THE ROOTING OF THE TREE OF LIFE: STILL AN OPEN QUESTION

This is the abstract of Pr. Patrick Forterre's lecture at the 11th International Conference on the Origine of Life, July 7-12 Orléans, France

Here is a text version of the abstract:
Otherwise take a look at the graphic display of the abtract

THE ROOTING OF THE TREE OF LIFE: STILL AN OPEN QUESTION

Patrick Forterre Institut de Génétique et Microbiologie, Université Paris-Sud, Centre d'Orsay, 91405, Orsay Cedex, France, Fax: 33 169 4178 08, e-mail: forterre@igmors.u-psud.fr
Comparison of the molecular biology and central metabolism between Bacteria, Archaea and Eucarya is expected to help in reconstituting the last common ancestor to all extant cellular life, the cenancestor. Fulfilling this expectation requires a rooted universal tree of life, in order to determine if common traits shared specifically by two domains are primitive features, absent in the third one, or shared derived traits, testifying for a common history of two domains to the exclusion of the third one. Two studies, based on phylogenetic analyses of pairs of paralogous proteins (elongation factors, V and F-ATPases), have rooted the universal tree in the bacterial branch. This rooting has been widely acclaimed and is now systematically used in review papers and even textbooks. It fits well with the procaryotic dogma (eucaryotes emerged from procaryotes), it explains the finding of eucaryal features in Archaea (the two groups are in the same clade) and it supports the hot origin of life hypothesis.
However, I argued previously that ATPase phylogenies might have been confused unrecognized paralogy, whereas the bacterial rooting obtained with the elongation factor data set was not supported by a strict cladistic (character) analysis (Forterre et al., 1993a,b). More recently, the bacterial rooting has also been obtained using paralogous amino-acyl tRNA synthetases (Brown and Doolittle, 1995). The quality of this data set is much better than previous ones, i.e. more amino-acids can be aligned and the geometry of the tRNA synthetase tree is not very different from the rRNA tree. I have now performed a cladistic analysis of the tRNA synthetase data set. Surprisingly, this analysis do not place the root in the bacterial branch, but identify more phylogenetically valid positions supporting the eucaryotic rooting. This result supports alternative scenarios to the hot origin of life hypothesis, such as the origin of procaryotes via thermoreduction (Forterre, 1995). However, character analysis shows that the eucaryotic rooting deduced from the tRNA synthetase data set is not strong, since it supposes a high level of homoplasy. Accordingly, this data set do not seem to contain sufficient phylogenetic information to give a robust rooting. However, its analysis by the strict cladistic method gives interesting results. First, it indicates that Archaea have an intermediate position in-between Eucarya and Bacteria, in agreement with the geometry of the rRNA tree and with the finding of a mixture of bacterial and eucaryal features in Archaea. Secondly, it allows to understand why it is so difficult to root the universal tree of life, and in which direction future work should be done.

Brown, J.R. and Doolittle, W.F. 1995, Proc. Natl. Acad. Sci. USA 92, 2441.
Forterre, P. 1995, C. R. Acad Sci, Paris, 318, 415.
Forterre, P. et al. 1993, Nature, 362, 795.
Forterre, P., et al., 1993, Biosystem, 28,15.