High diversity and host specificity observed among symbiotic dinoflagellates in reef coral communities from Hawaii

Transcription

1 Coral Reefs (2004) 23: DOI /s REPORT Todd C. LaJeunesse Æ Daniel J. Thornhill Evelyn F. Cox Æ Frank G. Stanton Æ William K. Fitt Gregory W. Schmidt High diversity and host specificity observed among symbiotic dinoflagellates in reef coral communities from Hawaii Received: 22 October 2003 / Accepted: 30 July 2004 / Published online: 11 September 2004 Ó Springer-Verlag 2004 Abstract The Hawaiian Islands represent one of the most geographically remote locations in the Indo- Pacific, and are a refuge for rare, endemic life. The diversity of symbiotic dinoflagellates (Symbiodinium sp.) inhabiting zooxanthellate corals and other symbiotic cnidarians from the High Islands region was surveyed. From the 18 host genera examined, there were 20 genetically distinct symbiont types (17 in clade C, 1 in clade A, 1 in clade B, and 1 in clade D) distinguished by internal transcribed spacer region 2 sequences. Most types were found to associate with a particular host genus or species and nearly half of them have not been identified in surveys of Western and Eastern Pacific hosts. A clear dominant generalist symbiont is lacking among Hawaiian cnidarians. This is in marked contrast with the symbiont community structures of the western Pacific and Caribbean, which are dominated by a few prevalent generalist symbionts inhabiting numerous host taxa. Geographic isolation, low host diversity, and a high proportion of coral species that directly transmit their symbionts from generation to generation are implicated in the formation of a coral reef community exhibiting high symbiont diversity and specificity. Communicated by H.R. Lasker T. C. LaJeunesse (&) Æ D. J. Thornhill Æ G. W. Schmidt Department of Plant Biology, University of Georgia, GA 30602, USA lajeunes@fiu.edu Tel.: Fax: T. C. LaJeunesse Æ D. J. Thornhill Æ W. K. Fitt Department of Ecology, University of Georgia, GA 30602, USA E. F. Cox Department of Zoology, Hawaii Institute of Marine Biology, USA F. G. Stanton Department of Biology, Leeward Community College, Hawaii, USA Introduction Encircled by the waters of the North American Countercurrent, the Hawaiian archipelago is among the most isolated subtropical regions in the Indo-Pacific (Simon 1987; Veron 1995). Due to its limited exchange with other central Pacific reef systems, Hawaii has the highest number of endemic corals in the Pacific (Hughes et al. 2002) and may potentially be a place of unusual symbiont-host combinations. Although, the Hawaiian High Islands experienced some coral bleaching in 1996, Hawaii s high latitude and proximity to up-welled cold eastern Pacific waters has mostly spared its corals from the severe bleaching and mortality accompanying abnormally warm seasurface temperatures associated with El Nin o-southern Oscillations (Jokiel and Brown 2004). Such events have severely impacted shallow reef systems throughout the Indo-Pacific and Caribbean (Hoegh-Guldberg 1999) and have led, in some cases, to community shifts among coral taxa and their symbionts (Glynn et al. 2001; Loya et al. 2001). Collectively, these geographic, environmental, and biological attributes prompted our investigation of coral-algal symbioses in this region of the tropics. We have learned much about the diversity, ecology, and evolution of Symbiodinium spp. and their host relations through the use of molecular genetic markers (e.g. Rowan and Powers 1991; Rowan et al. 1997; Baker and Rowan 1997; Coffroth et al. 2001; LaJeunesse 2002; Baker 2003; LaJeunesse et al. 2003; Santos et al. 2004). Numerous biological, hydrogeological, environmental, and historical factors determine the diversity of corals in a particular location (Veron 1995). For Symbiodinium spp., the taxonomic diversity and relative abundance of host taxa (i.e. habitat diversity and availability from the symbiont point of view) seem to be an additional axis governing their evolutionary diversification and ecological distribution (LaJeunesse 2002). Host communities are usually

2 597 dominated by few symbionts that occupy a broad range of host taxa; however, host-specific, regionally endemic, and/or ecologically rare lineages comprise the majority of symbiont diversity (>90%) (LaJeunesse 2002; LaJeunesse et al. 2003, 2004). These observations suggest that niche partitioning through host specialization and geographic isolation is important in symbiont speciation (Futuyma and Moreno 1988). The evaluation of Symbiodinium spp. diversity in Hawaii, as assessed by ribosomal internal transcribed spacer (ITS) DNA, was conducted under the same methods and criteria as carried out on other reefs (LaJeunesse 2002; LaJeunesse et al. 2003, 2004). The enumeration of the symbiont diversity and patterns of host association from this highly endemic, sub-province of the Pacific contributes to our understanding of Symbiodinium spp. biogeography, host specificity, and evolution. Moreover, the relatively low host diversity at this location allows a test of the hypothesis that biological interactions influence the apparent inverse relation between host and symbiont diversity. Materials and methods Collections Host Cnidarians (multiple individuals from each taxon) were collected using SCUBA or mask and snorkel from Kaneohe Bay on the southeastern side of Oahu, and from Haleiwa Trench, Wailua Bay on its north shore. Coral fragments of 3 5 cm 2 were stripped of tissue using an airbrush (Baker and Rowan 1997) and the resultant slurry of mucus, animal cell debris, and dinoflagellates was homogenized using a Tissue Tearor homogenizer (Biospec Ind.). After centrifugation at 800 1,000 g, the resulting algal pellets were transferred to a 1.5-ml microcentrifuge tube and preserved in 20% dimethylsulfoxide (DMSO), 0.25 M EDTA, sodium chloride-saturated water solution (Seutin et al. 1991). The symbionts from soft-bodied Actinarian, Zoanthidian, and Scyphozoan taxa were isolated as previously described (LaJeunesse 2002). Porites compressa and Montipora capitata (= verrucosa Maragos 1995), displaying wide morphological variation under identical environmental conditions, were sampled extensively to determine if any of these showed differences in their symbiont affiliations. DNA extractions, PCR-DGGE, and sequencing Nucleic acid extractions on 10 to 30 mg of dinoflagellate material were conducted using a modified Promega Wizard genomic DNA extraction protocol (LaJeunesse et al. 2003). The internal transcribed spacer region (ITS) 2 was amplified from each extract using primers ITS 2 clamp and ITSintfor 2 (LaJeunesse and Trench 2000) with the touch-down thermal cycle given in LaJeunesse (2002). Products from these PCR reactions were electrophoresed on denaturing gradient gels (45 80%) using a CBScientific system (Del Mar, CA). As described in detail by LaJeunesse (2002) and LaJeunesse et al. (2003), prominent bands characteristic of each unique fingerprint/profile were excised, reamplified, and sequenced. The redundancy PCR-DGGE and direct sequencing provides an important cross check of sequence identity with migration characteristics. PCR-DGGE also produces a fingerprint that shows dominant sequences attributed to either intragenomic variation across multiple ribosomal loci and/or the presence of multiple symbiont types (LaJeunesse 2002). A large proportion of ribosomal variants reported by investigators sequencing blindly appear to be artifacts, possibly resulting from PCR and cloning errors (Speksnijder et al. 2001). Perhaps a greater source of variation is from the sequencing of rare variants (occurring throughout the repeat region, but low in copy number) that are not indicative of a genome s entire ribosomal array (Diekmann et al. 2003; LaJeunesse et al., unpublished). Phylogenetic analyses Sequences were aligned manually using Sequence Navigator version 1.0 software (ABI, Division of Perkin-Elmer, Foster City, CA). Sister lineages Fr1 (sensu Pochon et al. 2001) and type F1 in clade F (Symbiodinium kawagutii (LaJeunesse 2001)) were used as outgroups (Genbank sequences AJ and AF333515). ITS sequences among clade C types have a high proportion of invariant characters (only 17 characters out of 308 are parsimony-informative). Base substitutions across the ITS rdna in these Symbiodinium spp. are far from approaching saturation. Therefore, maximum parsimony (under heuristic search mode), which does not assume a particular model of molecular evolution and allows for use of informative sequence gaps and insertions (considered a 5th character state), is thus a conservative method for phylogenetic inference of clade C. Maximum likelihood and neighbor-joining analyses (with gaps excluded) were also performed on the data using PAUP* V. 4.0b10 software (Swofford 1999). A fourth method of phylogenetic inference using Baysian statistics was implemented using the software MrBayes version 3.0b4 (Huelsenbeck and Ronquiest 2001). Under the HKY85 and general time reversible (GTR) models of sequence evolution, beginning with an unspecified tree topology, log probability of observing the data plateaued reaching stationarity early between 10,000 and 20,000 generations. Both one-hundred thousand and one million generations (with no discard, i.e. no burn-in, hence probabilities are conservative estimates) under repeated trials produced near identical values.

3 598 Results A total of 20 distinctive symbiont types (17 clade C, 1 clade B, 1 clade A and 1 clade D), characterized by PCR-DGGE fingerprinting and sequencing of the ITS 2 region (LaJeunesse 2001, 2002; LaJeunesse et al. 2003) (Fig. 1), were identified from 23 host species representing 18 genera (Table 1). As in other regions of the Pacific, including the Pacific coast of Panama, southern and central GBR, Okinawa, and Guam (Baker and Rowan 1997; Loh et al. 1998; Pochon et al. 2001; LaJeunesse et al. 2003, 2004), the community of symbiotic cnidarians maintained a disproportionate number of symbioses with clade C types, while symbioses involving members from clades A, B, and D were rare. Identical fingerprint DGGE-PCR profiles were consistently produced from replicate sampling of each host species surveyed (Table 1). Host-symbiont associations surveyed from each location on Oahu were the same. Fig. 1 PCR-DGGE fingerprinting of the ITS 2 region characterizing the diversity of Symbiodinium spp. found among Hawaiian corals. These analyses identified 17 different C-type signatures whose ribosomal gene arrays were either dominated by a single sequence or exhibited two or more co-dominant variants. Most types were found in a specific host genus (or a single species within that genus) whose identity is provided above. In fingerprints with multiple bands, the relative band intensities in each pattern did not change when observed in replicate host samples, indicating fixed intragenomic variation. Mixed fingerprint profiles suggesting the presence of two or more different symbiont populations were never detected within a host colony. An alphanumeric name for each symbiont type is given above the corresponding fingerprint profile; uppercase letters indicate lineage (clade), numbers represent ITS type, and lowercase letters denote a definitive rdna co-dominant paralog, when one was clearly evident. Examples of heteroduplexes are indicated The occurrence of a particular symbiont in the ecosystem was therefore dependent on the presence of a particular host taxon. Because a majority of host species contained their own distinctive Symbiodinium type, the community composition for these reefs lacked a dominant host generalist symbiont (Fig. 2). The external physical environment was a factor in influencing certain partner combinations for several corals. In the host species, Montipora capitata (=verrucosa), M. flabellata, andporites compressa, a change in the symbiont fingerprint was attributed to depth of collection. Montipora capitata is morphologically plastic and has two color morphs. The common brown variety and all its morphological variants associated with C31 above 15 m, but hosted type C26a at greater depths. The less common and less palatable shallow-dwelling ruddy orange form of Montipora capitata associated with type D1a in clade D (Table 1). M. patula colonies above 10 m also hosted D1a (c.f. Rowan and Powers 1991). The symbiont described as Symbiodinium kawagutii from clade F (LaJeunesse 2001), originally cultured by R. Kinzie, and now maintained in labs around the world, is most probably not the true symbiont of M. capitata (= verrucosa) but more likely an opportunistic contaminant from the initial isolation process (c.f. Santos et al. 2001; LaJeunesse 2002). Type C15, observed previously in Porites spp. from the west Pacific and western Indian Ocean (LaJeunesse et al. 2003; Baker and LaJeunesse, unpublished), was also identified in each of the five Porites spp. from Hawaii (Table 1). Morpho-types of Porites compressa, sampled below 10 m contained C15b, a symbiont also found in the encrusting octocoral, Sarcothelia sp., living at similar depths. Interestingly, a purple-blue morph colony of P. compressa that was originally collected

4 599 Table 1 Host species, symbiont type and collection depth (meters) of samples from Oahu Hawaii, north central Pacific. Numerals in parentheses next to host names indicate the number of colonies independently sampled Host Depth (meters) Symbiont type (Genbank accession) Scleractinia Acroporidae Montipora capitata (orange color morph) (4) Montipora capitata (various colony morphologies) (14) Montipora capitata D1a C31 (AY258496) 20.0 C26/26a (AY239378/AY258501) (deep encrusting form) (2) Montipora flabellata 2.0 C32 (AY258498) Montipora flabellata C32a Montipora patula (2) D1a Montipora patula 20.0 C31 Pocilloporidae Pocillopora damicornis (dark morph) (4) C1d (AY258488) Pocillopora meandrina (4) C1c Pocillopora molokensis (5) C34 (AY258499) Pocillopora eydouxi (2) 20.0 C1c Pocillopora ligulata 20.0 C1g (AY589730) Poritidae Porites compressa (various colony morphologies) (19) C15 Porites compressa (3) C15b (AY258491) Porites compressa (10-year transplant ) 1.0 (>10.0) C15b Porites evermanni (4) C15 Porites lobata (5) C15 Porites rus (3) C15c (AY258492) Porites brighami (2) 20.0 C15 Siderastreidae Coscinaraea wellsi (1) C1 Psammocora spp. (4) C1f (AY258490) Fungiidae Fungia scutaria (5) C1f Fungia scutaria (2) (acanthicauli) 3.0 C1f Cycloseris vaughani (1) 20.0 C1 Agariciidae Pavona varians (7) C27 Pavona duerdeni (4) C27 Leptoseris incrustans (2) C1 Faviidae Leptastrea purpurea (6) C1f Leptastrea bottae (3) 2.0 C1f Cyphastrea ocellina (3) 2.0 C1f Actiniaria Aiptasia pulchella (2) 1.0 B1 Boloceroides mcmurrichi 2.0 C3 Zoanthidea Palythoa sp. (2) C1 Palythoa sp C3 Protopalythoa sp. (3) 1.0 C3m (AY258497) Zoanthus pacificus (3) 1.0 C29 (AY258494) Alcyonacea Sarcothelia ( edmondsoni ) sp. (2) C15b Scyphozoa Cassiopeia sp. 3.0 A3 from a depth below 10 m off the Island of Hawaii and transplanted to a shallow patch reef in Kaneohe Bay (C. Hunter personal communication), possessed the deep-water symbiont, C15b. That the symbiont population in the transplanted P. compressa did not undergo a shift to C15 after 10 years raises fundamental questions regarding the flexibility of this symbiosis. Phylograms, derived by maximum parsimony, comparing clade C ITS 2 sequences both rooted (Fig. 3a) and unrooted (Fig. 3b), have identical topologies. Neighbour-joining, Maximum Likelihood, and Bayesian inference methods also yielded very similar trees. Because gap data were not included, posterior probabilities were not calculated for certain internal branch nodes (Fig. 3a). C3 is most basal among clade C sequences (c.f. LaJeunesse 2002). Of potential ecological significance, host generalists in the west pacific and Caribbean, C1 and C3, were found in host taxa uncommon or rare around Oahu. The fingerprint of C1f, identified from several Hawaiian coral taxa (Fig. 1), is not presently

5 600 Molecular-based surveys of Symbiodinium spp. diversity in various regions around the world, including Hawaii, reveal much about the genetic relatedness, host specificity, and geographic distribution of these organisms. These phylogeographic data and ecological observations are essential in understanding the ecological and evolutionary processes exhibited by these animal-algal endosymbioses. Diversity and community structure of Hawaii s Symbiodinium spp. Fig. 2 Symbiont diversity (separated by clade) and community structure based on each symbiont s prevalence among numbers of host genera for a Carrie Bow Cay, Belize Caribbean (LaJeunesse and Warner, unpublished); b Heron Island, southern GBR (LaJeunesse et al. 2003) and c Oahu, Hawaii. Hawaii lacks a clear host generalist that dominates the host community as found in other reef systems. Thus, Hawaii s symbiont diversity is nearly equivalent to the southern GBR reef with only half the number of genera and one-third the number of species surveyed. The number of host genera surveyed from each region is given in the upper right hand corner of each panel known in western regions of the Pacific, and yet was recently found in Psammocora spp. from the northeastern Pacific (LaJeunesse et al. 2003, 2004; LaJeunesse and Reyes-Bonilla, unpublished). Some of the new symbiont fingerprints observed (Fig. 1) might be localized, or endemic, to this region of the Pacific. Discussion The majority of zooxanthellae diversity surveyed in Hawaii and on other Pacific reefs (Baker and Rowan 1997; Pochon 2001; LaJeunesse et al. 2003, 2004), belongs to clade C. These Symbiodinium spp. appear to represent the most ecologically important group in the Indo-Pacific (Baker 2003). The community structures of Symbiodinium spp. characterized from Oahu s host assemblage contrasts with Caribbean and western Pacific reefs where a greater percentage of host taxa share a common symbiont type (Fig. 2) (LaJeunesse 2002; LaJeunesse et al. 2003; LaJeunesse and Warner, unpublished). Absence of a dominant generalist combined with a high number of symbioses involving specialized symbionts, accounts for this proportionately high diversity relative to other reef systems around the world (Fig. 4). Two main factors probably contribute to high relative symbiont diversity among Hawaiian corals. Firstly, Hawaii has a high composition of host taxa in the genera Porites, Montipora, and Pocillopora, corals that directly transfer their symbionts to the next generation via vertical transmission. Secondly, for coral communities displaying low species richness but high species evenness over evolutionary time scales, selection pressures may favor greater host-symbiont specialization in species that acquire their symbionts from environmental pools (Law 1985). In many symbiotic systems where host offspring acquire their symbionts directly from the parent (vertical transmission), genetic isolation and natural selection promotes symbiont specialization (Douglas 1998; Loh et al. 2001). Most corals in the west Pacific broadcast spawn (Richmond and Hunter 1990), producing offspring that must acquire symbionts from environmental sources (horizontal transmission). Porites and Montipora are common coral genera in the Indo-Pacific and are especially abundant in Hawaii, which mostly lacks the elsewhere abundant the Acropora spp. While species from these genera typically broadcast spawn, they all release eggs containing vertically transmitted symbionts (Richmond and Hunter 1990). Depending on geographic location, Pocillopora spp. brood and/or spawn. Both larvae and eggs, when released, possess symbionts obtained from their parents (Kinzie 1996). Together, these three ecologically dominant genera maintain specific associations with separate sub-cladal lineages within clade C Symbiodinium (Fig. 3). The number of specific types from each of these sub-clades accounts for a large proportion of the Symbiodinium spp. diversity found in Hawaii. While direct symbiont transfer from generation to generation favors the evolution of specialist symbiont lines (Douglas 1998), high specificity and hence stability

6 601 Fig. 3 ITS 2 phylogenies inferred from maximum parsimony of clade C Symbiodinium spp. identified from Hawaiian cnidarians. a Phylogeny rooted with the outgroup taxa Fr1 and Fr5 (S. kawagutii, in clade F type 1; in group Fr 5 sensu Pochon et al. 2001) indicating that C3 is most basal among clade C sequences. Values indicated for each internal branch node are bootstrap estimates (first number), bootstrap with resampling doubled to compensate for the high proportion of invariant characters (underlined), and Bayesian posterior probabilities (parentheses). Internal nodes that lack posterior probabilities are based on insertion/deletions, not assessed by Bayesian methods (b). Clade C taxa only (outgroup taxa omitted) were reanalyzed and presented as an unrooted polytomy. The topology is identical to the boldfaced portion of the rooted phylogeny in panel a. A reanalysis of bootstrap support was conducted and the values shown are as described previously. Coral genera whose offspring acquire symbionts through vertical transmission, Porites, Montipora, and Pocillopora, associate with Symbiodinium spp. that form discrete sub-clades (shaded portions). Solid squares are symbiont taxa presently recorded only from Hawaii. Type C1d also occurs in P. damicornis from the eastern Pacific (open square), but is possibly absent from the western Pacific (LaJeunesse et al. 2003, 2004). Remaining branch termini without symbols are known from hosts in the western Pacific (LaJeunesse et al. 2003, 2004). Intra-genomic variants (in parentheses), not used to distinguish a different symbiont, were included to show their phylogenetic positions relative to sister sequences from the same genome. Neighbour-joining, Maximum Likelihood, and Baysian based analyses (insertions/deletion data removed) yielded near identical trees are also observed in hosts that acquire symbionts from the environment (Wilkinson 2001). Adults and juveniles of the mushroom coral Fungia scutaria, a broadcaster with horizontal symbiont transmission, sampled from Kaneohe Bay and the North Shore possessed symbiont C1f regardless of depth. This symbiont when compared against types C15, Cld, and C31, originating from P. compressa, P. damicornis, and M. capitata respectively, infects with greater success and achieves higher growth rates in Fungia scutaria larvae (Weis et al. 2001; Rodriguez-Lanetty et al. 2004). Most zoanthids surveyed from the Pacific possess symbiont types not found in other hosts (LaJeunesse et al. 2003, 2004). This group, with one known exception, consists primarily of broadcast spawners that do not vertically transmit their symbionts (Ryland 1997). Clearly, aspects other than mode of symbiont transmission are important in the evolution of specific host-symbiont associations. The host cell environment, modified by the external physical environment, must promote niche diversification among symbiont taxa (Futuyma and Moreno 1988; LaJeunesse 2002). Ecological and evolutionary trade-offs exist between symbionts associating with numerous host taxa versus those specializing with a narrow host range (Futuyma and Moreno 1988). Selection pressures supporting the maintenance of a generalist or the evolution of special-

7 602 Fig. 4 Symbiont diversity versus host diversity for various Pacific and Caribbean reefs. For each location, the number of Symbiodinium ITS types were divided by the number of host genera surveyed. This standardized value was then graphed against total host generic diversity. Hawaii (1) possesses the lowest host diversity, yet exhibits proportionately high symbiont diversity. The host assemblages surveyed at Feather (10) and Rib reefs (9) (Central GBR, LaJeunesse et al. 2004) contain a low number of symbionts in proportion to their host diversity (>50 genera). Caribbean reefs near Lee Stocking, Bahamas (2) (LaJeunesse, unpublished) and Carrie Bow Cay, Belize (3) (LaJeunesse and Warner, unpublished) also exhibit a relatively high Symbiodinium spp. diversity. Reef communities from Key Largo Florida, USA (4), and Curac ao Island (inshore Central GBR) (6), which are exposed to more variable and/or stressful environments, have low host and symbiont diversities. Other symbiont-to-host valuations are recorded for Zamami Island, Okinawa, Japan (7) (LaJeunesse et al. 2004), Heron Island, Australia (southern GBR) (8) (LaJeunesse et al. 2003), and Puerto Morelos, Mexico (5) (LaJeunesse 2002). Open circles around numerals indicate Caribbean reefs, while solid squares indicate Pacific reefs ists may shift depending on host taxa abundances (habitat availability) and diversity (habitat heterogeneity) (Law 1985). The High Islands began their emergence during the Miocene-Pliocene transition (Clague and Dalrymple 1987), as coral reef communities from older islands experienced major turnover (Jokiel 1987; Kay and Paulumbi 1987). Since then, isolation, low colonization rates, and periodic extinctions have suppressed host diversity in this remote region, a situation that may promote symbiont specialization for even those corals that acquire symbionts from environmental pools at the beginning of each generation. Biogeography Various Symbiodinium spp, like their coral hosts, show differences in geographic distribution (LaJeunesse 2002; LaJeunesse et al. 2004). Some are distributed over great distances, even to different oceans, while many are regionally localized and possibly endemic. A comparison of symbionts from Hawaii with those found in hosts from the GBR and Okinawa indicates that the Central Pacific province contains many types not found in the west Pacific (Fig. 3) (LaJeunesse et al. 2003, 2004). Surveys of coral reefs nearest to Hawaii and throughout the rest of the central Pacific will be necessary to determine if any of these new types are endemic to Hawaii. The geographic distribution of different symbiont types are potentially useful for investigating genetic connectivity between coral reef systems, even over limited spatial scales such as the Caribbean (Santos et al. 2004). Genetically similar symbionts from each of the pocilloporid, montiporid and poritid sub-clades illustrated in Fig. 3 are also present in regions from the West Pacific. While some share exact identity with those from Hawaii, many others have slight sequence differences that distinguish them regionally. For example, type C1c has an extensive geographic distribution and is found in Pocillopora spp. from the GBR and Okinawa (LaJeunesse et al. 2003, unpublished), and in Pavona spp. from the eastern Pacific (IglesiasPrieto et al. 2004). This is in contrast to the distribution of a closely-related type, C1d, that occurs in P. damicornis from Hawaii and the eastern Pacific but, appears not to be present in the western Pacific (LaJeunesse et al. 2003, 2004). Types C1 g and C34, known only from Hawaiian pocilloporids, may have more restricted geographic distributions and are possibly endemic (Fig. 3). Many coral-algal symbioses, even those with flexible partnerships, remain stable over wide geographic distances as long as the external environment remains similar (LaJeunesse and Trench 2000; Coffroth et al. 2001; Loh et al. 2001; LaJeunesse 2002). Entire symbiont communities appear to be maintained over geographic ranges even as great as that encompassed by the barrier reef along the western Caribbean (LaJeunesse et al., unpublished). While the high islands of Hawaii, Kauai, and Maui, among others, were not surveyed in this study, they probably have similar symbioses as observed on Oahu. These islands have similar coral communities and probably constitute a connected system (Veron 1995; Jokiel 1987). Hostsymbiont combinations described on the north shore and southeast side of Oahu were identical, and the P. compressa transplant from the Island of Hawaii possessed the same symbiont type as P. compressa from similar depths on Oahu. These data suggest that the symbioses characterized on Oahu also are present nearby islands. Acknowledgments The authors thank the participants in the Hawaii Institute of Marine Biology s Molecular biology of corals, Edwin W. Pauley summer program of CBS Scientific generously loaned their DGGE electrophoresis system for this course. Cynthia L. Hunter for helpful discussions and field assistance. T.C.L thanks Michael P. Lesser and Jo-Ann Leong for facilitating his course participation and stay on Coconut Island. This work was supported by the Edwin W. Pauley foundation, National Science Foundation grant OCE to WK Fitt and GW Schmidt, a

8 603 grant from the Smithsonian Center for Caribbean Research, and a NSF graduate research fellowship to D. J. Thornhill. References Baker AC (2001) Corals bleach to survive change. Nature 411: Baker AC (2003) Flexibility and specificity in coral-algal symbiosis: diversity, ecology and biogeography of Symbiodinium. Ann Rev Ecol Evol Syst 34: Baker AC, Rowan R (1997) Diversity of symbiotic dinoflagellates (zooxanthellae) in scleractinian corals of the Caribbean and eastern Pacific. Proc 8th Int Coral Reef Symp 2: Buddemeier RW, Fautin DG (1993) Coral bleaching as an adaptive mechanism. BioScience 43: Clague DA, Dalrymple GB (1987) In: Volcanism in Hawaii. Decker RW, Wright TL, Stauffer PH (eds) US Govt Printing Office pp 5 54 Coffroth MA, Santos SR, Goulet TL (2001) Early ontogenic expression of specificity in a cnidarian-algal symbiosis. Mar Ecol Prog Ser 222:85 96 Diekmann OE, Olsen JL, Stam WT, Bak RPM (2003) Genetic variation within Symbiodinium clade B from the coral genus Madracis in the Caribbean (Netherlands Antilles) Coral Reefs 22:29 33 Douglas AE (1998) Host benefit and the evolution of specialization in symbiosis. Heredity 81: Futuyma DJ, Moreno G (1988) The evolution of ecological specialization. Ann Rev Ecol Syst 19: Glynn PW, Mat JL, Baker AC, Calderon MO (2001) Coral bleaching and mortality in Panama and Ecuador during the El Nino-southern oscillation event: spatial/temporal patterns and comparisons with the event. Bull Mar Sci 69: Hoegh-Guldberg O (1999) Climate change, coral bleaching and the future of the world s coral reefs. Mar Freshwater Res 50: Huelsenbeck JP, Ronquist F (2001) MrBays: Bayesian inference of phylogenetic trees. Bioinformatics 17: Hughes TP, Bellwood DR, Connolly SR (2002) Biodiversity hotspots, centers of endemicity, and the conservation of coral reefs. Ecol Lett 5: Hunter CL (1988) Genotypic diversity and population structure of the Hawaiian reef coral Porites compressa. Dissertation. University of Hawaii, 136 pp Iglesias-Prieto R, Trench RK (1997) Photoadaptation, photoacclimation and niche diversification in invertebrate-dinoflagellate symbioses. Proc 8th Int Coral Reef Symp 2: Iglesias-Prieto R, Beltra n VH, LaJeunesse TC, Reyes-Bonilla H, Thome PE (2004) Different algal symbionts explain the vertical distribution of dominant reef corals in the eastern Pacific. Proc R Soc Lond 271: Jokiel PL (1987) Ecology, biogeography and evolution of corals in Hawaii. TREE 2: Jokiel PL, Brown EK (2004) Global warming, regional trends and inshore environmental conditions influence coral bleaching in Hawaii. Global Change Biol Jokiel PL, Coles SL (1990) Response of Hawaiian and other IndoPacific reef corals to elevated temperature. Coral Reefs 8: Kay AE, Paulumbi SR (1987) Endemism and evolution in Hawaiian marine invertebrates. TREE 2: Kinzie RA (1996) Modes of speciation and reproduction in archaeocoeniid corals. Galaxea 13:47 64 LaJeunesse TC (2001) Investigating the biodiversity, ecology, and phylogeny of endosymbiotic dinoflagellates in the genus Symbiodinium using the internal transcribed spacer region: in search of a species level marker. J Phycol 37: LaJeunesse TC (2002) Diversity and community structure of symbiotic dinoflagellates from Caribbean coral reefs. Mar Biol 141: LaJeunesse TC, Trench RK (2000) The biogeography of two species of Symbiodinium (Freudenthal) inhabiting the intertidal anemone, Anthopleura elegantissima (Brandt). Biol Bull 199: LaJeunesse TC, Loh WKW, van Woesik R, Hoegh-Guldberg O, Schmidt GW, Fitt WK (2003) Low symbiont diversity in southern Great Barrier Reef corals relative to those of the Caribbean. Limnol Oceanogr 48: LaJeunesse TC, Bhagooli R, Hidaka M, Done T, devantier L, Schmidt GW, Fitt WK, Hoegh-Guldberg (2004) Closely-related Symbiodinium spp. differ in relative dominance within coral reef host communities across environmental, latitudinal, and biogeographic gradients. Mar Ecol Prog Ser Law R (1985) Evolution in a mutualistic environment. In: Boucher DH (ed) The biology of mutualism: ecology and evolution. Oxford University Press, New York, pp Loh WK, Loi T, Carter D, Hoegh-Guldberg O (2001) Genetic variability of the symbiotic dinoflagellates from the wide ranging coral species Seriatopora hystrix and Acropora longicyathus in the Indo-West Pacific. MEPS 222: Loya Y, Sakai K, Yamazato K, Nakano Y, Sambali H, Van Woesik R (2001) Coral bleaching: the winners and the losers. Ecol Lett 4: Maragos JE (1995) Revised checklist of extant shallow-water coral species from Hawaii (Cnidaria: Anthozoa: Scleractinia). Bishop Museum Occ Papers 42:54 55 Pochon X, Pawlowski J, Zaninetti L, Rowan R (2001) High genetic diversity and relative specificity among Symbiodinium -like endosymbiotic dinoflagellates in soritid foraminiferans. Mar Biol 139: Richmond RH, Hunter CL (1990) Reproduction and recruitment of corals: comparisons among the Caribbean, the tropical Pacific and the Red Sea. MEPS 60: Rodriguez-Lanetty M, Krupp D, Weis VM. (2004) Distinct ITS types of Symbiodinium in clade C correlate to cnidarian/dinoflagellate specificity during symbiosis onset. Mar Ecol Prog Ser 275: Rowan R, Powers DA (1991) A molecular genetic classification of zooxanthellae and the evolution of animal-algal symbiosis. Science 251: Rowan R, Knowlton N, Baker A, Jara J (1997) Landscape ecology of algal symbionts creates variation in episodes of coral bleaching. Nature 388: Ryland JS (1997) Reproduction in zoanthidea (Anthozoa: Hexacorallia) Invert. Reprod Dev. 31: Santos SR, Taylor DJ, Coffroth MA (2001) Genetic comparisons of freshly isolated vs. cultured symbiotic dinoflagellates: implications for extrapolating to the intact symbiosis. J Phycol 37: Santos SR, Shearer TL, Hannes AR, Coffroth, MA (2004) Finescale diversity and specificity in the most prevalent lineage of symbiotic dinoflagellates (Symbiodinium, Dinophyceae) of the Caribbean. Molec Ecol 13: Seutin G, White BN, Boag PT (1991) Preservation of avian blood and tissue samples for DNA analyses. Can J Zool 69:82 92 Simon C (1987) Hawaiian evolutionary biology: an introduction. TREE 2: Speksnijder AGCL, Kowalchuk GA, De Jong S, Kline E, Stephan JR, Laanbroek HJ (2001) Microvariation artifacts introduced by PCR and cloning of closely related 16S rrna gene sequences. Appl Environ Microbiol 67: Swofford DL (1999) PAUP*, Phylogenetic Analysis Using Parsimony (*and other methods), Version 4.0b10, Sinauer, Sunderland, Mass. Veron JEN (1995) Corals in space and time. UNSW Press, Sydney, pp 321 Weis VM, Reynolds WS, deboer MD, Krupp DA (2001) Hostsymbiont specificity during onset of symbiosis between the dinoflagellates Symbiodinium spp. and planula larvae of the scleractinian coral Fungia scutaria. Coral Reefs 20: Wilkinson DM (2001) Horizontally acquired mutualisms, an unsolved problem in ecology? Oikos 92: