Fig.1. Tree of life based on 16-18S rRNA sequences.
Click here for a larger image of Fig.1 (5 kb)
Fred R. Opperdoes
Comparative analyses of rRNA sequences, initiated in the 1970s, suggest that the living world is divided into three domains: Eucarya, Archaea (formerly archaebacteria), and Bacteria (formerly eubacteria). In both 16S and 23S rRNA unrooted trees, the length of the archaeal branch (i.e., the branch connecting the last common ancestor of all archaea to the point of trifurcation) is strikingly shorter than the bacterial and eucaryal branches (see Fig. 1 below).
Fig.1. Tree of life based on 16-18S rRNA sequences.
Click here for a larger image of Fig.1 (5
kb)
Thus, Archaea may have more similarity to Bacteria or Eucarya than both of them have to each other, in good agreement with the finding that archaea exhibit a mixture of eucaryal and bacterial traits at the molecular level. This kind of tree often has been called "universal tree of life". According to this tree the earliest eukaryotic cells were Archaezoa that are amitochondriate organisms that were adapted to an anaerobic environment such as the extant Diplomonads (Giardia), Parabasalia (trichomonads) and the Microsporidia. Mitochondria then were acquired at a later stage from a bacterial endosymbiont belonging to the group of the alpha-proteobacteria. Also according to this tree the Kinetoplastida, comprising trypanosomatids, bodonids and euglenoids, were the first group to have acquired a mitochondrion by endosymbiosis.
However recent data using phylogenetic methods based on protein-coding genes have shown that classical methods of molecular phylogeny using rRNA genes may fail to delineate phylogenetic relationships between domains or between major lineages of these domains.
During the last five years, investigators have analysed an increasing number of protein sequences from Archaea, Bacteria and Eukarya, furnishing many more representative examples from each of the domains. At first, these new trees were expected to support the idea of amitochondriate Archaezoa to have preceded all mitochondriate eukaryotes. However, much of the new information does not support this idea. As data accumulate, a drastically different pictures emerges.
For example, while 18S rRNA analyses have suggested that Microsporidia are one of the most ancient eukaryotic lineages, recent analyses of their proteins performed by Keeling and Doolittle in Halifax (Canada) and by Hasegawa in Tokyo (Institute of Statistical Mathematics) indicate that they are close relatives of fungi. Similarly, Hervé Philippe of Orsay University in France noticed that two new bacterial ileu-tRNA synthetase sequences branch between eukaryotes and archaea, producing an additional "confused tree." He suggests that trees that do not yield topology of the three domains are "normal," considering that protein sequences are saturated, and that independent structural or functional changes in each domain are required for a tree to yield the three-domain topology.
Studies based on protein sequences of heatshock (HSP70) proteins and chaperonins have provided evidence that, firstly, the amitochondriate organisms all have had a mitochondrion which may have been lost secondarily, or that was converted from a mitochondrion to an anaerobic hydrogenosome as is the case in the trichomonads, belonging to the Parabasalia. In trichomonads, Giardia, Entamoeba and the Microsporidia, an HSP70 has been detected, which clusters with the mitochondrial HSP70s from other eukaryotes. Secondly, the microsporidian HSP70 clusters with the HSP70s of Fungi, and such an affiliation which is also supported by a phylogenetic analysis based on other protein sequences (see above). Similar analyses using protein-coding genes do not support the long branch of primitive eukaryotic evolution with many protists separating in a paraphyletic manner as shown in the rRNA tree (Fig.1), but suggest a massive and rapid radiation of protists, algae, fungi and animals almost simultaneously.
Apparently all amitochondriate protists have at one stage had mitochondria and lost them secondarily as an adaptation to an anaerobic environment. At present no known organisms exist that never have had a mitochondrion. This may suggest that the formation of the early eukaryotic cell was the result of the acquisition of the mitochondrion, although it cannot be excluded that truly amitochondriate eukaryotes have existed, but that no extant representative exist at present.
Clearly, one of the problems associated with the universal tree of life, as proposed by Sogin, is that the trunk of early eukaryotic evolution as well as many of the protist branches are extremely long. This indicates that rapid evolution must have taken place in the rRNA probably due to the increase in size of the eukaryotic ribosomes from 70S to 80S and the simultaneous increase in size of the 16S small subunit rRNA to its 18S homologue. This may lead to the long branch attraction phenomenon, placing long branches preferentially towards the bottom of the tree.
The above section has illustrated some of the problems related to the use of rRNA sequences. Although rRNAs have been extremely useful in unravelling the phylogenetic relationships of organisms, certain problems cannot be solved. A valuable alternative would be the use of protein-coding sequences for the construction of phylogenetic trees and the above section has already illustrated that interesting answers can be obtained from proteins as well.
In the following sections a number of arguments are being developed as to why it may be advantageous to use protein-coding sequences rather than nucleic acid sequences for phylogenetic analyses.
Arguments in favour of protein rather
than the DNA