NEW YORK, May 17 – New findings published in the May 18 issue of Science call into question the Human Genome Project’s assertion that genes have been laterally transferred from bacteria to a human vertebrate ancestor.
Steven Salzberg, director of bioinformatics at The Institute for Genomic Research, and his colleagues examined the Ensembl database of genes, focusing on 223 genes that the HGP said were found in bacteria and human genomes but not those of other sequenced organisms. For comparison’s sake, they also looked at protein products of genes in Celera’s database.
The HGP postulated in its February 15 Nature paper, "Initial Sequencing and Analysis of the Human Genome," that the most “parsimonious” explanation for these shared genes “was that these genes entered the vertebrate (or prevertebrate) lineage by horizontal transfer from bacteria.”
This finding made national headlines when the Nature paper was released. But when Salzberg’s group reexamined the genes in HGP’s Ensembl set, along with Celera’ s comparable set of proteins, they found that 182 of the putative bacteria-vertebrate transfer genes could be matched to genes in other eukaryotes or could be explained away by other means.
Some genes had been removed from the subsequent October release of Ensembl, one was a contaminant, and a number of others could be matched to other eukaryotic genes once they were matched up with similar genes from the other set—either Ensembl or Celera.
Even though these remaining genes—41 in Ensembl and 46 in the Celera set—have no known homologs in species other than bacteria and humans, Salzburg said it is unlikely that any will have jumped directly from germ to human germline.
“The curve is heading rapidly towards zero,” said Salzberg. “Once we have five or 10 more invertebrate genomes, that will get rid of those last 40 or so genes.”
Salzberg and his colleagues conducted further study on these remaining genes by constructing phylogenetic maps of different species. There were none for which it appeared the phylogenetic tree pointed to a direct bacteria-human link, Salzberg said.
“We wanted to be as thorough as we could. We couldn’t find any evidence of lateral transfer,” said Salzberg. “As lateral transfer is an extraordinary event, we would like to see strong statistical evidence, and it’s just not there.”
An accompanying commentary in the May 18 issue of Science, written by Jan Andersson and Camilla Nesbo of the Canadian Institute for Advanced Research, provides theoretical support for Salzberg’s claims. “It is reasonable to assume that the downward trend might continue” in the number of genes shared exclusively between humans and bacteria as more nonvertebrate genomes become available for screening,” they wrote.
But Andersson and Nesbo point out that shrinking the pool of putative bacteria-human transfers does not rule out the possibility of their occurrence. “The statistical arguments of the sort Salzberg, et al present can never eliminate the possibility that some of these [bacteria vetebrate transfer] candidates (or others they eliminated with their parsimonious broad brush approach) really are true bacteria vertebrate transfers.”
Salzberg is making the database of 41 genes from Ensembl and 46 from Celera’s database available to researchers through the Science website.
“Researchers should try to test these genes to see if any can truly be explained by bacteria-human transfer, and they will, but I don’t think they will succeed,” he said.
Andersson and Nesbo prepared phylogenetic trees for seven genes in Salzberg’s set, and found one that looked like it was transferred from bacteria to vertebrates. They suggested that further research in which researchers construct molecular phylogenetic trees and to look thoroughly for human genes nested within trees of bacteria.Meanwhile, Salzberg said he hoped the publication of his paper would help allay fears that the idea of large-scale bacterial-human gene transfer might engender. “I hope the message gets out there that bacteria haven’t, and are not in the process of inserting their genes into our germline," he said.