Molecular diversity of coral reef-associated zoanthids off Qeshm Island, northern Persian Gulf

The Persian Gulf, a semi-enclosed sea in the subtropical northwest of the Indian Ocean, is noted for its unique biodiversity under its extreme ecological conditions. Despite high biodiversity levels, many groups of marine invertebrates in this area have remained uninvestigated. The order Zoantharia (zoanthids) is one of these taxonomically neglected groups. In this study, diversity of shallow water zoanthids off the Qeshm Island, the largest island in the Persian Gulf, was investigated for the first time. Using in situ field examination integrated with 16S rDNA sequencing and phylogenetic analysis, the presence of three zoanthid species in the inter-tidal and shallow water zone of Qeshm Island were demonstrated: Zoanthus sansibaricus (n = 12) with five morphotypes, Palythoa cf. mutuki (n = 10) with two morphotypes and Palythoa tuberculosa (n = 4) with just one morphotype. In addition to species identification, molecular examination determined phylogenetic relationships of specimens with other previously reported zoanthid species. While Zoanthus sansibaricus and Palythoa tuberculosa are two known zoanthid species, based on molecular data, Palythoa cf. mutuki is potentially a novel undescribed species. However, due to lack of data on zoanthid research and distribution for the entire Persian Gulf, further investigation is needed to clearly ascertain this matter.


Introduction
The order Zoantharia (zoanthid) is a group of benthic colonial anthozoans which belongs to the subclass Hexacorallia. They are characterized by having two rows of tentacles and one ventral siphonoglyph (Haddon and Shackleton 1891). These cnidarians are found in many marine ecosystems and are particularly common in coral reef ecosystems worldwide (Burnett et al. 1994(Burnett et al. , 1995Reimer et al. 2011). Although zoanthid genera, Zoanthus and Palythoa are common in shallow subtropical and tropical waters, their taxonomy and identification to species level was historically confusing due to the intraspecific variation in polyp shape, size, color, oral disk color (Burnett et al. 1997;Reimer et al. 2004) and the absence of exact morphological description of species in the past literature. In addition, most zoanthids are known for being encrusted by sand and other detritus to enhance their structure, which make their histological sectioning very difficult .
Recently, taking advantage of molecular phylogenetic methods, the reorganization process of the classification of this taxonomically neglected group has begun (Reimer and Fujii 2010;Sinniger et al. 2010). The mitochondrial markers have been successfully utilized in examining zoanthid taxa (Sinniger et al. 2010). Applying mt16S rDNA as a DNA marker has aided in identification and reorganization of zoanthids species, genera and families (Reimer et al. 2006a(Reimer et al. , 2006b(Reimer et al. , 2006cReimer and Todd 2009;Sinniger and Haussermann 2009;Sinniger et al. 2010;Swain 2009).
Coral reefs in the Persian Gulf are part of the complex and unique inter-tidal and sub-tidal habitats. Due to isolation from the open ocean, the Persian Gulf experiences extreme environmental conditions. It is one of the warmest and the most saline waters on earth and this has imposed a harsh condition on the marine organisms (Coles and Fadlallah 1991;Reynolds 1993) and as a result, a particular and unique marine fauna has evolved in this environment. In the northern Persian Gulf, there are 17 islands off the Iranian coastline with fringing coral reefs (Sheppard and Sheppard 1991). Although coral reef communities have been described at most of the Persian Gulf Islands (Fatemi and Shokri 2001;Namin and Van Ofwegen 2009;Rezai et al. 2010;Kavousi et al. 2011;Mostafavi et al. 2007;Samiei et al. 2013), there are no data on zoanthids in this area.
The purpose of this study was morphological observation and molecular identification of shallow water zoanthids off Qeshm Island. Upon collection, zoanthid specimens were grouped based on their external appearance. Then all the grouped species were molecularly identified. In this paper chromatic variability of all the collected specimens were provided and the accuracy of initial morphological identification was discussed.

Sample collection and preliminary identification
Twenty-six colonies of zoanthids were photographed and collected between December 2011 and March 2013 by wading or SCUBA diving in the inter-tidal zones of seven locations off Qeshm Island (Fig. 1).
Depth of sampling is shown in Table 1. Specimens were stored in 100 % ethanol at -20°C. Initial identification was carried out based on in situ photographs ( Fig. 2) with the support of published literature (Burnett et al. 1997;Ryland and Lancaster 2003;Reimer 2007;Reimer et al. 2006a).

DNA extraction, PCR amplification and sequencing
Zoanthid fragments were crushed in DNAB (0.4 M NaCl, 50 mM EDTA, pH 8.0) buffer and then DNA was extracted using the CTAB-chloroform method (Baker 1999). Mitochondrial 16S rDNA was PCR amplified using zoanthid-specific primers 16Sant1a and 16SbmoH (Sinniger et al. 2005) with the following thermal cycle conditions: 2 min at 94°C then 35 cycles: 30 s at 94°C, 1 min at 52°C, 90 s at 72°C, followed by 7 min final extension at 72°C. The amplified products were analyzed by 1.5 % agarose gel electrophoresis. Original PCR products from zoanthids specimens were sent to the Macrogen Company in South Korea and directly sequenced with ABI-3730XL analyzer by the capillary system method.

Phylogenetic analysis
Sequences obtained from this study were deposited in GenBank and their accession numbers are shown in Table 1. The nucleotide sequences obtained in this study were aligned with 16S rDNA sequences available from Genbank (accession numbers are shown in Fig. 3) using the software CLUSTALW (Thompson et al. 1994). The outgroup sequence for 16S rDNA tree was Parazoanthus gracilis.
The alignment dataset was analyzed using maximum likelihood (ML), maximum parsimony (MP) and Bayesian methods. The most appropriate model selection for ML and Bayesian analyses was performed using Akaike Information Criterion (AIC) in MODELTEST 2.3 (Nylander 2004). The general time-reversible model (Rodrigues et al. 1990) with invariable sites (GTR ? I) gave the best fit to the 16S rDNA data. ML and MP analyses were conducted using the PAUP beta version 4.0b10 (Swofford 2003). For ML, a heuristic search was carried out with 100 random additions of taxa, followed by tree-bisection-reconnection (TBR) branchswapping rearrangements with a maximum of 100 optimal trees kept in each replicate. Maximum likelihood clade support was assessed by nonparametric bootstrapping with 1,000 replicates and the same heuristicsearch parameters. The MP analysis, also carried out by the heuristic-search method, consisted of 100 runs of stepwise random taxon additions. MP clades were assessed with 1,000 bootstrap replicates (excluding uninformative characters), with 100 random additions of taxa for each replicate.
The Bayesian analysis was implemented in MrBayes 2.3 (Ronquist and Huelsenbeck 2003) and was based on the model selected by MODELTEST above. Starting from random trees, four Markov chains (with one cold and three heated chains) were run simultaneously to sample trees using the Markov Chain Monte Carlo (MCMC) principle which approximates the posterior probability (PP) of trees. After the burn-in phase (the first 5 million generations was discarded), every 100th tree out of 20 6 was considered. The phylogenetic trees generated in all analyses were visualized using TREEVIEW (Page 1996).

Morphological preliminary observation
The 26 collected zoanthid specimens were first divided into three morphological groups. Fourteen specimens were identified as belonging to the genus Palythoa due to their heavy sand encrustation, external brown or In situ images of zoanthid specimens collected in this study. Each image is representative of one morphological group listed in Table 2 cream coloration (Table 2; Fig. 2) and the tentacle numbers which were almost fit with the numbers reported from Palythoa species of Indian and Pacific Ocean (Herberts 1972;Ryland and Lancaster 2003). Of these, four specimens (QeSu2, QeSu6, QeAu3 and QeAu6) were distinguished as Palythoa tuberculosa by their densely packed immersae polyps, ranging from almost light to dark brown. The other ten specimens were identified as Palythoa mutuki based on their elongated, cylindrical polyps with underdeveloped coenchyme (liberae polyps) and green or brown oral disk coloration. These polyps were joined basally, but their closeness and flaring columns obscured the bases (Fig. 2). The remaining morphological group (Table 2) contained 12 specimens which were clearly from the genus Zoanthus, as they had smooth, non-encrusted, liberae polyps. Since during the present study only chromatic variability of each specimen was monitored, species-level assignment was mostly impossible and they were just identifiable at the genus level. All morphological characters of each specimen (sand encrustation, polyp shape and color, oral groove, oral zone and oral disk colors, tentacle number and color) are summarized in Table 2.

Utility of external morphological characteristics and molecular identification of zoanthids
In comparing morphological versus molecular identification results, we demonstrated that zoanthid identification based on the 16S ribosomal mitochondrial DNA marker is more accurate than using morphological characteristics. This is especially so when species with polymorphic characteristics are dealt with that is not consistently reliable.
During the preliminary identification process, all specimens were identified to the generic level using only external morphological criteria, but their species-level assignment were not possible or correct in most cases. The identification of Palythoa mutuki specimens remained somewhat ambiguous. Although morphologically, these specimens perfectly agreed with Palythoa mutuki, the mt 16S rDNA data presented unexpected two or three base pair of differences with Palythoa mutuki. While the mitochondrial markers are very conservative in anthozoans (Shearer et al. 2002), this low level of variation has previously indicated species-level differences in Palythoa species. For example, Palythoa mutuki and Palythoa tuberculosa can be distinguished by one base pair of difference over mt 16S rDNA data set. Therefore, with examination of only one marker, for now we designated these specimens as Palythoa cf. mutuki.
As mt 16S rDNA are considered to be one of the mitochondrial markers which are able to correctly place zoanthid specimens, especially for Zoanthidae and Spheopidae families, within a species group Reimer and Todd 2009;Swain 2009), applying this marker in the present study categorized these specimens into three species-level groups with a high level of confidence.

Zoanthid diversity in the Persian Gulf
All specimens during the present study have been categorized into three different species groups: Zoanthus sansibaricus from Zoanthidae family, Palythoa tuberculosa and Palythoa cf. mutuki from Sphenopidae family. Of these three species, two are identified species which have been widely distributed in the western and eastern Pacific Ocean; i.e., Great Barrier Reef (Burnett et al. 1997;Ryland and Lancaster 2003), New Caledonia (Sinniger 2006), Japan (Reimer et al. 2004(Reimer et al. , 2006aReimer 2007), Galapagos , Singapore (Reimer and Todd 2009) and Indian Ocean (Herberts 1972), while the remaining species, Palythoa cf. mutuki did not molecularly match with any of the previously reported sequences.
As mentioned earlier in this study, there is an overall lack of data on zoanthids from the Persian Gulf. Thus, it is somewhat difficult to assess if the Palythoa cf. mutuki is an endemic species, that was previously undescribed, or range extensions of species from other regions.
The Gulf is linked to the open ocean by the narrow Strait of Hormuz which limits water exchanges (Wilson et al. 2002). The three zoanthid species, Zoanthus sansibaricus, Palythoa tuberculosa and Palythoa mutuki seem to be very widespread in the Indo-Pacific, and they have been found in isolated locations like Ogasawara  and Galapagos  Islands. It has been demonstrated that Palythoa tuberculosa has a larval stage of over 2 weeks (Polak et al. 2011;Hirose et al. 2011). However, there are few data on the length of time of larval stages for zoanthid species in this study aside from Palythoa tuberculosa. We hypothesized that many of the zoanthids that are found in the Persian Gulf may have extended larval stages which have been reached and colonized as the first species in the relatively isolated Persian Gulf from the wider Indian Ocean.

Conclusion
This study demonstrated the presence of at least two families, two genera and three species of zoanthids in the Persian Gulf. Furthermore, based on these observations, the presence of a potential new species was demonstrated. Most importantly, the results of this study indicated a need for further sampling and investigation of zoanthid from more regions to help complete the knowledge of zoanthid diversity in this area.
Open Access This article is distributed under the terms of the Creative Commons Attribution License which permits any use, distribution, and reproduction in any medium, provided the original author(s) and the source are credited.