May 18, 2001
Link Between Human Genes and Bacteria Is Hotly Debated
By NICHOLAS WADE
n a fresh skirmish in the genome wars, a finding presented as a major
discovery by the consortium of academic centers that decoded the human genome
has come under attack from the camp of the consortium's rival, Celera
Genomics.
The finding, one of the most surprising in the consortium's report on the human genome this February, was that 223 of the 30,000 human genes appear to have been acquired directly from bacteria. The implication was that the vertebrate and human genomes might have been shaped not just by inheritance but by weird accidents like bacterial infection of the egg or sperm cells.
The consortium cited the finding as fourth in a list of 11 major discoveries about the human genome and gave it considerable emphasis in its press briefing. Celera, in its genome analysis, made no mention of bacterial genes in the human genome.
An article published on the Web this week by the journal Science gives little credence to the consortium's idea.
"When I saw this claim in the consortium's paper I was surprised," said Dr. Steven L. Salzberg, the lead author of the critique. "It's not been documented previously that bacterial genes got into the human genome. When I looked at their method, I was immediately struck by the fact it was likely to be an error, because the method was simply wrong."
Dr. Salzberg is senior director of bioinformatics at the Institute for Genomic Research, which was founded by Dr. J. Craig Venter, now president of Celera in Rockville, Md. Dr. Salzberg said that he had not discussed his critique with Dr. Venter and that his article was a spontaneous effort to set the record straight. Dr. Salzberg's institute issued a news release today drawing attention to his critique.
The lead author of the consortium's genome article, Dr. Eric Lander of the Whitehead Institute in Cambridge, Mass., recently criticized Dr. Venter's decoding of the human genome, saying that Celera's method was a "flop" and would not have worked without reliance on the consortium's method, a contention Dr. Venter denied.
Dr. Lander said that his conclusions about bacterial gene transfer had been tentative and that Dr. Salzberg's criticisms were "thoughtful" but he declined to discuss them further.
Dr. Eugene Koonin of the National Center for Biotechnology Information, who wrote the section of the consortium's article on bacterial genes with Dr. Lander, said he could not discuss Dr. Salzberg's paper without Dr. Lander's assent. Later Dr. Koonin said he had arrived at an agreement with Dr. Lander that "the only thing I am going to say is that I fully support what he said."
The issue stems from the recent finding, made possible by the decoding of several bacterial genomes, that many genes in different species of bacteria are so similar that they appear to have been exchanged directly, perhaps in some act of engulfment or infection. This source of genes is called horizontal or lateral transfer to distinguish it from vertical descent, or inheritance from a common ancestor, which is the usual reason for different species to possess similar genes.
Humans have about a thousand genes similar to those of bacteria, presumably because the genes are so vital that their DNA structure has remained much the same over millions of years of descent from a common ancestor. But the consortium asserted that some of these genes were likely to have been acquired by lateral transfer. The reason was that the consortium could find no counterpart genes in the genomes of several species distantly related to humans, like the roundworm and the fruit fly. The consortium considered but rejected as less likely the possibility that these species, too, had inherited the genes and all lost them over the course of evolution.
Dr. Salzberg said that the consortium scientists had not tried hard enough to find counterpart genes in distant species and that he quickly turned up counterparts to a couple dozen of the supposed bacterial transfer genes in jellyfish. That means that humans and jellyfish probably acquired the genes the more common way, through inheritance, the critique says.
Indeed, Dr. Salzberg said, there were only 41 of the 223 genes on the list that he could not find in any other species. Even those 41 will dwindle to zero, he predicted, as more genomes are sequenced.
Dr. W. Ford Doolittle, an expert on bacterial evolution at Dalhousie University in Nova Scotia, said the lateral transfer of bacterial genes into the human genome was possible but the consortium's method of trying to prove it was "awfully incautious."
In a parallel battle, the journal Nature this week contains an article criticizing Celera's version of the fruit-fly genome, which appeared in Science last year. The article, by Dr. Samuel Karlin and colleagues at Stanford University, says there are many small discrepancies between the protein sequences that would be expected from Celera's fruit-fly genome and those of actual fruit-fly proteins recorded in SwissProt, a database of protein sequences.
-- Anonymous, May 18, 2001