondrial TR54503|c0_g1 Glutathione peroxidase 1 -4.37 0.72 1.92E-05 TR18639|c0_g1 Peroxiredoxin-2 -3.96 0.86 4.00E-04 TR26880|c0_g1 Heat shock protein 88 -3.22 0.93 2.84E-04 TR4879|c0_g1 Heat shock protein 82 -2.23 2.08 2.88E-05 TR56256|c0_g1 Heat shock cognate 90 kDa protein -2.90 2.88 3.05E-05 Thioredoxin domain-containing protein TR112961|c0_g1 -2.38 2.91 6.38E-05 5 TR98120|c0_g1 DnaJ protein P58IPK homolog -4.22 0.67 2.34E-05 TR52403|c0_g1 DnaJ homolog subfamily A member 2 -2.96 1.09 2.34E-05

44 2.7 Supplementary figures

Figure S1. Additional physiological measurements of Symbiodinium exposed to heat stress. Intact lines represent the 27°C temperature treatment, and dashed lines represent the 32°C temperature treatment. Before heating, all samples were kept at 27°C (values in the grey regions). (A) ΔF/Fm’ (mean ± s.e.m., n = 4). (B) Ek (mean ± s.e.m., n = 4). (C) Culture viability (mean ± s.e.m., n = 4). (D) Cell density (mean ± s.e.m., n = 4). No statistically significant (PERMANOVA) differences were detected between temperature treatments at p < 0.05. 45

Figure S2. Regulation of additional meiosis, ROS scavenging, and molecular chaperone genes. (A) The number of genes for gene types involved in sexual reproduction or thermal tolerance in the SM and MI transcriptomes. The number of DEGs from each gene type at 32°C (Trinity/edgeR: fold ≥ 4 and FDR ≤ 0.001 relative to 27°C) are shown for each population on (B) day 9 and (C) day 13. Gene types that had no DEGs in either population are excluded from (B) and (C). Gene abbreviations are as follows: meiotic recombination protein Spo11-1 (Spo11-1), alternative oxidase (Aox), glutaredoxin (Glrx), thioredoxin (Txn), cytochrome P450 (Cyp450), chaperone protein DnaJ (DnaJ). DEG annotation and differential expression details are provided in Tables S4-7.

Figure S3. Presence and sequence similarity of a Fe-Sod gene in type C1 Symbiodinium. (A) ITS2 and Fe-Sod PCR products amplified from purified SM or MI genomic DNA were imaged after electrophoresis in a 1% agarose gel. (B) Nucleotide sequence alignment with ClustalW (Thompson et al. 2002) revealed 47-61% identity between a full-length Fe-Sod gene from the MI population (TR20255|c0_g1, open reading frame: 674-78[-]) and partial- length Fe-Sod genes identified by Krueger et al. (2015) in types B1, E, and F1 Symbiodinium. Alignments were visualized with UCSF Chimera (Pettersen et al. 2004). Consensus bases across all sequences are red and capitalized. The degree of conservation for each base across all sequences is represented by grey bar height.

Figure S4. Validation of CellROX® Orange reagent usage with Symbiodinium culture media. The SM population was kept at 27°C for six days with a 12:12 hour light:dark cycle (50 µmol quanta m-2 s-1) for control samples or heated at 32°C for six days with constant light (100 µmol quanta m-2 s-1) for stressed samples. 2 mM DL-α lipoic acid (Calbiochem) or 100 µM ebselen (Sigma) was added to samples one hour before incubating media with CellROX® reagent. Asterisks indicate statistically significant (ANOVA) differences between experimental samples (stressed with or without ROS inhibitors; mean ± s.e.m, n = 3) and control samples (mean ± s.e.m, n = 3) at p < 0.05.

Figure S5. No detected degradation of Symbiodinium immediately snap frozen in media. Cells from the SM population were imaged on an Olympus fv1000 confocal microscope with the 60x objective, and extracted RNA was run on an Agilent 2100 bioanalyzer. (A) Live cells and a representative bioanalyzer report of RNA integrity. (B) Thawed cells from pelleting culture aliquots by centrifugation, removing media, and then snap freezing cells (~10 minute process) and a representative bioanalyzer report of RNA integrity. (C) Thawed cells that had been immediately snap frozen in media (within 10 seconds of removal from the experimental incubator) and a representative bioanalyzer report of RNA integrity. No sign of cell lysis or loss of RNA integrity between RNA preservation methods was observed. 49 2.8 Supplementary dataset Dataset S1A-M. Enriched (+) and depleted (-) GO categories (FDR < 0.05) for biological processes (BP), molecular functions (MF), and cellular components (CC) in the subsets of deregulated genes in each Symbiodinium population after 9 and 13 days at 32°C: https://academic.oup.com/mbe/article/33/9/2201/2579440/Sex-Scavengers-and- Chaperones-Transcriptome#supplementary-data

50 Chapter 3 Evidence for a role of viruses in the thermal sensitivity of coral photosymbionts

Publication II Levin RA, Voolstra CR, Weynberg KD, van Oppen MJH (2017a). Evidence for a role of viruses in the thermal sensitivity of coral photosymbionts. The ISME Journal 11: 808-812.

Declaration I certify that this publication was a direct result of my research towards this PhD and that reproduction in this thesis does not breach copyright regulations.

...... Rachel A. Levin

51 3.1 Abstract Symbiodinium, the dinoflagellate photosymbiont of corals, is posited to become more susceptible to viral infections when heat-stressed. To investigate this hypothesis, we mined transcriptome data of a thermo-sensitive and a thermo-tolerant type C1 Symbiodinium population at ambient (27°C) and elevated (32°C) temperatures. We uncovered hundreds of transcripts from nucleocytoplasmic large double-stranded DNA viruses (NCLDVs) and the genome of a novel positive-sense single-stranded RNA virus (+ssRNAV). In the transcriptome of the thermo-sensitive population only, +ssRNAV transcripts had remarkable expression levels in the top 0.03% of all transcripts at 27°C, but at 32°C, expression levels of +ssRNAV transcripts decreased while expression levels of antiviral transcripts increased. In both transcriptomes, expression of NCLDV transcripts increased at 32°C, but thermal-induction of NCLDV transcripts involved in DNA manipulation was restricted to the thermo-sensitive population. Our findings reveal that viruses infecting Symbiodinium are affected by heat stress and may contribute to Symbiodinium thermal sensitivity.

52 3.2 Main text Tropical reef-building corals form a multi-partite symbiosis with the photosynthetic dinoflagellate Symbiodinium and a diverse microbial community that includes bacteria, archaea, fungi, protists, and viruses; collectively termed the coral holobiont (Rohwer et al. 2002). The Symbiodinium-coral symbiosis can be disrupted by heat stress, which results in the loss of Symbiodinium cells from coral tissues, i.e., coral bleaching (Hoegh-Guldberg 1999). Heat stress has been alleged to promote lytic viral infections of Symbiodinium, as virus-like particles have been found in heat-stressed Symbiodinium from the temperate sea anemone Anemonia viridis (Wilson et al. 2001) and from the corals Pavona danai, Acropora formosa, and Stylophora pistillata (Davy et al. 2006, Wilson et al. 2005). Additionally, following heat shock at 31°C, expression of a protein with homology to a eukaryotic viral protein increased > 100-fold in a Symbiodinium-enriched fraction of Stylophora pistillata tissue (Weston et al. 2012).

Using transcriptome data generated by Levin et al. (2016) for two type C1 Symbiodinium populations cultured at 27°C and 32°C (n = 4), we explore the effect of heat stress on viruses associated with Symbiodinium and Symbiodinium antiviral responses. The Symbiodinium populations were originally isolated from the coral Acropora tenuis at South Molle Island (SM) and Magnetic Island (MI) (Great Barrier Reef, Australia). The thermo- sensitive SM population was found to suffer physiological damage in culture and to bleach in hospite at 32°C, while the thermo-tolerant MI population was unaffected (Howells et al. 2012, Levin et al. 2016).

Since many viral RNAs are polyadenylated (Priet et al. 2015, Wilson et al. 2000), they were retained in the poly(A)+ purified Symbiodinium RNA samples used for RNA-Seq (Levin et al. 2016). Viral transcripts in each de novo transcriptome were identified through a robust BLASTx bit score approach adapted from Boschetti et al. (2012), which calculates the difference between the highest viral and the highest non-viral bit score to determine if a transcript is from a virus or the host, followed by GC content, tri-nucleotide frequency,

53 and codon usage analyses (Supplementary materials and methods, Figure S1, Datasets 1- S2). Symbiodinium antiviral transcripts were identified by searching the transcriptomes for transcripts encoding antiviral gene types found in corals and for transcripts with antiviral gene ontology (Supplementary materials and methods).

The thermo-sensitive SM and thermo-tolerant MI transcriptomes contain 306 and 238 viral transcripts (Supplementary discussion), as well as 62 and 65 antiviral transcripts, respectively (NCBI GEO accession: GSE77911). In both transcriptomes, viral transcripts show homology to genes from NCLDVs Mimiviridae and Phycodnaviridae and the major capsid protein (MCP) gene from the dinoflagellate-specific +ssRNAV Alvernaviridae (Dinornavirus), which is in agreement with previous sequencing and transmission electron microscopy findings (Correa et al. 2016, Correa et al. 2013a). The +ssRNAV transcripts in the thermo-sensitive SM transcriptome share 94% nucleotide (nt) identity (TR74740|c13_g1_i1, 5202 nt; TR74740|c13_g1_i2, 2154 nt). The +ssRNAV transcript in the thermo-tolerant MI transcriptome is much shorter (TR97578|c0_g1_i1, 475 nt) but shares 100% nt identity with TR74740|c13_g1_i1. Successful PCR amplification of the MCP genes from complementary DNA reverse-transcribed from RNA - but not from genomic DNA - of both Symbiodinium populations supports that they are from +ssRNAVs (Supplementary materials and methods, Figure S2). However, phylogenetic analysis of the translated MCP gene sequences revealed that they are highly divergent from the Dinornavirus MCP gene and previously identified partial-length Dinornavirus-like MCP genes (Supplementary discussion, Table S1, Figure S3).

Both +ssRNAV transcripts in the thermo-sensitive SM transcriptome contain a putative viral internal ribosomal entry site (IRES; Figure S4), which is related to the IRES of the +ssRNA cricket paralysis virus, directly upstream from the MCP gene. The longer +ssRNAV transcript, TR74740|c13_g1_i1, also encodes an un-annotated open reading frame (ORF) determined to be a +ssRNAV RNA replicase (RNA-dependent RNA- polymerase) polyprotein based on protein structure modelling (Figure S5). The RNA

54 replicase gene precedes the IRES and MCP gene, giving the full transcript a markedly similar arrangement to the Dinornavirus and cricket paralysis virus complete genomes (Nagasaki et al. 2005, Wilson et al. 2000). Thus, we conclude TR74740|c13_g1_i1 to be the RNA genome of a novel +ssRNAV, making this the first discovered genome of any virus infecting Symbiodinium (Figure 1A). Surprisingly, the conserved dinoflagellate spliced leader (dinoSL), which is present on > 95% of Symbiodinium mRNAs (Zhang et al. 2013), is at the 5’ end of TR74740|c13_g1_i1. Although, if the +ssRNAV is dinoflagellate-specific like Dinornavirus, incorporation of the dinoSL in the viral RNA genome is likely a case of molecular mimicry, a well-documented viral strategy to evade host immune responses that detect foreign nucleic acids (Elde and Malik 2009), or for efficient cap-dependent translation by host polysomes (Zeiner et al. 2003).

Figure 1. Genome of the novel +ssRNAV and its expression in Symbiodinium transcriptomes. (A) Genome model of the +ssRNAV infecting Symbiodinium (TR74740|c13_g1_i1, GenBank accession: KX538960) and its similarities to the RNA genomes of the dinoflagellate-specific Dinornavirus (Heterocapsa circularisquama RNA virus strain 34, NCBI accession: AB218608) and cricket paralysis virus (NCBI accession: AF218039). (B) TMM-normalized FPKM at 27°C (averaged across replicates for each transcript on day -1, n = 8) for the +ssRNAV transcripts (TR74740|c13_g1_i1, 5757 FPKM; TR74740|c13_g1_i2, 1996 FPKM; TR97578|c0_g1_i1, 1 FPKM), NCLDV transcripts, and non-viral transcripts in the thermo-sensitive SM and thermo-tolerant MI transcriptomes. Black lines mark the mean FPKM + standard deviation for each subset of transcripts.

Viral transcripts had similar expression levels to many non-viral transcripts at 27°C with the exception of the two +ssRNAV transcripts in the thermo-sensitive SM transcriptome (Figure 1B). Both +ssRNAV transcripts in the thermo-sensitive SM transcriptome maintained average expression levels > 1,300 fragments per kb of transcript per million mapped reads (FPKM) on all sampling time points at 27°C, whereas the +ssRNAV transcript in the thermo-tolerant MI transcriptome had an average expression level <2 FPKM on all sampling time points. The vastly dissimilar expression levels of +ssRNAV transcripts between the transcriptomes suggest that the thermo- sensitive SM Symbiodinium population was experiencing a severe viral infection.

Differential expression analysis (Supplementary materials and methods) confirmed that no viral or antiviral transcripts were differentially expressed between experimental groups of either population on day -1 (pre-heating; all samples acclimated at 27°C) (Figure 2A-E). In both transcriptomes, up-regulation of NCLDV transcripts was induced at 32°C and increased from day 9 to day 13 (Figure 2B and D, Tables S2-9). However, the up- regulated NCLDV transcripts in the thermo-sensitive SM transcriptome encoded for a greater diversity of genes (F-box and FNIP repeat-containing proteins, resolvase, transposase, ankyrin repeat protein) compared to those in the thermo-tolerant MI transcriptome (only F-box and FNIP repeat-containing proteins). Viral F-box and FNIP repeat-containing proteins and ankyrin repeat proteins play possible roles in degrading proteins through protein-protein interactions, countering host defences, and exploiting the host's ubiquitin-proteasome system to create an appropriate cellular environment for viral replication (Correa et al. 2013b, Fischer et al. 2010, Sonnberg et al. 2008, Suhre 2005); whereas resolvases and transposases directly participate in DNA manipulation (Iyer et al. 2001, Schroeder et al. 2009).

In the thermo-sensitive SM transcriptome, heat stress resulted in down-regulation of the highly expressed +ssRNAV transcripts and up-regulation of antiviral transcripts (Figure 2A and C, Tables S2-3 and S6-7). Conversely, the thermo-tolerant MI transcriptome

57 showed no differential expression of its lowly expressed +ssRNAV transcript and down- regulation of antiviral transcripts at 32°C (Figure 2E, Tables S8-9). Symbiodinium antiviral responses may therefore become activated from increased DNA manipulation by NCLDVs or initial up-regulation of highly expressed +ssRNAV transcripts at 32°C that was mitigated by the sampling time points on days 9 and 13.

Figure 2. Viral infections and antiviral responses of Symbiodinium under heat stress. Differential expression at 32°C of (A) thermo-sensitive SM transcriptome +ssRNAV transcripts, (B) thermo-sensitive SM transcriptome NCLDV transcripts, (C) thermo- sensitive SM transcriptome antiviral transcripts, (D) thermo-tolerant MI transcriptome NCLDV transcripts, and (E) thermo-tolerant MI transcriptome antiviral transcripts. Only transcripts with ≥2-fold expression between 27°C and 32°C treatments on at least one sampling time point were analyzed. Transcripts were considered to have no differential expression on a sampling time point where the false discovery rate (FDR) was > 0.001. Up- regulated transcripts at 32°C are shown in yellow. Down-regulated transcripts at 32°C are shown in purple. The grey regions represent the pre-heating sampling time point on day - 1 when all replicates were still at 27°C. On day 0, replicates were ramped to 32°C (n = 4) or maintained at 27°C (n = 4) for the duration of the study (Supplementary materials and methods). Differential expression and transcript annotation results are detailed in Tables S2-9. 59

Our study exemplifies how RNA-Seq data can be utilized to gain valuable insight into resident viruses. Our results indicate that only the thermo-sensitive SM Symbiodinium population experienced an extreme +ssRNAV infection and thermally induced NCLDV DNA manipulation. Thus, viral infections may factor into Symbiodinium thermal sensitivity, and consequently, coral bleaching.

3.3 Acknowledgements The Australian Institute of Marine Science supplied the Symbiodinium strains (aims-aten-C1-WSY, aims-aten-C1-MI) used in this study. Yi Jin Liew provided assistance with the bit score analysis. Rhys T. Graham provided technical support for MATLAB. The Centre for Marine Bio-Innovation at The University of New South Wales, the Joyce W. Vickery Scientific Research Fund Grant awarded to Rachel A. Levin from the Linnean Society of New South Wales, and Future Fellowship #FT100100088 awarded to Madeleine J. H. van Oppen from the Australian Research Council contributed financial support. Raw, processed, and annotated sequencing data are available through NCBI GEO (accession: GSE77911). The novel +ssRNAV genome (TR74740|c13_g1_i1) and highly related, partial +ssRNAV genome (TR74740|c13_g1_i2) discovered in this study have been deposited in GenBank (accession: KX538960 and KX787934, poly(A) tails were trimmed from RNA-Seq reads prior to de novo assembly of transcripts).

3.4 Supplementary materials and methods

3.4.1 Generation of transcriptomes from type C1 Symbiodinium populations We queried transcriptome data of a thermo-sensitive type C1 Symbiodinium population (SM, aims-aten-C1-WSY) and a thermo-tolerant type C1 Symbiodinium population (MI, aims-aten-C1-MI) exposed to 27°C and 32°C in culture. Culture conditions, genotyping, experimental design, RNA extraction, transcriptome assembly, and transcriptome annotation are detailed by Levin et al. (2016). Briefly, samples were

60 collected for transcriptome sequencing on day -1 (preheating; all samples acclimated to 27°C, n = 8 per population). On day 0, four samples per population were ramped at 0.5°C/hour to 32°C for the heat stress temperature treatment, while four samples per population remained at 27°C for the ambient temperature treatment. Samples (n = 4 per population per temperature treatment) were collected for transcriptome sequencing on days 9 and 13. The sampling time points for transcriptome sequencing were determined by physiological measurements that detected thermal damage (reduced photosynthetic efficiency and increased reactive oxygen species leakage from cells) to only the thermo- sensitive SM population (Levin et al. 2016). Extracted RNA from each sampling time point was poly(A)+ purified, sequenced with an Illumina HiSeq2500 (100-bp single-end, strand- specific reads; ~1 x 107 reads per sample), and assembled with Trinity version 2.0.6 (Grabherr et al. 2011b, Haas et al. 2013) into de novo transcriptomes for each population. Transcripts in each de novo transcriptome that shared 99% sequence identity were collapsed into the longest representative transcript using cd-hit-est (Huang et al. 2010) to remove redundancy. Transcriptomes were annotated with BLAST+ against SwissProt (e- value ≤ 10-5), UniRef90/TrEMBL (e-value ≤ 10-5), and Pfam-A databases (HMMR, domain noise cut-off). In the absence of a BLAST+ hit in SwissProt, annotation from UniRef90/TrEMBL was used, followed by Pfam-A.

3.4.2 Identification of viral and antiviral transcripts A BLASTx bit score approach adapted from (Boschetti et al. 2012) was performed to measure “foreignness” of transcripts for determining viral origin. Assembled transcripts from the de novo transcriptomes of the thermo-sensitive SM and thermo-tolerant MI populations that received top BLAST+ hits to genes from eukaryotic viruses were aligned with BLASTx against partitioned SwissProt (curated) databases for archaea, bacteria, eukaryotes, and viruses. For eukaryotic virus transcripts that only received BLAST+ hits in UniRef90/TrEMBL, partitioned UniRef90/TrEMBL (non-curated) databases were used instead. For each sequence, the difference in bit score between the top BLASTx hit from the viral database and the top BLASTx hit from any non-viral database (h) was calculated.

61 Only sequences with h ≥ 30 (Boschetti et al. 2012) and a bit score ≥50 (Roux et al. 2011, Weynberg et al. 2014, Wood-Charlson et al. 2015) were considered to be of viral origin. The threshold of 30 was set for h as it provides the optimal trade-off between sensitivity and specificity in the detection of foreign genes (Boschetti et al. 2012). To account for Symbiodinium genes that may be missing from the SwissProt or UniRef90/TrEMBL databases as well as to remove viral transcripts that have been integrated into the Symbiodinium genome from horizontal gene transfer events (keep only those that are from ongoing viral infections), BLASTx was run against both published Symbiodinium genomes (types B1 and F1) (Lin et al. 2015, Shoguchi et al. 2013). Again, only viral sequences with h ≥ 30 were retained.

GC content and tri-nucleotide sequence statistics (http://www.genomatix.de/cgi- bin/tools/tools.pl, last accessed July 2016) for the subset of viral transcripts and subset of non-viral transcripts in each transcriptome were converted from raw counts to frequencies (% occurrence) for subset comparisons (Dataset S1). The program TransDecoder (http://transdecoder.github.io/, last accessed July 2016) was used to extract ORFs from the subset of viral transcripts and subset of non-viral (host) transcripts in each transcriptome. The EMBOSS tool CUSP (Rice et al. 2000) was used to analyze codon usage frequencies of the ORFs in each subset (Dataset S2). Divergences in tri- nucleotide frequency matrices and codon usage matrices between subsets were determined by calculating the correlation coefficient for each matrix comparison in MATLAB (MathWorks, Inc.) (Figure S1).

Symbiodinium antiviral transcripts were identified by searching the annotation results for transcripts encoding known antiviral genes found in corals (Dunlap et al. 2013) and for transcripts assigned gene ontology categories for defense response to virus, negative regulation of viral genome, and negative regulation of viral genome replication. gene ontology categories were assigned based on SwissProt annotation. In the absence of SwissProt gene ontology assignment, Pfam-A gene ontology assignment was used.

62 3.4.3 Differential expression analysis Differential expression analysis for all transcripts in each population was conducted with RSEM (Li and Dewey 2011) and edgeR (Chen et al. 2014) as instructed in the standard Trinity pipeline (https://github.com/trinityrnaseq/trinityrnaseq/wiki/Trinity- Differential-Expression, last accessed July 2016) (Grabherr et al. 2011b, Haas et al. 2013). Only transcripts with at least 2-fold differential expression at a maximum FDR of 0.001 between temperature treatments on at least one sampling time point were considered. Transcripts were not collapsed into Trinity ‘genes’ (transcript clusters based on shared sequence content) as not all transcripts in the same cluster always received annotation despite one or more of the transcripts in the same cluster aligning to a known viral gene - likely due to the limited amount of marine virus sequence information in databases (Brum and Sullivan 2015, Wood-Charlson et al. 2015). Transcript annotation and bit score was required for the HGT index approach.

3.4.4 PCR amplification of the +ssRNAV MCP gene Total RNA was extracted and purified by Levin et al. (2016) using the RNeasy Plant Minikit (Qiagen) with an added on-column DNase treatment (Qiagen) and stored at -80°C. RNA was reverse-transcribed into complementary DNA (cDNA) with the High-Capacity cDNA Reverse Transcription Kit (Thermo Fisher Scientific). To obtain purified genomic DNA (gDNA), Symbiodinium cells (~5 x 106) from the thermo-sensitive SM and thermo-tolerant MI populations were washed with sterile media and then processed with the PowerPlant® Pro DNA Isolation Kit (MO BIO), which includes RNase-treatment of gDNA. Using ITS2 PCR primers from Stat et al. (2011), the ITS2 region was successfully amplified from the gDNA of the thermo-sensitive SM and thermo-tolerant MI populations, verifying that the gDNA was high quality and amplifiable.

MCP primer set 1 was designed for a long region of the MCP genes (970 nt) based on conserved sites in TR74740|c13_g1_i1 and TR74740|c13_g1_i2 (1F: 5’ GCA CTT ACA AGG AAG CAG CGA GC 3’, 1R: 5’ CAT GAG CTC CAG CCT CCA ATG 3’). MCP primer set 2 was

63 designed for a short region of the MCP genes (128 nt) based on conserved sites in TR74740|c13_g1_i1, TR74740|c13_g1_i2, and TR97578|c0_g1_i1 (2F: 5’ TAT CAC CCG AGG TCC GTT TGG TTC 3’, 2R: 5’ TGA CTC TTG TCG TCC GAA TGA CTG TG 3’). Phusion® High-Fidelity PCR Master Mix with HF Buffer (New England BioLabs) was used as directed for 35 rounds of PCR with a primer annealing temperature of 63°C to amplify the long and short regions of the MCP genes from gDNA and cDNA of the thermo-sensitive SM and thermo-tolerant MI populations. Both primer sets successfully amplified the MCP genes from only cDNA (Figure S2A). PCR products were sequenced by the Australian Genome Research Facility and confirmed to be from the MCP genes.

3.5 Supplementary discussion Viral transcript diversity in Symbiodinium transcriptomes The first discovery of potential sequences from pathogenic viruses (including Mimiviridae, Phycodnaviridae, Poxviridae, Iridoviridae, Baculoviridae, and Herpesviridae) in corals indicated that a diverse group of viruses infect the coral holobiont (Wegley et al. 2007). Transcriptionally up-regulated viral A-type inclusion genes in heat stressed Montastraea faveolata and transcriptionally up-regulated transposon, pol-like protein, and reverse transcriptase genes in bleached Montastraea faveolata (DeSalvo et al. 2008) suggested that coral-associated viruses are responsive to heat stress. Over 900 transcripts in each of our type C1 Symbiodinium transcriptomes received a top BLAST+ hit from at least one database to a gene from a eukaryotic virus. Similarly to findings in corals (Correa et al. 2016, Wegley et al. 2007), the transcripts shared homology with genes from Mimiviridae, Phycodnaviridae, Poxviridae, Iridoviridae, Marseilleviridae, Asfarviridae, Baculoviridae, Herpesviridae, Polyomaviridae, Retroviridae, and Alvernaviridae (Dinornavirus). However, the majority of these transcripts were not sufficiently divergent from non-viral sequences and Symbiodinium genomic sequences (h ≥ 30) (Boschetti et al. 2012) to confidently assign them as true viral transcripts.

64 Following the BLASTx bit score approach (Supplementary materials and methods), 306 and 238 transcripts were concluded to be from viruses infecting the thermo-sensitive SM and thermo-tolerant MI populations, respectively. Viral transcripts expressed in both transcriptomes shared BLASTx homology with Mimiviridae, Phycodnaviridae, and Dinornavirus genes, though dominantly (> 80%) with Mimiviridae FNIP repeat-containing genes. The diverse transcripts containing FNIP repeats had unknown functions since leucine-rich repeats (e.g., FNIP repeats) are common protein recognition motifs found in a broad variety of proteins (Kobe and Kajava 2001). None of the FNIP repeat-containing transcripts within each transcriptome were identical. Furthermore, we used USEARCH (Edgar 2010) with a minimum identity cut-off of 80% to collapse similar sequences and found reductions of only 254 to 172 and 211 to 155 transcripts in the thermo-sensitive SM and thermo-tolerant MI transcriptomes, respectively.

Genes containing FNIP repeats are prevalent in the genomes of Mimivirus and its relative Cafeteria roenbergensis virus (Fischer et al. 2010, Suhre 2005), thus an abundance of viral FNIP repeat-containing transcripts in our transcriptomes is not entirely surprising. Furthermore, the relatively high proportion of viral FNIP repeat-containing transcripts compared to other Mimiviridae-like transcripts may be overestimated as a result of the limited amount of marine virus sequence information in databases (Brum and Sullivan 2015, Wood-Charlson et al. 2015) preventing annotation of all viral transcripts in the transcriptomes and/or an inability to confirm viral origin of many virally annotated transcripts in the transcriptomes. Phylogenetic analysis strongly indicates that a large number of genes were acquired by Mimivirus from its hosts as well as from bacteria infecting the same hosts (Moreira and Brochier-Armanet 2008). Consequently, many Mimiviridae-like transcripts in our transcriptomes may not be sufficiently divergent (h ≥ 30) from non-viral sequences and therefore would not have been retained for further analysis. Moreover, although all Mimivirus genes are polyadenylated (Byrne et al. 2009), certain dsDNA marine viruses contain a mix of poly(A)+ and poly(A)- genes (Tsai et al. 2004). Accordingly, the breadth of gene types that can be retained through poly(A)+

65 purification cannot be determined without knowledge of the polyadenylation rate exhibited by the specific viruses present in our study.

Other Mimiviridae-like transcripts detected in both transcriptomes shared homology with ankyrin-repeat protein genes, bifunctional E2/E3 enzyme genes, putative serine/threonine-protein kinase/receptor genes, a band 7 family protein gene, and uncharacterized protein genes. Mimiviridae-like transcripts uniquely detected in the thermo-sensitive SM transcriptome shared homology with a deoxynucleotide monophosphate kinase gene, an uncharacterized glycosyltransferase gene, a resolvase gene, and a transposase gene. A Mimiviridae-like gene detected only in the thermo- tolerant MI transcriptome was a N-glycosidase gene. One transcript with similarity to a Phycodnaviridae transcription factor gene was expressed in the thermo-sensitive SM transcriptome, and two transcripts with similarity to Phycodnaviridae uncharacterized protein genes were expressed in the thermo-tolerant MI transcriptome. Also, one transcript expressed in the thermo-sensitive SM transcriptome had similarity to an Iridoviridae uncharacterized protein gene.

As discussed in the main text, two transcripts in the thermo-sensitive SM transcriptome and one transcript in the thermo-tolerant MI transcriptome shared BLASTx homology with the Dinornavirus MCP gene. However, phylogenetic analysis revealed that the Dinornavirus-like MCP genes in our transcriptomes of type C1 Symbiodinium from South Molle Island and Magnetic Island (Levin et al. 2016), partial Dinornavirus-like MCP genes in the metagenome of Acropora tenuis from Orpheus Island (Weynberg et al. 2014), and partial Dinornavirus-like MCP genes in the virome of Montastraea cavernosa from Key West (Correa et al. 2013a) are substantially divergent from the known Dinornavirus MCP genes (Nagasaki et al. 2005) as well as from one another (Table S1, Figure S3). The Dinornavirus-like MCP genes from the separate studies may therefore be members of different, unknown Alvernaviridae genera, potentially with distinctive host and/or geographic specificities.

66 3.6 Supplementary tables Table S1. The known Dinornavirus MCP genes compared to Dinornavirus-like MCP genes found in ‘omics datasets from corals and Symbiodinium.

Sequence Complete ORF Sequence name Sequence source Dinoflagellate host Location Accession number Reference description gene? length Dinornavirus Dinornavirus Heterocapsa Tomaru et al., 2004; HcRNAV34 HcRNAV34 MCP Western Japan Yes 1080 nt NCBI: AB218608 HcRNAV34 circularisquama Nagasaki et al., 2005 gene Dinornavirus Dinornavirus Heterocapsa Tomaru et al., 2004; HcRNAV109 HcRNAV109 MCP Western Japan Yes 1080 nt NCBI: AB218609 HcRNAV109 circularisquama Nagasaki et al., 2005 gene Homology to Thermo-sensitive GenBank: Type C1 South Molle Island, Great TR74740|c13_g1_i1 Dinornavirus MCP SM Symbiodinium Yes 1077 nt KX538960, NCBI Levin et al., 2016 Symbiodinium Barrier Reef, QLD, AU gene transcriptome GEO: GSE77911 Homology to Thermo-sensitive GenBank: Type C1 South Molle Island, Great TR74740|c13_g1_i2 Dinornavirus MCP SM Symbiodinium Yes 1077 nt KX787934, NCBI Levin et al., 2016 Symbiodinium Barrier Reef, QLD, AU gene transcriptome GEO: GSE77911 Homology to Thermo-tolerant MI Type C1 Magnetic Island, Great NCBI GEO: TR97578|c0_g1_i1 Dinornavirus MCP Symbiodinium No 471 nt Levin et al., 2016 Symbiodinium Barrier Reef, QLD, AU GSE77911 gene transcriptome Homology to Hypothesized: Acropora tenuis Orpheus Island, Great NCBI SRA run: 729699.2 Dinornavirus Type C3 No 297 nt Weynberg et al., 2014 metagenome Barrier Reef, QLD, AU SRR1207979 MCP gene Symbiodinium Homology to Hypothesized: Acropora tenuis Orpheus Island, Great NCBI SRA run: 527285.2 Dinornavirus Type C3 No 279 nt Weynberg et al., 2014 metagenome Barrier Reef, QLD, AU SRR1210582 MCP gene Symbiodinium Homology to Hypothesized: Acropora tenuis Orpheus Island, Great NCBI SRA run: 503847.2 Dinornavirus Type C3 No 258 nt Weynberg et al., 2014 metagenome Barrier Reef, QLD, AU SRR1210582 MCP gene Symbiodinium Homology to Heat-stressed Hypothesized: NCBI SRA run: GAIR4WK03FYXY4 Dinornavirus Montastraea Clade C or D Key West, Florida, USA No 375 nt Correa et al., 2013 SRR493108 MCP gene cavernosa virome Symbiodinium Homology to Heat-stressed Hypothesized: NCBI SRA run: GAIR4WK03GFJLN Dinornavirus Montastraea Clade C or D Key West, Florida, USA No 345 nt Correa et al., 2013 SRR493108 MCP gene cavernosa virome Symbiodinium Homology to Heat-stressed Hypothesized: NCBI SRA run: GAIR4WK03F1XL6 Dinornavirus Montastraea Clade C or D Key West, Florida, USA No 339 nt Correa et al., 2013 SRR493108 MCP gene cavernosa virome Symbiodinium

Table S2. Up-regulated (+) and down-regulated (-) viral transcripts in the thermo-sensitive SM transcriptome after 9 days at 32°C (fold ≥ 2 relative to 27°C on at least one sampling time point, FDR ≤ 0.001).

h (against h (against

Transcript log2(fold) FDR CPM Annotation Virus Bit score partitioned Symbiodinium databases) genomes) TR74718|c0_g2_i1 +3.29 4.55E-11 1.93 Putative resolvase R771 Mimiviridae 129.80 49.69 129.80 TR74718|c0_g1_i1 +2.87 7.48E-16 3.89 Putative transposase R104 Mimiviridae 113.62 55.47 113.62 TR38522|c0_g3_i1 +1.86 3.47E-26 13.34 Uncharacterized protein L728 Mimiviridae 228.78 228.78 116.78 TR38522|c0_g1_i1 +1.77 1.83E-20 9.68 Uncharacterized protein L728 Mimiviridae 228.01 228.01 116.01 Putative F-box and FNIP repeat- TR72194|c2_g1_i2 +1.72 4.85E-05 3.09 Mimiviridae 670.06 356.21 532.06 containing protein L60 Putative F-box and FNIP repeat- TR73034|c0_g1_i2 +1.36 7.34E-10 8.10 Mimiviridae 726.02 342.03 262.02 containing protein L60 Putative F-box and FNIP repeat- TR73034|c0_g1_i1 +1.33 8.45E-06 5.20 Mimiviridae 414.42 153.30 43.42 containing protein L60 TR53463|c0_g1_i1 -0.82 2.25E-08 38.18 Uncharacterized protein R871 Mimiviridae 222.96 222.96 72.96 TR74740|c13_g1_i1 -2.84 3.01E-14 16378.33 Major viral capsid protein Dinornavirus 179.10 179.10 179.10 TR74740|c13_g1_i2 -2.87 9.75E-15 2305.35 Major viral capsid protein Dinornavirus 191.04 191.04 191.04

Table S3. Up-regulated (+) and down-regulated (-) viral transcripts in the thermo-sensitive SM transcriptome after 13 days at 32°C (fold ≥ 2 relative to 27°C on at least one sampling time point, FDR ≤ 0.001).

h (against h (against

Transcript log2(fold) FDR CPM Annotation Virus Bit score partitioned Symbiodinium databases) genomes) Putative F-box and FNIP repeat- TR74370|c1_g3_i1 +5.77 6.61E-06 0.66 Mimiviridae 498.66 338.86 341.66 containing protein L60 TR74718|c0_g2_i1 +3.03 1.83E-08 1.69 Putative resolvase R771 Mimiviridae 129.80 49.69 129.80 TR74718|c0_g1_i1 +2.81 2.19E-11 3.39 Putative transposase R104 Mimiviridae 113.62 55.47 113.62

Putative F-box and FNIP repeat- TR42922|c0_g1_i1 +2.43 6.34E-06 1.14 Mimiviridae 537.24 184.48 334.24 containing protein L60 TR38522|c0_g1_i1 +2.05 1.78E-26 11.85 Uncharacterized protein L728 Mimiviridae 228.01 228.01 116.01 Putative F-box and FNIP repeat- TR66569|c0_g1_i2 +1.93 1.82E-06 1.80 Mimiviridae 635.88 323.14 343.88 containing protein L60 Putative F-box and FNIP repeat- TR72673|c23_g1_i1 +1.76 1.07E-04 1.44 Mimiviridae 527.27 39.68 196.27 containing protein L60 TR38522|c0_g3_i1 +1.74 4.50E-21 15.99 Uncharacterized protein L728 Mimiviridae 228.78 228.78 116.78 Putative F-box and FNIP repeat- TR73034|c0_g1_i1 +1.65 7.68E-10 5.01 Mimiviridae 414.42 153.30 43.42 containing protein L60 TR73589|c0_g1_i1 +1.64 2.91E-05 2.39 Putative ankyrin repeat protein R784 Mimiviridae 53.91 53.91 53.91 Putative F-box and FNIP repeat- TR72194|c2_g1_i2 +1.58 3.03E-06 3.51 Mimiviridae 670.06 356.21 532.06 containing protein L60 Putative F-box and FNIP repeat- TR130476|c0_g1_i1 +1.57 3.51E-04 1.51 Mimiviridae 515.70 350.48 211.70 containing protein L60 Putative FNIP repeat-containing TR31628|c0_g2_i1 +1.49 1.21E-05 2.17 Mimiviridae 470.23 85.87 222.23 protein R636 Putative F-box and FNIP repeat- TR73034|c0_g1_i2 +1.41 1.29E-13 8.47 Mimiviridae 414.42 153.30 262.02 containing protein L60 Putative F-box and FNIP repeat- TR74062|c9_g1_i2 +1.11 3.27E-04 3.07 Mimiviridae 516.45 155.22 314.45 containing protein L60 Putative F-box and FNIP repeat- TR37111|c0_g2_i1 -0.98 1.18E-04 4.48 Mimiviridae 571.47 308.83 451.47 containing protein L60 TR53463|c0_g1_i1 -1.05 2.40E-13 33.57 Uncharacterized protein R871 Mimiviridae 222.96 222.96 72.96 Putative F-box and FNIP repeat- TR37111|c0_g1_i1 -1.05 4.65E-04 2.65 Mimiviridae 529.52 243.00 386.52 containing protein L60 Putative F-box and FNIP repeat- TR65978|c2_g4_i1 -1.14 2.48E-04 2.53 Mimiviridae 454.38 138.63 361.58 containing protein L60 Putative F-box and FNIP repeat- TR12211|c0_g1_i1 -1.14 1.10E-04 2.62 Mimiviridae 225.26 225.26 136.26 containing protein L60 Putative F-box and FNIP repeat- TR22857|c0_g1_i1 -1.91 1.75E-04 1.05 Mimiviridae 583.82 241.46 426.82 containing protein L60 Putative F-box and FNIP repeat- TR73567|c8_g16_i1 -2.37 1.17E-04 0.82 Mimiviridae 709.72 394.73 578.72 containing protein L60 TR74740|c13_g1_i1 -3.05 4.57E-20 9895.89 Major viral capsid protein Dinornavirus 179.10 179.10 179.10 TR74740|c13_g1_i2 -3.10 1.63E-19 1389.39 Major viral capsid protein Dinornavirus 191.04 191.04 191.04

Table S4. Up-regulated (+) and down-regulated (-) viral transcripts in the thermo-tolerant MI transcriptome after 9 days at 32°C (fold ≥ 2 relative to 27°C on at least one sampling time point, FDR ≤ 0.001).

h (against h (against

Transcript log2(fold) FDR CPM Annotation Virus Bit score partitioned Symbiodinium databases) genomes) Putative F-box and FNIP repeat- TR51820|c0_g4_i1 +2.53 1.94E-08 1.45 Mimiviridae 684.03 245.73 349.03 containing protein L60 Putative F-box and FNIP repeat- TR61729|c0_g1_i1 +2.03 2.93E-17 6.46 Mimiviridae 644.37 525.74 169.37 containing protein L60 Putative F-box and FNIP repeat- TR62954|c0_g4_i1 +1.78 1.03E-11 4.13 Mimiviridae 590.00 132.50 404.00 containing protein L60 Putative F-box and FNIP repeat- TR61729|c0_g1_i2 +1.69 5.35E-12 5.41 Mimiviridae 486.05 189.12 118.05 containing protein L60 Putative F-box and FNIP repeat- TR61287|c9_g2_i1 +1.51 2.66E-04 1.98 Mimiviridae 668.61 265.76 327.61 containing protein L60 Putative F-box and FNIP repeat- TR18026|c0_g1_i1 +1.49 9.47E-05 1.80 Mimiviridae 465.98 78.58 318.98 containing protein L60 Putative F-box and FNIP repeat- TR63607|c7_g16_i7 +1.32 3.60E-04 1.67 Mimiviridae 651.17 385.07 532.17 containing protein L60 Putative F-box and FNIP repeat- TR13909|c1_g1_i1 +1.21 2.49E-04 2.70 Mimiviridae 781.43 383.20 558.43 containing protein L60 Putative F-box and FNIP repeat- TR10186|c0_g1_i1 +1.21 5.27E-05 4.24 Mimiviridae 731.71 183.31 565.71 containing protein L60 Putative F-box and FNIP repeat- TR62747|c9_g3_i2 +1.02 1.26E-05 4.72 Mimiviridae 607.00 298.13 246.00 containing protein L60 Putative F-box and FNIP repeat- TR61287|c8_g2_i1 +0.96 1.43E-04 6.22 Mimiviridae 457.93 90.12 174.93 containing protein L60 Putative F-box and FNIP repeat- TR63204|c5_g6_i1 +0.88 6.26E-04 6.83 Mimiviridae 612.01 273.86 291.01 containing protein L60 TR46628|c0_g1_i1 -2.02 2.54E-07 2.83 Uncharacterized protein R883 Mimiviridae 80.88 80.88 83.17

TR105785|c0_g1_i1 -3.00 3.20E-05 0.79 Uncharacterized protein R617 Mimiviridae 127.10 71.26 42.40

Table S5. Up-regulated (+) and down-regulated (-) viral transcripts in the thermo-tolerant MI transcriptome after 13 days at 32°C (fold ≥ 2 relative to 27°C on at least one sampling time point, FDR ≤ 0.001).

h (against h (against

Transcript log2(fold) FDR CPM Annotation Virus Bit score partitioned Symbiodinium databases) genomes) Putative F-box and FNIP repeat- TR51820|c0_g4_i1 +2.21 1.38E-04 1.05 Mimiviridae 684.03 245.73 349.03 containing protein L60 Putative F-box and FNIP repeat- TR79579|c0_g1_i1 +2.20 3.79E-04 0.96 Mimiviridae 366.65 69.70 58.65 containing protein L60 Putative F-box and FNIP repeat- TR61729|c0_g1_i2 +1.91 1.25E-13 4.61 Mimiviridae 486.05 189.12 118.05 containing protein L60 Putative F-box and FNIP repeat- TR61729|c0_g1_i1 +1.91 1.62E-21 7.44 Mimiviridae 644.37 525.74 169.37 containing protein L60 Putative F-box and FNIP repeat- TR10186|c0_g1_i1 +1.84 1.20E-12 4.30 Mimiviridae 731.71 183.31 565.71 containing protein L60 Putative F-box and FNIP repeat- TR63722|c16_g3_i4 +1.69 5.05E-04 1.24 Mimiviridae 591.87 214.10 479.87 containing protein L60 Putative F-box and FNIP repeat- TR17872|c0_g1_i1 +1.64 6.01E-09 3.37 Mimiviridae 617.32 335.82 478.32 containing protein L60 Putative F-box and FNIP repeat- TR62954|c0_g4_i1 +1.57 1.50E-07 3.69 Mimiviridae 590.00 132.50 404.00 containing protein L60 Putative F-box and FNIP repeat- TR65676|c0_g1_i1 +1.53 3.12E-06 3.19 Mimiviridae 390.52 90.51 127.52 containing protein L60 Putative F-box and FNIP repeat- TR63623|c3_g3_i8 +1.35 8.74E-04 1.79 Mimiviridae 594.59 191.01 478.59 containing protein L60 Putative F-box and FNIP repeat- TR61287|c8_g2_i1 +1.19 3.40E-08 6.53 Mimiviridae 457.93 90.12 174.93 containing protein L60 Putative F-box and FNIP repeat- TR62747|c9_g3_i2 +1.14 4.51E-06 4.46 Mimiviridae 607.00 298.13 246.00 containing protein L60 Putative F-box and FNIP repeat- TR27291|c2_g1_i3 +1.08 1.23E-04 4.18 Mimiviridae 773.28 467.90 637.28 containing protein L60 Putative F-box and FNIP repeat- TR63204|c5_g6_i1 +1.07 1.19E-04 6.29 Mimiviridae 612.01 273.86 291.01 containing protein L60 Putative F-box and FNIP repeat- TR60935|c0_g1_i1 -1.63 6.21E-10 4.48 Mimiviridae 729.10 345.11 408.10 containing protein L60 TR46628|c0_g1_i1 -3.01 7.16E-07 1.39 Uncharacterized protein R883 Mimiviridae 83.17 83.17 83.17

Table S6. Up-regulated (+) and down-regulated (-) antiviral transcripts in the thermo- sensitive SM transcriptome after 9 days at 32°C (fold ≥ 2 relative to 27°C on at least one sampling time point, FDR ≤ 0.001).

Transcript log2(fold) FDR CPM Annotation

TR5690|c0_g1_i1 +1.43 2.26E-04 2.15 Baculoviral IAP repeat-containing protein 3 TR128451|c0_g1_i1 +1.36 1.15E-08 9.07 Influenza virus NS1A-binding protein TR5690|c0_g1_i2 +1.28 1.71E-04 2.50 Baculoviral IAP repeat-containing protein 3 TR73376|c0_g1_i1 +1.24 5.58E-04 2.79 Influenza virus NS1A-binding protein

TR122761|c0_g1_i1 +1.08 2.35E-08 8.83 Influenza virus NS1A-binding protein homolog A TR76594|c0_g1_i1 +0.93 7.68E-10 16.94 5'-3' exoribonuclease 2 homolog TR61542|c0_g1_i1 +0.80 1.40E-04 9.24 Interferon-induced helicase C domain-containing protein 1

Table S7. Up-regulated (+) and down-regulated (-) antiviral transcripts in the thermo- sensitive SM transcriptome after 13 days at 32°C (fold ≥ 2 relative to 27°C on at least one sampling time point, FDR ≤ 0.001).

Transcript log2(fold) FDR CPM Annotation

TR5690|c0_g1_i1 +1.77 9.56E-08 2.83 Baculoviral IAP repeat-containing protein 3 TR5690|c0_g1_i2 +1.62 2.28E-08 3.56 Baculoviral IAP repeat-containing protein 3 TR128451|c0_g1_i1 +1.48 1.19E-08 11.61 Influenza virus NS1A-binding protein TR122761|c0_g1_i1 +1.45 2.19E-07 10.68 Influenza virus NS1A-binding protein homolog A TR61542|c0_g1_i1 +1.42 6.72E-11 9.51 Interferon-induced helicase C domain-containing protein 1 TR76594|c0_g1_i1 +1.19 8.01E-15 20.88 5'-3' exoribonuclease 2 homolog TR73376|c0_g1_i1 +1.07 4.98E-04 3.16 Influenza virus NS1A-binding protein TR24409|c0_g1_i1 +1.07 3.29E-06 5.15 Antiviral helicase SKI2

TR61681|c0_g1_i1 -2.46 3.67E-04 0.68 Baculoviral IAP repeat-containing protein 6

Table S8. Up-regulated (+) and down-regulated (-) antiviral transcripts in the thermo- tolerant MI transcriptome after 9 days at 32°C (fold ≥ 2 relative to 27°C on at least one sampling time point, FDR ≤ 0.001).

Transcript log2(fold) FDR CPM Annotation

TR61737|c0_g1_i1 -0.58 1.06E-03 8.29 Influenza virus NS1A-binding protein TR68891|c0_g1_i1 -3.24 9.20E-10 2.98 Baculoviral IAP repeat-containing protein 7

72 Table S9. Up-regulated (+) and down-regulated (-) antiviral transcripts in the thermo- tolerant MI transcriptome after 13 days at 32°C (fold ≥ 2 relative to 27°C on at least one sampling time point, FDR ≤ 0.001).

Transcript log2(fold) FDR CPM Annotation

TR59840|c0_g2_i1 +2.82 2.07E-06 1.04 Baculoviral IAP repeat-containing protein 8

TR61737|c0_g1_i1 -1.08 7.43E-07 8.59 Influenza virus NS1A-binding protein TR45815|c0_g1_i1 -1.40 1.42E-04 2.40 Ski2p TR68891|c0_g1_i1 -3.16 1.45E-04 1.05 Baculoviral IAP repeat-containing protein 7

73 3.7 Supplementary figures

Figure S1. Sequence characteristics diverge between virus and host transcripts. The length of the line between the subsets of viral and non-viral (host) transcripts in the thermo- sensitive SM and thermo-tolerant MI transcriptomes represents the extent of divergence in (A) tri-nucleotide frequency and (B) codon usage. Correlation coefficients are reported for each comparison.

Figure S2. The novel +ssRNAV MCP gene. The MCP genes were successfully PCR amplified from cDNA, but not gDNA, of the thermo-sensitive SM and thermo-tolerant MI populations with MCP primer sets 1 and 2, supporting that transcripts TR74740|c13_g1_i1, TR74740|c13_g1_i2, and TR97578|c0_g1_i1 are of RNA virus and not host origin. The larger region of the MCP gene was amplified from cDNA of both the thermo-sensitive SM and thermo-tolerant MI population, confirming that the short length of TR97578|c0_g1_i1 (475 nt) was a result of incomplete de novo assembly due to low MCP transcript expression in the thermo-tolerant MI transcriptome.

Figure S3. Molecular phylogenetic analysis of +ssRNAV MCP genes. Relationships between in-frame nt sequences of the Dinornavirus MCP genes and Dinornavirus-like MCP genes detailed in Table S1 were assessed. (A) Evolutionary history was inferred by using the Maximum Likelihood method based on the Tamura-Nei model (Tamura and Nei 1993). The tree with the highest log likelihood (-2,158.3565) is shown. The percentage of replicate trees in which the associated taxa clustered together in the bootstrap test (1,000 replicates) is shown next to the branches (Felsenstein 1985). Initial tree(s) for the heuristic search were obtained automatically by applying Neighbor-Join and BioNJ algorithms to a matrix of pairwise distances estimated using the Maximum Composite Likelihood (MCL) approach, and then selecting the topology with superior log likelihood value. A discrete Gamma distribution was used to model evolutionary rate differences among sites (5 categories (+G, parameter = 2.9108)). The rate variation model allowed for some sites to be evolutionarily invariable ([+I], 1.8733% sites). The tree is drawn to scale, with branch lengths measured in the number of substitutions per site. Codon positions included were

76 1st+2nd+3rd+Noncoding. All positions with less than 80% site coverage were eliminated. That is, fewer than 20% alignment gaps, missing data, and ambiguous bases were allowed at any position. There were a total of 258 positions in the final dataset. Evolutionary analyses were conducted in MEGA7 (Kumar et al. 2016). (B) A percent identity matrix was generated based on nt sequence alignments produced using MUSCLE (Edgar 2004, McWilliam et al. 2013).

Figure S4. The novel +ssRNAV IRES. Secondary structure of the novel +ssRNAV IRES in TR74740|c13_g1_i1 and TR74740|c13_g1_i2 from the thermo-sensitive SM transcriptome was predicted using VIPS and is similar to the cricket paralysis virus IRES (R score > 1.61, P < 0.001) (http://140.135.61.250/vips/, last accessed July 2016) (Hong et al. 2013). The structure is colored by positional entropy.

Figure S5. The novel +ssRNAV RNA replicase. The protein encoded by ORF-1 in TR74740|c13_g1_i1, which did not receive annotation from BLAST+, was modelled using Phyre2 in intensive modelling mode (http://www.sbg.bio.ic.ac.uk/~phyre2, last accessed July 2016) (Kelley et al. 2015). The top ten results all matched RNA replicases (RNA- dependent RNA-polymerases) and, in all cases when specified, are from +ssRNAVs (results 2, 3, 4, 5, 6, 10).

78 3.8 Supplementary datasets Dataset S1. Mono- and tri-nucleotide statistics for viral transcripts and non-viral transcripts in each Symbiodinium transcriptome: http://www.nature.com/ismej/journal/v11/n3/suppinfo/ismej2016154s1.html?url=/ismej /journal/v11/n3/full/ismej2016154a.html

Dataset S2. Codon usage of viral transcripts and non-viral transcripts in each Symbiodinium transcriptome: http://www.nature.com/ismej/journal/v11/n3/suppinfo/ismej2016154s1.html?url=/ismej /journal/v11/n3/full/ismej2016154a.html

80 Chapter 4 Expanding the Symbiodinium (Dinophyceae, Suessiales) toolkit through protoplast technology

Publication III Levin RA, Suggett DJ, Nitschke MR, van Oppen MJH, Steinberg PD (2017b). Expanding the Symbiodinium (Dinophyceae, Suessiales) toolkit through protoplast technology. Journal of Eukaryotic Microbiology doi: 10.1111/jeu.12393.

Declaration I certify that this publication was a direct result of my research towards this PhD and that reproduction in this thesis does not breach copyright regulations.

...... Rachel A. Levin

81 4.1 Abstract

Dinoflagellates within the genus Symbiodinium are photosymbionts of many tropical reef invertebrates, including corals, making them central to the health of coral reefs. Symbiodinium have therefore gained significant research attention, though studies have been constrained by technical limitations. In particular, the generation of viable cells with their cell walls removed (termed protoplasts) has enabled a wide range of experimental techniques for bacteria, fungi, plants, and algae such as ultrastructure studies, virus infection studies, patch clamping, genetic transformation, and protoplast fusion. However, previous studies have struggled to remove the cell walls from armored dinoflagellates, potentially due to the internal placement of their cell walls. Here we produce the first Symbiodinium protoplasts from three genetically and physiologically distinct strains via incubation with cellulase and osmotic agents. Digestion of the cell walls was verified by a lack of Calcofluor White fluorescence signal and by cell swelling in hypotonic culture medium. Fused protoplasts were also observed, motivating future investigation into intra- and inter-specific somatic hybridization of Symbiodinium. Following digestion and transfer to regeneration medium, protoplasts remained photosynthetically active, regrew cell walls, regained motility, and entered exponential growth. Generation of Symbiodinium protoplasts opens exciting, new avenues for researching these crucial symbiotic dinoflagellates, including genetic modification.

82 4.2 Introduction Photosynthetic dinoflagellates are globally important primary producers that sustain unique ecological pathways and biogeochemical processes throughout Earth’s aquatic biospheres (Murray et al. 2016). Species within the genus Symbiodinium are extensively studied due to their fundamental role as endosymbionts of reef-building corals and many other marine invertebrates (Fabina et al. 2013, LaJeunesse 2002). Symbiodinium not only drive coral productivity and reef growth (Muscatine 1990, Muscatine and Porter 1977), but their diverse genetic backgrounds and distinct physiologies can also determine the thermal bleaching thresholds of their coral host (Berkelmans and van Oppen 2006, Fitt et al. 2001, Howells et al. 2012, Levin et al. 2016, Loram et al. 2007, Yuyama et al. 2012). With coral reefs currently under extreme threat from climate change (Ainsworth et al. 2016, Baker et al. 2008), there is a pressing need to enhance our understanding of Symbiodinium biology. Despite the extensive research focus on Symbiodinium, almost all the knowledge gained to date has relied on studies of wild-type (not genetically modified) cells (e.g., Howells et al. 2012, Levin et al. 2016, Parkinson et al. 2016, Rosic et al. 2015, Suggett et al. 2015, Suggett et al. 2008, Xiang et al. 2015). Only two studies have validated successful genetic transformation of Symbiodinium (Ortiz-Matamoros et al. 2015b, ten Lohuis and Miller 1998), and only one study has reported artificial selection of Symbiodinium (Huertas et al. 2011). Our current inability to advance research in genetics and cell biology of Symbiodinium is due to a lack of tools that facilitate genetic modification of Symbiodinium (Davy et al. 2012). Consequently, understanding of this ecologically critical dinoflagellate still remains in its infancy.

One such missing tool is the ability to generate Symbiodinium protoplasts. The first protoplasts were produced from plant cells over a century ago through mechanical disruption of the cell wall (Klercker 1892). In the mid-1900s, introduction of enzymes to digest cells walls (Bachmann and Bonner 1959, Cocking 1960, Eddy and Williamson 1957, Weibull and Bergström 1958) led to high protoplast viability and yields (Davey et al. 2005). Protoplasts have since unlocked major research areas of bacteria, fungi, plants, and algae

83 – notably genetic modification strategies (Bravo and Evans 2011, Carlson 1973, Davey et al. 2005, Gietz and Woods 2001, Hopwood 1981, Reddy et al. 2007) that have lead to economically valuable agricultural improvements (Bajaj 2012, Wang et al. 2013). Absence of the cell wall enables alternative methods for intracellular delivery of foreign DNA into cells such as polyethylene glycol (PEG)- and liposome-mediated transformation that can improve genetic transformation efficiency in certain species (Caboche 1990, Hansen and Wright 1999, Mathur and Koncz 1998, Rakoczy-Trojanowska 2002) and intracellular delivery of RNA-protein complexes for CRISPR-Cas9 genome editing (Woo et al. 2015). Furthermore, removal of the cell wall has allowed for protoplast fusion (i.e., somatic hybridization), another method of genetic modification in which two protoplasts are joined to form one hybrid cell containing two genomes (one from each parent). Protoplast fusion can even be used to hybridize cells of sexually incompatible species such as potatoes and tomatoes (Grosser et al. 1990, Kito et al. 1998, Melchers et al. 1978, Sagadevan et al. 2009). When capable of cell division, the hybrid cell can be mitotically propagated to produce genetically enhanced plants expressing novel combinations of traits controlled by individual genes and/or entire gene networks inherited from each parent cell (Bravo and Evans 2011, Davey et al. 2005, Peberdy 1980).

Generation of viable protoplasts has not been accomplished for many microalgae strains, due to the complexity of microalgal cell walls and diversity of cell wall types across different species (Coll 2006). Particularly, armored dinoflagellates, unlike plants, have an intricate cell covering composed of an internal cell wall (a cellulose-enforced pellicle and cellulosic thecal plates) positioned between membranous layers (Markell et al. 1992, Morrill and Loeblich 1981, Wakefield et al. 2000). The membranous outer layer and thecal vesicle membrane serve as potential obstacles in dinoflagellate protoplast generation by reducing access to the internal cell wall. Several prior attempts have been made to produce protoplasts of armored dinoflagellates (Adamich and Sweeney 1976, Kwok et al. 2007, Pozdnyakov et al. 2014, Trench and Blank 1987), but success has been limited. Adamich and Sweeney (1976) reported the first protoplast generation of Gonyaulax

84 polyedra using the detergent Liquinox, but trials of the Liquinox method failed to produce Symbiodinium protoplasts (Jit Ern Chen, pers. comm.). Cellulase was also considered to be unsuccessful at digesting cell walls in live Symbiodinium cells (Trench and Blank 1987), despite cellulase fully digesting isolated cell walls extracted from Symbiodinium cells (Markell et al. 1992). However, cellulase has been indicated to digest the cell walls of the dinoflagellate Lingulodinium (Wang et al. 2005). Cell walls of Crypthecodinium cohnii have been reduced, but not removed, by growing cells on agar plates with polyethylene glycol (PEG) (Kwok et al. 2007), and finally, treatment of Prorocentrum minimum with a cellulose synthesis inhibitor had low efficacy in preventing cellulose production (Pozdnyakov et al. 2014).

Here we produce protoplasts of Symbiodinium that are subsequently capable of regenerating their cell walls and returning to a healthy physiological status. Additionally, we explore the potential for fusion of Symbiodinium protoplasts, which may unlock the gateway for investigating intra- and inter-specific somatic hybridization of genetically distinct Symbiodinium isolates.

4.3 Materials and methods

4.3.1 Culture maintenance and genotyping ITS2-type A3 Symbiodinium (University of Technology, Sydney, Climate Change Cluster culture collection: monoclonal strain CS73; host: Tridacna maxima; geographic origin: Heron Island, Great Barrier Reef) (Suggett et al. 2015), ITS2-type C1 Symbiodinium (Australian Institute of Marine Science culture collection: monoclonal strain SCF055; host: Acropora tenuis; geographic origin: Magnetic Island, Great Barrier Reef), and a mixed population of Symbiodinium (University of Technology, Sydney, Climate Change Cluster culture collection: undefined combination of ITS2-types from strains described in Suggett et al. (2015) were maintained at 25 °C in culture medium composed of filtered seawater supplemented with Diago’s IMK (Wako Pure Chemical Industries Ltd., Osaka, Japan) at pH 7.9. Different Symbiodinium ITS2-types are considered different species. Light was

85 provided at an irradiance of 50 µmol photons m−2 s−1 (Philips TLD 18W/54 fluorescent tubes, 10,000 K) on a 12 h light : 12 h dark cycle starting at 9:00 am each day. Genomic DNA was extracted from each strain with the PowerPlant Pro DNA Isolation Kit (MoBio, Carlsbad, CA). The ITS2 region from genomic DNA of each strain confirmed by amplification with ITS2 primers from Stat et al. (2011) and subsequent Sanger sequencing by the Australian Genome Research Facility.

4.3.2 Enzymatic digestion and regeneration of the cell wall For digestion, aliquots (~107 cells) of each Symbiodinium culture in exponential growth phase were pelleted at 2,000 × g for 5 min. Culture medium was decanted, and the cell pellet was re-suspended in 10 ml of digestion solution composed of 0.5 M D- sorbitol (Sigma-Aldrich, St. Louis, MO) dissolved in culture medium along with either 1.5 kilounits (KU) or 3 KU of cellulase from Trichoderma sp. (catalog number C1794, Sigma- Aldrich, St. Louis, MO). The cells in digestion solution were then incubated at 29-30 °C (to increase cellulase activity) in the dark on a shaking platform at 100 rpm for 36-48 h.

After digestion, protoplasts were pelleted at 200 × g for 5 min. Digestion solution was replaced with 10 ml of protoplast wash solution containing of 0.5 M D-sorbitol, 0.5 M sucrose, and 25 mM CaCl2 (Sigma-Aldrich, St. Louis, MO) dissolved in culture medium supplemented with 100 µg/ml kanamycin. The protoplasts were then incubated at 25 °C in the dark on a shaking platform at 100 rpm for 3 h, after which they were pelleted and the wash step was repeated. Kanamycin was included in the protoplast wash solution to prevent overgrowth of Symbiodinium-associated bacteria due to the high concentration of sucrose that serves as a metabolically active osmoticum and carbon source. D-sorbitol is also important as an inert osmoticum, and CaCl2 promotes protoplast membrane stability (Gamborg 1977). After washing, protoplasts were pelleted, suspended in 10 ml of regeneration medium (0.5 M D-sorbitol and 25 mM CaCl2 dissolved in culture medium, pH 7.0), and returned to Symbiodinium maintenance conditions (25 °C, 50 μmol photons m−2 s−1, 12 h light : 12 h dark cycle).

Removal and regeneration of cell walls from incubation in digestion solution was assessed by cellulose staining with Calcofluor White (Sigma-Aldrich, St. Louis, MO). Culture aliquots (200 µl) were pelleted, resuspended in 20 µl of regeneration medium, and gently pipetted onto the 10 mm well of a FluoroDish (World Precision Instruments, Sarasota, FL). In the dish, cells were pre-treated with 20% dimethyl sulfoxide (DMSO) and a drop of 10% potassium hydroxide (BD Biosciences, Franklin Lakes, NJ) to ensure rapid permeablization of the cell membranes and wall then stained with 2 µl of Calcofluor White. After 5 min, cells were imaged on an Olympus FV1000 inverted confocal microscope (Tokyo, Japan) using the oil-immersion PL SAPO 100×/1.4 objective in a sequential scan with a 405 nm laser (dichroic mirror 490 nm, emission filter 425/50 nm) to detect stained cellulose and a 543 nm laser (dichroic mirror 560 nm, emission filter LP560) to detect chlorophyll fluorescence of cells.

To further verify protoplast generation, control cells and cellulase-treated cells were exposed to a range of osmotic pressures: culture medium alone, culture medium + 0.5 M D-sorbitol, culture medium + 1 M D-sorbitol. After 4 h of incubation, bright-field images of cells were captured with a Nikon Eclipse Ni upright microscope on the 20× objective. The cross-sectional area of imaged cells was calculated with the equation Area = π(length/2)(width/2), generalizing cell shape as an ellipsoid.

To evaluate protoplast fusion, culture aliquots (750 µl) were fixed with 1% paraformaldehyde for 24 h and stained with 1 µM DAPI (4',6-diamidino-2-phenylindole) for 20 min. The 100× (oil-immersion) objective of a Nikon Eclipse Ni upright microscope was used to capture bright-field, autofluorecence (excitation: 460-490 nm, emission: 500- ∞ nm), and DAPI fluorescence (excitation: 340-380 nm, emission: 435-485 nm) images of cells.

87 4.3.3 Protoplast physiology The study contained four phases: Pre-digestion (day -2), digestion (4:00 pm on day -2 to 8:00 am on day 0), washing (8:00 am to 2:00 pm on day 0) and cell wall regeneration (2:00 pm on day 0 to day 28). Physiological assessments of Symbiodinium cultures (n = 3) were conducted on day -2 (pre-digestion) and on days 1, 3, 7, 14, and 28 (post-digestion, cell wall regeneration phase) to track cell health and morphology. Each physiological measurement was recorded at the same time on each sampling day. On days 2 and 6, cells were pelleted, and 15 ml of fresh regeneration medium was added. On days 13 and 27, cells were pelleted, and regeneration medium was replaced with 40 ml of culture medium.

Two active chlorophyll a fluorescence approaches were used to assay the photophysiological status of cells after 10-15 min of low-light acclimation in order to fully oxidise the photosynthetic electron transport chain (Suggett et al. 2015). First, cells were assessed using pulse-amplitude modulated imaging (PAM) microfluorometry (Imaging- PAM M-series Chlorophyll Fluorometer, Heinz Walz GmbH, Effeltrich, Germany) to yield a relative description of photosystem II (PSII) physiological heterogeneity amongst each culture at the single cell level. In additionally, cells were subjected to Fast Repetition Rate fluorescence analysis (FRRf; Soliense Inc.) to determine absolute quantification of PSII photosynthetic performance.

For the Imaging-PAM, the following configuration was used to deliver a multiple turnover (MT) excitation protocol, following the manufactures guidelines (Imaging-PAM M-series Chlorophyll Fluorometer, Instrument Description and Information for Users, 2.152 / 07.06, 5. Revised Edition: March 2014): An IMAG-RGB LED Lamp Module (Heinz Walz GmbH, Effeltrich, Germany) mounted on a Zeis Axio A1 microscope (Zeiss, Göttingen, Germany) with a Zeiss Fluar 20× objective, RG665 detector filter, and a 420-640 nm dichroic mirror, allowed for computer-assisted measuring of maximum quantum yield of

PSII (Fv/Fm, dimensionless). The Special SP-routine of the Imaging-WIN software (Heinz

88 Walz GmbH, Effeltrich, Germany, version 2.41a) was used for increased sensitivity at low levels of excitation intensity. After a non-biological fluorescence standard was used to normalize measurements of the maximum fluorescence yield (Fm), 10 µl of low-light acclimated cells were then mounted onto a glass slide on the microscope. Once the steady state fluorescence yield (F0) stabilized, a saturating pulse of blue (460 nm) light was applied and FV/FM ([Fm-F0]/Fm) was imaged (measuring light intensity = 2, gain = 5, frequency = 8, damping = 2, F0 averaging n = 3).

Next, low-light acclimated culture aliquots (50 µl) were diluted (1 : 30) in regeneration or culture medium and placed inside the Soliense FRRf optical head. The sample was subjected to 10 consecutive single turnover (ST) induction protocols (each separated by 150 ms) with an excitation sequence of 1.0 μs × 100 flashlets (2.5 μs interval) followed by a relaxation sequence of 50 × 1.0 μs flashlets (20 μs interval) as previously described (Schuback et al. 2015, Suggett et al. 2015). The final induction transient acquired for each sample was the average of the 10 × ST inductions, which was then fit against the KPF model (Kolber et al. 1998) to yield values for minimum and maximum fluorescence (F0, Fm; and hence Fv/Fm, dimensionless), PSII effective absorption cross 2 -1 section (σPSII, Å quanta ) and the electron turnover time of the primary PSII quinone acceptor, QA (τQA, μs), using custom software (Zbigniew Kolber, pers. comm.). It should be noted that all induction protocols were generated using only the LED with a 478 nm peak excitation, hence all σPSII values reported here are specific to this wavelength only. Values for τQA are represented by the initial of three decay constants (t) from the relaxation phase of the induction. All induction curve fits accounted for any formation of P680 triplet fluorescence quenching. Finally, all fits were performed relative to a sample blank consisting of an aliquot of regeneration or culture medium.

To measure motility, culture aliquots (100 µl) were added to wells of a 96 well plate, and cells were allowed to settle for 15 min. The total number of motile cells out of 100 counted cells was recorded for each aliquot through bright-field observations with a

89 Nikon Eclipse Ti inverted microscope (Tokyo, Japan) using the 20× objective. The total number of cells in each culture was quantified using a hemocytometer. Morphology of cells was assessed from bright-field images captured with a Nikon Eclipse Ni upright microscope (Tokyo, Japan) using the 100× (oil-immersion) objective.

4.4 Results and discussion Three KU of cellulase per digestion of 107 cells was necessary for reliable protoplast generation within 36-48 h at 29-30 °C, as 1.5 KU only resulted in partial digestion of the cell walls (Figure 1A). Exposure to hypotonic culture medium caused protoplasts to exhibit swelling compared to control cells (Figure 1B) since digestion of the cell wall (lack of cell wall pressure) leads to increased turgor pressure. Addition of D- sorbitol to culture medium improved protoplast isotonicity (Figure 1B). Complete regeneration of cell walls, indistinguishable from cells pre-digestion, occurred by day 14 post-digestion (Figure 1A). Our results are consistent with previous findings that cellulose is the major structural component of Symbiodinium cell walls (Markell et al. 1992) and with previous evidence that internal cellulosic dinoflagellate cell walls can be digested with cellulase (Wang et al. 2005). Furthermore, our study overcomes a previous shortcoming where the use of cellulase failed to digest cell walls of live Symbiodinium (Trench and Blank 1987). Unfortunately, Trench and Blank (1987) did not detail their cellulase protocol, but an explanation for their contrasting results may be that their cellulase concentrations had been too low and/or their digestion times had been too short. In our study, cellulase was able to partially or completely digest Symbiodinium cell walls despite their internal placement, possibly because the Symbiodinium outer “membrane” is a thin layer composed of polysaccharides or glycoproteins (Blank 1987, Trench and Blank 1987, Wakefield et al. 2000), substances through which enzymes and other proteins can diffuse (Kingshott and Griesser 1999, McArthur et al. 2000, Olmsted et al. 2001), and the thecal vesicle membrane is delicate causing it to easily rupture (Wakefield et al. 2000), which would allow cellulase access to the thecal plates. Moreover, Symbiodinium undergo ecdysis of their cell covering layers during the cell cycle and under

90 stress (Bricheux et al. 1992, Pozdnyakov and Skarlato 2012, Wakefield et al. 2000), which might increase contact of the digestion solution and the underlying cellulose. Turgor pressure and osmotic stress due to disruption of the cell wall can also reduce membrane integrity (Fellows and Boyer 1976), potentially granting the digestion solution increased internal access. However, the mode through which cellulase contacted the internal cellulose was not directly investigated here.

Figure 1. Cell wall digestion confirmed by cellulose staining and by swelling of protoplasts in hypotonic culture medium. (A) Cellulose staining of type C1 Symbiodinium cells with Calcofluor White on day -2 pre-digestion (control cell wall), on day 1 after exposure to 1.5 KU of cellulase (partially digested cell wall), on day 1 after exposure to 3 KU of cellulase (digested cell wall), and on day 14 after exposure to 3 KU of cellulase (regenerated cell wall). Scale bar (white, bottom left) = 10 μm. (B) Box-and-whisker plot showing the size (cross-sectional cell area) distribution (10th-90th percentile) of type C1 Symbiodinium cells (n = 25) exposed to varying concentrations of D-sorbitol in culture medium.

Pearl chain assemblies (Figure 2A), which create fusion interfaces between protoplasts (Zimmermann 1982), formed without exposure to a chemical fusogen or electric current. Protoplast fusion (Figure 2B) occurred across all strains in < 1% of cells, which is still notable considering that no separate protocol to induce fusion (e.g., chemical

91 fusion or electrofusion) was applied (Anne and Peberdy 1976, Zimmermann and Scheurich 1981). Yet, fusion was not necessarily spontaneous as the presence of salts and calcium can induce fusion (Boss et al. 1984, Kameya and Takahashi 1972). Furthermore, some protoplasts still exhibited swelling when exposed to 0.5 M D-sorbitol (Figure 1B), which enhances efficiency of protoplast fusion through disruption of the membrane skeleton (Bhojwani 2013, Ahkong and Lucy, 1986). Massive size, irregular shape, and unsegregated chloroplasts clearly differentiated fused protoplasts from protoplasts of resting phase, interphase, or mitotic phase cells (Figure 2B, Video S1). Fusion was further confirmed by detection of two nuclei in hybrid cells (Figure 2B) and by scanning multiple focal planes to ensure that a continuous membrane encased the hybrid cell as opposed to two separate protoplasts atop one another (Video S1). No cells with morphology equivalent to hybrid cells were observed in control Symbiodinium cultures. While hybrid cells were not isolated from cultures for long-term assessment, they were viable on day 1 post-digestion since uniform cytoplasmic streaming across the hybrid cells was apparent, which also showed that the cytoplasms from each parent protoplast had joined (Video S1). Protoplast fusion in other organisms has been applied to combine desired strains based on phenotypic traits such as increased thermal tolerance, increased viral tolerance, and increased antioxidant activity (Jbir-Koubaa et al. 2015, Pereira de Carvalho Costa et al. 2003, Zhang et al. 2012) – all of which are traits in Symbiodinium that have been implicated to reduce coral bleaching (Correa et al. 2016, Howells et al. 2012, Levin et al. 2016, Levin et al. 2017a, Ragni et al. 2010, Suggett et al. 2008). Moving forward, protocols for increasing fusion frequency (reviewed by Bajaj 2012, Sowers 2013) should be trialled to promote somatic hybridization, and the ability of hybrid cells to produce daughter cells should be evaluated to establish the feasibility of Symbiodinium breeding/genome shuffling.

Figure 2. Somatic hybridization (fusion) of Symbiodinium protoplasts. (A) Pearl chain assembly of type C1 Symbiodinium protoplasts preceding somatic hybridization. (B) Bright- field, autofluorescence, and DAPI fluorescence images of a hybrid cell compared to a dividing (mitotic phase) cell from the type C1 Symbiodinium strain after cellulase treatment. Scale bar (black, bottom left) = 10 μm. Focal plane scanning and cytoplasmic streaming of a live hybrid cell from the mixed Symbiodinium population after cellulase treatment are shown in Video S1.

Photosynthetic assessment of individually imaged protoplasts (Imaging-PAM, Figure 3) and culture aliquots (FRRf, Figure 4A-C) revealed that protoplasts remained photosynthetically active following digestion of cell walls. However, on day 1 post- digestion, Fv/Fm decreased by 4.9-32.1%, σPSII increased by 9.5-22.5%, and τQA decreased by 15.8-38.1% across all cultures compared to on day -2 pre-digestion (Figure 4A-C). Initially, photophysiology of type C1 was the least affected, correlating with its chloroplast structure being the least altered: bright-field imaging of entire type C1 protoplasts detected slight pigmentation loss (Figure 5) while confocal imaging of only autofluorescence in a single plane of type C1 protoplasts did not detect noticeable change (Figure 1A). The photophysiology trends observed across the cultures were not due to cell death, but rather the transient decline in health status of live protoplasts because

93 parameters of active photophysiology cannot be recorded from dead cells. Parallel decreased Fv/Fm and increased σPSII are often observed when PSII reaction centres (RCIIs) become increasingly deactivated via cell stress (Suggett et al. 2009), most likely induced from a combination of removal of the cell wall, increased temperature during cell wall digestion (Takahashi et al. 2008), and osmotic and pH differences (Berkowitz and Gibbs

1982) in regeneration medium verses culture medium. Transient reduction of τQA on day 1 post-digestion relative to day -2 pre-digestion suggests enhanced electron turnover by RCIIs despite net RCII inactivation, for example via compensatory photoprotection (Behrenfeld et al. 1998) or transient thylakoid membrane (and/or trans-thylakoid membrane pH) instability (Tchernov et al. 2004). Nonetheless, the exact mechanism at play cannot be confirmed from these data alone.

Figure 3. Single-cell photobiology throughout protoplast generation and cell wall regeneration. Individual Symbiodinium cells are shown according to a false-color spectrum where red indicates a relative maximum quantum yield of PSII approaching 0 and purple indicates a relative maximum quantum yield of PSII approaching 1. Panels from left to right indicate the pre-digestion sampling day -2 and post-digestion sampling days 1, 3, 7, 14, and 28 for type A3 (top row), type C1 (middle row), and the mixed population (bottom row). Scale bar (white, bottom left) = 20 μm.

Figure 4. Physiological status of cultures throughout protoplast generation and cell wall regeneration. (A) Maximum quantum yield of PSII (mean ± standard error, n = 3). (B) PSII effective absorption cross section (mean ± standard error, n = 3). (C) Electron turnover time of the primary PSII quinone acceptor (mean ± standard error, n = 3). (D) Cell motility (mean ± standard error, n = 3). (E) Growth rate (mean ± standard error, n = 3). The blue solid line on day 0 marks when fresh protoplasts were resuspended in regeneration medium. The blue dotted line on day 13 marks when cells were returned to culture medium.

Along with PSII inactivation, protoplasts were not motile on day 1 post-digestion (Figure 4D), potentially due to cell cycle arrest in the coccoid (flagella-lacking, dark-cycle) stage from 36-48 h of incubation in the dark (Pozdnyakov and Skarlato 2012, Wang et al. 2008) or incidental mechanical loss of flagella (Bricheux et al. 1992) during the cell wall digestion procedure. Total cell abundance on day 1 post-digestion was reduced to 51-85% of that on day -2 pre-digestion from lysis during cellulase treatment (Figure 4E). The decrease continued on day 3 post-digestion when abundance of cells ranged from 32.2- 79.0% relative to day -2 pre-digestion from an inability of some cells to recover from stress following cellulase treatment.

By day 7 post-digestion, protoplast cultures substantially improved photophysiological performance (Figure 3, 4A-C) and morphology (Figure 5), demonstrating that the stress experienced from exposure to cellulase and increased temperature during digestion was not sustained. Likewise, the potential effects of turgor pressure and osmotic stress on membrane ultrastructure, which can reduce PSII activity (Fellows and Boyer 1976), may have gradually decreased from regenerating wall pressure in conjunction with cells undergoing intracellular osmotic adjustments to prevent photosynthetic damage (Downton 1983). Uniquely, σPSII of only type C1 post-digestion continuously increased over time in regeneration medium, indicating that type C1 may be less capable of osmotic acclimation than type A3 and the mixed population. The differential photophysiological responses amongst types clearly highlight the broad phenotypic diversity apparent for the genus Symbiodinium (e.g., Suggett et al. 2015).

Complete morphological improvement, indistinguishable from cells pre-digestion, occurred by day 14 post-digestion (Figure 5). However, while Fv/Fm returned to a healthy state in all cultures, τQA of type A3 and type C1 decreased. Again, it is not possible to fully confirm the cause of these trends, but they coincide with the switch from regeneration medium to culture medium that occurred on day 13 post-digestion (Figure 4C). Type A3

97 and type C1 may therefore require more than one day to acclimate to the absence of osmotic agents and increased pH of the culture medium compared to the regeneration medium. Gradual dilution of regeneration medium with culture medium over several days could be trialled to alleviate this cell stress response (Davey et al. 2005). Regardless, by day 28 post-digestion, all cultures were in exponential growth phase and exhibited Fv/Fm,

σPSII, τQA, and motility equivalent to cells on day -2 pre-digestion (Figure 4A-E).

Figure 5. Bright-field imaging of cell morphology throughout protoplast generation and cell wall regeneration. Microscopy revealed morphological differences between Symbiodinium with cell walls and Symbiodinium protoplasts. Panels from left to right indicate the pre-digestion sampling day -2 and post-digestion sampling days 1, 3, 7, 14, and 28 for type A3 (top row), type C1 (middle row), and the mixed population (bottom row). Scale bar (black, bottom left) = 10 μm.

In summary, we report the first generation of Symbiodinium protoplasts and the first observations of protoplast fusion in any dinoflagellate. Protoplasts have led to major

98 breakthroughs in the biological understanding and genetic enhancement of a vast range of organisms (Bravo and Evans 2011, Carlson 1973, Davey et al. 2005, Gietz and Woods 2001, Hopwood 1981, Reddy et al. 2007), thus we anticipate our study may also significantly expand the experimental scope of Symbiodinium studies. Considering the recent calls for heightened focus on genetic modification of dinoflagellates (Murray et al. 2016) and genetic enhancement of Symbiodinium to sustain the health of coral reefs (van Oppen et al. 2015), Symbiodinium protoplast generation and fusion are valuable and timely technologies ready to be utilized.

4.5 Acknowledgements The Centre for Marine Bio-Innovation at The University of New South Wales contributed financial support for this study. The Biomedical Imaging Facility at The University of New South Wales covered expenses for the use of the Olympus FV1000 confocal microscope. Iveta Slapetova provided assistance with confocal imaging. Dave Hughes provided assistance with FRRf measurements. Caitlin Lawson provided assistance with Symbiodinium culturing.

4.6 Supplementary video Video S1. Cytoplasmic streaming across fused cytoplasms of a live, hybrid cell from the mixed Symbiodinium population after cellulase treatment. Bright-field video was recorded with a Nikon Eclipse Ni upright microscope using the 100× (oil-immersion) objective: http://onlinelibrary.wiley.com/store/10.1111/jeu.12393/asset/supinfo/jeu12393-sup- 0001-VideoS1.mp4?v=1&s=c823700fb43bead89f9ad3348d5eadde6ef5b111

100 Chapter 5 Engineering strategies to decode and enhance the genomes of coral symbionts

Publication IV Levin RA, Suggett DJ, Voolstra, CR, Agrawal S, Steinberg PD, Suggett DJ, van Oppen MJH (2017c). Engineering strategies to decode and enhance the genomes of coral symbionts. Frontiers in Microbiology 8: 1220.

Declaration I certify that this publication was a direct result of my research towards this PhD and that reproduction in this thesis does not breach copyright regulations.

...... Rachel A. Levin

101 5.1 Abstract Elevated sea surface temperatures from a severe and prolonged El Niño event (2014–2016) fueled by climate change have resulted in mass coral bleaching (loss of dinoflagellate photosymbionts, Symbiodinium spp., from coral tissues) and subsequent coral mortality, devastating reefs worldwide. Genetic variation within and between Symbiodinium species strongly influences the bleaching tolerance of corals, thus recent papers have called for genetic engineering of Symbiodinium to elucidate the genetic basis of bleaching-relevant Symbiodinium traits. However, while Symbiodinium has been intensively studied for over 50 years, genetic transformation of Symbiodinium has seen little success likely due to the large evolutionary divergence between Symbiodinium and other model eukaryotes rendering standard transformation systems incompatible. Here, we integrate the growing wealth of Symbiodinium next-generation sequencing (NGS) data to design tailored genetic engineering strategies. Specifically, we develop a testable expression construct model that incorporates endogenous Symbiodinium promoters, terminators, and genes of interest, as well as an internal ribosomal entry site from a Symbiodinium virus. Furthermore, we assess the potential for CRISPR/Cas9 genome editing through new analyses of the three currently available Symbiodinium genomes. Finally, we discuss how genetic engineering could be applied to enhance the stress tolerance of Symbiodinium, and in turn, coral reefs.

102 5.2 Introduction Photosynthetic dinoflagellates are critical primary producers in the aquatic environment, yet, their functional genomics are largely unexplored (Leggat et al. 2011a, Murray et al. 2016). Symbiodinium is considered one of the most important dinoflagellate genera given its role as the essential photosymbiont of many tropical reef invertebrates, notably reef-building corals (Trench and Blank 1987). Provision of photosynthetically derived metabolites from Symbiodinium to the coral host drives coral calcification and growth that forms the foundation of coral reef ecosystems (Kirk and Weis 2016, Muscatine 1990, Muscatine and Porter 1977). Thermal and light stress cause photosynthetic dysfunction of Symbiodinium and increased leakage of harmful reactive oxygen species from their cells, a process considered largely responsible for the dissociation of Symbiodinium from corals characterized as ‘coral bleaching’ (Levin et al. 2016, Suggett et al. 2008, Warner et al. 1999, Weis 2008). Symbiodinium has therefore become established as a major focus for research globally, and in effect, a model genus for dinoflagellates.

Dinoflagellates evolved an estimated 520 million years ago (Moldowan and Talyzina 1998) and exhibit substantial evolutionary divergence from model eukaryotic organisms including other microalgae such as Chlamydomonas and diatoms. Consequently, dinoflagellates possess unusual biological features that have hindered research progress, such as some of the largest known nuclear genomes (1.5-112 Gbp, typically exceeding the size of the human haploid genome), permanently condensed liquid-crystalline chromosomes, trans-splicing of polycistronic mRNAs, and plastid genomes that are divided up into minicircles (Lin et al. 2015, Murray et al. 2016, Shoguchi et al. 2013, Zhang et al. 2013). The Symbiodinium genus evolved an estimated 50 million years ago and is highly diverse, containing nine major evolutionary lineages or ‘clades’ (A- I) (Coffroth and Santos 2005, Pochon and Gates 2010, Pochon et al. 2006) with hundreds of genetically distinct ‘types/sub-clades’ considered to be different species (http://www.symbiogbr.org/, last accessed March 2017) (Tonk et al. 2013). Genetic

103 factors that promote differences in stress tolerance between Symbiodinium variants (both inter- and intra-specific) strongly influence coral gene expression and bleaching susceptibility (Baker 2001, Berkelmans and van Oppen 2006, DeSalvo et al. 2010, Howells et al. 2012, Oliver and Palumbi 2011, Rowan 2004, Yuyama et al. 2012). However, the capacity to fully explore Symbiodinium genetics is currently restricted by a lack of genetic engineering capability. Genetic engineering has been central to the study of gene function and phenotypic enhancement in organisms ranging from microbes to mammals and a key platform for socioeconomic industries and biotechnologies; yet only two cases of transgene expression in Symbiodinium have ever been validated (Ortiz-Matamoros et al. 2015b, ten Lohuis and Miller 1998):

In 1998, a type A1 strain was transformed at very low efficiencies using silicon carbide whiskers with plasmids encoding expression constructs with plant, plant-viral, and agrobacterial promoters (nos, CaMV 35S, and p1’2’) to drive transcription of antibiotic resistance genes (nptII and hptII) and a reporter gene (GUS) (ten Lohuis and Miller 1998); however, these results have yet to be reproduced. It was not until 2015 that another case of transgene expression in Symbiodinium was reported. Plasmids encoding expression constructs with plant and plant-viral promoters (nos and double CaMV 35S) to drive transcription of a herbicide resistance gene (bar) and a reporter gene (GFP) were introduced to type A1, B1, and F1 strains using glass beads. Whilst cells transiently exhibited improved herbicide resistance and suggestive GFP signal, transformations were not validated through DNA, RNA, or protein analysis (Ortiz-Matamoros et al. 2015a). Further transformation of these strains was attempted using Agrobacterium carrying plasmids with the same expression constructs, but the transformants were transient and unable to divide (Ortiz-Matamoros et al. 2015b). Of these studies, none attempted manipulation of ecologically-relevant genes thereby limiting new insight gained into Symbiodinium biology.

104 Therefore, in order to overcome the bottleneck that has become established in transforming Symbiodinium (and other dinoflagellates), we recommend a new approach that capitalizes on the recent surge in ‘omics’ breakthroughs (Figure 1). By evaluating the rapidly increasing supply of NGS data, we propose a genetic engineering framework for Symbiodinium that may markedly advance our understanding of these important dinoflagellates. Furthermore, genetic manipulation of Symbiodinium in order to reduce coral bleaching has been highlighted as a strategy to facilitate coral management as reefs continue to rapidly deteriorate under climate change (van Oppen et al. 2015). Combatting the impacts of climate change and conserving marine organisms are both key goals for sustainable development set forth by the United Nations (https://sustainabledevelopment.un.org/?menu=1300, last accessed March 2017). Thus, we believe genetic engineering of Symbiodinium may open a novel avenue to achieve these goals by protecting corals from climate change.

105

Figure 1. Breakthroughs in NGS of Symbiodinium. A timeline highlighting the key genomic (grey), transcriptomic (blue), and virus RNA (red) findings from recent NGS studies of Symbiodinium.

5.3 Tailoring a genetic engineering framework for Symbiodinium Fundamental components of Symbiodinium biology have recently been uncovered through a boom in NGS (Figure 1), particularly the assembly of the first Symbiodinium genomes (Aranda et al. 2016, Lin et al. 2015, Shoguchi et al. 2013) and transcriptomes (Barshis et al. 2014, Baumgarten et al. 2013, Bayer et al. 2012, Gierz et al. 2017, Levin et al. 2016, Parkinson et al. 2016, Rosic et al. 2015, Xiang et al. 2015, Zhang et al. 2013), direct correlation between Symbiodinium transcriptional and physiological states (Gierz et al. 2017, Levin et al. 2016, Xiang et al. 2015), and discovery of genes from viruses actively infecting Symbiodinium cells (Correa et al. 2013a, Levin et al. 2017a). Furthermore, NGS of

106 Symbiodinium has revealed genetic elements that may allow for transformation of Symbiodinium. In the following sections, we detail how unique Symbiodinium promoters, specific Symbiodinium genes underpinning important phenotypes, and a viral internal ribosomal entry site recognized by Symbiodinium ribosomes could be integrated to build expression constructs for Symbiodinium.

5.3.1 Transcriptional promoters and terminators Currently, dinoflagellate nuclear genome assemblies are all from the genus Symbiodinium (types A1, B1, and F1) (Aranda et al. 2016, Lin et al. 2015, Shoguchi et al. 2013), emphasizing the importance of Symbiodinium to dinoflagellate research. The assemblies have revealed the immense size of Symbiodinium genomes with 36,850-49,109 genes, unidirectional gene orientation, prevalent gene tandem arrays, microRNAs along with putative gene targets, and unique promoter architecture (Aranda et al. 2016, Lin et al. 2015, Shoguchi et al. 2013). Rather than the traditional TATA-box of eukaryotic promoters, Symbiodinium promoters appear to have a TTTT-box that is followed by a unique transcription start site (YYANWYY), branch point (YTNAY), and acceptor for the dinoflagellate spliced leader (AG) (Lin et al. 2015). Additionally, instead of the typical eukaryotic polyadenylation signal AAUAAA, dinoflagellate terminators use AAAAG/C (Bachvaroff and Place 2008). Hence, utilization of endogenous Symbiodinium promoters and terminators (as opposed to promoters and terminators from other organisms) would likely improve expression and stability of transgenes introduced into Symbiodinium. By chance, the CAMV 35S (plant-viral) promoter happens to contain all of the described Symbiodinium promoter elements, and the CAMV 35S (plant-viral) and nos (plant) terminators both contain the dinoflagellate polyadenylation signal; this may have contributed to their ability to drive transgene expression in Symbiodinium previously (Ortiz-Matamoros et al. 2015b, ten Lohuis and Miller 1998).

Recent transcriptomic studies have identified highly expressed Symbiodinium nuclear genes that can be genome-mapped to uncover strong, endogenous promoters

107 and their corresponding terminators. These promoters and terminators can be isolated from purified genomic DNA (gDNA) through PCR and incorporated into custom DNA expression constructs for Symbiodinium (Figure 2). Among the most highly expressed transcripts in Symbiodinium transcriptomes are genes for peridinin-chlorophyll a-binding protein, caroteno-chlorophyll a-c-binding protein, major basic nuclear protein 2, dinoflagellate viral nucleoprotein, and glyceraldehyde-3-phosphate dehydrogenase (Baumgarten et al. 2013, Levin et al. 2016, Parkinson et al. 2016); though all are multi- copy genes (Aranda et al. 2016, Lin et al. 2015, Shoguchi et al. 2013). Ideally, highly expressed nuclear genes chosen for promoter selection should not have high copy numbers, as their expression levels may largely be due to prevalence in the genome rather than strong promoters. Constitutively expressed genes are also desirable for selection of promoters that drive consistent transcription regardless of experimental conditions, and thus, drive reliable transgene expression.

To illustrate this approach of Symbiodinium promoter selection, we examined NGS data from a type A1 Symbiodinium strain for which the nuclear genome has been recently sequenced (Aranda et al. 2016) and the transcriptional responses to various conditions (temperatures, ionic stress, dark stress, and contrasting circadian rhythm time points) have been determined (Baumgarten et al. 2013). Locus 144 and Locus 1768 in the type A1 transcriptome, a subunit of a large neutral amino acids transporter and a putative ATP- binding cassette transporter gene, both show high expression across all conditions (average expression in the top 2% of all genes) (Baumgarten et al. 2013) and map tightly to the type A1 genome scaffolds 710 and 484, respectively. No significant open reading frames are found > 5 kilobases (kb) up- or down-stream of either gene, confirming that they are not part of tandem arrays. For each gene, all Symbiodinium promoter elements are within 1 kb of the start codon, and the dinoflagellate polyadenylation signal is found ~300 base pairs (bp) after the stop codon. These promoter and terminator regions could therefore be isolated and utilized to drive high and consistent expression of transgenes in a Symbiodinium expression construct.

108

5.3.2 Genes of interest Recent transcriptomic studies have been fundamental in the discovery of Symbiodinium nuclear genes that underpin phenotypic traits, such as those related to cell adhesion (e.g., GspB, Svep1, Slap1) (Xiang et al. 2015), sexual reproduction (e.g., Msh4, Msh5, Spo11-2) (Chi et al. 2014, Gierz et al. 2017, Levin et al. 2016), antiviral response (e.g., Birc3, Ns1bp, Ifih1) (Levin et al. 2017a), and antioxidant activity/thermal tolerance (e.g., Fe-sod, Mn-sod, Pxrd, Hsp70) (Gierz et al. 2017, Levin et al. 2016). Symbiodinium antioxidant genes are of particular interests because of their potential role in defining bleaching susceptibility of the coral host (Krueger et al. 2015, Levin et al. 2016). For instance, iron-type superoxide dismutase (Fe-sod) genes are believed to minimize thermally induced oxidative damage to photosynthetic apparatuses and leakage of harmful reactive oxygen species from type C1 Symbiodinium cells – determinants of coral bleaching (Weis 2008); however, these genes are not expressed at detectable levels in all Symbiodinium variants (Krueger et al. 2015, Levin et al. 2016). A Fe-sod gene could therefore be inserted after a strong Symbiodinium promoter in an expression construct to drive its over-expression for evaluation of its phenotypic influence on Symbiodinium. Endogenous genes of interest should be isolated through PCR of complementary DNA (cDNA) reverse transcribed from purified mRNA, since gDNA introns may prevent proper expression in constructs (Figure 2).

Expression of exogenous genes of interest in Symbiodinium could also greatly advance investigations of ecological processes central to coral reef health. For instance, documenting competition between Symbiodinium types, transmission and acquisition of Symbiodinium types by the coral host, and shuffling of Symbiodinium types within host tissues (Berkelmans and van Oppen 2006, Boulotte et al. 2016, Byler et al. 2013, Little et al. 2004, Toller et al. 2001, van Oppen et al. 2001) is currently reliant upon sequencing since it is not possible to visually differentiate many types. As a result, studies have been restricted to low temporal and spatial resolution relative to real-time imaging. Instead, the

109 ability to color-code Symbiodinium types through genetic transformation with various fluorescent proteins could illuminate these phenomena by enabling real-time imaging for visually differentiating types. Additionally, tagging endogenous genes of interest through fluorescent protein fusions would permit imaging of protein localization within Symbiodinium cells and potential protein secretion out of Symbiodinium cells (Xiang et al. 2015). When selecting appropriate fluorescent proteins, it will be imperative to consider the extreme autofluorescence of Symbiodinium (Shaner et al. 2005); for example, venus (excitation/emission: 515/528 nm), tdTomato (excitation/emission: 554/581 nm), and mCherry (excitation/emission: 587/610 nm) are promising candidates as their fluorescence properties are off-peak of the Symbiodinium excitation and emission spectra (Hennige et al. 2009, Jiang et al. 2012). Finally, codon optimization may be necessary for optimal exogenous gene expression in Symbiodinium since codon usage of Symbiodinium genes can be divergent from foreign genes (Levin et al. 2017a) and even between Symbiodinium nuclear and minicircle genes (Bayer et al. 2012).

5.3.3 Selectable marker genes Although antibiotics have previously been used to select transformed Symbiodinium (Ortiz-Matamoros et al. 2015b, ten Lohuis and Miller 1998), their use is problematic for two main reasons. Firstly, eliminating wild-type Symbiodinium in culture requires high concentrations of antibiotics (e.g., 3 mg/ml of G418 or hygromycin) (ten Lohuis and Miller 1998), making experimentation and long-term maintenance of transformed cell lines extremely costly. It is also important to note that natural antibiotic resistances are not uniform across all strains (Table S1), so dosage curves are necessary before conducting transformation trials. Secondly, dinoflagellates including Symbiodinium require symbiotic bacteria to grow optimally (Alavi et al. 2001, Croft et al. 2005, Miller and Belas 2006, Ritchie 2012). Since eukaryotic antibiotics can also be toxic to prokaryotes (Colanduoni and Villafranca 1986, Gonzalez et al. 1978, Pline et al. 2001, Vicens and Westhof 2003), bacterial communities in Symbiodinium cultures are removed during antibiotic selection.

110

To preserve symbiotic bacteria, alternatives to antibiotic selection markers should be considered, such as genes that provide growth advantages under specific conditions by increasing pathogen resistance, increasing thermal tolerance, or allowing for utilization of non-metabolized carbohydrates (Breyer et al. 2014). The precise functions of these alternative marker genes (e.g., Phosphomannose isomerase) are well defined and shown to be applicable to many photosynthetic species (Stoykova and Stoeva-Popova 2011), though their compatibilities with dinoflagellates are unknown. Discovery of endogenous selectable markers should therefore also be pursued. Recent Symbiodinium transcriptomic studies have uncovered genes involved in selection-relevant phenotypes like photosynthetic ability at unique light regimes (Parkinson et al. 2016) or tolerance to increased temperature regimes (Levin et al. 2016). These Symbiodinium genes could first be expressed in more easily transformed microalgae like Chlamydomonas and diatoms to gauge the potential for their up-regulation to grant a significant selectable advantage under specific conditions.

5.3.4 Viral elements Viral promoters and terminators, internal ribosome entry sites (IRES), and 2A peptides are staple regulatory elements incorporated in expression constructs since they have evolved to be recognized by eukaryotic machinery for efficient and stable foreign gene expression (Benfey and Chua 1990, Levin et al. 2014, Martıneź -Salas 1999). Symbiodinium transcriptomics have led to the discovery of genes, as well as an entire RNA genome, from novel eukaryotic viruses that infect Symbiodinium (Correa et al. 2013a, Levin et al. 2017a). A putative viral IRES, which allows cap-independent translation to produce separate proteins from one mRNA transcript, was found between the two open reading frames in the RNA genome of the +ssRNA virus infecting type C1 Symbiodinium (GenBank accession: KX538960 and KX787934) (Levin et al. 2017a). The +ssRNA virus transcripts were extremely abundant in a type C1 Symbiodinium transcriptome (Levin et

111 al. 2017a), and such rampant +ssRNA virus replication indicates that Symbiodinium ribosomes have high affinity to this IRES.

IRES enable the creation of polycistronic constructs transcriptionally controlled by a single promoter (Martıneź -Salas 1999). By permitting simultaneous expression of two independent proteins from one mRNA, a bicistronic construct can achieve long-term expression of a gene of interest because the gene of interest is transcriptionally fused to the selectable marker gene (Gurtu et al. 1996) (Figure 2). Conversely, in monocistronic constructs, the selectable marker gene often maintains expression, while the gene of interest becomes transcriptionally repressed over time if it does not increase fitness of the cell (Allera-Moreau et al. 2007). Therefore, the IRES from the Symbiodinium +ssRNA virus is a valuable viral element that is recognized by Symbiodinium ribosomes and may improve the stability of transgene expression in Symbiodinium. Moving forward, NGS data from coral holobiont metagenomes (Correa et al. 2016, Weynberg et al. 2014) and isolated Symbiodinium viruses (Weynberg et al. 2017) should be mined for promoter, terminator, and other regulatory elements of Symbiodinium viruses, given the proven benefits of viral elements to genetic engineering. Once assembled, the Symbiodinium expression construct (Figure 2) can be combined with the backbone of a standard cloning plasmid; added into an artificial, replicating minicircle (Karas et al. 2015, Nehlsen et al. 2006); or serve as a repair template for CRISPR/Cas9 genome editing (Cong et al. 2013).

112

Figure 2. Design of a tailored expression construct for Symbiodinium. Genetic elements that can be isolated from Symbiodinium cells: Symbiodinium genomic DNA (dark grey), Symbiodinium messenger RNA (blue), resident virus genomic DNA (light grey), resident virus genomic or messenger RNA (red). Solid lines (identified elements) and dashed or dotted lines (unidentified elements) are used to arrange the elements into a Symbiodinium expression construct. The pictured Symbiodinium cell (type C1) was stained with DAPI for imaging on a DeltaVision OMX Blaze microscope (excitation/emission: 405 nm/419-465 nm).

5.4 CRISPR/Cas9 genome editing and Symbiodinium Within the past five years, CRISPR/Cas9 has revolutionized genome editing by allowing precise changes to be made to target sites in the genome (Baek et al. 2016, Cong et al. 2013, Nymark et al. 2016). In short, a single guide RNA (sgRNA) is designed to recruit the Cas9 endonuclease protein and to match a specific, desired target site in the genome

113 that must be immediately followed by a protospacer adjacent motif (PAM) sequence (5’- NGG-3’). Once complexed with Cas9, the sgRNA guides Cas9 to the target genome site. Cas9 then interacts with the PAM sequence and creates a double-strand break in the target site. The cell can either repair the double stranded break through non-homologous end joining (NHEJ) or homology-directed repair (HDR) (Ran et al. 2013). NHEJ genome editing arises from introduction of a random mutation/insertion/deletion when the broken ends of DNA are directly ligated, which can cause the target gene to be knocked out (i.e., non-functional). Gene knockout provides insight into the role and criticality of a gene by assessing the effect of its absence. Alternatively, HDR genome editing uses a repair template flanked by 5’ and 3’ homologous arm sequences that match the up- and down-stream regions of the double-stranded break. The repair template can be designed for gene knockout, introduction of a specific mutation/insertion/deletion, or genomic integration of a transgene(s)/entire expression construct (Ran et al. 2013).

Symbiodinium exhibits an asexual haploid vegetative stage (Santos and Coffroth 2003) with sister chromatids developing in S-phase of the cell cycle (Watrin and Legagneux 2003), but HDR has yet to be directly observed in Symbiodinium. Therefore, CRISPR/Cas9 genome editing of Symbiodinium may be restricted to NHEJ. Ku70, Ku80, and DNA ligase IV (genes central to NHEJ, Chu et al. 2015) are all expressed in Symbiodinium transcriptomes (Levin et al. 2016). That said, some evidence does suggest Symbiodinium can enter a transient sexual diploid stage (Chi et al. 2014, Levin et al. 2016, Wilkinson et al. 2015), which has been documented in other dinoflagellates (Figueroa et al. 2015). In yeast, ploidy shifts the dominant double-stranded break repair mechanism - diploid cells favor HDR, while haploid cells favor NHEJ (Lee et al. 1999). Moreover, genes specific to meiosis, a process during which HDR occurs (Thacker and Keeney 2016), have been found in Symbiodinium genomes and transcriptomes (Chi et al. 2014, Levin et al. 2016, Lin et al. 2015, Rosic et al. 2015). Msh4, Msh5, and Spo11-2 are all highly up-regulated at elevated temperatures (Levin et al. 2016), suggesting that HDR pathways in Symbiodinium are activated. Brca2, a gene that controls HDR (Holloman 2011), is likewise up-regulated in

114 heat stressed Symbiodinium (SM population: TR74441|c0_g1; MI population: TR63986|c0_g1) (Levin et al. 2016). Hence, the potential for genomic integration of transgenes through HDR may improve if Symbiodinium are pre-stressed. HDR in Symbiodinium may also be increased by suppression of Ku70, Ku80, or DNA ligase IV (Chu et al. 2015).

The permanently condensed chromosomes of Symbiodinium could present an obstacle for CRISPR/Cas9 genome editing by possibly limiting access of sgRNAs to certain target sites. An additional challenge for genome editing is the abundance of multi-copy genes in the large Symbiodinium genomes. Gene redundancy can prevent knockout of gene function since the CRISPR/Cas9 system is not 100% efficient, meaning uncleaved functional gene copies can remain. Additionally, CRISPR/Cas9 targeting of genes with high copy numbers has been found to decrease cell proliferation and survival likely due to an increased frequency of DNA damage events (Aguirre et al. 2016). Also, design of sgRNAs requires a sequenced genome, but only three Symbiodinium genomes - each from a separate evolutionary lineage - are currently available.

As a first step to overcome some of these limitations, we analyzed the three published Symbiodinium genomes (types A1, B1, and F1) (Aranda et al. 2016, Lin et al. 2015, Shoguchi et al. 2013) to identify conserved single copy genes. We then predicted a target site in each conserved gene with high sgRNA efficiency and specificity across the genomes (Supplementary materials and methods). Conserved target sites may permit CRISPR/Cas9 genome editing of Symbiodinium types that have yet to be sequenced. Our analysis revealed 1792 conserved single copy orthologs, 261 of which have an optimal target site compatible with all genomes (Dataset S1A). The 261 single copy orthologs for CRISPR/Cas9 genome editing were enriched for a wide array of functional gene groups of interest including cellular components for photosynthesis and biological pathways for oxidation-reduction and for response to UV-B (Figure S1, Tables S2-4). Knockout of these genes would critically improve our understanding of Symbiodinium gene function, and if

115 HDR is present in Symbiodinium, these sgRNA target sites could also be used to introduce genes of interest or entire Symbiodinium expression constructs into the genome. Furthermore, we identified sgRNA target sites in the type A1 genome scaffolds 710 and 484 (Aranda et al. 2016) immediately downstream from the potentially strong, constitutive Symbiodinium promoters discussed earlier (Dataset S1B). Assuming HDR, reporter genes such as fluorescent proteins could be introduced at these sites to measure promoter activity.

The CRISPR/Cas9 system can be carried by plasmids that contain expression constructs for the Cas9, sgRNA, and in the case of HDR, the repair template with homologous arms. Target site cleavage is improved by increased CRISPR/Cas9 construct expression (Hsu et al. 2013), so strong endogenous promoters and terminators from Symbiodinium discussed earlier could be employed for to drive transcription of Cas9 by Symbiodinium. However, transcription of sgRNAs requires RNA polymerase III (Pol III) rather than RNA polymerase II. Therefore, promoters specifically recognized by Pol III (e.g. promoter of the U6 snRNA gene) are needed. Such promoters have been isolated from other eukaryotes for sgRNA transcription; but, as discussed earlier, they contain motifs (e.g., TATA-box) that Symbiodinium lack (Clarke et al. 2013, Goomer and Kunkel 1992). In Symbiodinium, 26 U6 snRNA gene copies have been identified (see Table S5 in Shoguchi et al. 2013), one of which is unusually located in a cluster with U1, U2, U4, U5, 5S, and spliced leader snRNA genes (type B1 genome scaffold 8131) (Shoguchi et al. 2013). Thus, genomic sequences found upstream and downstream of these Symbiodinium U6 snRNA genes could be isolated and trialed in sgRNA expression constructs as potential promoters and terminators recognized by Symbiodinium Pol III. Alternatively, the CRISPR/Cas9 system can be introduced to cells as pre-complexed sgRNA and purified Cas9 protein, which can achieve higher genome editing specificity by ~10-fold compared to CRISPR/Cas9 plasmids and also removes the need to optimize Cas9 codon usage or to find appropriate promoters that will express Cas9 or sgRNAs (Zuris et al. 2015).

116 5.5 Intracellular delivery of constructs and complexes Verified delivery of expression constructs into Symbiodinium was previously achieved using silicon carbide whiskers, which yielded very few transformants (ten Lohuis and Miller 1998), and with Agrobacterium, which produced transient transformants that were unable to divide (Ortiz-Matamoros et al. 2015b). Low efficiency foreign DNA delivery may be due to obstruction by the thick, multilayer Symbiodinium cell covering comprised of an external polysaccharide or glycoprotein layer atop an internal cell wall (thecal plates and the pellicle) then finally the plasma membrane (Markell et al. 1992, Wakefield et al. 2000). To overcome this barrier, methods including high-voltage electroporation, bioballistics, microinjection, and viral transduction should be trialed. Continued exploration into Symbiodinium viruses may facilitate development of a compatible transduction system. Additionally, the first method to produce viable Symbiodinium protoplasts (cells with their cell wall removed) was developed (Levin et al. 2017b). Protoplasts have been instrumental in genetic manipulation of cell-walled organisms through somatic hybridization as well as by allowing for alternate DNA delivery methods (Davey et al. 2005). Protoplast-dependent methods such as polyethylene glycol mediated transformation (Mathur and Koncz 1998) and liposome mediated transformation (Caboche 1990) may improve efficiency of construct delivery into Symbiodinium. Cell walls also serve as a barrier to RNA/protein complexes like pre-complexed sgRNA and Cas9 protein. Thus, genome editing of Symbiodinium with pre-complexed sgRNA and Cas9 protein may require the use of protoplasts (Woo et al. 2015). Polyethylene glycol mediated transformation (Woo et al. 2015), cationic lipid transformation (Zuris et al. 2015), and electroporation (Baek et al. 2016) have all been used to effectively deliver pre- complexed sgRNA and Cas9 protein through cell membranes of other eukaryotes that lacked cell walls.

5.6 Can we reduce coral bleaching with genetically enhanced Symbiodinium? Coral reefs are the most diverse marine habitat per unit area (Knowlton et al. 2010, Reaka-Kudla et al. 1996) and provide world economies with nearly US$30 billion in

117 net benefits from goods and services annually (Cesar et al. 2003). Climate change impact models predict that most reefs will be severely damaged or lost in this century unless immediate protection efforts are made (Carpenter et al. 2008, Donner 2009, Hoegh- Guldberg 1999, Hoegh-Guldberg et al. 2007, Hughes et al. 2017, Mora et al. 2016, Pandolfi et al. 2011) prompting calls for the development of novel mitigation and restoration approaches (Piaggio et al. 2016, Rinkevich 2014, van Oppen et al. 2017, van Oppen et al. 2015). Exceptional genetic variability naturally exists within the genus Symbiodinium, suggesting that seeding vulnerable corals with more climate-change tolerant Symbiodinium variants could provide a means to reduce bleaching susceptibility of corals (van Oppen et al. 2015). Although, uptake of non-native Symbiodinium variants by corals may not be widely achievable since many coral species only associate with specific Symbiodinium types (LaJeunesse et al. 2004b). Furthermore, shifts from innately less stress tolerant Symbiodinium types to more stress tolerant Symbiodinium types (e.g., from type C2 to D) can have negative impacts on a number of coral fitness traits including growth and fecundity (Jones and Berkelmans 2011, Little et al. 2004).

Environmental bioengineering is an alternative strategy to safeguard against climate change (Piaggio et al. 2016, Solé 2015). Microalgae, such as Symbiodinium, are clear and promising candidates for genetic engineering with the aim of regaining and preserving ecosystem-climate homeostasis (Solé 2015) because they can significantly influence the health of entire ecosystems (Berkelmans and van Oppen 2006, Howells et al. 2012, Kirk and Weis 2016, Murray et al. 2016, Rowan 2004). Genetic engineering to increase stress tolerance of the Symbiodinium variants that are naturally harbored by at- risk corals holds potential to reduce bleaching susceptibility without negatively impacting the fitness of the coral host since existing Symbiodinium-coral partnerships would be preserved. Fe-sod, Mn-sod Prxd, and Hsp70 genes from Symbiodinium (Gierz et al. 2017, Goyen et al. 2017, Levin et al. 2016) are standout candidates whose engineered up- regulation may enhance thermal and bleaching tolerance by reducing heat-induced

118 oxidative damage, but thorough evaluation of how this artificial up-regulation contributes to long term fitness and the Symbiodinium-coral symbiosis would be mandatory.

Application of genetic engineering to support environmental management practices has been gaining momentum. Notably, sterile male mosquitoes have been engineered to control mosquito-borne diseases (Gabrieli et al. 2014). Field releases of the sterile males significantly reduced wild mosquito populations, supporting their value to disease control (Harris et al. 2012). Similarly, fungus-resistance has been engineered in American chestnut trees in order to restore the natural population that was nearly eradicated from the spread of a foreign fungus. Introduction of these transgenic trees into the wild may receive federal approval in just the next few years, which would make them the first threatened plant species to be restored through genetic engineering (Jacobs et al. 2013, Powell 2014).

Considering the great promise shown by genetic engineering-based approaches to promote environmental health (Jacobs et al. 2013, Powell 2014) and human health (Gabrieli et al. 2014, Harris et al. 2012, Paine et al. 2005), as well as to sustain food security (Schroeder et al. 2013), it is logical for genetic engineering to be proposed as an important component of the growing repertoire of forward-looking coral reef management approaches (Piaggio et al. 2016, van Oppen et al. 2015). Due to the urgent need to protect coral reefs from climate change, the Symbiodinium research community must commit to an all-hands-on-deck attitude to achieve and extensively test genetic enhancement of Symbiodinium and other novel reef restoration strategies in the laboratory setting. In parallel, comprehensive cost-benefit-risk evaluation of the potential ecological and socioeconomic impacts from implementation of such strategies in the natural environment must be exhaustive before field-based trials are initiated. Additionally, transparent dialogs with policy makers, coral reef managers, and the general public need to be initiated now to begin the process of education and public acceptance of genetic engineering approaches for coral reef mitigation and restoration.

119

As we have discussed here, recent NGS breakthroughs have revealed natural genetic elements of Symbiodinium and their viruses (Figure 1). Based on these discoveries, we have developed a tailored genetic engineering framework for Symbiodinium that may also be applicable to other dinoflagellate genera. In doing so, we have opened a new prospective avenue to decode Symbiodinium functional genomics that may ultimately allow for engineering increased stress tolerance of Symbiodinium to reduce coral bleaching.

5.7 Acknowledgements We thank the Aranda and Voolstra lab for providing the type A1 (Symbiodinium microadriaticum) genome prior to its publication. Funding from the University of New South Wales and King Abdullah University of Science and Technology (KAUST) supported the analyses presented here.

5.8 Supplementary materials and methods

5.8.1 sgRNA design for functional genomics studies Orthologous genes (n=1,792) between three Symbiodinium types (A1, B1, and F1) were identified using the software Proteinortho v5.13 (Lechner et al. 2011) using genes predicted from their respective genomes (Aranda et al. 2016, Lin et al. 2015, Shoguchi et al. 2013). Genes found in type A1 (S. microadriaticum) that had orthologs in all three types and no paralogs in the intra- or inter-specific comparisons (i.e., 1-to-1 orthologs) were selected for designing sgRNAs using the R package CRISPRseek (Zhu et al. 2014). The sgRNA design and quality control for specificity and efficiency was determined as follows: CRISPRseek searched each ortholog for an optimal target sgRNA sequence approximately ~20 bp in length, the sequences of these potential sgRNA candidates were then compared to the entire type A1 genome to estimate specificity and efficiency of the sgRNA towards the target gene. The sgRNAs were ranked and ordered on the basis of their overall score and sgRNA efficiency. Out of the 1,792 genes, sgRNAs could be designed for 1,789 genes. 120

To ensure that sgRNAs for a particular gene in type A1 were specific to their corresponding orthologs in the other two species, BLASTn searches were carried out with all potential sgRNAs (n=351,691, orthologs=1,789) queried against an ortholog database from type B1 (S. minutum) and type F1 (S. kawagutii). Out of the 351,691 sgRNAs designed for 1,789 orthologs from type A1, 11,480 sgRNAs found hits to 1,208 orthologs in type B1. Similarly, 11,969 sgRNAs found hits for 1,115 orthologs in type F1. Given that our primary aim was to identify sgRNAs that can be used to target the same ortholog in all three species, 2,805 sgRNAs from 574 genes that had matches in all three types were used for further analyses. Besides type A1, these sgRNAs were scored for target specificity and efficiency against the genomes of type B1 and type F1 in separate runs of CRISPRseek with a maximum of four mismatches allowed between sgRNAs and the target sequence. Out of the initial 2,805 sgRNAs covering 574 orthologs, 2,757 (572 orthologs) could be successfully scored for type B1 and 2,782 sgRNAs (574 orthologs) could be successfully scored for type F1.

The BLASTn filtering above allowed filtering for sgRNA sequences that match the intended target gene, but did not preclude cases where short sequence stretches of sgRNAs had matches to non-target genes. To gauge the efficiency of sgRNA to the intended target gene, all flanking sequences identified for a particular sgRNA and species were run through BLAST (e-value < 10e-5, percent identity = 100%) and only those with a correct sgRNA-ortholog match were kept. This procedure resulted in 2,244 sgRNAs targeting 495 orthologs in type A1, 1,335 sgRNAs targeting 376 genes in type B1, and 1,582 sgRNAs targeting 419 orthologs in type F1. Among these, 825 sgRNAs targeting 261 orthologs were common to all three species and one sgRNA per ortholog (261 sgRNA for 261 orthologs) was retained by selection of the sgRNA with the highest average efficiency (Dataset S1A).

121 To further confirm that the designed sgRNAs target genes in the genome that are eventually transcribed, the flanking sequence of each sgRNA was BLASTed against the transcriptome of type A1 (S. microadriaticum) obtained from Baumgarten et al. (2013). Out of the 261 flanking sequences, 258 found a BLAST hit in the transcriptome (e-value < 10-5, percent identity = 100%). Furthermore, the flanking sequences were BLASTed against the genome of type A1 and 221 of the 261 sgRNAs were found to reside on different scaffolds, indicating that the designed sgRNAs are likely to target different areas of the genome.

Detection of significantly enriched gene ontology (GO) terms associated with the 261 orthologs was performed with topGO package (version 2.26.0) in R using Fisher's exact test and the "weight01" algorithm implemented in the package (Alexa and Rahnenfuhrer 2010). Significant gene ontology terms (p < 0.05) were identified (Tables S2- 4) and visualized using REVIGO (Supek et al. 2011) (Figure S1).

5.8.2 sgRNA design for promoter studies sgRNAs were designed to target upstream of the genes “Loci_144” and “Loci_1768” that were highly expressed across experimental conditions in the type A1 transcriptome (Baumgarten et al. 2013). The 1 kb sequence immediately preceding the start codon of each gene was extracted from the type A1 genome scaffolds 710 and 484 (Aranda et al. 2016) (matching Loci_144 and Loci_1768 in the type A1 transcriptome, respectively) (Baumgarten et al. 2013). These sequences were confirmed to contain all known Symbiodinium promoter elements (Lin et al. 2015), and the ends of the putative promoter regions were predicted using the method of Umarov and Solovyev (2017). The sgRNA with the highest efficiency and score located downstream from the potential splice leader acceptor (predicted promoter end) for each sequence was selected from a set of sgRNAs obtained by CRISPRseek (Zhu et al. 2014) (Dataset S1B).

122 5.9 Supplementary tables Table S1. Efficacy of G418 antibiotic capacity across diverse Symbiodinium variants. All Symbiodinium variants were cultured in the presence of 200 μg/ml kanamycin to remove bacteria. This ensured that the effects of G418 on Symbiodinium growth were not due to changes in the Symbiodinium-associated bacterial communities. Symbiodinium variants were exposed to four weeks of G418 treatment, with fresh G418 added biweekly. Growth at each G418 concentration was visually scored from - (no growth) up to +++ (growth equal to control).

0 mg/ml 1.25 mg/ml 2.5 mg/ml 5 mg/ml 10 mg/ml Culture identity ITS2 type G418 G418 G418 G418 G418 CCMP2464, rt61 A1 +++ - - - - CS159 A3 +++ + - - - CCMP2465, rt292 A3 +++ - - - - CS73 A3c +++ + - - - CCMP2456, rt379 A4 +++ - - - - CCMP2469, JS879, A13 +++ +++ + + + rt80 CCMP3345, B1 +++ ++ - - - CCMP2460, rt2 rt12 B1 +++ ++ ++ ++ ++ UTSB B1 +++ + - - - CCMP3364 B2 +++ ++ + + + HHIB C2 +++ ++ - - - rt203 C2 +++ ++ - - - PHMS T54 D1b D1–5 +++ + + - - rt401 DS1 +++ ++ + - - UTSD D1 +++ +++ + + +

CS156 F1 +++ +++ - - - UTSC F1 +++ ++ + + + CCMP2455, rt133 F2 +++ ++ + + +

123 Table S2. Significantly enriched (Fisher’s exact test, P < 0.05) biological process gene ontology categories for which there are single copy orthologs with optimal CRISPR/Cas9 target sites conserved across all three Symbiodinium genomes.

GO.ID Term Annotated Significant Expected topgoFisher GO:0006265 DNA topological change 38 3 0.37 0.0058 GO:0097040 phthiocerol biosynthetic process 13 2 0.13 0.0068 phenolic phthiocerol biosynthetic GO:0097041 13 2 0.13 0.0068 proces... GO:0000105 histidine biosynthetic process 13 2 0.13 0.0068 GO:0055114 oxidation-reduction process 1933 32 18.68 0.0072 GO:0016441 posttranscriptional gene silencing 61 3 0.59 0.0096 GO:0009410 response to xenobiotic stimulus 23 2 0.22 0.0096 GO:0035964 COPI-coated vesicle budding 1 1 0.01 0.0097 GO:0051684 maintenance of Golgi location 1 1 0.01 0.0097 GO:0090399 replicative senescence 1 1 0.01 0.0097 GO:0006007 glucose catabolic process 136 5 1.31 0.0104 ubiquitin-dependent protein GO:0006511 402 5 3.89 0.0113 catabolic pr... GO:0010224 response to UV-B 51 3 0.49 0.0132 GO:0006450 regulation of translational fidelity 21 2 0.2 0.0173 GO:0071770 DIM/DIP cell wall layer assembly 22 2 0.21 0.0189 GO:0006707 cholesterol catabolic process 2 1 0.02 0.0192 GO:0070980 biphenyl catabolic process 2 1 0.02 0.0192 plasmodesmata-mediated GO:0010497 2 1 0.02 0.0192 intercellular tra... plasma membrane ATP synthesis GO:0042777 2 1 0.02 0.0192 coupled pr... GO:0052696 flavonoid glucuronidation 2 1 0.02 0.0192 GO:0052697 xenobiotic glucuronidation 2 1 0.02 0.0192 'de novo' GDP-L-fucose biosynthetic GO:0042351 2 1 0.02 0.0192 proc... GO:0051661 maintenance of centrosome location 2 1 0.02 0.0192 GO:0007080 mitotic metaphase plate congression 2 1 0.02 0.0192 negative regulation of superoxide GO:0032929 2 1 0.02 0.0192 anion ... signal transduction involved in intra-S GO:0072428 2 1 0.02 0.0192 ... GO:0030157 pancreatic juice secretion 2 1 0.02 0.0192 negative regulation of interleukin-1 GO:0032691 2 1 0.02 0.0192 bet... DNA damage induced protein GO:0006975 2 1 0.02 0.0192 phosphorylati... negative regulation of nitric oxide GO:0045019 2 1 0.02 0.0192 bios... interspecies interaction between GO:0044419 284 2 2.74 0.0194 organis...

124 GO:0046898 response to cycloheximide 3 1 0.03 0.0287 GO:0019543 propionate catabolic process 3 1 0.03 0.0287 cellular response to thyroid hormone GO:0097067 3 1 0.03 0.0287 sti... GO:0006571 tyrosine biosynthetic process 3 1 0.03 0.0287 GO:0031100 organ regeneration 3 1 0.03 0.0287 GO:0006429 leucyl-tRNA aminoacylation 3 1 0.03 0.0287 negative regulation of tumor necrosis GO:0032720 3 1 0.03 0.0287 fa... positive regulation of interferon-beta GO:0032728 3 1 0.03 0.0287 p... GO:0048768 root hair cell tip growth 3 1 0.03 0.0287 GO:0009094 L-phenylalanine biosynthetic process 3 1 0.03 0.0287 GO:0071361 cellular response to ethanol 3 1 0.03 0.0287 negative regulation of mitochondrial GO:1901029 3 1 0.03 0.0287 out... negative regulation of interleukin-12 GO:0032695 3 1 0.03 0.0287 pr... GO:0034205 beta-amyloid formation 3 1 0.03 0.0287 GO:0009853 photorespiration 29 2 0.28 0.0318 regulation of potassium ion GO:1901016 75 2 0.72 0.038 transmembran... GO:0010125 mycothiol biosynthetic process 4 1 0.04 0.0381 GO:0071918 urea transmembrane transport 4 1 0.04 0.0381 mitochondrial electron transport, GO:0006122 4 1 0.04 0.0381 ubiqui... GO:0009102 biotin biosynthetic process 4 1 0.04 0.0381 GO:0032544 plastid translation 4 1 0.04 0.0381 GO:0006953 acute-phase response 4 1 0.04 0.0381 GO:0007059 chromosome segregation 155 3 1.5 0.0382 hydrogen ion transmembrane GO:1902600 122 4 1.18 0.0383 transport GO:0046854 phosphatidylinositol phosphorylation 77 3 0.74 0.0386 GO:0006397 mRNA processing 353 9 3.41 0.043 cytoplasmic mRNA processing body GO:0033962 5 1 0.05 0.0474 assembl... GO:0051552 flavone metabolic process 5 1 0.05 0.0474 GO:0043278 response to morphine 5 1 0.05 0.0474

125 Table S3. Significantly enriched (Fisher’s exact test, P < 0.05) molecular function gene ontology categories for which there are single copy orthologs with optimal CRISPR/Cas9 target sites conserved across all three Symbiodinium genomes.

GO.ID Term Annotated Significant Expected topgoFisher threonine-type endopeptidase GO:0004298 13 3 0.12 0.00022 activity GO:0003917 DNA topoisomerase type I activity 16 3 0.15 0.00043 GO:0050897 cobalt ion binding 6 2 0.06 0.00131 GO:0004129 cytochrome-c oxidase activity 8 2 0.08 0.00242 GO:0008266 poly(U) RNA binding 1 1 0.01 0.0095 tryptophan 5-monooxygenase GO:0004510 1 1 0.01 0.0095 activity phosphatidylcholine transporter GO:0008525 1 1 0.01 0.0095 activity GO:0050577 GDP-L-fucose synthase activity 1 1 0.01 0.0095 phenylalanine 4-monooxygenase GO:0004505 1 1 0.01 0.0095 activity GO:0071208 histone pre-mRNA DCP binding 1 1 0.01 0.0095 GO:0034617 tetrahydrobiopterin binding 1 1 0.01 0.0095 GO:0004664 prephenate dehydratase activity 1 1 0.01 0.0095 GO:0070463 tubulin-dependent ATPase activity 1 1 0.01 0.0095 dynein light intermediate chain GO:0051959 1 1 0.01 0.0095 binding GO:0008767 UDP-galactopyranose mutase activity 1 1 0.01 0.0095 glutamate-1-semialdehyde 2,1- GO:0042286 1 1 0.01 0.0095 aminomutase... GO:0047547 2-methylcitrate dehydratase activity 1 1 0.01 0.0095 GO:0000293 ferric-chelate reductase activity 1 1 0.01 0.0095 8-amino-7-oxononanoate synthase GO:0008710 1 1 0.01 0.0095 activity GO:0004399 histidinol dehydrogenase activity 1 1 0.01 0.0095 GO:0005244 voltage-gated ion channel activity 544 6 5.17 0.01393 GO:0002161 aminoacyl-tRNA editing activity 21 2 0.2 0.01673 GO:0004046 aminoacylase activity 2 1 0.02 0.0189 GO:0004783 sulfite reductase (NADPH) activity 2 1 0.02 0.0189 GO:0004771 sterol esterase activity 2 1 0.02 0.0189 GO:0004560 alpha-L-fucosidase activity 2 1 0.02 0.0189 GO:0000287 magnesium ion binding 288 7 2.74 0.02041 GO:0047830 D-octopine dehydrogenase activity 3 1 0.03 0.02822 GO:0047617 acyl-CoA hydrolase activity 3 1 0.03 0.02822 GO:0004823 leucine-tRNA ligase activity 3 1 0.03 0.02822 2-methylisocitrate dehydratase GO:0047456 3 1 0.03 0.02822 activity GO:0001972 retinoic acid binding 3 1 0.03 0.02822

126 GO:0004834 tryptophan synthase activity 3 1 0.03 0.02822 GO:0043394 proteoglycan binding 3 1 0.03 0.02822 GO:0005506 iron ion binding 355 8 3.37 0.03207 GO:0003756 protein disulfide isomerase activity 31 2 0.29 0.03484 phosphatidylinositol transporter GO:0008526 4 1 0.04 0.03745 activit... GO:0008986 pyruvate, water dikinase activity 4 1 0.04 0.03745 GO:0003796 lysozyme activity 4 1 0.04 0.03745 urea transmembrane transporter GO:0015204 4 1 0.04 0.03745 activity poly(ADP-ribose) glycohydrolase GO:0004649 4 1 0.04 0.03745 activity GO:0031210 phosphatidylcholine binding 4 1 0.04 0.03745 GO:0003727 single-stranded RNA binding 36 3 0.34 0.04309 3-oxoacyl-[acyl-carrier-protein] GO:0004315 36 2 0.34 0.04577 synthas... GO:0008705 methionine synthase activity 5 1 0.05 0.04659 methenyltetrahydrofolate GO:0004477 5 1 0.05 0.04659 cyclohydrolase ... phosphatidate cytidylyltransferase GO:0004605 5 1 0.05 0.04659 activ...

Table S4. Significantly enriched (Fisher’s exact test, P < 0.05) cellular component gene ontology categories for which there are single copy orthologs with optimal CRISPR/Cas9 target sites conserved across all three Symbiodinium genomes.

GO.ID Term Annotated Significant Expected topgoFisher GO:0048046 apoplast 51 5 0.57 0.00026 proteasome core complex, alpha- GO:0019773 6 2 0.07 0.00181 subunit c... GO:0030140 trans-Golgi network transport vesicle 13 2 0.15 0.00895 GO:0034081 polyketide synthase complex 14 2 0.16 0.01036 GO:0002102 podosome 1 1 0.01 0.01119 GO:0070765 gamma-secretase complex 1 1 0.01 0.01119 GO:0042589 zymogen granule membrane 1 1 0.01 0.01119 GO:0034715 pICln-Sm protein complex 1 1 0.01 0.01119 plasma membrane-derived GO:0042717 1 1 0.01 0.01119 chromatophore me... GO:0009941 chloroplast envelope 207 6 2.32 0.01556 GO:0009535 chloroplast thylakoid membrane 240 7 2.69 0.01842 GO:0022626 cytosolic ribosome 107 4 1.2 0.01855 GO:0070069 cytochrome complex 8 2 0.09 0.02212 GO:0036021 endolysosome lumen 2 1 0.02 0.02225 GO:0034719 SMN-Sm protein complex 2 1 0.02 0.02225

127 GO:0009337 sulfite reductase complex (NADPH) 2 1 0.02 0.02225 GO:0034709 methylosome 2 1 0.02 0.02225 GO:0030017 sarcomere 274 2 3.07 0.02254 GO:0016605 PML body 21 2 0.23 0.02273 GO:0009570 chloroplast stroma 185 6 2.07 0.02568 GO:0005730 nucleolus 381 9 4.26 0.0276 GO:0009534 chloroplast thylakoid 264 9 2.95 0.02776 GO:0005828 kinetochore microtubule 3 1 0.03 0.03319 nuclear outer membrane- GO:0042175 484 5 5.42 0.03349 endoplasmic retic… GO:0042470 melanosome 26 2 0.29 0.03392 GO:0005687 U4 snRNP 4 1 0.04 0.04401 GO:0071339 MLL1 complex 4 1 0.04 0.04401 GO:0097362 MCM8-MCM9 complex 4 1 0.04 0.04401

128 5.10 Supplementary figure

Figure S1. Functional gene groups for CRISPR/Cas9 genome editing of Symbiodinium. Significantly enriched gene ontology (GO) categories (Fisher’s exact test, P < 0.05) for which there are at least three single copy orthologs with optimal CRISPR/Cas9 target sites conserved across all three Symbiodinium genomes. Relationship graphs were generated separately for biological process (BP), cellular component (CC), and molecular function (MF) GO categories using REVIGO. Redundant categories (similarity > 0.9) were collapsed into the most common equivalent category in the UniProt database. Bubble size indicates the relative frequency of the category in the UniProt database. Darker bubble color denotes categories that contain more single copy orthologs with conserved CRISPR/Cas9 target sites. The width of the lines connecting categories represents the similarity between categories.

129 5.11 Supplementary dataset Dataset S1A-B. (A) sgRNAs targeting 261 orthologs common to all three Symbiodinium genomes. (B) sgRNAs targeting scaffolds 710 and 484 (type A1 Symbiodnium genome) for promoter studies: http://journal.frontiersin.org/article/10.3389/fmicb.2017.01220/full#supplementary- material

130 Chapter 6 General discussion

In this thesis, I applied genomic and microbiology techniques to explore Symbiodinium biology with a focus on thermal tolerance and the potential for its enhancement. This work has produced four published chapters. The major findings and impact of each chapter are summarized below.

In chapter 2, I compared the ex hospite physiological and transcriptional responses of two type C1 Symbiodinium populations that have divergent thermal tolerances, as well as in hospite bleaching thresholds. Under heat stress, both populations up-regulated meiosis-specific genes, but only the thermo-tolerant population up-regulated genes and functional gene groups involved in protein folding and reactive oxygen species scavenging. Accordingly, only the heat-stressed thermo-sensitive population suffered a decline in photosynthetic ability and an increase in reactive oxygen species production, while the heat-stressed thermo-tolerant population was unaffected. Notably, iron-type superoxide dismutase genes were only expressed – as well as up-regulated under heat stress – in the thermo-tolerant population, indicating their importance to Symbiodinium thermal tolerance. This study was the first to capture transcriptional evidence of a potential shift from asexual to sexual reproduction in Symbiodinium and was also the first to identify Symbiodinium genes and functional gene groups whose transcriptional regulation tightly aligned with both observed physiological responses and known in hospite bleaching responses to heat stress (Levin et al. 2016).

In chapter 3, I mined the transcriptome data generated in chapter 2 for expressed genes from viruses infecting the Symbiodinium populations and for Symbiodinium antiviral genes. Hundreds of genes from viruses were identified in each transcriptome, including the first genome of a Symbiodinium virus (GenBank accession: KX538960). Overall, heat stress led to more up-regulation than down-regulation of virus genes in both populations. 131 However, virus genes involved in DNA manipulation were only up-regulated in the heat- stressed thermo-sensitive population. Correspondingly, the heat-stressed thermo- sensitive population also up-regulated more antiviral genes than the heat-stressed thermo-tolerant population, suggesting that the heat-stressed thermo-sensitive population was combatting more extreme viral infections. The unique expression of virus genes and Symbiodinium antiviral genes in each population under heat stress provided the first direct support for an effect of differential viral infections and antiviral responses on Symbiodinium thermal tolerance, and thus, bleaching tolerance (Levin et al. 2017a).

In chapter 4, I digested the internal cell walls of live Symbiodinium cells to generate the first Symbiodinium protoplasts. Furthermore, I achieved fusion of Symbiodinium protoplasts. This study equips the Symbiodinium research community with a critical tool to pursue genetic modification of Symbiodinium since absence of a cell wall has been shown to enable efficient delivery of foreign DNA, RNA, and proteins into cells of other organisms for genetic engineering. Moreover, protoplast fusion has allowed for the genomes of different strains and species (even if sexually incompatible) to be combined, which can create progeny with novel genomes and traits. Generation and fusion of Symbiodinium protoplasts may therefore unlock the gateway for modifying Symbiodinium genomes to increase thermal tolerance (Levin et al. 2017b).

In chapter 5, I integrated the results of chapters 2-4 along with findings from other key Symbiodinium studies to develop a framework for genetically engineering Symbiodinium. Endogenous genes for Symbiodinium thermal tolerance (chapter 2), the internal ribosomal entry site from the genome of the Symbiodinium virus (chapter 3), and strong transcriptional promoters and terminators identified through new analyses of the three Symbiodinium genomes served as building blocks for the proposed Symbiodinium- tailored gene expression construct. Additionally, I recommended utilization of Symbiodinium protoplasts (chapter 4) to potentially enhance intracellular delivery of such expression constructs as well as CRISPR-Cas9 RNA/protein genome editing complexes.

132 Optimized genetic engineering of Symbiodinium is essential for advancing functional genomics knowledge. Furthermore, genetic manipulation of Symbiodinium may also be an applicable strategy to improve thermal tolerance and reduce coral bleaching (Levin et al. 2017c).

Moving forward, genetic investigation of Symbiodinium should be continued to further define the genes responsible for thermal tolerance. A critical area of exploration will be to determine the degree of conservation between genetic networks of different Symbiodinium types. While heat stress commonly prompts up-regulation of superoxide dismutase genes in some type C1, type A1, type D1-5, and clade F Symbiodinium (Gierz et al. 2017, Goyen et al. 2017, Levin et al. 2016); a thermo-tolerant type C1 up-regulates the Fe metalloform of superoxide dismutase (Levin et al. 2016), type A1 and type D1-5 up- regulate the Mn metalloform of superoxide dismutase (Goyen et al. 2017), and clade F up- regulates both the Mn and the Cu/Zn metalloforms of superoxide dismutase (Gierz et al. 2017). With possibly vast diversity in fundamental thermal tolerance genes across many Symbiodinium types (Krueger et al. 2014), a focus on host-generalist types may provide insights into the more typical Symbiodinium thermal tolerance genes that may influence coral bleaching susceptibility. For example, types C1 and/or C3 are found on over 70% of surveyed coral reefs (Baker 2003) and are harbored by over 50% of surveyed coral species (LaJeunesse et al. 2004a). Key thermal tolerance genes in these select Symbiodinium types could be synthetically up-regulated to create host-generalists with enhanced thermal tolerance that could be horizontally transferred to the majority of corals, and thus, may globally reduce bleaching. Still, it is critical to acknowledge that genetically enhanced Symbiodinium could pose a threat to the biodiversity of wild-type Symbiodinium. However, ecosystems rarely reach species saturation (Sax et al. 2007), so introduction of genetically enhanced Symbiodinium is unlikely to eradicate natural Symbiodinium variants. Furthermore, introduction of certain foreign species to ecosystems has been shown to benefit native species rather than harm them (e.g., introduction of Japanese seaweed to

133 North American mudflats to protect native amphipods) (Wright et al. 2014), supporting the goal of introducing genetically enhanced Symbiodinium to protect native corals

In addition to Symbiodinium genetic enhancement through somatic hybridization or genetic engineering, manipulation of associated virus and/or bacteria communities may reduce coral bleaching susceptibility. In chapter 2 of this thesis, the novel Symbiodinium +ssRNA virus was found at higher abundances in the thermo-sensitive Symbiodinium population than in the thermo-tolerant Symbiodinium population (Levin et al. 2017a). Reduction or elimination of viruses (Arias et al. 2014, Lin et al. 2014) like the Symbiodinium +ssRNA virus may improve Symbiodinium thermal tolerance, and accordingly, coral bleaching thresholds. Moreover, certain bacterial taxa (e.g., Alphaproteobacteria) have been linked to bleaching resistance in corals (Ziegler et al. 2017), which suggests that introduction of certain bacteria to thermally susceptible corals may also improve bleaching thresholds (Peixoto et al. 2017, van Oppen et al. 2017).

In conclusion, we have just begun to decipher the complex genetic networks that underpin Symbiodinium thermal tolerance. In order to move forward with more pro-active coral reef management, future investigations of Symbiodinium should continue to elucidate thermal tolerance genes, which can be used as markers for identifying naturally acclimated or adapted Symbiodinium variants and for genetic manipulation. With the majority of coral reefs projected to be severely damaged in the coming decades (Frieler et al. 2013, Hughes et al. 2017, van Hooidonk et al. 2016), extreme mitigation strategies – including but not limited to genetic enhancement of Symbiodinium thermal tolerance – may become necessary to save coral reefs from climate change.

134 References

Adamich M, Sweeney BM (1976). The preparation and characterization of Gonyaulax spheroplasts. Planta 130: 1-6.

Aguirre AJ, Meyers RM, Weir BA, Vazquez F, Zhang C-Z, Ben-David U, Cook A, Ha G, Harrington WF, Doshi MB (2016). Genomic copy number dictates a gene-independent cell response to CRISPR-Cas9 targeting. Cancer discovery 6: 914-929.

Ainsworth TD, Heron SF, Ortiz JC, Mumby PJ, Grech A, Ogawa D, Eakin CM, Leggat W (2016). Climate change disables coral bleaching protection on the Great Barrier Reef. Science 352: 338-342.

Alavi M, Miller T, Erlandson K, Schneider R, Belas R (2001). Bacterial community associated with Pfiesteria-like dinoflagellate cultures. Environmental Microbiology 3: 380-396.

Alexa A, Rahnenfuhrer J (2010). topGO: enrichment analysis for gene ontology. R package version 2.

Allera-Moreau C, Delluc-Clavières A, Castano C, Van den Berghe L, Golzio M, Moreau M, Teissié J, Arnal JF, Prats AC (2007). Long term expression of bicistronic vector driven by the FGF-1 IRES in mouse muscle. BMC Biotechnology 7: 74.

Anne J, Peberdy J (1976). Induced fusion of fungal protoplasts following treatment with polyethylene glycol. Microbiology 92: 413-417.

Aranda M, Li Y, Liew Y, Baumgarten S, Simakov O, Wilson M, Piel J, Ashoor H, Bougouffa S, Bajic VB (2016). Genomes of coral dinoflagellate symbionts highlight evolutionary adaptations conducive to a symbiotic lifestyle. Scientific Reports 6: 39734.

Arbeitman MN, Furlong EEM, Imam F, Johnson E, Null BH, Baker BS, Krasnow MA, Scott MP, Davis RW, White KP (2002). Gene expression during the life cycle of Drosophila melanogaster. Science 297: 2270-2275.

Argueso JL, Wanat J, Gemici Z, Alani E (2004). Competing crossover pathways act during meiosis in Saccharomyces cerevisiae. Genetics 168: 1805-1816.

Arias A, Thorne L, Goodfellow I (2014). Favipiravir elicits antiviral mutagenesis during virus replication in vivo. Elife 3: e03679.

135 Arif C, Daniels C, Bayer T, Banguera-Hinestroza E, Barbrook A, Howe CJ, LaJeunesse TC, Voolstra CR (2014). Assessing Symbiodinium diversity in scleractinian corals via next- generation sequencing-based genotyping of the ITS2 rDNA region. Molecular Ecology 23: 4418-4433.

Bachmann BJ, Bonner D (1959). Protoplasts from Neurospora crassa. Journal of Bacteriology 78: 550-556.

Bachvaroff TR, Place AR (2008). From stop to start: tandem gene arrangement, copy number and trans-splicing sites in the dinoflagellate Amphidinium carterae. PLoS One 3: e2929.

Baek K, Kim DH, Jeong J, Sim SJ, Melis A, Kim J-S, Jin E, Bae S (2016). DNA-free two-gene knockout in Chlamydomonas reinhardtii via CRISPR-Cas9 ribonucleoproteins. Scientific Reports 6: 30620.

Baird AH, Guest JR, Willis BL (2009). Systematic and biogeographical patterns in the reproductive biology of scleractinian corals. Annual Review of Ecology, Evolution, and Systematics 40: 551-571.

Bajaj YP (2012). Somatic hybridization in crop improvement I. Springer Science & Business Media.

Baker AC (2001). Ecosystems: reef corals bleach to survive change. Nature 411: 765-766.

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

Baker AC, Glynn PW, Riegl B (2008). Climate change and coral reef bleaching: An ecological assessment of long-term impacts, recovery trends and future outlook. Estuarine, Coastal Shelf Science 80: 435-471.

Baldi P, Hatfield GW (2002). DNA microarrays and gene expression: from experiments to data analysis and modeling. Cambridge University Press.

Barnes DDJ, Chalker BBE (1990). Calcification and photosynthesis in reef-building corals and algae. Coral Reefs Ecosystems of the World 25: 109-131.

Barshis DJ, Ladner JT, Oliver TA, Palumbi SR (2014). Lineage-specific transcriptional profiles of Symbiodinium spp. unaltered by heat stress in a coral host. Molecular Biology and Evolution 31: 1343-1352.

136 Baumgarten S, Bayer T, Aranda M, Liew YJ, Carr A, Micklem G, Voolstra CR (2013). Integrating microRNA and mRNA expression profiling in Symbiodinium microadriaticum, a dinoflagellate symbiont of reef-building corals. BMC Genomics 14: 704.

Bay LK, Doyle J, Logan M, Berkelmans R (2016). Recovery from bleaching is mediated by threshold densities of background thermo-tolerant symbiont types in a reef-building coral. Royal Society Open Science 3: 160322.

Bay RA, Palumbi SR (2015). Rapid acclimation ability mediated by transcriptome changes in reef-building corals. Genome Biology and Evolution 7: 1602-1612.

Bayer T, Aranda M, Sunagawa S, Yum LK, DeSalvo MK, Lindquist E, Coffroth MA, Voolstra CR, Medina M (2012). Symbiodinium transcriptomes: genome insights into the dinoflagellate symbionts of reef-building corals. PLoS One 7: e35269.

Becks L, Agrawal AF (2012). The evolution of sex is favoured during adaptation to new environments. PLoS Biology 10: e1001317.

Behrenfeld MJ, Prasil O, Kolber ZS, Babin M, Falkowski PG (1998). Compensatory changes in photosystem II electron turnover rates protect photosynthesis from photoinhibition. Photosynthesis Research 58: 259-268.

Benfey PN, Chua NH (1990). The cauliflower mosaic virus 35S promoter: combinatorial regulation of transcription in plants. Science 250: 959-966.

Berkelmans R, Willis B (1999). Seasonal and local spatial patterns in the upper thermal limits of corals on the inshore Central Great Barrier Reef. Coral Reefs 18: 219-228.

Berkelmans R, van Oppen MJH (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 of London B: Biological Sciences 273: 2305-2312.

Berkowitz GA, Gibbs M (1982). Effect of osmotic stress on photosynthesis studied with the isolated spinach chloroplast generation and use of reducing power. Plant Physiology 70: 1143-1148.

Bita CE, Gerats T (2013). Plant tolerance to high temperature in a changing environment: scientific fundamentals and production of heat stress-tolerant crops. Frontiers in Plant Science 4: 273.

Black MT, Karoly DJ (2016). Southern Australia’s warmest October on record: the role of ENSO and climate change. Bulletin of the American Meteorological Society 97: S118-S121.

137 Blank R (1987). Cell architecture of the dinoflagellate Symbiodinium sp. inhabiting the Hawaiian stony coral Montipora verrucosa. Marine biology 94: 143-155.

Bocker T, Barusevicius A, Snowden T, Rasio D, Guerrette S, Robbins D, Schmidt C, Burczak J, Croce CM, Copeland T (1999). hMSH5 a human MutS homologue that forms a novel heterodimer with hMSH4 and is expressed during spermatogenesis. Cancer Research 59: 816-822.

Bolger AM, Lohse M, Usadel B (2014). Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinformatics 30: 2114-2120.

Boschetti C, Carr A, Crisp A, Eyres I, Wang-Koh Y, Lubzens E, Barraclough TG, Micklem G, Tunnacliffe A (2012). Biochemical diversification through foreign gene expression in Bdelloid Rotifers. PLoS Genetics 8: e1003035.

Boss W, Grimes H, Brightman A (1984). Calcium-induced fusion of fusogenic wild carrot protoplasts. Protoplasma 120: 209-215.

Boulotte NM, Dalton SJ, Carroll AG, Harrison PL, Putnam HM, Peplow LM, van Oppen MJH (2016). Exploring the Symbiodinium rare biosphere provides evidence for symbiont switching in reef-building corals. The ISME Journal 10: 2693-2701.

Bourne DG, Morrow KM, Webster NS (2016). Insights into the coral microbiome: underpinning the health and resilience of reef ecosystems. Annual Review of Microbiology 70: 317-340.

Bravo I, Figueroa RI (2014). Towards an ecological understanding of dinoflagellate cyst functions. Microorganisms 2: 11-32.

Bravo J, Evans D (2011). Protoplast fusion for crop improvement. Plant Breeding Reviews 3: 193-218.

Breyer D, Kopertekh L, Reheul D (2014). Alternatives to antibiotic resistance marker genes for in vitro selection of genetically modified plants–scientific developments, current use, opperational access and biosafety considerations. Critical Reviews in Plant Sciences 33: 286-330.

Bricheux G, Mahoney DG, Gibbs SP (1992). Development of the pellicle and thecal plates following ecdysis in the dinoflagellate Glenodinium foliaceum. Protoplasma 168: 159-171.

Bruce TJA, Matthes MC, Napier JA, Pickett JA (2007). Stressful “memories” of plants: evidence and possible mechanisms. Plant Science 173: 603-608.

138 Brum JR, Sullivan MB (2015). Rising to the challenge: accelerated pace of discovery transforms marine virology. Nature Reviews Microbiology 13: 147-159.

Burke L, Reytar K, Spalding M, Perry A (2011). Reefs at risk revisited. World Resources Institute.

Byler KA, Carmi-Veal M, Fine M, Goulet TL (2013). Multiple symbiont acquisition strategies as an adaptive mechanism in the coral Stylophora pistillata. PLoS One 8: e59596.

Byrne D, Grzela R, Lartigue A, Audic S, Chenivesse S, Encinas S, Claverie J-M, Abergel C (2009). The polyadenylation site of Mimivirus transcripts obeys a stringent ‘hairpin rule’. Genome Research 19: 1233-1242.

Caboche M (1990). Liposome-mediated transfer of nucleic acids in plant protoplasts. Physiologia Plantarum 79: 173-176.

Cai W, Borlace S, Lengaigne M, Van Rensch P, Collins M, Vecchi G, Timmermann A, Santoso A, McPhaden MJ, Wu L (2014). Increasing frequency of extreme El Niño events due to greenhouse warming. Nature Climate Change 4: 111-116.

Cao L, Alani E, Kleckner N (1990). A pathway for generation and processing of double- strand breaks during meiotic recombination in S. cerevisiae. Cell 61: 1089-1101.

Carlson PS (1973). The use of protoplasts for genetic research. Proceedings of the National Academy of Sciences of the United States of America 70: 598-602.

Carpenter KE, Abrar M, Aeby G, Aronson RB, Banks S, Bruckner A, Chiriboga A, Cortés J, Delbeek JC, DeVantier L (2008). One-third of reef-building corals face elevated extinction risk from climate change and local impacts. Science 321: 560-563.

Castells-Roca L, García-Martínez J, Moreno J, Herrero E, Bellí G, Pérez-Ortín JE (2011). Heat shock response in yeast involves changes in both transcription rates and mRNA stabilities. PLoS One 6: e17272.

Cesar H, Burke L, Pet-Soede L (2003). The economics of worldwide coral reef degradation. Cesar Environmental Economics Consulting.

Chen Y, Lun ATL, Smyth GK (2014). Differential expression analysis of complex RNA-seq experiments using edgeR. In: Statistical Analysis of Next Generation Sequence Data. Springer.

Chi J, Parrow MW, Dunthorn M (2014). Cryptic sex in Symbiodinium (Alveolata, Dinoflagellata) is supported by an inventory of meiotic genes. Journal of Eukaryotic Microbiology 61: 322-327.

139

Chiba Y, Mineta K, Hirai MY, Suzuki Y, Kanaya S, Takahashi H, Onouchi H, Yamaguchi J, Naito S (2013). Changes in mRNA stability associated with cold stress in Arabidopsis cells. Plant & Cell Physiology 54: 180-194.

Chu VT, Weber T, Wefers B, Wurst W, Sander S, Rajewsky K, Kühn R (2015). Increasing the efficiency of homology-directed repair for CRISPR-Cas9-induced precise gene editing in mammalian cells. Nature Biotechnology 33: 543-548.

Clarke BD, Cummins DM, McColl KA, Ward AC, Doran TJ (2013). Characterization of zebrafish polymerase III promoters for the expression of short-hairpin RNA interference molecules. Zebrafish 10: 472-479.

Cocking E (1960). A method for the isolation of plant protoplasts and vacuoles. Nature 187: 962–963.

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

Colanduoni JA, Villafranca JJ (1986). Inhibition of Escherichia coli glutamine synthetase by phosphinothricin. Bioorganic Chemistry 14: 163-169.

Coll J (2006). Methodologies for transferring DNA into eukaryotic microalgae: a review. Spanish Journal of Agricultural Research 4: 316-330.

Cong L, Ran FA, Cox D, Lin S, Barretto R, Habib N, Hsu PD, Wu X, Jiang W, Marraffini LA (2013). Multiplex genome engineering using CRISPR/Cas systems. Science 339: 819-823.

Correa AM, Welsh RM, Thurber RLV (2013a). Unique nucleocytoplasmic dsDNA and +ssRNA viruses are associated with the dinoflagellate endosymbionts of corals. The ISME Journal 7: 13-27.

Correa AM, Ainsworth TD, Rosales SM, Thurber AR, Butler CR, Thurber RLV (2016). Viral outbreak in corals associated with an in situ bleaching event: atypical herpes-like viruses and a new megavirus infecting Symbiodinium. Frontiers in Microbiology 7: 127.

Correa RL, Bruckner FP, de Souza Cascardo R, Alfenas-Zerbini P (2013b). The role of F-Box proteins during viral infection. International Journal of Molecular Sciences 14: 4030-4049.

Croft MT, Lawrence AD, Raux-Deery E, Warren MJ, Smith AG (2005). Algae acquire vitamin B12 through a symbiotic relationship with bacteria. Nature 438: 90-93.

Cumbo V, Baird A, van Oppen MJH (2013). The promiscuous larvae: flexibility in the establishment of symbiosis in corals. Coral Reefs 32: 111-120.

140

D'Souza TG, Michiels NK (2010). The costs and benefits of occasional sex: theoretical predictions and a case study. Journal of Heredity 101: S34-S41.

Davey MR, Anthony P, Power JB, Lowe KC (2005). Plant protoplast technology: current status. Acta Physiologiae Plantarum 27: 117-130.

Davy S, Burchett S, Dale A, Davies P, Davy J, Muncke C, Hoegh-Guldberg O, Wilson W (2006). Viruses: agents of coral disease? Diseases of Aquatic Organisms 69: 101-110.

DeSalvo MK, Voolstra CR, Sunagawa S, Schwarz JA, Stillman JH, Coffroth MA, Szmant AM, Medina M (2008). Differential gene expression during thermal stress and bleaching in the Caribbean coral Montastraea faveolata. Molecular Ecology 17: 3952-3971.

DeSalvo MK, Sunagawa S, Fisher PL, Voolstra CR, Iglesias-Prieto R, Medina M (2010). Coral host transcriptomic states are correlated with Symbiodinium genotypes. Molecular Ecology 19: 1174-1186.

Diekmann O, Olsen J, Stam W, Bak R (2003). Genetic variation within Symbiodinium clade B from the coral genus Madracis in the Caribbean (Netherlands Antilles). Coral Reefs 22: 29-33.

Donner SD, Skirving WJ, Little CM, Oppenheimer M, Hoegh-Guldberg O (2005). Global assessment of coral bleaching and required rates of adaptation under climate change. Global Change Biology 11: 2251-2265.

Donner SD, Knutson TR, Oppenheimer M (2007). Model-based assessment of the role of human-induced climate change in the 2005 Caribbean coral bleaching event. Proceedings of the National Academy of Sciences of the United States of America 104: 5483-5488.

Donner SD (2009). Coping with commitment: projected thermal stress on coral reefs under different future scenarios. PLoS One 4: e5712.

Downs CA, Fauth JE, Halas JC, Dustan P, Bemiss J, Woodley CM (2002). Oxidative stress and seasonal coral bleaching. Free Radical Biology & Medicine 33: 533-543.

Downton W (1983). Osmotic adjustment during water stress protects the photosynthetic apparatus against photoinhibition. Plant Science Letters 30: 137-143.

Dunlap WC, Starcevic A, Baranasic D, Diminic J, Zucko J, Gacesa R, van Oppen MJH, Hranueli D, Cullum J, Long PF (2013). KEGG orthology-based annotation of the predicted proteome of Acropora digitifera: ZoophyteBase - an open access and searchable database of a coral genome. BMC Genomics 14: 509-509.

141 Eddy A, Williamson D (1957). A method of isolating protoplasts from yeast. Nature 179: 1252–1253.

Edgar RC (2004). MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucleic Acids Research 32: 1792-1797.

Edgar RC (2010). Search and clustering orders of magnitude faster than BLAST. Bioinformatics 26: 2460-2461.

Edmunds PJ, Pochon X, Levitan DR, Yost DM, Belcaid M, Putnam HM, Gates RD (2014). Long-term changes in Symbiodinium communities in Orbicella annularis in St. John, US Virgin Islands. Marine Ecology Progress Series 506: 129-144.

Elde NC, Malik HS (2009). The evolutionary conundrum of pathogen mimicry. Nature Reviews Microbiology 7: 787-797.

Fabina NS, Putnam HM, Franklin EC, Stat M, Gates RD (2013). Symbiotic specificity, association patterns, and function determine community responses to global changes: defining critical research areas for coral-Symbiodinium symbioses. Global Change Biology 19: 3306-3316.

Fellows R, Boyer J (1976). Structure and activity of chloroplasts of sunflower leaves having various water potentials. Planta 132: 229-239.

Felsenstein J (1985). Confidence limits on phylogenies: an approach using the bootstrap. Evolution 39: 783-791.

Figueroa RI, Rengefors K, Bravo I, Bensch S (2010). From homothally to heterothally: mating preferences and genetic variation within clones of the dinoflagellate Gymnodinium catenatum. Deep Sea Research Part II: Topical Studies in Oceanography 57: 190-198.

Figueroa RI, Dapena C, Bravo I, Cuadrado A (2015). The hidden sexuality of Alexandrium Minutum: an example of overlooked sex in Dinoflagellates. PLoS One 10: e0142667.

Fischer MG, Allen MJ, Wilson WH, Suttle CA (2010). Giant virus with a remarkable complement of genes infects marine zooplankton. Proceedings of the National Academy of Sciences of the United States of America 107: 19508-19513.

Fitt WK, Brown BE, Warner ME, Dunne RP (2001). Coral bleaching: interpretation of thermal tolerance limits and thermal thresholds in tropical corals. Coral Reefs 20: 51-65.

Frieler K, Meinshausen M, Golly A, Mengel M, Lebek K, Donner S, Hoegh-Guldberg O (2013). Limiting global warming to 2 °C is unlikely to save most coral reefs. Nature Climate Change 3: 165-170.

142

Gabrieli P, Smidler A, Catteruccia F (2014). Engineering the control of mosquito-borne infectious diseases. Genome Biology 15: 535.

Gamborg O (1977). Culture media for plant protoplasts. CRC Handbook series in nutrition and Foods 4: 415-422.

Gierz SL, Forêt S, Leggat W (2017). Transcriptomic analysis of thermally stressed Symbiodinium reveals differential expression of stress and metabolism genes. Frontiers in Plant Science 8: 271.

Gietz RD, Woods RA (2001). Genetic transformation of yeast. Biotechniques 30: 816-831.

Gill SS, Tuteja N (2010). Reactive oxygen species and antioxidant machinery in abiotic stress tolerance in crop plants. Plant Physiology and Biochemistry 48: 909-930.

Glynn P (1993). Coral reef bleaching: ecological perspectives. Coral Reefs 12: 1-17.

Gonzalez A, Jimenez A, Vazquez D, Davies J, Schindler D (1978). Studies on the mode of action of hygromycin B, an inhibitor of translocation in eukaryotes. Biochimica et Biophysica Acta (BBA)-Nucleic Acids and Protein Synthesis 521: 459-469.

Goomer R, Kunkel G (1992). The transcriptional start site for a human U6 small nuclear RNA gene is dictated by a compound promoter element consisting of the PSE and the TATA box. Nucleic Acids Research 20: 4903-4912.

Gordon BR, Leggat W (2010). Symbiodinium—invertebrate symbioses and the role of metabolomics. Marine Drugs 8: 2546-2568.

Goyen S, Pernice M, Szabó M, Warner ME, Ralph PJ, Suggett DJ (2017). A molecular physiology basis for functional diversity of hydrogen peroxide production amongst Symbiodinium spp. (Dinophyceae). Marine Biology 164: 46.

Grabherr MG, Haas BJ, Yassour M, Levin JZ, Thompson DA, Amit I, Adiconis X, Fan L, Raychowdhury R, Zeng Q (2011a). Full-length transcriptome assembly from RNA-Seq data without a reference genome. Nature Biotechnology 29: 644-652.

Grabherr MG, Haas BJ, Yassour M, Levin JZ, Thompson DA, Amit I, Adiconis X, Fan L, Raychowdhury R, Zeng Q, Chen Z, Mauceli E, Hacohen N, Gnirke A, Rhind N, di Palma F, Birren BW, Nusbaum C, Lindblad-Toh K, Friedman N, Regev A (2011b). Trinity: reconstructing a full-length transcriptome without a genome from RNA-Seq data. Nature Biotechnology 29: 644-652.

143 Granados-Cifuentes C, Neigel J, Leberg P, Rodriguez-Lanetty M (2015). Genetic diversity of free-living Symbiodinium in the Caribbean: the importance of habitats and seasons. Coral Reefs 34: 927-939.

Great Barrier Reef Marine Park Authority G (2016). Interim report: 2016 coral bleaching event on the Great Barrier Reef. GBRMPA, Townsville.

Grosser J, Gmitter Jr F, Tusa N, Chandler J (1990). Somatic hybrid plants from sexually incompatible woody species: Citrus reticulata and Citropsis gilletiana. Plant Cell Reports 8: 656-659.

Gurtu V, Yan G, Zhang G (1996). IRES bicistronic expression vectors for efficient creation of stable mammalian cell lines. Biochemical and Biophysical Research Communications 229: 295-298.

Haas BJ, Papanicolaou A, Yassour M, Grabherr M, Blood PD, Bowden J, Couger MB, Eccles D, Li B, Lieber M (2013). De novo transcript sequence reconstruction from RNA-Seq using the Trinity platform for reference generation and analysis. Nature Protocols 8: 1494-1512.

Hand SC, Hardewig I (1996). Downregulation of cellular metabolism during environmental stress: mechanisms and implications. Annual Review of Physiology 58: 539-563.

Hansen G, Wright MS (1999). Recent advances in the transformation of plants. Trends in Plant Science 4: 226-231.

Harb A, Krishnan A, Ambavaram MMR, Pereira A (2010). Molecular and physiological analysis of drought stress in Arabidopsis reveals early responses leading to acclimation in plant growth. Plant Physiology 154: 1254-1271.

Harris AF, McKemey AR, Nimmo D, Curtis Z, Black I, Morgan SA, Oviedo MN, Lacroix R, Naish N, Morrison NI (2012). Successful suppression of a field mosquito population by sustained release of engineered male mosquitoes. Nature Biotechnology 30: 828-830.

Hartung F, Wurz-Wildersinn R, Fuchs J, Schubert I, Suer S, Puchta H (2007). The catalytically active tyrosine residues of both SPO11-1 and SPO11-2 are required for meiotic double-strand break induction in Arabidopsis. The Plant Cell 19: 3090-3099.

Hennige S, Suggett DJ, Warner ME, McDougall K, Smith DJ (2009). Photobiology of Symbiodinium revisited: bio-physical and bio-optical signatures. Coral Reefs 28: 179-195.

Hennon GMM, Ashworth J, Groussman RD, Berthiaume C, Morales RL, Baliga NS, Orellana MV, Armbrust EV (2015). Diatom acclimation to elevated CO2 via cAMP signalling and coordinated gene expression. Nature Climate Change 5: 761-765.

144 Her C, Wu X, Griswold MD, Zhou F (2003). Human MutS homologue MSH4 physically interacts with von Hippel-Lindau tumor suppressor-binding protein 1. Cancer Research 63: 865-872.

Higgins JD, Armstrong SJ, Franklin FCH, Jones GH (2004). The Arabidopsis MutS homolog AtMSH4 functions at an early step in recombination: evidence for two classes of recombination in Arabidopsis. Genes & Development 18: 2557-2570.

Hill R, Schreiber U, Gademann R, Larkum AWD, Kühl M, Ralph PJ (2004). Spatial heterogeneity of photosynthesis and the effect of temperature-induced bleaching conditions in three species of corals. Marine Biology 144: 633-640.

Hirose M, Kinzie R, Hidaka M (2001). Timing and process of entry of zooxanthellae into oocytes of hermatypic corals. Coral Reefs 20: 273-280.

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

Hoegh-Guldberg O, Mumby PJ, Hooten AJ, Steneck RS, Greenfield P, Gomez E, Harvell CD, Sale PF, Edwards AJ, Caldeira K (2007). Coral reefs under rapid climate change and ocean acidification. Science 318: 1737-1742.

Hollingsworth NM, Ponte L, Halsey C (1995). MSH5, a novel MutS homolog, facilitates meiotic reciprocal recombination between homologs in Saccharomyces cerevisiae but not mismatch repair. Genes & Development 9: 1728-1739.

Holloman WK (2011). Unraveling the mechanism of BRCA2 in homologous recombination. Nature Structural & Molecular Biology 18: 748-754.

Hong J-J, Wu T-Y, Chang T-Y, Chen C-Y (2013). Viral IRES prediction system-a web server for prediction of the IRES secondary structure in silico. PLoS One 8: e79288.

Hopwood D (1981). Genetic studies with bacterial protoplasts. Annual Review of Microbiology 35: 237-272.

Howells EJ, Beltran VH, Larsen NW, Bay LK, Willis BL, van Oppen MJH (2012). Coral thermal tolerance shaped by local adaptation of photosymbionts. Nature Climate Change 2: 116- 120.

Hsu PD, Scott DA, Weinstein JA, Ran FA, Konermann S, Agarwala V, Li Y, Fine EJ, Wu X, Shalem O (2013). DNA targeting specificity of RNA-guided Cas9 nucleases. Nature Biotechnology 31: 827-832.

145 Huang Y, Niu B, Gao Y, Fu L, Li W (2010). CD-HIT Suite: a web server for clustering and comparing biological sequences. Bioinformatics 26: 680-682.

Huertas IE, Rouco M, López-Rodas V, Costas E (2011). Warming will affect phytoplankton differently: evidence through a mechanistic approach. Proceedings of the Royal Society of London B: Biological Sciences 278: 3534-3543.

Hughes TP, Kerry JT, Álvarez-Noriega M, Álvarez-Romero JG, Anderson KD, Baird AH, Babcock RC, Beger M, Bellwood DR, Berkelmans R, Bridge TC, Butler IR, Byrne M, Cantin NE, Comeau S, Connolly SR, Cumming GS, Dalton SJ, Diaz-Pulido G, Eakin CM, Figueira WF, Gilmour JP, Harrison HB, Heron SF, Hoey AS, Hobbs J-PA, Hoogenboom MO, Kennedy EV, Kuo C-y, Lough JM, Lowe RJ, Liu G, McCulloch MT, Malcolm HA, McWilliam MJ, Pandolfi JM, Pears RJ, Pratchett MS, Schoepf V, Simpson T, Skirving WJ, Sommer B, Torda G, Wachenfeld DR, Willis BL, Wilson SK (2017). Global warming and recurrent mass bleaching of corals. Nature 543: 373-377.

Iyer LM, Aravind L, Koonin EV (2001). Common origin of four diverse families of large eukaryotic DNA viruses. Journal of Virology 75: 11720-11734.

Jacobs DF, Dalgleish HJ, Nelson CD (2013). A conceptual framework for restoration of threatened plants: the effective model of American chestnut (Castanea dentata) reintroduction. New Phytologist 197: 378-393.

Jbir-Koubaa R, Charfeddine S, Ellouz W, Saidi MN, Drira N, Gargouri-Bouzid R, Nouri-Ellouz O (2015). Investigation of the response to salinity and to oxidative stress of interspecific potato somatic hybrids grown in a greenhouse. Plant Cell, Tissue and Organ Culture (PCTOC) 120: 933-947.

Jiang J, Zhang H, Kang Y, Bina D, Lo CS, Blankenship RE (2012). Characterization of the peridinin–chlorophyll a-protein complex in the dinoflagellate Symbiodinium. Biochimica et Biophysica Acta (BBA)-Bioenergetics 1817: 983-989.

Jones AM, Berkelmans R (2011). Tradeoffs to thermal acclimation: energetics and reproduction of a reef coral with heat tolerant Symbiodinium type-D. Journal of Marine Biology 2011.

Kameya T, Takahashi N (1972). The effects of inorganic salts on fusion of protoplasts from roots and leaves of Brassica species. The Japanese Journal of Genetics 47: 215-217.

Karas BJ, Diner RE, Lefebvre SC, McQuaid J, Phillips AP, Noddings CM, Brunson JK, Valas RE, Deerinck TJ, Jablanovic J (2015). Designer diatom episomes delivered by bacterial conjugation. Nature Communications 6: 6925.

146 Keeney S, Giroux CN, Kleckner N (1997). Meiosis-specific DNA double-strand breaks are catalyzed by Spo11, a member of a widely conserved protein family. Cell 88: 375-384.

Keeney S (2007). Spo11 and the formation of DNA double-strand breaks in meiosis. In: Recombination and meiosis. Springer.

Kelley LA, Mezulis S, Yates CM, Wass MN, Sternberg MJ (2015). The Phyre2 web portal for protein modeling, prediction and analysis. Nature Protocols 10: 845-858.

Kelly KO, Dernburg AF, Stanfield GM, Villeneuve AM (2000). Caenorhabditis elegans msh-5 is required for both normal and radiation-induced meiotic crossing over but not for completion of meiosis. Genetics 156: 617-630.

Kingshott P, Griesser HJ (1999). Surfaces that resist bioadhesion. Current Opinion in Solid State and Materials Science 4: 403-412.

Kirk NL, Weis NL (2016). Animal–Symbiodinium symbioses: foundations of coral reef ecosystems. In: The mechanistic benefits of microbial symbionts. Springer.

Kito H, Kunimoto M, Kamanishi Y, Mizukami Y (1998). Protoplast fusion between Monostroma nitidum and Porphyra yezoensis and subsequent growth of hybrid plants. Journal of Applied Phycology 10: 15-21.

Klercker JA (1892). Eine methods zur isoliering lebender Protoplasten. Vetenskaps Adad Stockholm 9:463–71.

Kneitz B, Cohen PE, Avdievich E, Zhu L, Kane MF, Hou H, Kolodner RD, Kucherlapati R, Pollard JW, Edelmann W (2000). MutS homolog 4 localization to meiotic chromosomes is required for chromosome pairing during meiosis in male and female mice. Genes & Development 14: 1085-1097.

Knowlton N, Brainard RE, Fisher R, Moews M, Plaisance L, Caley MJH (2010). Coral reef biodiversity. Life in the World’s Oceans: Diversity Distribution and Abundance. Wiley- Blackwell. pp 65-74.

Kobe B, Kajava AV (2001). The leucine-rich repeat as a protein recognition motif. Current Opinion in Structural Biology 11: 725-732.

Kolber ZS, Prášil O, Falkowski PG (1998). Measurements of variable chlorophyll fluorescence using fast repetition rate techniques: defining methodology and experimental protocols. Biochimica et Biophysica Acta (BBA)-Bioenergetics 1367: 88-106.

147 Krueger T, Becker S, Pontasch S, Dove S, Hoegh-Guldberg O, Leggat W, Fisher PL, Davy SK (2014). Antioxidant plasticity and thermal sensitivity in four types of Symbiodinium sp. Journal of Phycology 50: 1035-1047.

Krueger T, Fisher PL, Becker S, Pontasch S, Dove S, Hoegh-Guldberg O, Leggat W, Davy SK (2015). Transcriptomic characterization of the enzymatic antioxidants FeSOD, MnSOD, APX and KatG in the dinoflagellate genus Symbiodinium. BMC Evolutionary Biology 15: 48.

Kumar S, Stecher G, Tamura K (2016). MEGA7: molecular evolutionary genetics analysis version 7.0 for bigger datasets. Molecular Biology and Evolution 33: 1870-1874.

Kwok AC, Mak CC, Wong FT, Wong JT (2007). Novel method for preparing spheroplasts from cells with an internal cellulosic cell wall. Eukaryotic Cell 6: 563-567.

Ladner JT, Barshis DJ, Palumbi SR (2012). Protein evolution in two co-occurring types of Symbiodinium: an exploration into the genetic basis of thermal tolerance in Symbiodinium clade D. BMC Evolutionary Biology 12: 217.

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

LaJeunesse T, Bhagooli R, Hidaka M, Done T, Schmidt G, Fitt W, Hoegh-Guldberg O (2004a). 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, Thornhill DJ, Cox EF, Stanton FG, Fitt WK, Schmidt GW (2004b). High diversity and host specificity observed among symbiotic dinoflagellates in reef coral communities from Hawaii. Coral Reefs 23: 596-603.

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, Thornhill DJ (2011). Improved resolution of reef-coral endosymbiont (Symbiodinium) species diversity, ecology, and evolution through psbA non-coding region genotyping. PLoS One 6: e29013.

Le Quéré C, Moriarty R, Andrew RM, Canadell JG, Sitch S, Korsbakken JI, Friedlingstein P, Peters GP, Andres RJ, Boden TA (2015). Global carbon budget 2015. Earth System Science Data 7: 349-396.

Lechner M, Findeiß S, Steiner L, Marz M, Stadler PF, Prohaska SJ (2011). Proteinortho: detection of (co-) orthologs in large-scale analysis. BMC Bioinformatics 12: 124.

148

Lee SE, Pâques F, Sylvan J, Haber JE (1999). Role of yeast SIR genes and mating type in directing DNA double-strand breaks to homologous and non-homologous repair paths. Current Biology 9: 767-770.

Leggat W, Yellowlees D, Medina M (2011a). Recent progress in Symbiodinium transcriptomics. Journal of Experimental Marine Biology and Ecology 408: 120-125.

Leggat W, Seneca F, Wasmund K, Ukani L, Yellowlees D, Ainsworth TD (2011b). Differential responses of the coral host and their algal symbiont to thermal stress. PLoS One 6: e26687.

Levin RA, Felsen CN, Yang J, Lin JY, Whitney MA, Nguyen QT, Tsien RY (2014). An optimized triple modality reporter for quantitative in vivo tumor imaging and therapy evaluation. PLoS One 9: e97415.

Levin RA, Voolstra CR, Weynberg KD, van Oppen MJH (2017a). Evidence for a role of viruses in the thermal sensitivity of coral photosymbionts. The ISME Journal 11: 808-812.

Levin RA, Suggett DJ, Nitschke MR, van Oppen MJH, Steinberg PD (2017b). Expanding the Symbiodinium (Dinophyceae, Suessiales) Toolkit Through Protoplast Technology. Journal of Eukaryotic Microbiology doi: 10.1111/jeu.12393.

Levin RA, Suggett DJ, Voolstra, CR, Agrawal S, Steinberg PD, Suggett DJ, van Oppen MJH (2017c). Engineering strategies to decode and enhance the genomes of coral symbionts. Frontiers in Microbiology 8: 1220.

Lewis CL, Coffroth MA (2004). The acquisition of exogenous algal symbionts by an octocoral after bleaching. Science 304: 1490-1492.

Li B, Dewey CN (2011). RSEM: accurate transcript quantification from RNA-Seq data with or without a reference genome. BMC Bioinformatics 12: 323.

Lin S, Cheng S, Song B, Zhong X, Lin X, Li W, Li L, Zhang Y, Zhang H, Ji Z (2015). The Symbiodinium kawagutii genome illuminates dinoflagellate gene expression and coral symbiosis. Science 350: 691-694.

Lin S-R, Yang H-C, Kuo Y-T, Liu C-J, Yang T-Y, Sung K-C, Lin Y-Y, Wang H-Y, Wang C-C, Shen Y-C (2014). The CRISPR/Cas9 system facilitates clearance of the intrahepatic HBV templates in vivo. Molecular Therapy-Nucleic Acids 3: e186.

149

Little AF, van Oppen MJH, Willis BL (2004). Flexibility in algal endosymbioses shapes growth in reef corals. Science 304: 1492-1494.

Loram J, Trapido-Rosenthal H, Douglas A (2007). Functional significance of genetically different symbiotic algae Symbiodinium in a coral reef symbiosis. Molecular Ecology 16: 4849-4857.

Markell D, Trench R, Iglesias-Prieto R (1992). Macromolecules associated with the cell walls of symbiotic dinoflagellates. Symbiosis 12: 19-31.

Martıneź -Salas E (1999). Internal ribosome entry site biology and its use in expression vectors. Current Opinion in Biotechnology 10: 458-464.

Mathur J, Koncz C (1998). PEG-mediated protoplast transformation with naked DNA. Methods in Molecular Biology 82: 267–276.

McArthur SL, McLean KM, Kingshott P, St John HA, Chatelier RC, Griesser HJ (2000). Effect of polysaccharide structure on protein adsorption. Colloids and Surfaces B: Biointerfaces 17: 37-48.

McCarthy DJ, Chen Y, Smyth GK (2012). Differential expression analysis of multifactor RNA-Seq experiments with respect to biological variation. Nucleic Acids Research 40: 4288-4297.

McGinty ES, Pieczonka J, Mydlarz LD (2012). Variations in reactive oxygen release and antioxidant activity in multiple Symbiodinium types in response to elevated temperature. Microbial Ecology 64: 1000-1007.

McWilliam H, Li W, Uludag M, Squizzato S, Park YM, Buso N, Cowley AP, Lopez R (2013). Analysis tool web services from the EMBL-EBI. Nucleic Acids Research 41: W597-W600.

Melchers G, Sacristán MD, Holder AA (1978). Somatic hybrid plants of potato and tomato regenerated from fused protoplasts. Carlsberg Research Communications 43: 203-218.

Miller TR, Belas R (2006). Motility is involved in Silicibacter sp. TM1040 interaction with dinoflagellates. Environmental Microbiology 8: 1648-1659.

Modrich P, Lahue R (1996). Mismatch repair in replication fidelity, genetic recombination, and cancer biology. Annual Review of Biochemistry 65: 101-133.

Moldowan JM, Talyzina NM (1998). Biogeochemical evidence for dinoflagellate ancestors in the Early Cambrian. Science 281: 1168-1170.

150 Mora C, Graham NA, Nyström M (2016). Ecological limitations to the resilience of coral reefs. Coral Reefs 35: 1-10.

Moreira D, Brochier-Armanet C (2008). Giant viruses, giant chimeras: the multiple evolutionary histories of Mimivirus genes. BMC Evolutionary Biology 8: 12.

Morey JS, Van Dolah FM (2013). Global analysis of mRNA half-lives and de novo transcription in a dinoflagellate, Karenia brevis. PLoS One 8: e66347.

Morrill LC, Loeblich AR (1981). The dinoflagellate pellicular wall layer and its occurence in the division Pyrrhophyta. Journal of Phycology 17: 315-323.

Murray SA, Suggett DJ, Doblin MA, Kohli GS, Seymour JR, Fabris M, Ralph PJ (2016). Unravelling the functional genetics of dinoflagellates: a review of approaches and opportunities. Perspectives in Phycology 3: 37-52.

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

Muscatine L. (1990). The role of symbiotic algae in carbon and energy flux in reef corals. In: Ecosystems of the World 25: Coral Reefs. Elsevier.

Nagasaki K, Shirai Y, Takao Y, Mizumoto H, Nishida K, Tomaru Y (2005). Comparison of genome sequences of single-stranded RNA viruses infecting the bivalve-killing dinoflagellate Heterocapsa circularisquama. Applied and Environmental Microbiology 71: 8888-8894.

Nehlsen K, Broll S, Bode J (2006). Replicating minicircles: generation of nonviral episomes for the efficient modification of dividing cells. Gene Therapy and Molecular Biology 10: 233-244.

Normile D (2016). El Niño's warmth devastating reefs worldwide. Science 352: 15-16.

Novak JE, Ross-Macdonald PB, Roeder GS (2001). The budding yeast Msh4 protein functions in chromosome synapsis and the regulation of crossover distribution. Genetics 158: 1013-1025.

Nymark M, Valle KC, Brembu T, Hancke K, Winge P, Andresen K, Johnsen G, Bones AM (2009). An integrated analysis of molecular acclimation to high light in the marine diatom Phaeodactylum tricornutum. PLoS One 4: e7743.

Nymark M, Sharma AK, Sparstad T, Bones AM, Winge P (2016). A CRISPR/Cas9 system adapted for gene editing in marine algae. Scientific Reports 6.

151 Oliver TA, Palumbi SR (2011). Many corals host thermally resistant symbionts in high- temperature habitat. Coral Reefs 30: 241-250.

Olmsted SS, Padgett JL, Yudin AI, Whaley KJ, Moench TR, Cone RA (2001). Diffusion of macromolecules and virus-like particles in human cervical mucus. Biophysical Journal 81: 1930-1937.

Oreskes N (2004). The scientific consensus on climate change. Science 306: 1686-1686.

Ortiz-Matamoros MF, Villanueva MA, Islas-Flores T (2015a). Transient transformation of cultured photosynthetic dinoflagellates (Symbiodinium spp.) with plant-targeted vectors. Ciencias Marinas 41: 21-32.

Ortiz-Matamoros MF, Islas-Flores T, Voigt B, Menzel D, Baluška F, Villanueva MA (2015b). Heterologous DNA uptake in cultured Symbiodinium spp. aided by Agrobacterium tumefaciens. PLoS One 10: e0132693.

Paine JA, Shipton CA, Chaggar S, Howells RM, Kennedy MJ, Vernon G, Wright SY, Hinchliffe E, Adams JL, Silverstone AL (2005). Improving the nutritional value of Golden Rice through increased pro-vitamin A content. Nature Biotechnology 23: 482-487.

Pala C (2016). Corals tie stronger El Niños to climate change. Science 354: 1210.

Pandolfi JM, Connolly SR, Marshall DJ, Cohen AL (2011). Projecting coral reef futures under global warming and ocean acidification. Science 333: 418-422.

Parkinson JE, Baumgarten S, Michell CT, Baums IB, LaJeunesse TC, Voolstra CR (2016). Gene expression variation resolves species and individual strains among coral-associated dinoflagellates within the genus Symbiodinium. Genome Biology and Evolution 8: 665-680.

Peberdy JF (1980). Protoplast fusion—a tool for genetic manipulation and breeding in industrial microorganisms. Enzyme and Microbial Technology 2: 23-29.

Peixoto R, Rosado P, Leite D, Rosado AS, Bourne D (2017). Beneficial microorganisms for corals (BMC): proposed mechanisms for coral health and resilience. Frontiers in Microbiology 8: 341.

Pereira de Carvalho Costa M, Mendes B, Mourão Filho F (2003). Somatic hybridisation for improvement of citrus rootstock: production of five new combinations with potential for improved disease resistance. Australian Journal of Experimental Agriculture 43: 1151- 1156.

152 Pernice M, Meibom A, Van Den Heuvel A, Kopp C, Domart-Coulon I, Hoegh-Guldberg O, Dove S (2012). A single-cell view of ammonium assimilation in coral–dinoflagellate symbiosis. The ISME Journal 6: 1314-1324.

Pettersen EF, Goddard TD, Huang CC, Couch GS, Greenblatt DM, Meng EC, Ferrin TE (2004). UCSF Chimera—a visualization system for exploratory research and analysis. Journal of Computational Chemistry 25: 1605-1612.

Piaggio AJ, Segelbacher G, Seddon PJ, Alphey L, Bennett EL, Carlson RH, Friedman RM, Kanavy D, Phelan R, Redford KH (2016). Is it time for synthetic biodiversity conservation? Trends in Ecology & Evolution 32: 97-107.

Platt T, Gallegos CL, Harrison WG (1980). Photoinhibition of photosynthesis in natural assemblages of marine phytoplankton. Journal of Marine Research 38: 687-701.

Pline WA, Lacy GH, Stromberg V, Hatzios KK (2001). Antibacterial activity of the herbicide glufosinate on Pseudomonas syringae pathovar glycinea. Pesticide Biochemistry and Physiology 71: 48-55.

Pochon X, Montoya-Burgos JI, Stadelmann B, Pawlowski J (2006). Molecular phylogeny, evolutionary rates, and divergence timing of the symbiotic dinoflagellate genus Symbiodinium. Molecular Phylogenetics and Evolution 38: 20-30.

Pochon X, Gates RD (2010). A new Symbiodinium clade (Dinophyceae) from soritid foraminifera in Hawai’i. Molecular Phylogenetics and Evolution 56: 492-497.

Powell W (2014). The American Chestnut's Genetic Rebirth. Scientific American 310: 68- 73.

Pozdnyakov I, Skarlato S (2012). Dinoflagellate amphiesma at different stages of the life cycle. Protistology 7: 108-115.

Pozdnyakov I, Matantseva O, Negulyaev Y, Skarlato S (2014). Obtaining spheroplasts of armored dinoflagellates and first single-channel recordings of their ion channels using patch-clamping. Marine Drugs 12: 4743-4755.

Priet S, Lartigue A, Debart F, Claverie J-M, Abergel C (2015). mRNA maturation in giant viruses: variation on a theme. Nucleic Acids Research 7: 3776-3788.

Putnam HM, Mayfield AB, Fan TY, Chen CS, Gates RD (2013). The physiological and molecular responses of larvae from the reef-building coral Pocillopora damicornis exposed to near-future increases in temperature and pCO2. Marine Biology 160: 2157-2173.

153 Quigley KM, Davies SW, Kenkel CD, Willis BL, Matz MV, Bay LK (2014). Deep-sequencing method for quantifying background abundances of Symbiodinium types: exploring the rare Symbiodinium biosphere in reef-building corals. PLoS One 9: e94297.

Ragni M, Airs RL, Hennige SJ, Suggett DJ, Warner ME, Geider RJ (2010). PSII photoinhibition and photorepair in Symbiodinium (Pyrrhophyta) differs between thermally tolerant and sensitive phylotypes. Marine Ecology Progress Series 406: 57-70.

Rakoczy-Trojanowska M (2002). Alternative methods of plant transformation-a short review. Cellular and Molecular Biology Letters 7: 849-858.

Ralph P, Gademann R, Larkum A, Kühl M (2002). Spatial heterogeneity in active chlorophyll fluorescence and PSII activity of coral tissues. Marine Biology 141: 639-646.

Ran FA, Hsu PD, Wright J, Agarwala V, Scott DA, Zhang F (2013). Genome engineering using the CRISPR-Cas9 system. Nature Protocols 8: 2281-2308.

Reaka-Kudla ML, Wilson DE, Wilson EO (1996). Biodiversity II: understanding and protecting our biological resources. Joseph Henry Press.

Reaka-Kudla ML (2001). Known and unknown biodiversity, risk of extinction and conservation strategy in the sea. In: Waters in Peril. Springer Science & Business Media.

Reddy C, Gupta MK, Mantri VA, Jha B (2007). Seaweed protoplasts: status, biotechnological perspectives and needs. Journal of Applied Phycology 20: 619–632.

Rehman AU, Szabó M, Deák Z, Sass L, Larkum A, Ralph P, Vass I (2016). Symbiodinium sp. cells produce light-induced intra-and extracellular singlet oxygen, which mediates photodamage of the photosynthetic apparatus and has the potential to interact with the animal host in coral symbiosis. New Phytologist 212: 472-484.

Rice P, Longden I, Bleasby A (2000). EMBOSS: the European molecular biology open software suite. Trends in Genetics 16: 276-277.

Rinkevich B (2014). Rebuilding coral reefs: does active reef restoration lead to sustainable reefs? Current Opinion in Environmental Sustainability 7: 28-36.

Ritchie KB (2012). Bacterial symbionts of corals and Symbiodinium. In: Beneficial microorganisms in multicellular life forms. Springer.

Robinson MD, McCarthy DJ, Smyth GK (2010). edgeR: a Bioconductor package for differential expression analysis of digital gene expression data. Bioinformatics 26: 139- 140.

154 Rohwer F, Seguritan V, Azam F, Knowlton N (2002). Diversity and distribution of coral- associated bacteria. Marine Ecology Progress Series 243: 1-10.

Rosic N, Kaniewska P, Chan C-KK, Ling EYS, Edwards D, Dove S, Hoegh-Guldberg O (2014). Early transcriptional changes in the reef-building coral Acropora aspera in response to thermal and nutrient stress. BMC Genomics 15: 1052.

Rosic N, Ling EYS, Chan C-KK, Lee HC, Kaniewska P, Edwards D, Dove S, Hoegh-Guldberg O (2015). Unfolding the secrets of coral–algal symbiosis. The ISME Journal 9: 844-856.

Rosic NN, Hoegh-Guldberg O (2010). A method for extracting a high-quality RNA from Symbiodinium sp. Journal of Applies Phycology 22: 139-146.

Rosic NN, Pernice M, Dunn S, Dove S, Hoegh-Guldberg O (2010). Differential regulation by heat stress of novel cytochrome P450 genes from the dinoflagellate symbionts of reef- building corals. Applied and Environmental Microbiology 76: 2823-2829.

Rosic NN, Pernice M, Dove S, Dunn S, Hoegh-Guldberg O (2011). Gene expression profiles of cytosolic heat shock proteins Hsp70 and Hsp90 from symbiotic dinoflagellates in response to thermal stress: possible implications for coral bleaching. Cell Stress and Chaperones 16: 69-80.

Ross C (2014). Nitric oxide and heat shock protein 90 co-regulate temperature-induced bleaching in the soft coral Eunicea fusca. Coral Reefs 33: 513-522.

Ross-Macdonald P, Roeder GS (1994). Mutation of a meiosis-specific MutS homolog decreases crossing over but not mismatch correction. Cell 79: 1069-1080.

Rossouw D, Olivares-Hernandes R, Nielsen J, Bauer FF (2009). Comparative transcriptomic approach to investigate differences in wine yeast physiology and metabolism during fermentation. Applied and Environmental Microbiology 75: 6600-6612.

Roth MS (2014). The engine of the reef: photobiology of the coral–algal symbiosis. Frontiers in Microbiology 5: 422.

Roux S, Faubladier M, Mahul A, Paulhe N, Bernard A, Debroas D, Enault F (2011). Metavir: a web server dedicated to virome analysis. Bioinformatics 27: 3074-3075.

Rowan R (2004). Coral bleaching: thermal adaptation in reef coral symbionts. Nature 430: 742-742.

Sagadevan E, Subhankar C, Kannan N, Mathivanan N (2009). Inter-specific protoplast fusion in Trichoderma spp. for enhancing enzyme production and biocontrol activity. Journal of Phytology 1: 285-298.

155

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

Sax DF, Stachowicz JJ, Brown JH, Bruno JF, Dawson MN, Gaines SD, Grosberg RK, Hastings A, Holt RD, Mayfield MM (2007). Ecological and evolutionary insights from species invasions. Trends in Ecology & Evolution 22: 465-471.

Schmieder R, Edwards R (2011). Quality control and preprocessing of metagenomic datasets. Bioinformatics 27: 863-864.

Schroeder DC, Park Y, Yoon H-M, Lee YS, Kang SW, Meints RH, Ivey RG, Choi T-J (2009). Genomic analysis of the smallest giant virus—Feldmannia sp. virus 158. Virology 384: 223- 232.

Schroeder JI, Delhaize E, Frommer WB, Guerinot ML, Harrison MJ, Herrera-Estrella L, Horie T, Kochian LV, Munns R, Nishizawa NK (2013). Using membrane transporters to improve crops for sustainable food production. Nature 497: 60-66.

Schuback N, Schallenberg C, Duckham C, Maldonado MT, Tortell PD (2015). Interacting effects of light and iron availability on the coupling of photosynthetic electron transport and CO 2-assimilation in marine phytoplankton. PLoS One 10: e0133235.

Sekine H, Ferreira RC, Pan-Hammarström Q, Graham RR, Ziemba B, De Vries SS, Liu J, Hippen K, Koeuth T, Ortmann W (2007). Role for Msh5 in the regulation of Ig class switch recombination. Proceedings of the National Academy of Sciences of the United States of America 104: 7193-7198.

Shaner NC, Steinbach PA, Tsien RY (2005). A guide to choosing fluorescent proteins. Nature Methods 2: 905-909.

Shodhan A, Lukaszewicz A, Novatchkova M, Loidl J (2014). Msh4 and Msh5 function in SC- independent chiasma formation during the streamlined meiosis of Tetrahymena. Genetics 198: 983-993.

Shoguchi E, Shinzato C, Kawashima T, Gyoja F, Mungpakdee S, Koyanagi R, Takeuchi T, Hisata K, Tanaka M, Fujiwara M (2013). Draft assembly of the Symbiodinium minutum nuclear genome reveals dinoflagellate gene structure. Current Biology 23: 1399-1408.

Shoguchi E, Shinzato C, Hisata K, Satoh N, Mungpakdee S (2015). The large mitochondrial genome of Symbiodinium minutum reveals conserved noncoding sequences between dinoflagellates and apicomplexans. Genome Biology and Evolution 7: 2237-2244.

156 Simão FA, Waterhouse RM, Ioannidis P, Kriventseva EV, Zdobnov EM (2015). BUSCO: assessing genome assembly and annotation completeness with single-copy orthologs. Bioinformatics 31: 3210-3212.

Singh A, Grover A (2008). Genetic engineering for heat tolerance in plants. Physiology and Molecular Biology of Plants 14: 155-166.

Snowden T, Acharya S, Butz C, Berardini M, Fishel R (2004). hMSH4-hMSH5 recognizes Holliday Junctions and forms a meiosis-specific sliding clamp that embraces homologous chromosomes. Molecular Cell 15: 437-451.

Solé R (2015). Bioengineering the biosphere? Ecological Complexity 22: 40-49.

Sonnberg S, Seet BT, Pawson T, Fleming SB, Mercer AA (2008). Poxvirus ankyrin repeat proteins are a unique class of F-box proteins that associate with cellular SCF1 ubiquitin ligase complexes. Proceedings of the National Academy of Sciences of the United States of America 105: 10955-10960.

Sowers AE (2013). Cell Fusion. Springer Science & Business Media.

Srivastava HS, Singh RP (1987). Role and regulation of L-glutamate dehydrogenase activity in higher plants. Phytochemistry 26: 597-610.

Stacey NJ, Kuromori T, Azumi Y, Roberts G, Breuer C, Wada T, Maxwell A, Roberts K, Sugimoto-Shirasu K (2006). Arabidopsis SPO11-2 functions with SPO11-1 in meiotic recombination. The Plant Journal 48: 206-216.

Stat M, Pochon X, Cowie ROM, Gates RD (2009). Specificity in communities of Symbiodinium in corals from Johnston Atoll. Marine Ecology Progress Series 386: 83-96.

Stat M, Bird CE, Pochon X, Chasqui L, Chauka LJ, Concepcion GT, Logan D, Takabayashi M, Toonen RJ, Gates RD (2011). Variation in Symbiodinium ITS2 sequence assemblages among coral colonies. PLoS One 6: e15854.

Stojic L, Brun R, Jiricny J (2004). Mismatch repair and DNA damage signalling. DNA Repair 3: 1091-1101.

Stoykova P, Stoeva-Popova P (2011). PMI (manA) as a nonantibiotic selectable marker gene in plant biotechnology. Plant Cell, Tissue and Organ Culture (PCTOC) 105: 141-148.

Strychar KB, Sammarco PW (2009). Exaptation in corals to high seawater temperatures: low concentrations of apoptotic and necrotic cells in host coral tissue under bleaching conditions. Journal of Experimental Marine Biology and Ecology 369: 31-42.

157 Suggett DJ, Warner ME, Smith DJ, Davey P, Hennige S, Baker NR (2008). Photosynthesis and production of hydrogen peroxide by Symbiodinium (Pyrrophyta) phylotypes with different thermal tolerances. Journal of Phycology 44: 948-956.

Suggett DJ, Vázquez-Dominguez E, Berman-Frank I (2009). Nitrogen and phosphorus limitation of oceanic microbial growth during spring in the Gulf of Aqaba. Aquatic Microbial Ecology 56: 227-239.

Suggett DJ, Goyen S, Evenhuis C, Szabó M, Pettay DT, Warner ME, Ralph PJ (2015). Functional diversity of photobiological traits within the genus Symbiodinium appears to be governed by the interaction of cell size with cladal designation. New Phytologist 208: 370- 381.

Suhre K (2005). Gene and genome duplication in Acanthamoeba polyphaga mimivirus. Journal of Virology 79: 14095-14101.

Supek F, Bošnjak M, Škunca N, Šmuc T (2011). REVIGO summarizes and visualizes long lists of gene ontology terms. PLoS One 6: e21800.

Takahashi S, Whitney S, Itoh S, Maruyama T, Badger M (2008). Heat stress causes inhibition of the de novo synthesis of antenna proteins and photobleaching in cultured Symbiodinium. Proceedings of the National Academy of Sciences of the United States of America 105: 4203-4208.

Takahashi S, Whitney SM, Badger MR (2009). Different thermal sensitivity of the repair of photodamaged photosynthetic machinery in cultured Symbiodinium species. Proceedings of the National Academy of Sciences of the United States of America 106: 3237-3242.

Tamburini BA, Tyler JK (2005). Localized histone acetylation and deacetylation triggered by the homologous recombination pathway of double-strand DNA repair. Molecular and Cellular Biology 25: 4903-4913.

Tamura K, Nei M (1993). Estimation of the number of nucleotide substitutions in the control region of mitochondrial DNA in humans and chimpanzees. Molecular Biology and Evolution 10: 512-526.

Tang L, Kwon S-Y, Kim S-H, Kim J-S, Choi JS, Cho KY, Sung CK, Kwak SS, Lee HS (2006). Enhanced tolerance of transgenic potato plants expressing both superoxide dismutase and ascorbate peroxidase in chloroplasts against oxidative stress and high temperature. Plant Cell Reports 25: 1380-1386.

Tchernov D, Gorbunov MY, de Vargas C, Yadav SN, Milligan AJ, Häggblom M, Falkowski PG (2004). Membrane lipids of symbiotic algae are diagnostic of sensitivity to thermal

158 bleaching in corals. Proceedings of the National Acadamy of Sciences of the United States of America 101: 13531-13535. ten Lohuis MR, Miller DJ (1998). Genetic transformation of dinoflagellates (Amphidinium and Symbiodinium): expression of GUS in microalgae using heterologous promoter constructs. The Plant Journal 13: 427-435.

Tercé-Laforgue T, Clément G, Marchi L, Restivo FM, Lea PJ, Hirel B (2015). Resolving the role of plant NAD-glutamate dehydrogenase: III. Overexpressing individually or simultaneously the two enzyme subunits under salt stress induces changes in the leaf metabolic profile and increases plant biomass production. Plant and Cell Physiology 56: 1918-1929.

Thacker D, Keeney S (2016). Homologous recombination during meiosis. In: DNA replication, recombination, and repair. Springer.

Thomas CD, Cameron A, Green RE, Bakkenes M, Beaumont LJ, Collingham YC, Erasmus BF, De Siqueira MF, Grainger A, Hannah L (2004). Extinction risk from climate change. Nature 427: 145-148.

Thomas L, Kendrick G, Kennington W, Richards Z, Stat M (2014). Exploring Symbiodinium diversity and host specificity in Acropora corals from geographical extremes of Western Australia with 454 amplicon pyrosequencing. Molecular Ecology 23: 3113-3126.

Thompson JD, Gibson T, Higgins DG (2002). Multiple sequence alignment using ClustalW and ClustalX. Current Protocols in Bioinformatics. Chapter 2: Unit 2.3.

Toller WW, 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: 360-373.

Tompkins JD, Wu X, Chu Y-L, Her C (2009). Evidence for a direct involvement of hMSH5 in promoting ionizing radiation induced apoptosis. Experimental Cell Research 315: 2420- 2432.

Tonk L, Bongaerts P, Sampayo EM, Hoegh-Guldberg O (2013). SymbioGBR: a web-based database of Symbiodinium associated with cnidarian hosts on the Great Barrier Reef. BMC Ecology 13: 7.

Trench RK, Blank RJ (1987). Symbiodinium microadriaticum Freudenthal, S. goreauii Sp. Nov., S. kawagutii sp. nov. and S. pilosum sp. nov.: gymnodinioid dinoflagellate symbionts of marine invertebrates. Journal of Phycology 23: 469-481.

159 Tsai JM, Wang HC, Leu JH, Hsiao HH, Wang AHJ, Kou GH, Lo CF (2004). Genomic and proteomic analysis of thirty-nine structural proteins of shrimp white spot syndrome virus. Journal of Virology 78: 11360-11370.

Tsubouchi T, Roeder GS (2005). A synaptonemal complex protein promotes homology- independent centromere coupling. Science 308: 870-873.

Umarov RK, Solovyev VV (2017). Recognition of prokaryotic and eukaryotic promoters using convolutional deep learning neural networks. PLoS One 12: e0171410.

Valencia-Sanchez MA, Liu J, Hannon GJ, Parker R (2006). Control of translation and mRNA degradation by miRNAs and siRNAs. Genes & Development 20: 515-524. van Breusegem F, Slooten L, Stassart J-M, Moens T, Botterman J, Van Montagu M, Inzé D (1999). Overproduction of Arabidopsis thaliana FeSOD confers oxidative stress tolerance to transgenic maize. Plant & Cell Physiology 40: 515-523. van Hooidonk R, Maynard J, Tamelander J, Gove J, Ahmadia G, Raymundo L, Williams G, Heron SF, Planes S (2016). Local-scale projections of coral reef futures and implications of the Paris Agreement. Scientific Reports 6: 39666. van Oppen MJH, Gates RD, Blackall LL, Cantin N, Chakravarti LJ, Chan WY, Cormick C, Crean A, Damjanovic K, Epstein H (2017). Shifting paradigms in restoration of the world's coral reefs. Global Change Biology doi: 10.1111/gcb.13647. van Oppen MJH, Palstra FP, Piquet AMT, Miller DJ (2001). Patterns of coral-dinoflagellate associations in Acropora: significance of local availability and physiology of Symbiodinium strains and host-symbiont selectivity. Proceedings of the Royal Society of London Series B- Biological Sciences 268: 1759-1767. van Oppen MJH, Oliver JK, Putnam HM, Gates RD (2015). Building coral reef resilience through assisted evolution. Proceedings of the National Academy of Sciences of the United States of America 112: 2307-2313.

Veron J, Hoegh-Guldberg O, Lenton T, Lough J, Obura D, Pearce-Kelly P, Sheppard C, Spalding M, Stafford-Smith M, Rogers A (2009). The coral reef crisis: The critical importance of < 350ppm CO2. Marine Pollution Bulletin 58: 1428-1436.

Vicens Q, Westhof E (2003). Crystal structure of geneticin bound to a bacterial 16S ribosomal RNA A site oligonucleotide. Journal of Molecular Biology 326: 1175-1188.

Vierling E (1991). The roles of heat shock proteins in plants. Annual Review of Plant Biology 42: 579-620.

160 Wakefield TS, Farmer MA, Kempf SC (2000). Revised description of the fine structure of in situ" zooxanthellae" genus Symbiodinium. The Biological Bulletin 199: 76-84.

Wang J, Jiang J, Wang Y (2013). Protoplast fusion for crop improvement and breeding in China. Plant Cell, Tissue and Organ Culture (PCTOC) 112: 131-142.

Wang LH, Liu YH, Ju YM, Hsiao YY, Fang LS, Chen CS (2008). Cell cycle propagation is driven by light–dark stimulation in a cultured symbiotic dinoflagellate isolated from corals. Coral Reefs 27: 823-835.

Wang Y, Qin J (2003). MSH2 and ATR form a signaling module and regulate two branches of the damage response to DNA methylation. Proceedings of the National Academy of Sciences of the United States of America 100: 15387-15392.

Wang Y, MacKenzie T, Morse D (2005). Purification of plastids from the dinoflagellate Lingulodinium. Marine Biotechnology 7: 659-668.

Warner ME, Fitt WK, Schmidt GW (1999). Damage to photosystem II in symbiotic dinoflagellates: a determinant of coral bleaching. Proceedings of the National Academy of Sciences of the United States of America 96: 8007-8012.

Watrin E, Legagneux V (2003). Introduction to chromosome dynamics in mitosis. Biology of the Cell 95: 507-513.

Wegley L, Edwards R, Rodriguez-Brito B, Liu H, Rohwer F (2007). Metagenomic analysis of the microbial community associated with the coral Porites astreoides. Environmental Microbiology 9: 2707-2719.

Weibull C, Bergström L (1958). The chemical nature of the cytoplasmic membrane and cell wall of Bacillus megaterium, strain M. Biochimica et Biophysica Acta 30: 340-351.

Weis VM (2008). Cellular mechanisms of cnidarian bleaching: stress causes the collapse of symbiosis. Journal of Experimental Biology 211: 3059-3066.

Weston AJ, Dunlap WC, Shick JM, Klueter A, Iglic K, Vukelic A, Starcevic A, Ward M, Wells ML, Trick CG (2012). A profile of an endosymbiont-enriched fraction of the coral Stylophora pistillata reveals proteins relevant to microbial-host interactions. Molecular & Cellular Proteomics 11: M111.015487.

Weynberg KD, Wood-Charslon EM, Suttle C, van Oppen MJH (2014). Generating viral metagenomes from the coral holobiont. Frontiers in Microbiology 5: 206.

161 Weynberg KD, Neave M, Clode PL, Voolstra CR, Brownlee C, Laffy P, Webster NS, Levin RA, Wood-Charlson EM, van Oppen MJH (2017). Prevalent and persistent viral infection in cultures of the coral algal endosymbiont Symbiodinium. Coral Reefs doi: 10.1007/s00338- 017-1568-7.

Wijffels S, Roemmich D, Monselesan D, Church J, Gilson J (2016). Ocean temperatures chronicle the ongoing warming of Earth. Nature Climate Change 6: 116-118.

Wilkinson C (2002). Status of coral reefs of the world: 2002. Australian Institute of Marine Science.

Wilkinson CR (1996). Global change and coral reefs: impacts on reefs, economies and human cultures. Global Change Biology 2: 547-558.

Wilkinson SP, Fisher PL, van Oppen MJH, Davy SK (2015). Intra-genomic variation in symbiotic dinoflagellates: recent divergence or recombination between lineages? BMC Evolutionary Biology 15: 46.

Wilson JE, Powell MJ, Hoover SE, Sarnow P (2000). Naturally occurring dicistronic cricket paralysis virus RNA is regulated by two internal ribosome entry sites. Molecular and Cellular Biology 20: 4990-4999.

Wilson W, Dale A, Davy J, Davy S (2005). An enemy within? Observations of virus-like particles in reef corals. Coral Reefs 24: 145-148.

Wilson WH, Francis I, Ryan K, Davy SK (2001). Temperature induction of viruses in symbiotic dinoflagellates. Aquatic Microbial Ecology 25: 99-102.

Wilusz CJ, Wilusz J (2004). Bringing the role of mRNA decay in the control of gene expression into focus. Trends in Genetics 20: 491-497.

Woo JW, Kim J, Kwon SI, Corvalán C, Cho SW, Kim H, Kim SG, Kim ST, Choe S, Kim JS (2015). DNA-free genome editing in plants with preassembled CRISPR-Cas9 ribonucleoproteins. Nature Biotechnology 33: 1162-1164.

Wood-Charlson EM, Weynberg KD, Suttle CA, Roux S, van Oppen MJH (2015). Metagenomic characterization of viral communities in corals: mining biological signal from methodological noise. Environmental Microbiology 17: 3440-3449.

Wright JT, Byers JE, DeVore JL, Sotka EE (2014). Engineering or food? mechanisms of facilitation by a habitat-forming invasive seaweed. Ecology 95: 2699-2706.

162 Wu L, Fan J, Belasco JG (2006). MicroRNAs direct rapid deadenylation of mRNA. Proceedings of the National Acadamy of Sciences of the United States of America 103: 4034-4039.

Xiang T, Nelson W, Rodriguez J, Tolleter D, Grossman AR (2015). Symbiodinium transcriptome and global responses of cells to immediate changes in light intensity when grown under autotrophic or mixotrophic conditions. The Plant Journal 82: 67-80.

Yamashita H, Koike K (2013). Genetic identity of free-living Symbiodinium obtained over a broad latitudinal range in the Japanese coast. Phycological Research 61: 68-80.

Yampolsky LY, Zeng E, Lopez J, Williams PJ, Dick KB, Colbourne JK, Pfrender ME (2014). Functional genomics of acclimation and adaptation in response to thermal stress in Daphnia. BMC genomics 15: 859.

Young MD, Wakefield MJ, Smyth GK, Oshlack A (2010). Method gene ontology analysis for RNA-Seq: accounting for selection bias. Genome Biology 11: R14.

Yuyama I, Harii S, Hidaka M (2012). Algal symbiont type affects gene expression in juveniles of the coral Acropora tenuis exposed to thermal stress. Marine Environmental Research 76: 41-47.

Zeiner GM, Sturm NR, Campbell DA (2003). The Leishmania tarentolae spliced leader contains determinants for association with polysomes. Journal of Biological Chemistry 278: 38269-38275.

Zhang H, Zhuang Y, Gill J, Lin S (2013). Proof that dinoflagellate spliced leader (DinoSL) is a useful hook for fishing dinoflagellate transcripts from mixed microbial samples: Symbiodinium kawagutii as a case study. Protist 164: 510-527.

Zhang M, Xiao Y, Zhu R, Zhang Q, Wang S-L (2012). Enhanced thermotolerance and ethanol tolerance in Saccharomyces cerevisiae mutated by high-energy pulse electron beam and protoplast fusion. Bioprocess and Biosystems Engineering 35: 1455-1465.

Zhu LJ, Holmes BR, Aronin N, Brodsky MH (2014). CRISPRseek: a bioconductor package to identify target-specific guide RNAs for CRISPR-Cas9 genome-editing systems. PLoS One 9: e108424.

Ziegler M, Seneca FO, Yum LK, Palumbi SR, Voolstra CR (2017). Bacterial community dynamics are linked to patterns of coral heat tolerance. Nature Communications 8: 14213.

Zimmermann U, Scheurich P (1981). High frequency fusion of plant protoplasts by electric fields. Planta 151: 26-32.

163 Zuris JA, Thompson DB, Shu Y, Guilinger JP, Bessen JL, Hu JH, Maeder ML, Joung JK, Chen Z-Y, Liu DR (2015). Cationic lipid-mediated delivery of proteins enables efficient protein- based genome editing in vitro and in vivo. Nature Biotechnology 33: 73-80.

164 Appendix A Sex, scavengers, and chaperones: transcriptome secrets of divergent Symbiodinium thermal tolerances

Declaration I certify that this publication was a direct result of my research towards this PhD and that reproduction in this thesis does not breach copyright regulations.

...... Rachel A. Levin

165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 Appendix B Evidence for a role of viruses in the thermal sensitivity of coral photosymbionts

Publication II Levin RA, Voolstra CR, Weynberg KD, van Oppen MJH (2017a). Evidence for a role of viruses in the thermal sensitivity of coral photosymbionts. The ISME Journal 11: 808-812.

Declaration I certify that this publication was a direct result of my research towards this PhD and that reproduction in this thesis does not breach copyright regulations.

...... Rachel A. Levin

181 182 183 184 185

186 Appendix C Expanding the Symbiodinium (Dinophyceae, Suessiales) toolkit through protoplast technology

Declaration I certify that this publication was a direct result of my research towards this PhD and that reproduction in this thesis does not breach copyright regulations.

...... Rachel A. Levin

187 188 189 190 191 192 193 194 195 196 197 198 Appendix D Engineering strategies to decode and enhance the genomes of coral symbionts

Declaration I certify that this publication was a direct result of my research towards this PhD and that reproduction in this thesis does not breach copyright regulations.

...... Rachel A. Levin

199 200 201 202 203 204 205 206 207 208 209 210