Mixed heterolobosean and novel gregarine lineage genes from culture ATCC 50646: Long-branch artefacts, not lateral gene transfer, distort alpha-tubulin phylogeny

作者：Cavalier Smith Thomas^*

来源：European Journal of Protistology, 2015, 51(2): 121-137.

DOI：10.1016/j.ejop.2014.12.001

• Actinocephaloidea
• alpha-tubulin phylogeny
• Hsp90 phylogeny
• 18S rDNA phylogeny
• Percolozoa
• Terrestrial gregarines

Contradictory and confusing results can arise if sequenced 'monoprotisf samples really contain DNA of very different species. Eukaryote-wide phylogenetic analyses using five genes from the amoebofiagellate culture ATCC 50646 previously implied it was an undescribed percolozoan related to percolatean flagellates (Stephanopogon, Percolomonas). Contrastingly, three phylogenetic analyses of 18S rRNA alone, did not place it within Percolozoa, but as an isolated deep-branching excavate. I resolve that contradiction by sequence phylogenies for all five genes individually, using up to 652 taxa. Its 18S rRNA sequence (GQ377652) is near-identical to one from stained-glass windows, somewhat more distant from one from cooling-tower water, all three related to terrestrial actinocephalid gregarines Hoplorhynchus and Pyxinia. All four protein-gene sequences (Hsp90; alpha-tubulin; beta-tubulin; actin) are from an amoebofiagellate heterolobosean percolozoan, not especially deeply branching. Contrary to previous conclusions from trees combining protein and rRNA sequences or rDNA trees including Eozoa only, this culture does not represent a major novel deep-branching eukaryote lineage distinct from Heterolobosea, and thus lacks special significance for deep eukaryote phylogeny, though the rDNA sequence is important for gregarine phylogeny. a-tubulin trees for over 250 eukaryotes refute earlier suggestions of lateral gene transfer within eukaryotes, being largely congruent with morphology and other gene trees.

• 出版日期2015-4