Defining the core of nontransferable prokaryotic genes: The euryarchaeal core

If lateral gene transfer (LGT) has affected all genes over the course of pr okaryotic evolution, reconstruction of organismal phylogeny is compromised. However, if a core of genes is immune to transfer, then the evolutionary h istory of that core might be our most reliable guide to the evolution of or ganisms. Such a core should be preferentially included in the subset of gen es shared by all organisms, but where universally conserved genes have been analyzed, there is too little phylogenetic signal to allow determination o f whether or not they indeed have the same history (Hansmann and Martin 200 0; Teichmann and Mitchison 1999). Here we look at a more restricted set, 52 1 homologous genes (COGs) simultaneously present in four sequenced euryarch aeal genomes. Although there is overall little robust phylogenetic signal i n this data set, there is, among well-supported trees, strong representatio n of all three possible four-taxon topologies. "Informational" genes seem n o less subject to LGT than are "operational genes," within the euryarchaeot es. We conclude that (i) even in this collection of conserved genes there h as been extensive LGT (orthologous gene replacement) and (ii) the notion th at there is a core of nontransferable genes (the core hypothesis") has not been proven and may be unprovable.