Gorgonia mariae and Antillogorgia bipinnata populations inferred from compensatory base change analysis of the internal transcribed spacer 2

Revisiting the phylogeny of Zoanthidea (Cnidaria: Anthozoa): Staggered alignment of hypervariable sequences improves species tree inference

The recent rapid proliferation of novel taxon identification in the Zoanthidea has been accompanied by a parallel propagation of gene trees as a tool of species discovery, but not a corresponding increase in our understanding of phylogeny. This disparity is caused by the trade-off between the capabilities of automated DNA sequence alignment and data content of genes applied to phylogenetic inference in this group. Conserved genes or segments are easily aligned across the order, but produce poorly resolved trees; hypervariable genes or segments contain the evolutionary signal necessary for resolution and robust support, but sequence alignment is daunting. Staggered alignments are a form of phylogeny-informed sequence alignment composed of a mosaic of local and universal regions that allow phylogenetic inference to be applied to all nucleotides from both hypervariable and conserved gene segments. Comparisons between species tree phylogenies inferred from all data (staggered alignment) and hypervariable-excluded data (standard alignment) demonstrate improved confidence and greater topological agreement with other sources of data for the complete-data tree. This novel phylogeny is the most comprehensive to date (in terms of taxa and data) and can serve as an expandable tool for evolutionary hypothesis testing in the Zoanthidea. Spanish language abstract available in Text S1. Translation by L. O. Swain, DePaul University, Chicago, Illinois, 60604, USA.

Genetic diversity, paraphyly and incomplete lineage sorting of mtDNA, ITS2 and microsatellite flanking region in closely related Heliopora species (Octocorallia)

Examining genetic diversity and lineage sorting of different genes in closely related species provide useful information for phylogenetic analyses and ultimately for understanding the origins of biodiversity. In this study, we examined inter- and intraspecific genetic variation in internal transcribed spacer 2 (ITS2), partial mitochondrial gene (mtMutS), and nuclear microsatellite flanking region in two closely related octocoral species (Heliopora coerulea, HC-A and HC-B). These species were recently identified in a population genetic study using microsatellite markers. The two species have different reproductive timing, which ecologically promotes lineage sorting. In this study, we examined whether species boundaries could be detected by the commonly used nuclear ITS2 and mtMutS, as well as by possibly neutral microsatellite flanking sequences. Haplotype network analysis of microsatellite flanking region revealed that a possible ancestral haplotype was still shared between the two species, indicating on-going lineage sorting. Haplotype network analysis of ITS2 and microsatellite flanking region revealed shared haplotypes between the two lineages. The two species shared fewer ITS2 sequences than microsatellite flanking region sequences. The almost fixed point mutation at the tip of helix 3 of ITS2 was not associated with the secondary structure or compensatory base changes (CBCs). The phylogenetic tree of ITS2 showed paraphyly and that of the microsatellite flanking region indicated that lineage sorting for the two species may be incomplete. Much higher intra- and inter-individual variation of ITS2 was observed in HC-B than that in HC-A, highlighting the importance of examining ITS2 from multiple individuals to estimate genetic diversity. The mitochondrial mtMutS gene sequences from 39 individuals, including both species collected from Japan and Taiwan, showed no variation because of slow rates of mitochondrial nucleotide substitution. This study suggests caution is warranted when reciprocal monophyly in a phylogenetic tree is used as the criterion for delimiting closely related octocoral species based on ITS2 or mtMtuS sequences. Detection of boundaries between closely related species requires multi-locus analysis, such as genetic admixture analysis using multiple individuals.