I heard a provocative seminar by Peter Gogarten of U. Conn today. His work is funded by the NASA Exobiology program and by NSF’s Tree of Life program. The title of his seminar was “Phylogenetics in Light of Horizontal Gene Transfer”. His main thesis is that horizontal gene transfers, regarded by most evolutionary biologists as a tremendous hindrance to deducing phylogeny, can instead provide clear signals that assist in phylogenetic reconstruction. However, one must first abandon the idea that evolution of taxa can best be represented as a bifurcating tree. Rather, as proposed by Darwin himself, a “coral of life” metaphor is more apt, because the base of the branches of life consists of dead lineages. Moreover, horizontal gene transfers cause reticulation rather than a straight bifurcating tree.
For me, with an interest in evolution of plants, the most interesting part of his talk dealt with dozens of genes found in red algae and green plants that appear to be of chlamydial origin. Most of these proteins appear to function in the plastid, many as transporters. Surprisingly, the cyanobacterial homologues of these genes are relatively distant on molecular phylogenetic trees. How can we account for these genes of apparent chlamydial origin, rather than the expected cyanobacterial origin, especially since chlamydia are not known to infect extant plants or red algae?
One hypothesis could be that plants acquired these genes from chlamydia associated with insect feeders. However, these horizontal gene transfers must have occurred in the common ancestor of red algae and green plants, which died out long before insects evolved. The possibility that gene transfer took place from plants to chlamydia is ruled out because these genes in question clearly resemble bacterial genes, not eukaryotic genes. A third hypothesis is that these genes resemble chlamydial genes because cyanobacteria and chlamydia have a common ancestry that has been obscured. However, this fails to explain why these genes in red algae and green plants are more similar to chlamydial genes than cyanobacterial genes, if they came from cyanobacteria.
Peter Gogarten proposes a fourth hypothesis, that these genes indicate an ancient parasitic or symbiotic relationship between chlamydia and the common ancestor of red algae and green plants. This common ancestor would have had both chlamydial and cyanobacterial residents in the cytoplasm. Over millions of years, horizontal gene transfer took place from chlamydia to the nucleus of the host cell, and those genes that could be redirected to the cyanobacterial endosymbiont to provide transport functions and improve efficiency of the photosynthetic symbiosis were retained. The chlamydial symbiont was subsequently lost. Gogarten wonders if the tripartitite gene swapping helps explain why successful acquisition of endosymbionts is rare – it requires assistance from a third party.
Huang and Gogarten, 2007. Did an ancient chlamydial endosymbiosis facilitate the establishment of primary plastids? Genome Biol. 8:R99 doi: 10.1186/gb-2007-8-6-r99
Huang and Gogarten, 2008. Concerted gene recruitment in early plant evolution. Genome Biol. 9:R109. doi: 10.1186/gb-2008-9-7-r109