Figures 1-4 phylogeneticanalyses March2014 · 2016. 9. 12. · Officers for 2014(with terms of...

Phycolo

gia

VO

LUM

E 53

NU

MB

ER 4

PAG

ES 3

05–

408

4 Volume 53 Number 4

Phycologiacontents

Published by the International Phycological SocietyPrinted in USA by Allen Press, Inc.

Todd C. LaJeunesse, Drew C. Wham, D. Tye Pettay, John Everett Parkinson, Shashank Keshavmurthy and Chaolun Allen Chen. Ecologically differentiated stress-tolerant endosymbionts in the dinoflagellate genus Symbiodinium (Dinophyceae) Clade D are different species ................................................................ 305–319

Michael Schagerl and Julia Wukovits. Cultivation and inorganic carbon uptake of the rare desmid Oocardium stratum (Conjugatophyceae) ............................................................................................................ 320–328

Emily T. Johnston, Phaik-Eem Lim, Nurliah Buhari, Emily J. Keil, M. Iqbal Djawad and Morgan L. Vis. Diversity of freshwater red algae (Rhodophyta) in Malaysia and Indonesia from morphological and molecular data .. 329–341

Helena David, Aitor Laza-Martínez, Irati Miguel and Emma Orive. Broad distribution of Coolia monotis and restricted distribution of Coolia cf. canariensis (Dinophyceae) in the Atlantic coast of the Iberian Peninsula ....................................................................................................................................... 342–352

Alejandra V. González, Jessica Beltrán and Bernabé Santelices. Colonization and growth strategies in two Codium species (Bryopsidales, Chlorophyta) with different thallus forms ............................................. 353–358

Teofil Nakov, Elizabeth C. Ruck, Yuri Galachyants, Sarah A. Spaulding and Edward C. Theriot. Molecular phylogeny of the Cymbellales (Bacillariophyceae, Heterokontophyta) with a comparison of models for accommodating rate-variation across sites ....................................................................................... 359–373

Mitsunobu Kamiya and John A. West. Cryptic diversity in the euryhaline red alga Caloglossa ogasawaraensis (Delesseriaceae, Ceramiales) .......................................................................................................... 374–382

Conxi Rodríguez-Prieto, D. Wilson Freshwater and Max H. Hommersand. Morphology and phylogenetic systematics of Ptilocladiopsis horrida and proposal of the Ptilocladiopsidaceae fam. nov. (Gigartinales, Rhodophyta) ................................................................................................................................... 383–395

Mónica B.J. Moniz, Michael D. Guiry and Fabio Rindi. tufA phylogeny and species boundaries in the green algal order Prasiolales (Trebouxiophyceae, Chlorophyta) ..................................................................... 396–406

Corrigendum ......................................................................................................................................... 407

July 2014 Vol. 53 No. 4 ISSN 0031-8884

Phycologia

Editor-in-Chief Robert A. Andersen

Officers for 2014 (with terms of membership of the Board of Directors)

President (2014-2015): M. A. Borowitzka, Algae R&D Center, Biological Sciences & Biotechnology, Murdoch University, Murdoch, WA, Australia. E-mail: [email protected]

Past-President (2014-2015): G. C. Zuccarello, School of Biological Sciences, Victoria University of Wellington, Wellington, New Zealand. E-mail: [email protected]

Vice President/President-Elect (2014-2015): J.A. Brodie, Natural History Museum, Department of Life Sciences, London, UK. E-mail: [email protected]

Secretary (2012-2014): C. F. Gurgel, School of Earth & Environmental Sciences, The University of Adelaide, Adelaide, SA, Australia. E-mail: [email protected]

Treasurer (2012-2014): J. W. LaClaire II, Department of Molecular Biosciences, University of Texas, Austin, TX, USA. E-mail: [email protected]

Other Members of the Executive Council

R.A. Andersen (Editor-in-Chief, Phycologia 2012-2014), Friday Harbor Laboratories, University of Washington, Friday Harbor, WA, USA. E-mail: [email protected]

W. D. Freshwater (2013-2015), Center for Marine Science, University of North Carolina-Wilmington, NC, USA. E-mail: [email protected]

M. Giordano (2013-2015), Dipartimento di Scienze della Vita e dell’Ambiente, Univer-sità Politecnica delle Marche, Ancona, Italy. E-mail: [email protected]

K. Hoef-Emden (2012-2014), Botanical Institute, University of Cologne, Köln, Germany. E-mail: [email protected]

S.-M. Lin (2012-2014), Institute of Marine Biology, National Taiwan Ocean University, Taiwan, Republic of China. E-mail: [email protected]

N. Lundholm (2014-2016), Natural History Museum of Denmark, University of Copen-hagen, Copenhagen, Denmark. E-mail: [email protected]

D.G. Mann (IOC Chairperson 2014-2021), Royal Botanic Garden Edinburgh, Edin-burgh, Scotland, UK. E-mail: [email protected]

W. A. Nelson (2013-2015), National Institute of Water and Atmospheric Research, Wel-lington, New Zealand. E-mail: [email protected]

C. Odebrecht (2013-2015), Oceanography Institute, Federal University of Rio Grande Rio Grande, RS, Brazil. E-mail: [email protected]

M. Poulin (2012-2014), Division of Research, Canadian Museum of Nature, Ottawa, Canada. E-mail: [email protected]

A.R. Sherwood (2014-2016), Department of Botany, University of Hawaii, Honolulu, HI, USA. E-mail: [email protected]

Editor-in-Chief

R.A. Andersen, Friday Harbor Laboratories, University of Washington, Friday Harbor, WA 98250, USA. E-mail: [email protected] (Managing Editor: Ed Hiscock, [email protected])

Board of Associate Editors

Andrew Alverson, University of Arkansas, USA. E-mail: [email protected]

John Beardall, Monash University, Australia. E-mail: [email protected]

Kerstin Hoef-Emden, University of Cologne, Germany. E-mail: [email protected]

Qiang Hu, Institute of Hydrobiology, Chinese Academy of Sciences, China. E-mail: [email protected]

John Huisman, Murdoch University and Western Australian Herbarium, Australia. E-mail: [email protected]

Ken-ichiro Ishida, University of Tsukuba, Japan. E-mail: [email protected]

J. Patrick Kociolek, University of Colorado, USA. E-mail: [email protected]

Gerald T. Kraft, University of Melbourne, Australia. E-mail: [email protected]

Frederik Leliaert, Ghent University, Belgium. E-mail: [email protected]

Renhui Li, The Chinese Academy of Sciences, China. E-mail: [email protected]

Allen J. Milligan, Oregon State University, USA. E-mail: [email protected]

Mariana Cabral de Oliveira, University of São Paulo, Brazil. E-mail: [email protected]

Nicole Poulsen, Technische Universität Dresden, Germany. E-mail: [email protected]

Hwan Su Yoon, Sungkyunkwan University, Korea. E-mail: [email protected]

Webmaster

Michael D. Guiry, National University of Ireland. E-mail: [email protected]

INTERNATIONAL PHYCOLOGICAL SOCIETYDedicated to the promotion of phycology and to the encouragement of international

cooperation among phycologists and phycological institutes (http://www.intphycsoc.org).

Phycologia (0031-8884) is owned and published bimonthly (one volume per annum) in January, March, May, July, September and November by the International Phycological Society, 810 East 10th Street, Lawrence, KS 66044. Periodicals postage paid at Lawrence, KS and additional mailing offices. A subscription to Phycologia is provided to regular ($65), student ($35), retired ($35), and lifetime ($1,100) members of the society; associate members ($15) do not receive the journal. Institution and library price as of 2014 for print and online is $788 for the United States, Canada and Mexico, $830 for other countries, and $723 for online only. The journal is dispatched outside North America by means of International Surface Air Lift. POSTMASTER: Send address changes to Phycologia, Business Office, 810 East 10th Street, P.O. Box 1897, Lawrence, KS 66044-8897.

Orders, Dues, Address Changes, Membership Applications: Phycologia, Business Office, 810 East 10th Street, P.O. Box 1897, Lawrence, KS 66044-8897. Telephone 800-627-0326 (in US and Canada), 785-843-1234 (all others), Fax 785-843-6153, E-mail [email protected].

Claims for missing issues of Phycologia: Claims for copies lost in the mail must be received within 90 days (180 days foreign) of the issue date to ensure replacement at no charge.

Copyright 2014 by the International Phycological Society. For information and permission concerning copyright, contact Allen Press, 810 East 10th Street, Lawrence, KS 66044. Printed in USA by Allen Press, Inc.

Front cover: Population genetic evidence was used for the first time by LaJeunesse et al. (pp. 305–319) to describe new algal species: three new Symbiodinium species (Dinophyceae) from the morphologically cryptic Clade D. These stress-tolerant symbionts impart physiological resilience to reef-building corals in harsher environments such as Nikko Bay, Palau (shown on cover). Photo: Allison Lewis.

The Publishers' policy is to use acid-free permanent paper, to the draft standard ISO/DIS 9706, and the American Standard NISO ANSI Z39.48-1992.

Ecologically differentiated stress-tolerant endosymbionts in the dinoflagellate genus

Symbiodinium (Dinophyceae) Clade D are different species

TODD C. LAJEUNESSE1*, DREW C. WHAM

1, D. TYE PETTAY1, JOHN EVERETT PARKINSON

1, SHASHANK KESHAVMURTHY2

AND CHAOLUN ALLEN CHEN2

1Department of Biology, The Pennsylvania State University, University Park, PA 16802, USA2Biodiversity Research Center and Taiwan International Graduate Program (TIGP) – Biodiversity, Academia Sinica, Taipei,

115, Taiwan

ABSTRACT: We used an integrative genetics approach using sequences of (1) nuclear ribosomal rDNA (internal transcribedspacers and partial large subunit rDNA), (2) single-copy microsatellite nuclear DNA, (3) chloroplast-encoded 23S rDNA,(4) mitochondrial cytochrome b, and (5) repeat variation at eight microsatellite markers, to test the hypothesis that thestress-tolerant, ‘morphologically cryptic’ Clade D Symbiodinium (Dinophyceae) was composed of more than one species.Concordant phylogenetic and population genetic evidence clearly differentiate separately evolving, reproductivelyisolated lineages. We describe Symbiodinium boreum sp. nov. and S. eurythalpos sp. nov., two symbionts known to occur incolonies of the zebra coral, Oulastrea crispata (Scleractinia), which lives in turbid, marginal habitats extending fromequatorial Southeast Asia to the main islands of Japan in the temperate northwest Pacific Ocean. Symbiodinium boreumwas associated with O. crispata in temperate latitudes and S. eurythalpos was common to colonies in the tropics. Thegeographical ranges of both symbiont species overlapped in the subtropics where they sometimes co-occurred in the samehost colony. Symbiodinium trenchii sp. nov. is also described. As a host-generalist symbiont, it often occurs in symbiosiswith various species of Scleractinia possessing open (horizontal) modes of symbiont acquisition and is common to reefcoral communities thriving in warm turbid reef habitats in the western Pacific Ocean, Indian Ocean, Arabian/PersianGulf, Red Sea and western Atlantic (Caribbean). As is typical for dinoflagellates, S. boreum and S. eurythalpos werehaploid, but microsatellite loci from field-collected and cultured S. trenchii often possessed two alleles, implying that agenome-wide duplication occurred during the evolution of this species. The recognition that Clade D Symbiodiniumcontains species exhibiting marked differences in host specificity and geographical distribution will yield greater scientificclarity about how stress-tolerant symbionts function in the ecological response of coral–dinoflagellate symbioses to globalclimate change.

KEY WORDS: Clade D, Coral symbionts, Dinoflagellates, Speciation, Symbiodinium, Taxonomy

INTRODUCTION

Reef-building corals contain dense populations of ‘‘morpho-logically cryptic’’ dinoflagellates assigned to the genusSymbiodinium (Freudenthal 1962). These eukaryotic micro-bial symbionts comprise numerous genetically divergentlineages, or ‘Clades’, several of which are particularlycommon to various shallow-water cnidarians (living mostlyin the photic zones of tropical and subtropical coastalhabitats) including Clades A, B, C and D (Coffroth & Santos2005). These clades are separated by a level of gene sequencedivergence that often defines higher-level taxonomic ranks inother dinoflagellates (Rowan & Powers 1992; LaJeunesse2001; Stern et al. 2010). Detailed genetic analyses, usingmore rapidly evolving markers of samples gathered overlarge geographic scales from various host species, find thateach Symbiodinium clade contains diverse groups of phylo-genetically distinct lineages referred to by researchers as‘subclades’, ‘types’, ‘species’, or ‘strains’ (LaJeunesse 2002;Fabricious et al. 2004; LaJeunesse et al. 2010a; LaJeunesse &Thornhill 2011; Thornhill et al. 2014), with the exception ofClade E, which may be monospecific (Jeong et al. 2014).

These numerous ‘‘subcladal’’ lineages have different physi-ological and ecological attributes, further revealing that theyare genetically and functionally distinct entities (LaJeunesse2002; Rodriguez-Lanetty et al. 2004; Frade et al. 2008;Thornhill et al. 2008; Hennige et al. 2009; Sampayo et al.2008, 2009), and efforts necessary to formally describe themas separate species have been initiated (LaJeunesse et al.2012).

The nonrecognition, or the incorrect assignment, ofspecies boundaries has hindered progress in our basicunderstanding of coral–dinoflagellate symbioses, particularlyin judging their capacity to respond to climate change(LaJeunesse et al. 2012). Speculations about the present andfuture significance of thermally tolerant endosymbionts havefocused largely on Symbiodinium Clade D (Baker et al. 2004;Berkelmans & van Oppen 2006; Jones et al. 2008; LaJeunesseet al. 2009; 2010b), yet the merit of such discussions dependson the accuracy of interpreting the genetic variationobserved among Clade D populations and their correspond-ing ecology and host specificity. Such differences inperception relate to underlying problems with taxonomicresolution and how to delimit species of ‘‘morphologicallycryptic’’ Symbiodinium (Blank & Trench, 1985; LaJeunesse2001; Correa & Baker 2009; Sampayo et al. 2009; Stat &Gates 2011; LaJeunesse et al. 2012), many of which do notpersist in culture media.

* Corresponding author ([email protected]).DOI: 10.2216/13-186.1� 2014 International Phycological Society

Phycologia Volume 53 (4), 305–319 Published 12 June 2014

305

Corals with symbionts that tolerate high temperatures andlive in extreme environments are less susceptible to thenegative effects of acute thermal stress (Rowan 2004;Berkelmanns & van Oppen 2006; Warner et al. 2006), havea higher bleaching threshold (Berkelmans & van Oppen2006), and survive warm and cold water anomalies betterthan colonies with more sensitive symbionts (Jones et al.2008; Sampayo et al. 2008; LaJeunesse et al. 2009, 2010b).Many cnidarians living in the warmest nearshore turbidhabitats of the tropical Indo-Pacific commonly harborsymbionts in Clade D (Fabricius et al. 2004; Mostafavi etal. 2007; Hennige et al. 2010; LaJeunesse et al. 2010a). CladeD is evolutionarily distant from other Symbiodinium spp.(Rowan & Powers 1992; LaJeunesse 2001; Pochon et al.2006; Stern et al. 2010), but the nucleotide sequences ofconserved DNA markers resolve little or no diversity withinthis group. More rapidly evolving internal transcribed spacer(ITS) ribosomal rDNA markers identify ecologically differ-entiated entities associated with specific Cnidaria (LaJeu-nesse et al. 2004a, b, 2010a; Lien et al. 2013). For thisreason, a provisional nomenclature that combined letters(signifying the clade of Symbiodinium) and numbers wasused to delimit different biological entities, or types (e.g. D1¼ Clade D, type 1; see LaJeunesse et al. 2010a for furtherdetails on this conditional naming scheme). Preliminaryanalysis of recently developed microsatellite loci showed thatthese various Clade D ITS types (e.g. D1–4, D5, D2–4–5, D8)were composed of many individuals identified by distinctmultilocus genotypes (MLGs), and that the range of allelesizes were similar within but differed significantly betweeneach type (Pettay and LaJeunesse 2009; LaJeunesse et al.2010a; Wham et al. 2011). Conclusions from this worksuggested that phylogenetic groupings separated by seem-ingly equivocal differences in nuclear nrDNA were repro-ductively isolated and, therefore, defined by the biologicalspecies concept as distinct species.

Among the genetic types currently comprising Symbiodi-nium Clade D (n ~ 18), the most common and geograph-ically widespread is type D1a (¼ D1–4; sensu LaJeunesse etal. 2010a). This D type is provisionally referred to asSymbiodinium ‘trenchi’, but lacks a formal species descrip-tion (i.e. nomen nudum; see LaJeunesse et al. 2005, 2009,2010a; Wham et al. 2011). It associates with a wide diversityof mostly scleractinian taxa and is common in coralcommunities from warm equatorial, turbid environmentsand rarely occurs at northern and southern latitudes higherthan the Tropics of Cancer and Capricorn, respectively (FigsS1–S5; LaJeunesse et al. 2004a, 2010a; Silverstein et al. 2011;but see Mostafavi et al. 2007). This host generalist is the onlyClade D found in the Atlantic Ocean where it associates withsome coral colonies exposed to acute or chronic physiolog-ical stress (Toller et al. 2001; LaJeunesse et al. 2009). It isalso one of two Clade D types known to grow successfully inartificial culture media (held at the National Center ofMarine Algae and Microbiota and the personal collection ofMary Alice Coffroth, SUNY Buffalo, USA).

Most Clade D types have restricted ecological andgeographic distributions and often fail to proliferate outsideof the host in culture media. For example, the zebra coralOulastrea crispata (Fig. S6) is an opportunistic colonizer andone of a small number of coral species that can persist in

extreme, marginal, high-sediment intertidal habitats (Lam2000). Because of this ecology, its symbionts must thereforeendure extreme latitudinal, seasonal, or daily changes intemperature and irradiance. Throughout its distributionbetween the continents of Asia and Australia and extendinginto high nonreef temperate latitudes around Japan (368N),O. crispata harbors potentially more than one kind of CladeD Symbiodinium (Chen et al. 2003, 2005; Lien et al. 2007,2013; Fig. S7). A recent analysis of ITS1 and ITS2 resolvedrDNA lineages with different latitudinal distributions (Lienet al. 2013). Type D8 previously identified by LaJeunesse andcolleagues (2010a) was exclusive to O. crispata colonies fromwarm-water, tropical habitats around Southeast Asia, typeD13 (and another type characterized by a second codomi-nant ITS2 sequence, designated D12–13) was common inpopulations from the subtropics, especially at locationsalong the coast of China, whereas type D15 occurred in O.crispata populations extending from the subtropics to colder-water, northern temperate environments around Japan (Lienet al. 2013). These Oulastrea-associated Clade D types haveyet to be detected in other scleractinians and appear to berelatively host specific (Chen et al. 2005; LaJeunesse et al.2010a; Lien et al. 2012).

To resolve whether the Clade D Symbiodinium associatedwith Oulastrea are composed of separately evolving specieslineages, we tested genetic concordance between multicopyribosomal ITS1-5.8S-ITS2 and partial large-subunit (LSU)sequence data and ‘single’ copy microsatellite nuclear(D1Sym88, and D1Sym93; Wham et al. 2011), chloroplast23S ribosomal (cp23S), and mitochondrial cytochrome b(cob) DNA markers by assessing reciprocal monophyly(Avise and Wollenberg 1997). We also examined gene flowand reproductive isolation using population genetic datafrom eight microsatellite loci (Pettay & LaJeunesse 2009;Wham et al. 2011). We combined these results withcomplementary evidence obtained from more widely distrib-uted populations of S. ‘trenchi’. These findings were thencombined with available ecological and geographic distribu-tions, and average cell size, to formally describe three CladeD species. Using this integrative approach, species areformally described that satisfy the core principles of majorspecies concepts (Sites & Marshall 2004; de Queiroz 2007).Finally, we briefly discuss the underlying selection pressuresmost likely affecting Symbiodinium spp. diversification.

MATERIAL AND METHODS

Tissue collections (n ¼ 4 to 57 per location) of variousScleractinia harbouring Symbiodinium ‘trenchi’ (¼ type D1a¼type D1–4) were obtained from collections throughout theIndo-Pacific region including Zanzibar, in the Republic ofTanzania, Thailand in the Andaman Sea, rock islandhabitats of Palau, the Red Sea, Western Australia, thecentral Great Barrier Reef, Taiwan, and the Phoenix Islands(Fig. S7; Table S1). Samples of Oulastrea crispata wereacquired from locations across the northeast Indian Oceanand West Pacific region encompassing 5800 km, from low-latitude tropical Thailand (~ 58N) to high-latitude temperateJapan (. 358N), and represent samples used in previous

306 Phycologia, Vol. 53 (4)

Figs 1–4. Unrooted phylogenetic reconstructions (maximum parsimony) comparing new Symbiodinium spp. in Clade D. Numbers abovebranches indicate bootstrap support for maximum parsimony (500 replicates) and posterior probabilities, respectively.

Fig. 1. ITS1-5.8S-ITS2 rDNA and partial LSU phylogeny based on an alignment matrix of 1186 bases.Fig. 2. Partial chloroplast 23S-rDNA phylogeny based on an alignment matrix of 658 bases.Fig. 3. Phylogeny based on flanking region sequence data (an alignment matrix of 279 bases) for microsatellite D1Sym88.Fig. 4. Phylogeny based on flanking region sequence data (an alignment matrix of 145 bases) for microsatellite D1Sym93.

LaJeunesse et al.: Clade D Symbiodinium species 307

studies including additional collections from Penghu Island,Taiwan (e.g. Fig. S7; Table S1).

Nine combined years of monthly means (2003–2011) forsea-surface temperatures (SSTs) were calculated for sam-pling locations in tropical, subtropical and temperateregions, respectively, using data obtained from the Giovannionline data system (http://gdata1.sci.gsfc.nasa.gov/daac-bin/G3/gui.cgi?instance_id¼mairs_monthly_hres), developedand maintained by National Aeronautics and Space Admin-istration Goddard Earth Sciences Data and InformationSystem. Approximately 24 km2 around each samplinglocation were selected and data retrieved from MODIS-Aqua (SST) databases. The mean annual temperatures andseasonal variation (6 standard deviation) at each of eightlocations were also calculated.

Nucleic acid extractions were conducted as described byLaJeunesse et al. (2003). Amplifications of ITS1, ITS2, cp23Sand cob were performed in 25-ll reaction volumes containing2.5 ll of 2.5 mM deoxynucleotide triphosphates (dNTPs), 2.5ll of 25mM MgCl2, 2.5 ll of standard Taq buffer (NewEngland Biolabs, Ipswich, Massachusetts, USA), 0.13 ll of5U ll�1 Taq DNA polymerase (New England Biolabs), 1 llof each forward and reverse primer at 10 lM, and 1 ll of 5–50-ng DNA template (see Sampayo et al. 2009 for details onprimer sequence and annealing temperatures). New Clade Dprimers were developed to amplify loci D1Sym88 andD1Sym93 (Wham et al. 2011) for sequencing the flankingregions of each microsatellite for phylogenetic analyses.Primers CladeD_93for1 (CAAAGGAGCTCTAGGGGTAG), CladeD_93rev1 (GATTTGGCTTCCTCTTTTGT),CladeD_88for1 (AATGCTCCATGTATGCTCAC) and Cla-deD_88rev1 (AAAAGACGTCATGGGATTCT) amplifyeach locus, respectively, using the following thermal cycleprofile: 948C for 3 min, 40 cycles of 948C for 30 s, 548C for 30s and 728C for 30 s, followed by one final cycle at 728C for 10min. Amplified polymerase chain reaction (PCR) productswere directly sequenced on an Applied Biosciences sequencer(Applied Biosciences, Foster City, California, USA) at thePennsylvania State University nucleic acids facility.

Amplifications of the ITS1 and ITS2 were first electro-phoresed on denaturing gradient gels (50–90% for ITS1 and45–80% for ITS2) using a CBScientific system (Del Mar,California, USA). This screening process is used to identify,excise and sequence the numerically dominant sequencevariant(s) comprising the ribosomal array in the genome ofthe resident symbiont in each sample (for protocol details aswell as primer sequences, see LaJeunesse 2002, LaJeunesse etal. 2008). PCR-denaturing gradient gel electrophoresis(DGGE) provides a necessary step that filters out low-copy-number variants that are uninformative for purposes ofidentification (Thornhill et al. 2007). This protocol rapidlydiagnoses whether a genome is represented by a singlenumerically dominant sequence (one prominent band), orthat codominant variants exist within the genome (multiplebanding creates a characteristic PCR-DGGE ‘‘fingerprint’’profile), or that multiple symbionts are present in the host(fingerprints from two or more species detected in onesample; LaJeunesse 2002; LaJeunesse et al. 2008).

Chromatograms were checked and sequences alignedusing SeqNavigator v. 1.0 (Applied Biosystems) or GeneiousPro software v. 6.0 (Biomatters Limited, Auckland, New

Zealand). Paup 4.0b10 (Swofford 2000) was used withdefault settings to perform phylogenetic analyses on alignedsequence data under the criteria of maximum parsimony(with indels included as a fifth character state), maximumlikelihood (using the HKY85þG substitution model) anddistance. We analyzed sequence data for each genetic markerusing ModelTest (Posada & Crandall 1998), implemented inthe Geneious Pro to calculate the most appropriate model ofnucleotide evolution. Both the hierarchical likelihood ratiotest (hLRT) and Akaike information criterion (AIC) selecteddifferent evolutionary models for loci D1Sym88 andD1Sym93 (hLRT: F81 and AIC: TVMþG for locusD1Sym88 and hLRT: F81 and AIC: GTR for locusD1Sym93); therefore, we evaluated trees based on bothmodels to test concordance. Additionally, Bayesian posteriorprobabilities were calculated with the software Mr. Bayes v.3.2 (Huelsenbeck & Ronquist 2001), using substitutionmodels listed above for loci D1Sym88 and D1Sym93. But,because of low sequence divergence in the data sets of ITS,LSU, cp23S and cob, no optimal model nucleotide evolutioncould be calculated; and, therefore, HKY85þG was used asthe default model. Bayesian computations were conductedwith Markov chain Monte Carlo chain lengths of 1,000,000(four chains) sampled every 500 iterations and 500 sampledtrees discarded for the burn-in. Phylogenies among indepen-dent loci were compared visually to determine reciprocalmonophyly.

Eight polymorphic microsatellite loci developed for CladeD Symbiodinium were utilized (D1Sym14, D1Sym17,D1Sym35, D1Sym67, D1Sym87, D1Sym88, D1Sym92 andD1Sym93; Pettay & LaJeunesse 2009; Wham et al. 2011).Each locus was amplified in 10-ll reaction volumescontaining 1 ll of 10 mM dNTPs, 0.2 U of Taq DNApolymerase (New England Biolabs), 1 ll of standard Taqbuffer (New England Biolabs), 1 ll of 25 mM MgCl2, 1 ll ofeach forward and reverse primer at 10 lM, and 1 ll ofapproximately 50-ng DNA template. PCR amplificationconditions consisted of an initial denaturing step of 948C for2 min, followed by a touchdown protocol using 10 cycles of948C for 15 s, an annealing temperature of 28C aboveindividually optimized annealing temperatures for each locus(see Pettay & LaJeunesse, 2009; Wham et al. 2011) for 15 s,and an extension step at 728C for 15 s; for each cycle theannealing temperature lowers 0.58C per cycle. These werefollowed by 22 cycles as described above at the primer-specific annealing temperature (see Pettay & LaJeunesse,2009; Wham et al. 2011) and ending in a final extension of728C for 5 min. After amplification, fragment sizes wereanalyzed on an ABI 3730 genetic analyzer (AppliedBiosystems) using a 500-base-pair (bp) standard (labeledwith a LIZ fluorophore) at the Penn State University NucleicAcid Facility. Fragment sizes were analyzed visually usingGeneMarker v.1.91 (SoftGenetics, State College, Pennsylva-nia, USA).

A single MLG was obtained for a majority of samples bycombining results from each locus. In samples wheremultiple genotypes co-occurred (observed in 30–60% ofsamples collected from regions where Clade D typesoverlapped, Table S1), alleles whose electropherogram peakswere one-third less than the intensity of the main peak werenoted as background and not scored. MLGs could not be


produced from samples containing codominant genotypesbecause it was unfeasible in these instances to correctlyassign alleles to one or another individual clone for eachlocus. For this reason, samples containing two or moreSymbiodinium genotypes were removed from the populationgenetic analyses of MLGs preformed with STRUCTURE.This conservative approach was used to lower the chances ofproducing MLGs that actually consisted of two divergent

lineages while preserving the possibility of observingindividuals of mixed ancestry. These samples were insteadused to assess the percentage of colonies with mixtures ofSymbiodinium from the same putative species lineage or ofdifferent Clade D species by scoring the diagnostic alleles ofSym14 at sites where species ranges overlapped. Sampleswith a single diagnostic peak were scored as pure, whereassamples containing both a peak in the size range consistentwith type D15 (165 bp) and a peak in the size rangeconsistent with D13 (D12–13) (177–180 bp) were scored asmixtures regardless of the relative size of the peak.

To analyze MLGs from the set of samples correspondingto Symbiodinium ‘trenchi’ (formalized herein to S. trenchii,see below), the dual-allele loci were converted to haploid.This was accomplished by randomly removing one of thetwo alleles at each locus in the statistics program R v2.15.0(R Core Team 2013). The resulting haploid genotypes werethen analyzed by principle coordinate analysis in theBayesian clustering program STRUCTURE (Pritchard et

Fig. 5. Double allele peaks (e.g. at loci D1Sym88 and D1Sym92) across independently sorting microsatellite loci were found in most samplesof Symbiodinium trenchii (¼D1–4 or D1a) relative to S. boreum and S. eurythalpos from Oulastrea crispata. These double alleles were observedin isoclonal cultures and were a distinct feature of S. trenchii’s genome.

Table 1. UPT values between populations calculated frommicrosatellite genotype data of each species are given on lowerleft (bold). Probability values based on 999 permutations are shownin upper right.

Symbiodiniumtrenchii S. boreum S. eurythalpos

S. trenchii 0.001 0.001S. boreum 0.442 0.001S. eurythalpos 0.288 0.328


al. 2000). A pair-wise, individual-by-individual geneticdistance matrix was generated using the software packageGenAlEx v. 6.1 (Peakall & Smouse 2006). For each pair-wisecomparison, loci with the same state were assigned a value of0, whereas loci that differed were given a value of 1. Thesevalues were then calculated across multiple loci. Principlecoordinate analysis was performed in R, using the ADE4(Chessel et al. 2004) and ADEGENET (Jombart 2008)packages. The analysis in STRUCTURE was performedwithout a location prior, under an admixture and uncorre-lated alleles model with a burn-in of 100,000 iterations and1,000,000 iterations of analysis at K ¼ 1–7. The optimum Kvalue was determined by the Delta-K method (Evanno et al.2005). For each resulting population, a UPT value (a haploidequivalent to FST) was calculated pair-wise for eachpopulation using GenAlEx.

Both field-preserved and cultured coccoid cells werephotographed under bright-field illumination at a magnifi-cation of 3400–1000 using an Olympus BX51 compoundmicroscope (Olympus Corp., Tokyo, Japan) with a JenoptikProgRes CF Scan digital camera (Jenoptik, Jena, Germany).An autoexposure setting within ProgRes Capture Pro 2.8software (Jenoptik) was used to expose and capture the cellimages. Measurements taken from live cultures offered a wayto assess the effect of preservation on cell size. Two cultured

isolates of type D1-4, A001 (CCMP3408) and MTB4(CCMP3409), growing in the artificial culture mediumASP-8A (Ahles 1967) were photographed during log phaseof growth under 80–120 lmol quanta�m�2�s�1 photosynthet-ically active radiation on a 14:10 (light:dark) photoperiod.To avoid the effect of age on appearance and size, cells werephotographed during the middle of logarithmic growth inculture (between days 10 to 15 after reinoculation into freshmedia). Cell sizes for at least 40 individuals per culture werecalculated with the program ImageJ (Abramoff et al. 2004).Cell volume was calculated on the basis of the dimensions ofan ellipsoid [volume ¼ (4/3)p 3 abc, where a, b, and c areequal to half the length, width, and height. Cell height wasassumed to be identical to cell width]. Size and volumedifferences between putative species were assessed viaunivariate type III ANOVA (factor: ‘Species’) on acombined data set of all maximum length measurements ofcultured and natural material belonging to each group. Datawere cube-root transformed to achieve normality andhomoscedasticity and statistical significance was assessedwith ANOVA at P ¼ 0.05 in the software SPSS Statistics v.19.0 (IBM, Armonk, New York, USA). Cell size averageswere plotted in original units for clarity along with 95%confidence intervals.

Fig. 6. Differences in alleles and relative frequencies for eight microsatellite loci observed among Clade D Symbiodinium spp. proposed asformal species. Circle color relates to symbiont species and size corresponds to relative allele frequencies in the samples analyzed for S.trenchii (n ¼ 97 distinct genotypes), S. eurythalpos (n ¼ 58 distinct genotypes), and S. boreum (n¼ 33 distinct genotypes).


RESULTS

Symbiodinium trenchii LaJeunesse sp. nov.Fig. 9

DIAGNOSIS: Coccoid cells, typically 9.8 to 10.4 lm in maximumdiameter; nucleotide sequences of the chloroplast cp23S-rDNA(KF220382), mitochondrial cob (KF193523), microsatellite flankingsequence of loci D1Sym88 and D1Sym93 (Figs 3, 4, DRYAD entrydoi:10.5061/dryad.3np1m), nuclear ribosomal ITS1/5.8S/ITS2(KJ019889) and partial LSU (KF740689) and fragment size ranges(allelic variants) of eight microsatellites distinct. Diallele peaks atnearly all microsatellite loci.

HOLOTYPE: USNM 1231333 deposited in the National Museum ofNatural History, Smithsonian Institution, Washington, DC, USA.

AUTHENTIC STRAINS: CCMP3408 (¼ strain A001) and CCMP3409(¼ strain MTB4) deposited at the Provasoli-Guillard National Centerfor Marine Algae and Microbiota, East Boothbay, Maine, USA.

TYPE LOCALITY: Okinawa, Japan (26839009.39 00N, 127851023.14 00E).

ETYMOLOGY: Named for Professor Robert K. Trench, a phycolog-ical pioneer who first recognized species diversity in Symbiodinium.

Symbiodinium boreum LaJeunesse & Chen sp. nov.Fig. 10

DIAGNOSIS: Coccoid cells, typically from 9.2 to 9.7 lm at maximumdiameter. Nucleotide sequences of the chloroplast cp23S-rDNA(KF220381), mitochondrial cob (KF193522), microsatellite flankingD1Sym88 and D1Sym93 (Figs 3, 4, DRYAD entry doi:10.5061/dryad.3np1m), nuclear ribosomal ITS1/5.8S/ITS2 (KJ019893) andpartial LSU (KF740688), and fragment size ranges (allelic variants) ofeight microsatellites unique.


PARATYPES: ASIZC0000966-70, deposited in the BiodiversityResearch Center Academia Sinica (BRCAS), Museum, ZoologicalCollection, Taipei, Taiwan. Samples were collected in the KochiPrefecture, Otsuki Town, Nishidomari, Japan (32846 042.82 00N,132843057.40 00E).

TYPE LOCALITY: Penghu Island, Taiwan (23831012 00N, 11983300 00E).

HABITAT: Marine, associated with the scleractinian, Oulastreacrispata.

ETYMOLOGY: From the Greek boreas meaning north and refers tothe northern subtropical and temperate environments where thisspecies occurs in association with Oulastrea crispata.

Symbiodinium eurythalpos LaJeunesse & Chen sp. nov.Fig. 11

DIAGNOSIS: Coccoid cells, typically 8.3 to 9.3 lm in maximumdiameter. Nucleotide sequences of the chloroplast cp23S-rDNA(KF220379, KF220380), mitochondrial cob (KF193520/1), microsat-ellite flanking D1Sym88 and D1Sym93 (Figs 3, 4, DRYAD entrydoi:10.5061/dryad.3np1m), nuclear ribosomal ITS1/5.8S/ITS2(KJ019890, KJ019891, KJ019892), partial LSU (KF740686/7) andfragment size ranges (allelic variants) of eight microsatellites distinct.


Fig. 7. Population genetic structure exhibited among threegenetically distinct lineages of Clade D Symbiodinium. Principlecoordinate analysis based on Euclidian distances showing clusteringof genotypes corresponding to phylogenetic lineages (see above).Axes 1 and 2 explain 45.5% and 19.7% of variance, respectively.

Fig. 8. Population genetic structure exhibited among three genetically distinct lineages of Clade D Symbiodinium. Bayesian analyses using thesoftware STRUCTURE assigned multilocus data to one of three genetically distinct populations irrespective of geographic origin. A fourthpopulation was delimited (K ¼ 4) corresponding to populations of Symbiodinium eurythalpos in the upper Gulf of Thailand.


PARATYPES: Accession ID numbers S22029–S22036 of specimensdeposited at the Ocean Genome Legacy (OGL), NortheasternUniversity, Nahant, Massachusetts, USA. Samples were collectedfrom Khang Kao Island, Thailand, in the Gulf of Thailand (13.118N,100.818E) and from Cape Panwa, Phuket Island, Thailand (7.8158N,98.4048E).

TYPE LOCALITY: Penghu Island, Taiwan (23831012 00N, 11983300 00E).

HABITAT: Marine, associated with the scleractinian, Oulastreacrispata.

ETYMOLOGY: From the Greek eurus (¼wide) combined with thalpos(¼ heat, warmth) and designated for its broad latitudinal distributionspanning tropical and subtropical environments.

Genetic evidence

Sequence data from nuclear-encoded rDNA (ITS1/5.8S/ITS2and LSU, 1186-base alignment matrix; Fig. 1), partial

chloroplast (cp23S-rDNA, 658-base alignment matrix; Fig.2) and single-copymicrosatellite flanking regions (D1Sym88, a279-base alignment matrix, Fig. 3 and D1Sym93, 145-basealignment matrix, Fig. 4) produced concordant phylogeneticpatterns consistent with the existence of at least threegenetically cohesive, evolutionarily divergent lineages. Se-quence alignments for the two microsatellite markers areavailable at DRYAD (entry doi:10.5061/dryad.3np1m).Maximum parsimony, distance and Bayesian approachesproduced concordant phylogenies (comparison not shown).Chloroplast cp23S-rDNA sequences for Symbiodinium bor-eum and S. eurythalpos were identical, but differed from S.trenchii by one deletion and three base substitutions in thehypervariable domain V region (Fig. 2). Noncoding, indepen-dently sorting microsatellite flanking sequences providedconcordant phylogenetic resolution of all three taxa (Figs 3,4). A fixed base substitution in the mitochondrial cob sequence(918 bases analyzed) that differentiated only S. boreum (Fig.S8) results in a nonsynonymous amino acid substitution(converting phenylalanine to tyrosine at amino acid position167) in the functionally conserved gene region containing theQo redox center (Fig. S8).

Our analysis of eight microsatellite loci found that mostsamples of Symbiodinium eurythalpos and S. boreum producedelectropherograms with single peaks for each locus withminimal stuttering and nonspecific fragment amplification

Figs 9–11. Coccoid cell morphology for three Clade D Symbiodinium spp.Fig. 9. Light micrograph of S. trenchii sp. nov. (cultured isolate A001); scale bar¼ 10 lm.Fig. 10. Light micrograph of S. boreum sp. nov. freshly isolated from host tissues collected off of Penghu Island, Taiwan. Size bar¼ 10 lm.Fig. 11. Light micrograph of S. eurythalpos sp. nov. freshly isolated from host tissues collected off of Penghu Island, Taiwan. Size bar¼ 10lm.

Fig. 12. Coccoid cell size differences among three Clade DSymbiodinium spp. A comparison of mean cell diameters amongovate coccoid cells obtained from various samples and theircorrespondence to different species of Clade D Symbiodinium.Cultured isolates (light brown squares) and freshly isolated cells ofS. trenchii (dark brown squares) acquired from coral genera,including Favia sp. and Goniastrea spp., from Phuket, Thailand andPalau. Freshly isolated cells of S. boreum and S. eurythalpos wereobtained from colonies of Oulastrea crispata from Penghu Island,Taiwan, and the Gulf of Thailand. Error bars represent standarderror calculated from . 40 measurements.

Fig. 13. Coccoid cell volumetric differences among three Clade DSymbiodinium spp. Mean cell volume and 95% confidence intervalsfor combined samples (from Fig. 12) representing each Clade Dspecies. Letters identify statistically distinct groups (ANOVA, P ,0.05).


(Fig. 5). This allowed for the construction of MLGs, whichindicated that both Oulastrea-associated Symbiodinium werehaploid, and that most colonies possessed a dominant clonalsymbiont population. Double, nonstutter peaks were ampli-fied at most loci in all samples of S. trenchii, includingmonoclonal cultured isolates (Fig. 5). For each microsatellitelocus, we found differences in allele frequency and size rangethat were characteristic of each species (Fig. 6; Table S2).Calculations of genetic (Euclidian) distance between geno-types (Jombart 2008) and Bayesian analysis of 227 MLGsusing the program STRUCTURE (Pritchard et al. 2000)identified three reproductively isolated populations (K ¼ 3,Figs 7, 8, S10 and S11), each corresponding to phylogeneticgroupings based on shared nucleotide sequence similarity(Figs 1–4). Membership to a particular cluster was high (90–100%) among individuals. Pair-wise calculations ofUPT values(~ haploid FST) further showed that each species is geneticallywell differentiated (Table 1). Although sample sizes were smallin some locations (Table S1), geographic distance did notcontribute to detectable population substructure across thedistribution of a particular species.

MLGs from samples representative of a species tended todiffer at multiple loci and many recombinant genotypes wereidentified in linkage equilibrium (e.g. allele 1 from locus A waspaired randomly with alleles 1, 2, 3, etc. from locus B and viceversa; MLG data available at DRYAD entry doi:10.5061/dryad.3np1m). The mean numbers of allelic differencesbetween individual MLGs were 4.7 (6 1.5 standard deviation)of 8 for S. trenchii, 3.5 (6 1.4) for S. boreum, and 5.4 (6 1.9)for S. eurythalpos. The independence in pairing between allelesacross loci is a strong indication of sexual recombination. Instrictly clonal populations, multilocus haplotypes shouldproduce genealogies with allele combinations that are linkedacross many loci (Hudson and Kaplan 1985, Wakeley 2008).Instead, we observed many alleles unique to a particularspecies (Fig. 6) and unlinked (i.e. scrambled) allele combina-tions among genotypes within populations of each species.Such patterns can only be explained by genetic isolation andsexual recombination. The electropherograms in the fragmentanalysis of some O. crispata samples contained combinationsof different MLGs that were mostly the result of co-occurringspecies in colonies collected at latitudes transitioning betweensubtropical and temperate regions. The presence of twospecies was independently verified with DGGE fingerprintingof ITS2 rDNA (Fig. S9).

Morphological, ecological and biogeographic evidence

Light micrographs of coccoid cells of Symbiodinium trenchii inculture (including CCMP3408 isolated from Acropora sp. inOkinawa and CCMP3409 from Orbicella annularis in theFlorida Keys) and freshly isolated cells of S. eurythalpos and S.boreum taken at maximum resolution of 31000 exhibitedspherical, or slightly ovate, nondescript morphology similar intheir general appearance and typical of most Symbiodinium(Figs 9–11; LaJeunesse 2001). CladeD Symbiodinium showed arange of cell diameters (Fig. 11). Mean cell volumes weredifferent for each species (F2,1077¼ 117.4, P , 0.001). Amongsamples obtained fromOulastrea crispata, S. boreum cells wereslightly larger than cells ofS. eurythalpos. Samples ofS. trenchii

obtained from several different hosts, or from culture, werelarge relative to S. boreum and S. eurythalpos (Figs. 12, 13).

Symbiodinium boreum and S. eurythalpos differed inlatitudinal distribution among colonies of O. crispata thatrelate to major differences in seasonal and average annualtemperatures (Figs 14, S12). The distribution of S. boreumoverlaps with S. eurythalpos in subtropical environments (Fig.14). At sites in this region, approximately 50% of colonies ofO. crispatawere codominated byMLGs representing differentsymbiont species (Table S1). There were no apparentpopulation genetic subdivisions detected across the geograph-ic range examined for each species (Fig. 8). At locations wherethese species distributions overlapped, populations of eachwere genetically distinct (Figs 8, 14).

DISCUSSION

Concordant phylogenetic, population genetic, and ecologicalevidence, as well as differences in geographical distributionand cell size, support the formal designation of these threeSymbiodinium lineages as separate species in Clade D.Whereas clonal evolution could explain concordant phylog-enies from numerous and geographically widespread samples(Figs 1–4; Avise and Wollenberg 1997), our analysis ofindependently sorting microsatellite loci provided a test forwhether genetic exchange occurred within or betweenlineages identified as phylogenetically distinct (Figs 1–4).Populations of each proposed Symbiodinium species con-tained unique alleles and significant differences in allelefrequencies, revealing a lack of genetic exchange betweenpopulations representing different species lineages (Figs 6–8;Table 1). Within each species, most clonal populationsappeared to be the products of genetic recombination whereallele combinations found among individual multilocusmicrosatellite genotypes were frequently scrambled acrossmany loci (i.e. exhibiting linkage equilibrium among loci;initially calculated in Pettay and LaJeunesse 2009; Wham etal. 2011). These observations support the concept thatSymbiodinium species comprise numerous clonally propa-gated cell lines (i.e. individuals) that periodically undergo sexto produce new genetically recombinant clonal lineages(LaJeunesse 2001).

The strong genetic partitioning between sympatric popu-lations of Symbiodinium boreum and S. eurythalpos validatesthat they are different species under the biological speciesconcept and not geographically isolated populations of thesame species exhibiting genetic structure over a broadlatitudinal range (Martin and McKay 2004). When geneflow occurs through hybridization, alleles that are diagnosticof one species should be found in combination with thealleles of another species. Sympatric populations of S.eurythalpos and S. boreum in Okinawa and Taiwan showedno evidence of hybridization or allele introgression, suggest-ing that effective cross-hybridization does not occur. (Fig. 8).Even low amounts of genetic exchange in regions ofpopulation sympatry, over a few generations, woulddissipate the strong genetic partitioning found in ourstatistical analysis using STRUCTURE; and eliminateconcordance among DNA phylogenies produced by inde-


pendently sorting nuclear and plastid loci (Figs 1–4, 8;Wright 1931).

Although intrinsic reproductive barriers maintain geneticboundaries at locations where species populations overlap,population genetic data indicate that gene flow occurs withsufficient frequency to genetically homogenize the popula-tions of each species over hundreds and even thousands ofkilometres (Figs 8, 14; Table 1). Widespread gene flow occursamong populations of S. trenchii separated by thousands ofkilometres from the western Indian Ocean to the PacificOcean (Fig. 8). The apparent dispersal capability of S.trenchii could be important to this species’ ecologicalresponse to climate change and its potential for rangeexpansion (LaJeunesse et al. 2009). Additional populationsof S. trenchii, especially from the Red Sea, Arabian/PersianGulf, and Atlantic Ocean (Caribbean), should be analysed todetermine whether any genetic structure exists in thiswidespread species.

ITS nrDNA is commonly used to diagnose the speciesidentity of various organisms (Seifert 2009; Moniz &Kaczmarska 2010; Yao et al. 2010), yet intragenomic andinterindividual variation can confound species resolutionamong Symbiodinium (Sampayo et al. 2009; Thornhill et al.2014). Tropical populations of S. eurythalpos were charac-terized by strains with a single numerically dominant ITS2sequence (type D8), whereas others possessed a secondancestral codominant variant in the genomes’ ribosomalarray, the D12 sequence (type D8–12). In temperate regions,

however, most strains (clones) of S. eurythalpos possessed asingle dominant sequence (type D13), or also in combinationwith the D12 sequence (type D12–13; Fig. S9; Lien et al.2013). The geographical patterns of these ITS genotypessuggest that sequences of the nrDNA operon are at differentstages of lineage sorting (from ancestral to derived sequencevariants) across populations of S. eurythalpos (Dover 1982).In contrast, a single sequence (D15) was numericallydominant in the genomes of all samples of S. boreum thatwere analysed and indicates that the rDNA operon is wellhomogenised across individuals of this species. The genomesof S. trenchii are consistently characterised by codominantITS2 sequences (the 1 and 4, or ‘a’, bands viewed on DGGEgels) that persist intragenomically in similar proportionswherever populations of this species occur.

Analyses of microsatellite markers substantiate the initialconclusions of Blank and Trench (1985) that Symbiodiniumcells exist predominantly in the haploid phase (Santos &Coffroth 2003; see also Pettay & LaJeunesse 2007, Andraset al. 2011, Pettay et al. 2011; Pinzon et al. 2011); and is ageneral characteristic feature of the Dinophyceae (Coats2002). Symbiodinium eurythalpos and S. boreum possessedonly single alleles at each locus (Fig. 5), except in sampleswhere mixed genotypes were clearly present. In contrast,two alleles were routinely scored at most loci for samples ofS. trenchii, including several independently acquired iso-clonal cultures (Fig. 5; Pettay & LaJeunesse 2009; Wham etal. 2011). The dual alleles characteristic of microsatellite

Fig. 14. Known latitudinal distributions and dominance of Symbiodinium boreum (blue) and S. eurythalpos (yellow) associated withpopulations of the zebra coral, Oulastrea crispata (inset upper left). Symbiodinium eurythalpos overlaps with S. boreum in Taiwan andOkinawa where they often co-occurred in the same colony. The orange shading shows the location of a genetically distinct population of S.eurythalpos in the upper Gulf of Thailand. The graph (inset lower right) depicts mean annual temperatures based on satellite data from eightcollection sites over a latitudinal range spanning 58N to . 358N (error bars correspond to standard deviations). White, grey, and blackshaded symbols correspond to the tropical, subtropical, and temperate locations, respectively.


loci from S. trenchii must be explained by differences ingenome ploidy, and not from reoccurring mixtures of twogenotypes with allelic compositions that are different atevery locus. At this time we reserve speculation on thesignificance of this finding and whether the apparentduplicated genome of this species explains its ecologicalbreadth and whether it enhances the adaptive potential ofS. trenchii populations.

The featureless spherical forms and relatively small sizesof Symbiodinium (typically 7–13 lm in mean diameter)preclude use of a morphological diagnostic trait for visualidentification. Although not intended to be diagnostic, cellsize analysis provides some morphological evidence forspecies differentiation (Schoenburg & Trench 1980; Trench& Blank 1987; LaJeunesse et al. 2012). Samples of S.trenchii obtained from a diversity of host genera collected inthe Pacific and Indian oceans were similar in average celldiameter and similar to isolates under long-term artificialculture (Fig. 12). Despite some interindividual variation,cell size in this and other Symbiodinium spp. appearstrongly influenced by genetic factors (LaJeunesse 2001;LaJeunesse et al. 2012). Cell size, or volume (Fig. 13), fromthe perspective of functionality, can substantially affectrespiration, growth rate, and photosynthesis of microalgae(Banse 1976). The larger cell size of S. boreum, relative to S.eurythalpos, may contribute to the physiological traits thatultimately explain its higher latitudinal distribution (seebelow).

Natural selection in the struggle for resource acquisitionand reproductive success often leads to population subdivi-sion and ecological specialization (Futuyma & Moreno 1988;Schluter 2009). For endosymbiotic dinoflagellates, the cni-darian host represents a stable habitat where growth andreproductive fitness are likely maximized (Knowlton &Rohwer 2003). It is therefore not surprising that host identitydetermines the distribution of most Symbiodinium spp.(Finney et al. 2010) and that the evolution of host–symbiontspecificity, as a form of ecological specialization, drives thepopulation subdivision and diversification of Symbiodiniumas mutualistic endosymbionts (LaJeunesse 2005; Thornhill etal. 2014). As sister species in Clade D, S. boreum and S.eurythalpos share a common ecological trait in theirsymbioses with Oulastrea crispata. This host is one of a smallnumber of Indo-Pacific corals that broods larvae (Lam 2000)and is prone to vertically transmit symbiont cells betweengenerations. Many host-specific lineages of Symbiodiniumcorrespond to animals that vertically transmit (LaJeunesse2005), and it is possible that the common ancestor to S.boreum and S. eurythalpos first evolved by the process ofecological specialization to Oulastrea (Thornhill et al. 2014).

The geographical partitioning of Symbiodinium boreumand S. eurythalpos suggest that long-standing selectionpressure from different latitudinal environments created asecondary axis of niche diversification that initiated lineagebifurcation from a common ancestor (Figs 14, S12). Meanannual temperatures were recently shown by Thomas et al.(2012) to have a large influence in thermal adaptation amongphytoplankton obtained at locations encompassing 1508 oflatitude. Their analyses of the optimum temperatures forgrowth compiled from over 130 species representing themajor phytoplankton groups suggested that populations of

microalgae were adapted to local environmental conditions(especially temperature) despite the potential for long-distance dispersal (Thomas et al. 2012). Colonies ofOulastrea crispata and their resident Symbiodinium experi-ence extreme differences in mean temperature and lightacross the latitudes of their distribution (Fig. 14; Lien et al.2013). At northernmost locations, average annual watertemperatures are approximately 20–228C (Fig. 14), andseasonal temperatures oscillate 108C, or more, over thecourse of a year (Fig. S12). Shallow inshore colonies mayexperience cold water conditions below 128C (Yajima et al.1986). Subtropical locations also experience wide seasonaloscillations in temperature with summer highs similar to thetropics (29–308C; Fig. S12). Both S. boreum and S.eurythalpos must therefore have physiological adaptationsto live with O. crispata in these fluctuating environments, buthow they tolerate such cold or warm extremes awaitsthorough physiological study.

The presence of a genetically distinct population ofSymbiodinium eurythalpos in the upper Gulf of Thailandmay be the product of severe selection by regionalenvironmental conditions (Figs 8, 14, S1; e.g. Pettay &LaJeunesse 2013). Surface currents into and out of the Gulfare confluent with the rest of the South China Sea anddiscounts the possibility that these distinct populations arethe product of long-term isolation (Akhir 2012). Instead thestrong genetic structure in this most northern portion of theGulf of Thailand may represent shifts in genetic diversity inresponse to changes in temperature, sedimentation andsalinity created by seasonal water stratification in the Gulfand the discharge of fresh water from numerous rivers in thearea (Windom et al. 1984; Yanagi et al. 2001; Cheevaporn &Menasveta 2003). A similar pattern of strong geneticsubdivision was recently shown for another Clade DSymbiodinium (S. ‘glynni’ nomen nudum) associated withpopulations of the coral Pocillopora from the eastern Pacific.Populations of this symbiont were genetically distinct at highlatitudes despite no evidence of population subdivision in thehost, whose larvae already possess symbionts when plank-tonic (Pettay & LaJeunesse 2013). The wide annual ranges intemperature and light in the Gulf of California, relative tomore equatorial locations, are obvious factors influencingstrong genetic differentiation among these S. ‘glynni’populations. These few examples involving the symbiosesof Oulastrea and Pocillopora indicate that environmentalgradients can also have a strong effect on the geneticdiversification of Symbiodinium.

We considered an alternate hypothesis that host geneticpartitioning by latitude, or the presence of cryptic species ofOulastrea, may instead be influencing the distributions ofSymbiodinium boreum and S. eurythalpos (Lien et al. 2013).However, our genetic evidence (ITS1 sequence comparison),in contrast to the interpretations of Lien et al. (2013),indicates that O. crispata comprises a single species, albeitwith some interindividual nrDNA sequence variation (Fig.S13). Population genetic markers developed and applied tothis question may reveal fine-scale genetic partitioning acrosslatitudinal populations of this coral. On the condition thathost populations are confluent, we presume that environ-mental factors, primarily light and temperature, largelyexplain the evolution of geographically distinct Oulastrea-


specific Symbiodinium. In turn, we propose that themaintenance of distinct associations with putatively ‘cold’-and ‘heat-tolerant’ species of Clade D Symbiodinium, in part,explains why O. crispata persists in marginal habitats over alarge latitudinal gradient (Lien et al. 2013).

Finally, preliminary ITS2 rDNA data diagnostic ofSymbiodinium eurythalpos suggest that this symbiont mayoccur in populations of Zoanthus sansibaricus (unpubl.) andthe aeolid nudibranch Pteraeolida ianthina (Ishikura et al.2004) at high latitudes (. 308N) around mainland Japan andat similar latitudes where Oulastrea harbours S. boreum. Ifthese ecological and geographical patterns are substantiated,it indicates that S. eurythalpos associates with other non-scleractinian hosts beyond its normal latitudinal distributionwith Oulastrea. Although apparently present in these higher-latitude regions, S. eurythalpos may be competitively inferiorto S. boreum when growing in Oulastrea colonies from colderenvironments.

In addition to the three species described here, Clade DSymbiodinium apparently comprises numerous other sepa-rate entities, each exhibiting differences in ecology andphysiology. This cautions against making broad assumptionsconcerning the contribution of this group as a whole in theecological or evolutionary response of coral–dinoflagellatesymbioses to climate change (Baker et al. 2004; Oliver &Palumbi 2009; Stat & Gates 2011). The host specificityexhibited by many in this clade suggests that their ecologicalsignificance is limited to few or one host species (LaJeunesseet al. 2010a). In contrast, broadly distributed populations ofS. trenchii are of potential interest given that this speciesassociates with a wide diversity of coral genera.

This work is the first to utilize population genetics indelimiting species of microalgae (proposed as a generalprocedure to objectively define species of eukaryote,Hausdorf & Hennig 2010; Hey & Pinho 2012). Althoughno individual can be representative of variation within aspecies population, the examination of numerous individualsis consistent with evolutionary biology’s view of the species.For most eukaryotic microbes, rarely is it possible to testspecies hypotheses of microalgae using the biological speciesconcept [Mayr 1942; but see the meticulous work ofColeman (2001), Rynearson & Armbrust (2004), and Adamset al. (2009), who worked with numerous cultures]. Theclonal cell lines (individual genotypes) that dominated mostof these coral samples made feasible the application ofpopulation genetic analyses on genotype diversity acquiredover different geographic scales encompassing differentregional environments. Because of their abundance and highclonality in host tissues, Symbiodinium may represent amongthe few kinds of dinoflagellates where this level of geneticstudy is feasible, and a model system for examininggenotypic diversity, gene flow, and dispersal that relate tothe macro- and microevolution of eukaryotic microbes(Andras et al. 2011; Pettay et al. 2011, 2013; Wham et al.2011; Thornhill et al. 2014).

ACKNOWLEDGEMENTS

We thank Yoshikatsu Nakano, Sakanan Plathong, andNarinratana Kongjandtre for obtaining Oulastrea samples.

Christian Voolstra at King Abdullah University of Scienceand Technology facilitated and supported collections ofSymbiodinium trenchii in the Red Sea and Arabian Gulf. Wealso thank the Penn State Microscopy and CytometryFacility, University Park, Pennsylvania, USA. Finally, weappreciated the comments of two anonymous reviewers,which improved the quality of the paper. This research wassupported by the USA National Science Foundation (OCE-0928764 and IOS-1258058 to TCL, DEG-0750756 to JEP),Penn State University, and Intergovernmental Oceanograph-ic Commission-United Nations Educational, Scientific andCultural Organization–World Bank targeted working groupon coral bleaching. SK was supported by the AcademiaSinica postdoctoral fellowship (2010–2012). CAC wassupported by Academia Sinica Thematic Grants (2005–2010) and National Science Council grants (2006–2010),Taiwan.

SUPPLEMENTARY DATA

Supplementary data associated with this article can be foundonline at http://dx.doi.org/10.2216/13–186.1.s1.

REFERENCES

ABRAMOFF M.D., MAGALHAES P.J. & RAM S.J. 2004. Imageprocessing with ImageJ. Biophotonics International 11: 36–42.

ADAMS N.G., TRAINER V.L., ROCAP G., HERWIG R.P. & HAUSER L.2009. Genetic population structure of Pseudo-nitzschia pungens(Bacillariophyceae) from the Pacific Northwest and the NorthSea. Journal of Phycology 45: 1037–1045.

AHLES M.D. 1967. Some aspects of the morphology and physiology ofSymbiodinium microadriaticum. Ph.D thesis. Fordham Univer-sity, New York. 158 pp.

AKHIR M.F.M. 2012. Surface circulation and temperature distribu-tion of southern South China Sea from global ocean mode.(OCCAM). Sains Malaysiana 41: 701–714.

ANDRAS J.P., KIRK N.L. & HARVELL C.W. 2011. Range-widepopulation genetic structure of Symbiodinium associated with theCaribbean sea fan coral, Gorgonia ventalina. Molecular Ecology20: 2525–2542.

AVISE J.C. & WOLLENBERG K. 1997. Phylogenetics and the origin ofspecies. Proceedings of the National Academy of Sciences of theUnited States of America 94: 7748–7755.

BAKER A.C., STARGER C.J., MCCLANAHAN T.R. & GLYNN P.W.2004. Corals’ adaptive response to climate change. Nature 430:741.

BANSE K. 1976. Rates of growth, respiration and photosynthesis ofunicellular algae as related to cell size – a review. Journal ofPhycology 12: 135–140.

BERKELMANS R. & VAN OPPEN M.J.H. 2006. Flexible partners incoral symbiosis: a ‘nugget of hope’ for coral reefs in an era ofclimate change. Proceedings of the Royal Society of London B:Biological Sciences 273: 2305–2312.

BLANK R.J. & TRENCH R.K. 1985. Speciation and symbioticdinoflagellates. Science 229: 656–658.

CHEEVAPORN V. & MENASVETA P. 2003. Water pollution and habitatdegradation in the Gulf of Thailand. Marine Pollution Bulletin47:43–51.

CHEN C.A., LAM K.K., NAKANO Y. & TSAI W.S. 2003. A stableassociation of the stress-tolerant zooxanthellae, SymbiodiniumClade D, with the low-temperature-tolerant coral, Oulastreacrispata (Scleractinia: Faviidae) in subtropical non-reefal coralcommunities. Zoological Studies 42: 540–550.


CHEN C.A., YANG Y.W., WEI N.V., TSAI W.S. & FANG L.S. (2005).Symbiont diversity in scleractinian corals from tropical reefs andsubtropical non-reef communities in Taiwan. Coral Reefs, 24, 11–22.

CHESSEL D., DUFOUR A.B. & THIOULOUSE J. 2004. The ade4package-I – one-table methods. R News 4: 5–10.

COATS D.W. 2002. Dinoflagellate life-cycle complexities. Journal ofPhycology 38: 417–419.

COFFROTH M.A. & SANTOS S.R. 2005. Genetic diversity of symbioticdinoflagellates in the genus Symbiodinium. Protist 156:19–34.

COLEMAN A.W. 2001. Biogeography and speciation in the Pandor-ina/Volvulina (Chlorophyta) superclade. Journal of Phycology 37:836–851.

CORREA A.M.S. & BAKER A.C. 2009. Understanding diversity incoral–algal symbioses: a cluster-based approach to interpretingfine-scale genetic variation in the genus Symbiodinium. CoralReefs 28: 81–93.

DE QUEIROZ K. 2007. Species concepts and species delimitation.Systematic Biology 56: 879–886.

DOVER G.A. 1982. Molecular drive: a cohesive mode of speciesevolution. Nature 299: 111–117.

EVANNO G., REGNAUT S. & GOUDET J. 2005. Detecting the numberof clusters of individuals using the software STRUCTURE: asimulation study. Molecular Ecology 14: 2611–2620.

FABRICIUS K.E., MIEOG J.C., COLIN P.L., IDIP D. & VAN OPPEN

M.J.H. 2004. Identity and diversity of coral endosymbionts(zooxanthellae) from three Palauan reefs with contrastingbleaching, temperature and shading histories. Molecular Ecology13: 2445–2458.

FINNEY J.C., PETTAY T., SAMPAYO E.M., WARNER M.E., OXENFORD

H. & LAJEUNESSE T.C. 2010. The relative significance of host-habitat, depth, and geography on the ecology, endemism andspeciation of coral endosymbionts. Microbial Ecology 60: 250–263.

FRADE P.R., ENGELBERT N., FARIA J., VISSER P.M. & BAK R.P.M.2008. Distribution and photobiology of Symbiodinium types indivergent light environments for three colour morphs of the coralMadracis pharensis: is there more to it than total irradiance? CoralReefs 27: 913–925.

FREUDENTHAL H.D. 1962. Symbiodinium gen. nov. and Symbiodi-nium microadriaticum sp. nov., a zooxanthella: taxonomy, lifecycle and morphology. Journal of Protozoology 9: 45–52.

FUTUYMA D.J. & MORENO G. 1988. The evolution of ecologicalspecialization. Annual Review of Ecology and Systematics 19: 207–233.

HAUSDORF B. & HENNIG C. 2010. Species delimitation usingdominant and codominant multilocus markers. SystematicBiology 59: 491–503.

HENNIGE S.J., SUGGETT D.J., WARNER M.E., MCDOUGALL K.E. &SMITH D.J. 2009. Photobiology of Symbiodinium revisited: bio-physical and bio-optical signatures. Coral Reefs 28: 179–195.

HENNIGE S.J., SMITH D.J., WALSH S.J., MCGINLEY M.P., WARNER

M.E., & SUGGETT D.J. 2010. Acclimation and adaptation ofscleractinian coral communities along environmental gradientswithin an Indonesian reef system. Journal of Experimental MarineBiology and Ecology 391: 143–152.

HEY J. & PINHO C. 2012. Population genetics and objectivity inspecies diagnosis. Evolution 66: 1413–1429.

HUDSON R.R. & KAPLAN N.L. 1985 Statistical properties of thenumber of recombination events in the history of a sample ofDNA sequences. Genetics 111: 147–164.

HUELSENBECK J. & RONQUIST F. 2001. MrBayes: a program for theBayesian inference of phylogeny. Bioinformatics 17: 754–755.

ISHIKURA M., HAGIWARA K., TAKISHITA K., HAGA M., IWAI K. &MARUYAMA T. 2004. Isolation of new Symbiodinium strains fromtridacnid giant clam (Tridacna crocea) and sea slug (Pteraeolidiaianthina) using culture medium containing giant clam tissuehomogenate. Marine Biotechnology 6: 378–385.

JEONG H.J., LEE S.Y., KANG N.S., YOO Y.D., LIM A.S., LEE M.J.,KIM H.S., YIH W., YAMASHITA H. & LAJEUNESSE T.C. Geneticsand morphology characterize the dinoflagellate Symbiodiniumvoratum, n. sp. (Dinophyceae), as the sole representative of

Symbiodinium Clade E. Journal of Eukaryotic Microbiology. 61:75–94.

JOMBART T. 2008. Adegenet: a R package for the multivariateanalysis of genetic markers. Bioinformatics 24: 1403–1405.

JONES A., BERKELMANS R., VAN OPPEN M.J., MIEOG J. & SINCLAIR W.2008. A community change in the algal endosymbionts of ascleractinian coral following a natural bleaching event: fieldevidence of acclimatization. Proceedings of the Royal Society B:Biological Sciences 275: 1359–1365.

KNOWLTON N. & ROHWER F. 2003. Multispecies microbial mutual-isms on coral reefs: the host as habitat. American Naturalist 162:S51–S62.

LAJEUNESSE T.C. 2001. Investigating the biodiversity, ecology, andphylogeny of endosymbiotic dinoflagellates in the genus Symbio-dinium using the ITS region: in search of a ‘‘species’’ level marker.Journal of Phycology 37: 866–880.

LAJEUNESSE T.C. 2002. Diversity and community structure ofsymbiotic dinoflagellates from Caribbean coral reefs. MarineBiology 141: 387–400.

LAJEUNESSE T.C. 2005. Species radiations of symbiotic dinoflagel-lates in the Atlantic and Indo-Pacific since the Miocene–Pliocenetransition. Molecular Biology and Evolution 22: 570–581.

LAJEUNESSE T.C., LOH W.K.W., VAN WOESIK R., HOEGH-GULDBERG

O., SCHMIDT G.W. & FITT W.K. 2003. Low symbiont diversity insouthern Great Barrier Reef corals relative to those of theCaribbean. Limnology and Oceanography 48: 2046–2054.

LAJEUNESSE T.C., BHAGOOLI R., HIDAKA M., DEVANTIER L., DONE

T., SCHMIDT G.W., FITT W.K. & HOEGH-GULDBERG O. 2004a.Closely related Symbiodinium spp. differ in relative dominance incoral reef host communities across environmental, latitudinal andbiogeographic gradients. Marine Ecology Progress Series 284:147–161.

LAJEUNESSE T.C., THORNHILL D.J., COX E., STANTON F., FITT W.K.& SCHMIDT G.W. 2004b. High diversity and host specificityobserved among symbiotic dinoflagellates in reef coral commu-nities from Hawaii. Coral Reefs 23: 596–603.

LAJEUNESSE T.C., LAMBERT G., ANDERSEN R.A., COFFROTH M.A. &GALBRAITH D.W. 2005. Symbiodinium (Pyrrhophyta) genomesizes (DNA content) are smallest among dinoflagellates. Journalof Phycology 41: 880–886.

LAJEUNESSE T.C., REYES-BONILLA H., WARNER M.E., WILLS M.,SCHMIDT G.W. & FITT W.K. 2008. Specificity and stability in highlatitude eastern Pacific coral-algal symbioses. Limnology andOceanography 53: 719–727.

LAJEUNESSE T.C., FINNEY J.C., SMITH R.T. & OXENFORD H. 2009.Outbreak and persistence of opportunistic symbiotic dinoflagel-lates during the 2005 Caribbean mass coral ‘bleaching’ event.Proceedings of the Royal Society B: Biological Sciences 276: 4139–4148.

LAJEUNESSE T.C., PETTAY D.T., SAMPAYO E.M., PHONGSUWAN N.,BROWN B., OBURA D., HOEGH-GULDBERG O. & FITT W.K. 2010a.Long-standing environmental conditions, geographic isolationand host–symbiont specificity influence the relative ecologicaldominance and genetic diversification of coral endosymbionts inthe genus Symbiodinium. Journal of Biogeography 37: 785–800.

LAJEUNESSE T.C., SMITH R., WALTHER M., PINZON J., PETTAY D.T.,MCGINLEY M., ASCHAFFENBURG M., MEDINA-ROSAS P., CUPUL-MAGANA A.L., PEREZ A.L., REYES-BONILLA H. & WARNER M.E.2010b. Host–symbiont recombination versus natural selection inthe response of coral–dinoflagellate symbioses to environmentaldisturbance. Proceedings of the Royal Society B: BiologicalSciences 277: 2925–2934.

LAJEUNESSE T.C. & THORNHILL D.J. 2011. Improved resolution ofreef-coral endosymbiont (Symbiodinium) species diversity, ecolo-gy, and evolution through psbA non-coding region genotyping.PLoS ONE 6:e29013.

LAJEUNESSE T.C., PARKINSON J.E. & REIMER J.D. 2012. A genetics-based description of Symbiodinium minutum sp. nov. and S.psygmophilum sp. nov. (Dinophyceae), two dinoflagellates sym-biotic with Cnidaria. Journal of Phycology 48: 1380–1391.

LAM K.K. 2000. Sexual reproduction of a low-temperature tolerantcoral Oulastrea crispata (Scleractinia, Faviidae) in Hong Kong,China. Marine Ecology Progress Series 205: 101–111.


LIEN Y.-T., NAKANO Y., PLATHONG S., FUKAMI H., WANG J.T. &CHEN C.A. 2007. Occurrence of the putatively heat-tolerantSymbiodinium phylotype D in high-latitudinal outlying coralcommunities. Coral Reefs 26: 35–44.

LIEN Y.-T., FUKAMI H., & YAMASHITA Y. 2012. Symbiodinium CladeC dominates zooxanthellate corals (Scleractinia) in the temperateregion of Japan. Zoological Science 29: 173–180.

LIEN Y.-T., KESHAVMURTHY S., NAKANO Y., PLATHONG S., HUANG

H., HSU C.-M., FUKAMI H., YAMASHITA Y., HSIEH H.J., WANG J.-T. & CHEN C.A. 2013. Host genetic connectivity and diversity ofSymbiodinium D types in a stress-tolerant scleractinian coral,Oulastrea crispata, in the West Pacific. Marine Ecology ProgressSeries 473: 163–177.

MARTIN P.R. & MCKAY J.K. 2004. Latitudinal variation in geneticdivergence of populations and the potential for future speciation.Evolution 58: 938–945.

MAYR E. 1942. Systematics and the origin of species, from theviewpoint of a zoologist. Harvard University Press, Cambridge,MA.. 334 pp.

MONIZ M.B.J & KACZMARSKA I. 2010. Barcoding of diatoms:nuclear encoded ITS revisited. Protist 161: 7–34.

MOSTAFAVI P.G., FATEMI S.M.R., SHAHHOSSEINY M.H., HOEGH-GULDBERG O. & LOH W.K.W. 2007. Predominance of Clade DSymbiodinium in shallow-water reef-building corals off Kish andLarak Islands (Persian Gulf, Iran). Marine Biology 153: 25–34.

OLIVER T.A. & PALUMBI S.R. 2009. Distributions of stress-resistantcoral symbionts match environmental patterns but not regionalscales. Marine Ecology Progress Series 378: 93–103.

PEAKALL R. & SMOUSE P.E. 2006. GENALEX 6: genetic analysis inExcel. Population genetic software for teaching and research.Molecular Ecology Notes 6: 288–295.

PETTAY D.T. & LAJEUNESSE T.C. 2007. Microsatellites from Clade BSymbiodinium spp. specialized for Caribbean corals in the genusMadracis. Molecular Ecology Notes 7: 1271–1274.

PETTAY D.T. & LAJEUNESSE T.C. 2009. Microsatellite loci forassessing genetic diversity, dispersal and clonality of coralsymbionts in the ’stress-tolerant’ Symbiodinium Clade D. Molec-ular Ecology Resources 9: 1022–1025.

PETTAY D.T. & LAJEUNESSE T.C. 2013. Long-range dispersal andhigh-latitude environments influence the population structure of a‘‘stress-tolerant’’ dinoflagellate endosymbiont. PLoS ONE 8:e79208.

PETTAY D.T., WHAM D.C., PINZON J.H. & LAJEUNESSE T.C. 2011.Genotypic diversity and spatial–temporal distribution of Symbio-dinium clones in an abundant reef coral. Molecular Ecology 20:5197–5212.

PINZON J.H., DEVLIN-DURANTE M.K., WEBER X.M., BAUMS I.B. &LAJEUNESSE T.C. 2011. Microsatellite loci for Symbiodinium A3(S. fitti), a common algal symbiont among Caribbean Acropora(stony corals) and Indo-Pacific giant clams (Tridacna). Conserva-tion Genetics Resources 3: 45–47.

POCHON X., MONTOYA-BURGOS J.I., STADELMANN B. & PALOWSKI J.2006. Molecular phylogeny, evolutionary rates, and divergencetiming of the symbiotic dinoflagellate genus Symbiodinium. MolPhyl. Evol. 38: 20–30.

PRITCHARD J., STEPHENS M. & DONNELLY P. 2000. Inference ofpopulation structure using multilocus genotype data. Genetics155: 945–959.

R Foundation for Statistical Computing, Vienna Austria. URLhttp://www.R-project.org/.

RODRIGUEZ-LANETTY M., KRUPP D.A. & WEIS V.M. 2004. DistinctITS types of Symbiodinium in Clade C correlate with cnidarian/dinoflagellate specificity during onset of symbiosis. MarineEcology Progess Series 275: 97–102.

ROWAN R. 2004. Coral bleaching: thermal adaptation in reef coralsymbionts. Nature 430: 742.

ROWAN R. & POWERS D.A. 1992. Ribosomal-RNA sequences andthe diversity of symbiotic dinoflagellates (zooxanthellae). Pro-ceedings of the National Academy of Sciences of the United Statesof America 89: 3639–3643.

RYNEARSON T.A. & ARMBRUST E.V. 2004. Genetic differentiationamong populations of the planktonic marine diatom Ditylumbrightwellii (Bacillariophyceae). Journal of Phycology 40: 34–43.

SAMPAYO E.M., RIDGWAY T., BONGAERTS P. & HOEGH-GULBERG O.2008. Bleaching susceptibility and mortality of corals aredetermined by fine-scale differences in symbiont type. Proceedingsof the National Academy of Sciences of the United States ofAmerica 105: 10444–10449.

SAMPAYO E., DOVE S. & LAJEUNESSE T.C. 2009. Cohesive moleculargenetic data delineate species diversity in the dinoflagellate genusSymbiodinium. Molecular Ecology 18: 500–519.

SANTOS S.R. & COFFROTH M.A. 2003. Molecular genetic evidencethat dinoflagellates belonging to the genus SymbiodiniumFreudenthal are haploid. Biological Bulletin 204: 10–20.

SCHLUTER D. 2009. Evidence for ecological speciation and itsalternative. Science 323: 737–741.

SCHOENBERG D.A. & TRENCH R.K. 1980. Genetic variation inSymbiodinium (¼Gymnodinium) microadriaticum Freudenthal,and specificity in its symbiosis with marine invertebrates. II.Morphological variation in S. microadriaticum. Proceedings ofthe Royal Society of London B: Biological Sciences 207: 429–444.

SEIFERT K.A. 2009. Progress towards DNA barcoding of fungi.Molecular Ecology Resources 9: 83–89

SILVERSTEIN R.N., CORREA A.M.S., LAJEUNESSE T.C. & BAKER A.C.2011. Novel algal symbiont (Symbiodinium spp.) diversity in reefcorals of Western Australia. Marine Ecology Progress Series 422:63–75.

SITES J.W. & MARSHALL J.C. 2004. Operational criteria fordelimiting species. Annual Review of Ecology Evolution andSystematics 35: 199–227.

STAT M. & GATES R.D. 2011. Clade D Symbiodinium inscleractinian corals: a ‘‘nugget’’ of hope, a selfish opportunist,an ominous sign, or all of the above? Journal of Marine Biology2011: 730715.

STERN R.F., HORAK A., ANDREW R.L., COFFROTH M.A., ANDERSEN

R.A., KUPPER F.C., JAMESON I., HOPPENRATH M., VERON B.,KASAI F., BRAND J., JAMES E.R. & KEELING P.J. 2010.Environmental barcoding reveals massive dinoflagellate diversityon marine environments. PLoS ONE 5: e13991

SWOFFORD D. 2000. PAUP*, phylogenetic analysis using parsimony(* and other methods), Version 4.0b10. Sinauer Associates,Sunderland, MA.

THOMAS M.T., KREMER C.T., KLAUSMEIER C.A. & LITCHMAN E.2012. A global pattern of thermal adaptation in marinephytoplankton. Science 338: 1085–1088.

THORNHILL D.J., LAJEUNESSE T.C. & SANTOS S.R. 2007. MeasuringrDNA diversity in eukaryotic microbial systems: how intra-genomic variation, pseudogenes, and PCR artifacts confoundbiodiversity estimates. Molecular Ecology 16: 5326–5340.

THORNHILL D.J., KEMP D.W., BRUNS B.U., FITT W.K., & SCHMIDT

G.W. 2008. Correspondence between cold tolerance and temper-ate biogeography in a western Atlantic Symbiodinium (Dinophy-ta) lineage. Journal of Phycology 44: 1126–1135.

THORNHILL D.J., LEWIS A.M, WHAM D.C. & LAJEUNESSE T.C. 2014.Host-specialist lineages dominate the adaptive radiation of reefcoral endosymbionts. Evolution DOI: 10.1111/evo.12270

TOLLER W.W., ROWAN R. & KNOWLTON N. 2001. Zooxanthellae ofthe Montastraea annularis species complex: patterns of distribu-tion of four taxa of Symbiodinium on different reefs and acrossdepths. Biological Bulletin 201: 348–359.

TRENCH R.K. & BLANK R.J. 1987. Symbiodinium microadriaticumFreudenthal, S. goreauii sp. nov., S. kawagutii sp. nov. and S.pilosum sp. nov.: gymnodinioid dinoflagellate symbionts ofmarine invertebrates. Journal of Phycology 23: 469–481.

WAKELEY J. 2008. Coalescent theory: an introduction. Roberts &Company Publishers, Greenwood Village, CO, p. 352.

WHAM D.C., PETTAY D.T. & LAJEUNESSE T.C. 2011. Microsatelliteloci for the host-generalist ‘‘zooxanthella’’ Symbiodinium trenchiand other Clade D Symbiodinium. Conservation Genetics Resourc-es 3: 541–544.

WINDOM H.L., SILPIPAT S., CHANPONGSANG A., SMITH R.G. &HUNGSPREUGS M. 1984. Trace metal composition of andaccumulation rates of sediments in the upper Gulf of Thailand.Estuary and Coastal Shelf Science 19: 133–142.


WRIGHT S. 1931. Evolution in Mendelian populations. Genetics16:97–159.

YAJIMA T., SANO O., OKAMOTO T., SHIRAI Y., SHINYA T. & MATADA

M. 1986. Ecological distribution of the reef coral, Oulastreacrispata (Lamarck) at the shore region in the vicinity of TukumoBay. Bulletin of the Japan Sea Research Institute 18: 21–36.

YANAGI T., SACHOEMAR S.I., TAKAO T. & FUJIWARA S. 2001.Seasonal variation of stratification in the Gulf of Thailand.Journal of Oceanography 57: 461–470.

YAO H., SONG J., LIU K., HAN J., LI Y. PANG X., XU H., XIAO P. &CHEN S. 2010. Use of ITS2 region as the universal DNA barcodefor plants and animals. PLoS ONE 5: e13102.

Received 26 May 2013; accepted 14 January 2014

Associate Editor: Kerstin Hoef-Emden


Ecologically differentiated–stress tolerant–endosymbionts in the Dinoflagellate

Genus Symbiodinium (Dinophyceae), clade D, are different species

TODD C. LAJEUNESSE1*, DREW C. WHAM1, DANIEL TYE PETTAY1, JOHN EVERETT

PARKINSON1, SHASHANK KESHAVMURTHY2 AND CHAOLUN ALLEN CHEN2

1Department of Biology, The Pennsylvania State University, University Park, PA 16802, United States

2Biodiversity Research Center and Taiwan International Graduate Program (TIGP)-Biodiversity, Academia Sinica, Taipei, 115, Taiwan

Table S1. Samples locations for colonies of Oulastrea crispata across ~ 4000 km from the southern Andaman Sea to Japan. For each sample, ITS rDNA was analyzed and then a subset of samples received further analysis using the cp23S, and microsatellite flankers Sym88 and Sym93. The number of multilocus microsatellite genotypes acquired using 8 loci are also listed. In addition, locations are given for samples of S. trenchii (= D1-4 or D1a).

Location Latitude Longitude

Number of Samples

analyzed for rDNA ITS type*

Number of multilocus genotypes generated with 8 loci

Oulastrea crispata

Japan

Ishikawa (ISHI) 36˚38'N 136˚34'E 10

9

Shimoda (SHIM) 35˚38'N 140˚17'E 7

7

Okayama (OKAY) 34˚56'N 133˚48'E 10

5

Wakayama (WAKA) 34˚10'N 135˚22'E 10

9

Kumamoto (KUMA) 32˚30'N 130˚48'E 12

11

Kagoshima (KAGO) 31˚53'N 130˚37'E 4

3

Okinawa

Ishigaki (OKIN) 24˚22'N 124˚10'E 12

11

Miyagi (OKIN) 24˚22'N 124˚10'E 10

3

Toguchi (OKIN) 24˚22'N 124˚10'E 6

1

Taiwan

Wa-t'ung (WATU) 23˚38'N 119˚34'E 10

5

Penghu (PENG) 23°31'N 119°33'E 25

0

China

Hong Kong (HONG) 22˚28'N 114˚19'E 5

4

Weijhou Island (WEIJ) 21˚210'N 109˚15'E 6

5

Hainan Island (HAIN) 20˚50'N 110˚19'E 9

1

Thialand

Si-Chang (Si-CH) 13˚08'N 100˚51'E 11

7

Khang Kao Island (KANG) 13˚07'N 100˚48'E 19

19

KoBu-Lon Le (Bu-LO) 06˚50'N 99˚30'E 3

1

Sa-Tun (Sa-Tu 06˚30''N 100˚02'E 10

3

Phuket Island (PUKT) 07˚48''N 98˚24'E 4

3

Symbiodinium trenchi (D1-4, D1a)

Zanzibar (TAN) 06˚07''S 39˚10'E 16

13

06˚09''S 39˚09'E

Andaman Sea (AND) 07˚39''N 98˚36'E 124

57

07˚48''N 98˚24'E

Palau (PAL) 07˚19''N 134˚29'E 166

31

07˚15''N 134˚23'E

Res Sea (RS) 16˚36''N 41˚55'E 15 0

Arabian/Persian Gulf (AG) 27˚21''N 49˚54'E 4 0

Western Australia (WA) 20˚32"S 116˚38"E 15 0

Taiwan (TAI) 21˚57"N 120˚45"E 3 0

GBR, Australia (GBR) 18˚28"S 146˚52"E 5 0

17˚32"N 146˚23"E

Pheonix Islands (PHO) 4 0

* The LSU, cp23S, cob, and flanking sequences of microsatellite loci 88 and 93 were sequenced from a subset of these samples.

Table S2. Summary statistics on allele number and evenness among multi-locus genotypes representing Symbiodinium clade D species. Standard errors are listed in smaller font below each summary statistic. S. trenchi S. boreum S. eurythalpos Ntotal 101 54 72 Nunique 97 33 58 Na 5.500 3.625 5.625 0.423 0.905 1.164 Ae 2.426 2.296 2.971

0.252 0.571 0.519 Diagnostic loci Sym_93 Sym_17 Sym_67

Na = Average no. of alleles per locus Ae = No. of effective alleles per locus [1 / (Sum pi^2)]

Supporting Figures for online publication only

Figs S1-S5. Diverse host assemblages from rock island habitats in Palau associated with Symbiodinium trenchii.

Fig. S1. Colony of Lobophyllia hemprichi. Fig. S2. Mycedium elephantotus.

Fig. S3. Ctenactis echinata Fig. S4. Goniastrea edwardsi

Fig. S5. Favia palida Fig. S6 Colonies of the zebra coral, Oulastrea crispata, from an intertidal fringing reef in Phuket (left), Thailand, and high sediment sub-tidal habitat from Singapore (right). Figure S7. Collection sites throughout the Indo-Pacific where reef-coral species harboring S. trenchii (type D1-4 or D1a) were obtained (white circles); and locations where colonies of Oulastrea crispata were sampled (black circles). Red circle outlines indicate three locations, Zanzibar, Tanzania (TAN); Andaman Sea, Thailand (AND), and Palau (PAL) where samples of S. trenchii were genotyped using microsatellites for population genetic analyses. Other collection locations for S. trenchii include the Red Sea (RS), Arabian Gulf (AG), Western Australia (WA), Great Barrier Reef (GBR), Taiwan (TAI), and Phoenix Islands (PHO). The known distribution of O. crispata is indicated by the yellow shading. The Tropics of Cancer and Capricorn are represented by dashed grey lines. Fig. S8. Amino acid (aa) variability in Symbiodinium cytochromes b. In this structural model of a portion of the Qo reaction center, conservative invariant aa residues are solid circles. Variable residues are indicated by color that corresponds to and are possibly diagnostic of Symbiodinium from different clades (inset maximum parsimony phylogeny based on cob sequences; 918 bp). The amino acid at position 167 (colored red) is encoded as a polar Tyrosine in the species S. boreum and is uniquely different compared to all other Symbiodinium examined which instead possess a Phenylalanine (nonpolar) at this residue. Fig. S9. Denaturing gradient gel showing fingerprint profiles similar to those produced by Lien et al. (2013) of various Symbiodinium Clade D ITS2 types recovered from samples of Oulastrea crispata collected over a broad geographic and latitudinal range (Table 1; from Lien et al. 2013). This method screens for the numerically dominant and evolutionarily stable sequence variants in the ribosomal array to diagnose ecologically distinct Symbiodinium sp. (LaJeunesse 2002; Sampayo et al. 2009; LaJeunesse and Thornhill 2011). Fig. S10. Evanno’s Delta K showing the change in likelihood between K values using the software STRUCTURE. Maximum Delta K indicated by the red arrow.

Fig. S11. The probability of the data at each of the proposed K values in STRUCTURE. The change in likelihood between K values reaches is greates between K=2 and K=3 and the probability is maximized at K=4. The genetic distinctiveness of S. eurythalpos populations in the upper Gulf of Thailand explains this graphical pattern. Fig. S12. Average monthly temperatures from January 2003 to December 2011 inferred from satellite data. showing seasonal variation for three representative temperate, sub-tropical, and tropical locations where colonies of O. crispata were dominated by different Symbiodinium species. White, grey, and black correspond to the Tropical, Sub-Tropical, and Temperate locations indicated in Fig. 14, respectively. Fig. S13. Oulastrea ITS1 sequence data resolved no apparent phylogeographic structuring over the sample range indicating that the latitudinal partitioning of S. boreum, S. eurythalpos, and S. calidum are influenced by external environmental factors.

Figures S1-S6

Figure S7

Figure S8

Figure S9

Figures S10 and S11

S10

S11

Del

ta K

K

0

2 3 4 5 6 7 8 9

Mea

n of

est

. Ln

prob

of d

ata

K

2 4 6 8 10

-2800

-2600

-2400

-2200

-2000

-1800

-1600L(K) (mean + -SD)

DeltaK = mean ([L”(K)])/sd(L(K))

Figure S12

14

16

18

20

22

24

26

28

30

32

J F M A M J J A S O N D

Temperate

WAKAKUMASHIM

Tropical

Bu-LOSi-CH/KANG

Sub-tropical

HAINWEIJHONG

Tem

pera

ture

˚C

Figure S13

1

11

11

1

1

2

1

1

1

1

1

1

Oul 2299B (Shimoda) D15

Oul Phu07 441 (Phuket) D8

Oul 3278B (Sa-Tun) D8-12

Ou 3272A (Bu Lon) D8-12

Oul 2216 (Kumamoto) D15

Oul 2314A (Miyagi) D8

Oul 2314B (Miyagi) D8

Oul 2308B (Miyagi) D15 & D8

Oul 2236A (Wakayama) D15

Oul 3275 (Sa-Tun) D8

Oul 3272B (Bu-Lon) D8-12

Oul Phu07 579 (Phuket) D8

Oul 2299A (Shimoda) D15

Oul 3298 (Watung) D15

Oul 3301A (Watung) D8

Oul 3414A (Si Chiang) D8

Oul 2266 (Ishigaki) D8

Oul 3278A (Sa-Tun) D8-12

Oul 3301B (Watung) D8

Oul 3413 (Weijhou) D13-14

Oul 3409B (Weijhou) D12-13

Oul 3414 (Si-Chiang) D8

Oul 3409A (Weijhou) D12-13

Oul 2308A (Miyagi) D15 & D8

Oul 2298A (Shimoda) D15

Oul 2270B (Ishigaki) D8

Oul 2270A (Ishigaki) D8

Oul 2223 (Okyama) D15

S. eurythalpos

S. boreum

Oulastrea crispata ITS 1

Figures 1-4 phylogeneticanalyses March2014 · 2016. 9. 12. · Officers for 2014(with terms of...

Documents

Transcript of Figures 1-4 phylogeneticanalyses March2014 · 2016. 9. 12. · Officers for 2014(with terms of...