Diversity and Distribution of Symbiodinium Associated with Seven Common Coral Species in the Chagos Archipelago, Central Indian Ocean

The Chagos Archipelago designated as a no-take marine protected area in 2010, lying about 500 km south of the Maldives in the Indian Ocean, has a high conservation priority, particularly because of its fast recovery from the ocean-wide massive coral mortality following the 1998 coral bleaching event. The aims of this study were to examine Symbiodinium diversity and distribution associated with scleractinian corals in five atolls of the Chagos Archipelago, spread over 10,000 km 2. Symbiodinium clade diversity in 262 samples of seven common coral species, Acropora muricata, Isopora palifera, Pocillopora damicornis, P. verrucosa, P. eydouxi, Seriatopora hystrix, and Stylophora pistillata were determined using PCR-SSCP of the ribosomal internal transcribed spacer 1 (ITS1), PCR-DDGE of ITS2, and phylogenetic analyses. The results indicated that Symbiodinium in clade C were the dominant symbiont group in the seven coral species. Our analysis revealed types of Symbiodinium clade C specific to coral species. Types C1 and C3 (with C3z and C3i variants) were dominant in Acroporidae and C1 and C1c were the dominant types in Pocilloporidae. We also found 2 novel ITS2 types in S. hystrix and 1 novel ITS2 type of Symbiodinium in A. muricata. Some colonies of A. muricata and I. palifera were also associated with Symbiodinium A1. These results suggest that corals in the Chagos Archipelago host different assemblages of Symbiodinium types then their conspecifics from other locations in the Indian Ocean; and that future research will show whether these patterns in Symbiodinium genotypes may be due to local adaptation to specific conditions in the Chagos.

Overall composition of Symbodinium clades in the seven coral species sampled from all four locations in the Chagos Archipelago revealed the presence of only two Symbiodinium clades; C and A ( Fig. 2 and Fig. S1).
Among the divergent lineages of Symbiodinium, clade C is ecologically abundant and diverse among reef corals in both the Indo-Pacific and western Atlantic, whereas clade A is relatively common in shallow water scleractinians in the Caribbean and Indian Ocean [22,27,34] but is rare in the Pacific (reviewed in [5], [45], but see [46]). Although surveys have shown that clade C is the dominant symbiont associated with scleractinian corals in the Pacific Ocean and West Indian Ocean [1,7,19,22,[47][48], a recent study on the same sampling scale in the north-eastern Indian Ocean (Andaman Sea) showed a significant proportion of clade D and a diversity of Symbiodinium types in different coral hosts [22]. Although some regions of the Indian Ocean have been surveyed [7,22,49], there is no information of Symbiodinium diversity from the central Indian Ocean either at regional or local scale, which limits our understanding of the relationship between coral host and Symbiodinium on a broad biogeographical scale.
The Chagos Archiphelago, designated as a no-take marine protected area in 2010 [50][51], is located in the central Indian Ocean on the Chagos-Laccadive Ridge that extends as far south as 20 u S. It covers 550,000 km 2 with .60,000 km 2 shallow limestone platform and reefs [51]. Chagos is a valuable location for studying corals, as it contains .25% of the Indian Ocean reef area [51] in a condition and habitat that is largely unaffected by direct, local human impacts, although during the El Niñ o event of 1998, the Chagos suffered severe coral bleaching and mass mortality especially among large colonies of tabular Acropora [52]. However, significant and rapid recovery was observed afterwards [24]. Also, coral diseases are extremely low in the Chagos reefs [51]. The role of the Chagos Archiphelago in the Indian Ocean is very important for a number of reasons, including conservation of commercial fish stocks [50] and supports densities of coral reef fishes one to two orders of magnitude greater than in other Indian Ocean locations [51]. It also provides a scientific reference point for several aspects of Indian Ocean research and for global studies of reef condition and reef restoration. Further, the Chagos archipelago appears to act as a marine biodiversity corridor between the north-eastern and western Indian Ocean [51], such that knowledge of its coral and coral reefs is very important for conservation efforts. This has led to its designation as the world's largest marine protected area [50,53].
isIn the wake of global climate change and future sea surface warming predicted for this region and others [24], we conducted this study, 1. To document the diversity of Symbiodinium in seven The numbers in the brackets denote sample obtained from each site, for each coral species and number of clade C and A. doi:10.1371/journal.pone.0035836.t001 common scleractinian species in the Chagos Archipelago and 2.
To know whether such remote reefs contribute to some unique Symbiodinium clades/types specific to corals present and, if so, discuss the implication of such associations. Under the scenario of temperature and turbidity in determining the regional diversity of Symbiodinium [22], we expected proportionately less clade D relative to clade C in our coral samples, as predicted by ecological characteristics of the Chagos where water clarity is high.

Ethics Statement
Coral tissues were collected and exported from British Indian Ocean Territory with permission granted by the Administrator of the Territory, Foreign and Commonwealth Office, London, United Kingdom.

Sample sites and field collection
Seven common scleractinian coral species (Isopora palifera, Acropora muricata, Pocillopora damicornis, P. eydouxi, P. verrucosa, Stylophora pistillata and Seriatopora hystrix) were sampled at depths of 2 m to 7 m from five parts of the Chagos Archipelago in March 2006 (Fig 1). Sampled sites were Salomon, Peros Banhos and Diego Garcia atolls, and from Eagle Islands and Three Brothers from the Great Chagos Bank. In all 262 samples (Table 1) collected from 7 coral species were preserved in 95% ethanol for DNA extraction.

Molecular analysis of Symbiodinium clade and types
Use of combinations of genetic analyses better resolves the Symbiodinium identity in a given sample [54]. Thus, we used three techniques to analyze Symbiodinium clades/types from the extracted DNA. DNA extraction followed the protocol of [55] followed by amplification of DNA using the nuclear ribosomal internal transcribed spacer 1 (ITS1) regions and the nuclear ribosomal internal transcribed spacer 2 (ITS2) regions (see below). Before analyzing the samples using ITS1 and ITS2, DNA from all the samples were screened for Symbiodinium clades using large subunit ribosomal DNA (lsur-DNA) restriction fragment length polymorphism (lsurDNA-RFLP) and further confirmed with small subunit ribosomal DNA-RFLP (ssurDNA-RFLP) (Fig. S1). RFLP analysis was carried out following the protocols of [42].

ITS1 and Single Strand Confirmation Polymorphism (SSCP)
The ITS1 region was amplified using the primer set with a fluorescent Hex-labelled forward primer 'SymITSFP (59-CTCAGCTCTGGACGTTGYGTTGG-39) and 'SymITSRP' . Single-stranded conformation polymorphism (SSCP) was performed following the protocol described in [56]. SSCP were preformed using Gelscan 3000 (Corbett Research). PCR products were first denatured at 95 u C for 3 min and on ice immediately for 3 min. Then PCR products were loaded onto the 4% nondenaturing TBE-polyacrylamide gel (1.5 ml of 40% acrylamide / bis-acrylamide (37.5:1), 11.445 ml of ddH 2 O, 0.3 ml of 100% glycerol, 0.9 ml 10x TBE, 30 ml of 25% APS and 30 ml of TEMED), pulsed into gel for 25s and flushed. This was followed by a run in 1200 V, 22 u C with 0.6x TBE buffer for 40 min.
The method used for assigning the ITS2-DGGE fingerprint followed [22]. Prominent bands of each fingerprint were sent for direct sequencing then matched with the sequences from Genbank (Table S1).

Direct sequencing
For samples of Acroporidae for which ITS2-DGGE did not successfully detect the presence of clade A (this might be due to the low copy number of clade A), the region between ITS1 (primer SymITSFP) to lsurDNA (primer D1/D2 R) was amplified, cloned and three clones for each sample were sequenced. The molecular cloning and sequencing methods were followed as per [55].

Results
Symbiodinium nomenclature depends on the method of molecular analysis used. For example, types C and Ca in SSCP is designated as C1 and C3 in DGGE. Since we have utilized both SSCP and DGGE for analysis, to avoid confusion, throughout the text we will follow the nomenclature used for DGGE (see Table 2 for more details).

PCR-DGGE of ITS2
Results of ITS2 analysis also showed Symbiodinium clade C: C1, C1c, C3, C3i, C3z, C3kk, C132a, C132 and C40 (Fig. 3B). Specific types of C found in Acroporidae were C1, C3, C3z and C40. In Pocillopora spp, DGGE analysis showed the presence of C1c (Fig. 3B) and C 2 , while in S. pistillata and S. hystrix this method revealed the presence of C1 together with an unknown type Cs. The results also showed the presence of two novel ITS2 types, C132a and C132 in S. hystrix (Fig. 3B) Although clade A was not detected by DGGE (this might be due to low copy number of clade A), results from SSCP, lsurDNA and ssurDNA-RFLP showed the presence of clade A in Acropora species. Analysis of ITS2 sequences further indicated that the Clade A in Isopora and Acropora from Chagos corresponds to Symbiodinium A1 from other regions in the Indian Ocean and the Pacific (Fig. S2) [22,46].

Discussion
Several studies in the past few years have advanced our knowledge on the biogeographic distribution of Symbiodnium diversity in corals and have contributed to our understanding of the coral-Symbiodinium association with respect to their specificity to regional environments [17][18][19][20][21][22]. Notably, [22] compared symbiosis patterns in the Indian Ocean (north-east and west Indian Ocean) with the Great Barrier Reef. This showed the occurrence of high incidences of clade D Symbiodinium in coral hosts in high temperature, turbid waters of the Andaman Sea (Thailand) compared with clade C in the Great Barrier Reef and Zanzibar. It was suggested that high temperature and turbidity might explain, in part, the ecological success of clade D in these areas [22].
Samples form the Chagos that we analyzed did not show any presence of clade D, even though the composition of Symbiodinium varied among reefs (Fig. 2). The absence of clade D in this survey could be because insufficient taxa were sampled. However, not one of the 262 specimens of seven species of major and abundant coral species in the Chagos showed the presence of Symbiodinium clade D. Our results indicate high incidence of Symbiodinium clade C types with occasional occurrence of Symbiodinium A1 in the samples analyzed, which is different form the previously investigated locations in the Indian Ocean.
The type of coral-Symbiodinium association in the Chagos observed through this study may be due to the fact that the central Indian Ocean atolls are bathed by water very low in dissolved organic matter (CDOM) with much less pelagic chlorophyll than the northeast Indian Ocean (Andaman Sea) and west Indian Ocean (Zanzibar) (Fig. 4 see also [22,58]). Water quality not only affects the distribution of corals but also the Symbiodinium composition [22,59]. Also, the mass mortality in 1998 in the Chagos archipelago was triggered as least as much by high irradiance as by warming of seawater; the trade winds that year did not develop [60] leading to prolonged calm sea surface conditions conducive to greater light penetration [61]. This is illustrated by many examples of coral colonies whose shaded sides survived while top surfaces exposed to light did not [62]. The bleaching in the central Indian Ocean in recent years appears to be as much a consequence of very stable water layers and thermoclines given extended periods of low winds, leading to enhanced light penetration, as it was to raised temperatures [61,63]. However, the average sea surface temperatures in the Chagos atolls are around 28uC, with the warmest being about 30uC [64], which is cooler than in the north-east Indian Ocean (Andaman Sea) where average sea surface temperatures exceed 31uC in summer (Fig. 4). All the above-described conditions can be conducive to the presence of Symbiodinum clade C and A in the corals of the Chagos Archipelago. But these correlations needs to be explored with caution since presence of corals and their association with a particular Symbiodnium types dose not necessarily correlate with their physiological adaptation to the prevailing environmental conditions in a given area.
The variety of C types present in I. palifera may be due to vertical transmission of Symbiodinium from parent corals [19]. A. muricata hosted type C1 and C3, C3kk (novel Symbiodinium type) together with Symbiodinium A1 (29% of total samples analyzed), otherwise it was mainly associated with C1. The Presence of Symbiodinium A1 in Acroporidae in the Chagos is consistent with previous reports of these associations at various locations in the Pacific [46]. Similarly, two Symbiodinium types C40 and C3z found in the Acroporidae have been previously recorded in the west Pacific Ocean, eastern Indian Ocean off the coast of Australia and at several locations in Thailand respectively [19,22,66]. This indicates that Acroporidae may be flexible in its association with different Symbiodinium clades, however, whether such flexibility is related to local environmental conditions in the Chagos is a matter of future investigations. Symbiodinium type Cs was associated with both S. hystrix and S. pistillata, whereas Csh, C132a, C132 were only found in S. hystrix (Fig. 3B). It has been observed that association of C1b-c in Pocillopora species is generally in the clear water reefs in the West Pacific Ocean [59]. This might explain the presence of type C1c (close related with C1b-c) in Pocillopora spp in the Chagos archipelago (Fig. 3B). C1c is a common Symbiodinium type that has been found in Pocilloridae in the Pacific [19]. However, we could not detect the presence of Symbiodinium type C1h that is associated with Pocillipora spp. in Zanzibar [22].

What makes Symbiodinium clade C dominant in the corals of the Chagos Archipelago?
Patterns of Symbiodinium associated with corals sampled from Chagos can be explained by the different environmental conditions For ITS2-DGGE analysis, samples from all representative Symbiodinium types are shown on the gel (B). Sample number is shown on each lane gel with markers denoted by the letter M on each gel. Symbiodinium type name is shown at the bottom of the lane. Acropora muricata -Lane 1-7; Isopora palifera -Lane 1-17 (inside red box); Pocillopora damicornis -Lane 1-4; Pocillopora verrucosa -Lane 5-7; Pocillopora eudouxi -Lane 8-10; Seriatopora hystrix -Lane 1 and 2 and Stylophora pistillata -Lane 1-5. The bands pointed with black arrows were cut and sequenced to confirm the Symbiodinium types. doi:10.1371/journal.pone.0035836.g003 that prevail in different parts of the Indian Ocean (see Fig. 4). This ocean lacks equatorial upwelling due to the climatological winds, which tend to be westerly rather than easterly. As a result of this, a warm pool is found in the Eastern Indian Ocean rather than in the west, as is the case in the tropical Atlantic and Pacific. This is attributed to the presence of large-scale, land-sea contrasts between the Asian land mass and the Indian Ocean, resulting in the monsoonal winds and seasonal reversing of currents north of ,20uS. Also lack of winds over the surface of the ocean observed in the central Indian Ocean results in less chlorophyll and CDOM (Fig. 4). The average sea surface temperature in the Chagos is usually ,30uC (Fig. 4, also see [67]) with temperatures exceeding 31uC on rare occasions, together with high PAR that results in coral bleaching [62,63]. This is conducive to the dominance of Symbiodinium clade C and presence of clade A to a certain extent in the Chagos corals. However, changing environmental conditions as a result of global climate change (see Fig. 4) shows an increasing trend in average sea surface temperature, which might also be working against such associations.

Conclusion
Time will reveal whether corals in the Chagos Archipelago adapt to the global climate change and increase in average sea surface temperatures. Presently there is no sign of high temperature resistant Symbiodinim clades (clade D) in the corals of the Chagos that we analyzed. The marine environment of the Chagos Archipelago is an exceptional place, which is a result of the absence of overfishing, pollution and minimal other human impacts. The reefs of the Chagos make up perhaps 50% of the total reef area in the Indian Ocean that remains in the least disturbed, low threat category. The view that reefs in the Chagos need strong protection is based not only on studies of the Chagos itself, but from knowledge and observation of the generally poor condition of reefs elsewhere in the Indian Ocean (from Madagascar to Kuwait, and from Sri Lanka westwards) and other oceans too. Now, since the Chagos Archipelago has been designated as a no-take MPA, it is suggested that more such investigations need to be undertaken to assess the overall situation of Symbiodinium diversity and distribution in corals of the Chagos, which would facilitate better understanding of the robustness of the corals in this archipelago and help in better management of this large marine protected area in the central Indian Ocean.

Future directions
Our study has shown that most common coral species in the Chagos archipelago are associated with Symbiodinium clade C and A. Whether such coral-Symbiodinium associations present in the Chagos are physiologically adapted or acclimated to environmental conditions prevailing in the Chagos Archipelago and environmental tolerances of these Symbiodinium types needs to be confirmed through laboratory based physiological experiments.