Understanding diversity in coral-algal symbiosis: a cluster-based approach to interpreting fine-scale genetic variation in the genus Symbiodinium

Reef corals associate with an extraordinary diversity of dinoflagellate endosymbionts (genus Symbiodinium), and this diversity has become critical to understanding how corals respond to environmental changes. A popular molecular marker for Symbiodinium diversity, the Internal Transcribed Spacer-2 (ITS-2) region of ribosomal DNA, has revealed hundreds of distinct variants that are generally interpreted as representing different species, even though many have not been systematically tested for functional or ecological differentiation. Many of these variants are only minimally divergent from one another (1 bp or less), and others occupy basal nodes of traditional species phylogenies ("living ancestors"), indicating that some Symbiodinium ITS-2 diversity may represent intraspecific sequence variation. This hypothesis was tested for Symbiodinium clades A-D (the dominant symbionts of reef corals) through the construction of statistical parsimony networks of ITS-2 sequence diversity, and identification of clusters of closely related sequences within these networks. Initial assessments indicated that ecological differentiation exists between, but not within, these clusters. This approach, although imperfect in its ability to identify species boundaries in all cases, nevertheless dramatically reduces "species" diversity in Symbiodinium (from similar to 175 to 35). This testable alternative hypothesis indicates that, in Symbiodinium, "species" consist of clusters of closely related ITS-2 sequences diverging from ancestral variants that are typically ecologically dominant. A cluster-based view of Symbiodinium ITS-2 diversity improves our ability to: (1) construct well-supported symbiont phylogenies; (2) establish functional niches for symbiont species; and (3) understand flexibility and specificity within coral-algal symbioses. This cluster-based approach can ultimately be integrated with emerging population-level datasets (microsatellites and microsatellite flanking regions) to improve understanding of species diversity in Symbiodinium.