Of archae and eo: what's in a name?

Abstract

On November 3, 1977, the Los Angeles Times carried a front page story reporting that a life form had been discovered. The report noted that the form arose earlier than the two previously known forms and was most like the common ancestor of all life on earth. What the article was describing was the finding by Carl Woese and his co-workers that a group of organisms, previously thought of as bacteria, actually formed a distinctly separate group when considered on the basis of their ribosomal RNA sequences (1). Woese called these organisms the archaebacteria (archae = primitive), in line with the notion that they resembled the most ancient form of life on earth. It was apparent, however, that although the phylogenetic trees constructed with ribosomal RNA clearly showed three distinct clusters-corresponding to the eubacteria, the archaebacteria, and the eukaryotes-the resemblance of any of them to a more ancestor could only be supposed and not rooted in any firm sense. In fact, the three groups appeared to be equidistant from each other (Fig. 1A). Earlier that same year, Woese and Fox (2) had coined the term to describe a hypothetical ancestor for prokaryotes and eukaryotes. Because there was microfossil evidence that suggested bacteria have remained virtually unchanged, in a morphological sense, over the course of 3.5 billion years, the progenote must have been very ancient. It was described as a cell with a rudimentary translation apparatus, the loosely assembled genome of which may have been wholly RNA-based (3). Unraveling the true evolutionary relationships of the three urkingdoms (ur = primary) and their earlier ancestral stock has proved vexingly difficult, however, and widely different scenarios have been postulated since the initial report (Fig. 1). Now in this issue of the Proceedings, James Brown and W. Ford Doolittle present findings that go a long way toward resolving the arguments (4). A review of some of the conflicting observations that has beset this field may help explain why their study is so important. The concept of three widely divergent urkingdoms was sustained as more ribosomal RNA sequences were reported. Initially, the new kingdom had comprised only methanogens, but as exploration continued, it was found to embrace halophilic and extremely thermophilic bacteria as well. As such, the group seemed to be composed of organisms whose lifestyles seemed in harmony with the presumed harsh conditions of an early earth. The notion of three ancient lineages was further confirmed by electron microscopy of ribosomes, each of the three kingdoms having recognizably different features (5). At least one characteristic prominence was common to the archaebacteria and eukaryotes but absent from eubacteria; nonetheless, these authors, sensitive to the fashionable rules of strict parsimony, demurred from concluding that eukaryotes evolved from archaebacteria, cautiously proposing instead that all three lineages sprang from a hypothetical ancestor called the paleocyte (ancient + cell). At about the same time, comparisons of 5S RNA sequences also suggested that the archaebacteria were more similar to eukaryotes, leading to the suggestion that the group would be better named metabacteria (6), a reckoning more in line with Fig. 1D. Not long thereafter, the electron microscopy group claimed even greater morphological resolution, the new details revealing that one subgroup of archaebacteria, hyperthermophiles like Sulfolobus, had ribosomes that were distinctive from other archaebacteria and significantly more similar to ribosomes from eukaryotes (7). Once again, adhering to the rules of strict parsimony, they concluded that what this group and eukaryotes had in common were primitive traits and that the two groups were older than, or at least as old as, other archaebacteria and eubacteria. They suggested the name eocyte (dawn + cell) for the subgroup, which they felt should be elevated to kingdom status. Subsequently, several newly determined protein sequences seemed to be at odds with a primordial archaebacteria interpretation, the archaebacterial sequences more often than not appearing more similar to eukaryotic homologues than to those from the eubacteria. The bulk of the evidence seemed to be shifting to a phylogeny like that described in Fig. 1D. The matter became increasingly disputatious, with arguments and counterarguments about the methodology of sequence comparison and phylogenetic tree construction often becoming heated. The usual problems of changing rates and arbitrary rooting were argued incessantly. Then, in 1989, two groups appeared to offer insurmountable evidence that the archaebacteria were really more akin to eukaryotes than eubacteria (Fig. 1D). Iwabe et al. (8) and, independently, Gogarten et al. (9) used the stratagem of