There seems to be general agreement that some of the dendrograms constructed
from sequence data are difficult to rationalize. If one is a believer in
the Three Domain hypothesis then rationalizations tend to exploit some
sort of horizontal gene transfer to explain odd dendrograms. However, it
is possible that the data refutes the Three Domain hypothesis and that
such rationalizations are not justified.
J.Peter Gogarten writes,
"I think that one of the main problems in understanding the early
evolution is the paradigm of a tree (Hilario and Gogarten, 1993).
The data reported by Zillig et al. (1992) and Gupta and Golding
(1993) suggest that not only the archaebacteria but also the
eubacteria made an early (i.e. before the mitochondria) and
significant contribution to the eukaryotic cell. There are many
potential cases for horizontal gene transfer (reviewed by Smith
and Doolittle, 1993); however, not only single genes and operons
were transferred between species, it is likely that the merger of
formerly independent lineages to form a new species is not restricted
to higher plants and the known cases of endosymbiosis, but the merger
of independent lineages occurred (and occurs) throughout the
evolution."
"The eukaryotes almost certainly received major contributions from
Archae (Gogarten et al. 1989, Iwabe et al. 1989) and Eubacteria
(Zillig et al. 1992, Gupta et al.1993) In addition, the glutamine
synthetase (Kumada et al. 1993, Tiboni et al. 1993) and the "HSP70"
data suggest that also the Archaebacteria (or at least the mesophilic
Archaebacteria) received genes from gram positive eubacteria). In
order to confirm that these horizontal transfers do not represent
separate events of horizontal gene transfer but the merger of
independent lineages more sufficiently resolved molecular phylogenies
would be needed."
I have a serious problem with some of the literature that discusses hori-
zontal gene transfers. For example, take the case of the glutamine synthetase
genes. In their recent review, Smith et al. (1992) point out that dendrograms
of the type II genes do not require any extraordinary explanations since
the bacterial sequences separate from the eukaryotic sequences in their
analysis. (There has been debate over whether bacteria have acquired a
eukaryotic gene.) However, they note that dendrograms of the type I genes
show some "anomalies". The two archaebacterial sequences cluster within
the eubacteria - Methanobacter is close to the gram positive bacteria and
Sulfolobus clusters with gram negatives. Now, those who have been following
this discussion will realize that one possible explanation is that the
archaebacteria do not form a separate monophyletic group (domain). But
Smith, Feng and Doolittle (1992) propose that the tree "...clearly suggest
one or more horizontal transfers". My point is that horizontal gene transfers
are offered as an explanation for "anomalies" but perhaps the so-called
"anomalies" are real and it is our understanding of deep phylogeny that is
wrong.
You referred to two other papers that analyze the glutamine synthetase
genes. In Kumada et al. (1993) their dendrogram clearly shows Methanococcus
clustered with gram positive bacteria (as in Smith et al. 1992). However,
Kumada et al. do not address this "anomaly". Instead they concentrate
on showing that a previous suggestion of horizontal transfer between plants
and bacteria is NOT supported by their data.
The paper by Tiboni et al. (1993) is very interesting. These workers examined
the glutamine synthetase gene from the archaebacterium Pyrococcus woesei.
The dendrograms show, in agreement with the results quoted above, that
Sulfolobus clusters with the gram negative bacteria and Methanococcus with
the gram positives. Pyrococcus seems to fall somewhere between the two
groups but closer to the gram positives. No two of the three archaebacterial
sequences map together and there is no evidence that any of the archaebacteria
fall outside the eubacterial cluster. The obvious conclusion is that the
Three Domain hypothesis is not in any way supported but these workers assume
that archaebacteria and eubacteria should be separate domains and they propose
a horizontal gene transfer to explain their data. In the discussion they quote
all of the results that support the Three Domain hypothesis but none of the
contrary evidence. Strange.
Gupta and Golding (1993) fall into the same trap. Rather than accept the
fact that their HSP70 trees do not show the archaebacteria as a separate
domain, they postulate horizontal gene transfers to "explain" their data.
Why not just state that the Three Domain hypothesis is in trouble - maybe
it is not correct?
I don't think that either the glutamine synthetase or the HSP70 data offer
any support for horizontal gene transfer. Maybe it is the EF-Tu and the
ATPase genes that were transferred from eukaryotes to archaebacteria? (-:
I'm not arguing against any horizontal gene transfers. There seem to be a
few good examples of transfer between eukaryotes and bacteria. This is in
addition to the transfer of chloroplast and mitochondrial genes to the nuclear
genome, which is well established. What I am disputing is the use of gene
transfer rationales to "explain" why data doesn't fit the Three Domain
model. (BTW, I still haven't read Hilario and Gogarten (1993) because I
can't find the journal. Maybe everything is explained in that paper.)
Laurence A. Moran (Larry)
Gupta, R.H., and Golding, G.B. (1993) Evolution of HSP70 gene and its
implications regarding relationships between archaebacteria, eubacteria and
eukaryotes. J. Mol. Evol. 37, 573-582.
Kumada, Y., Benson, D.R., Hillemann, D., Hosted, T.J., Rochford, D.A.,
Thompson, C.J., Wohlleben, W., Tateno, Y. (1993) Evolution of the glutamine
synthase gene, one of the oldest existing and functioning genes.
Proc. Natl. Acad. Sci. (USA) 90, 3009-3013.
Tiboni, O., Cammarano, P., Sanangelantoni, A.M. (1993) Cloning and sequencing
of the gene encoding glutamine synthase I from the archaeum Pyrococcus
woesei: Anomalous phylogenies inferred from analysis of archaeal and
bacterial glutamine synthase I sequences. J. Bacteriol. 175, 2961-2969.
Smith, M. W., Feng, D-F., and Doolittle, R.F. (1992) Evolution by acquisition:
the case for horizontal gene transfers. TIBS 17, 489-493.
