GLOSSARY:
Eukaryote: organism whose cells contain a nucleus that encloses the chromosomes.

Genome: entire library of genetic information in a species.

Lower eukaryote: plant, animal or fungus below humans in development.

Mitochondria: components responsible for converting energy inside cells.

Prokaryote: organism whose DNA is not contained in a nucleus.

Vicariate: science nerd who lives through the intellectual accomplishments of others.

(CHART AT RIGHT):
1: When the same protein appears in prokaryotes (bacteria), lower eukaryotes (like worms or plants), and humans, the gene came probably from a common ancestor.

2: Ditto when the human and lower eukaryote protein are identical, and both are similar to the bacterial protein.

3: When the human protein is identical to the prokaryote protein, the gene may seem to have moved horizontally from bacterium to human. BUT genes mutate at different rates in various organisms, so the lower eukaryote might once have been identical.

4: When the lower eukaryote has no similar gene, we might assume horizontal transfer. BUT genes disappear, so the lower eukaryote may once have had the gene.

It was perfect grist for the Why Files weird-science mill. We're talking about the notion that bacteria that once infected our ancestors stuck their genes into our chromosomes. This creepy idea got a lot of play earlier this year when scientists reported in Nature that 223 genes had jumped from lowly, disgusting, dishonorable, nasty, slimy, single-celled bacteria to noble, honorable, intelligent and kind Homo sapiens (see "International Human Genome," below). Talk about chutzpah! imagine an organism too primitive to hide its DNA inside a nucleus sticking us with its genes!

Amidst the hoopla about unraveling the human genome, we missed the news. But that may be just as well, since, according to Steven Salzberg, a bioinformatics wizard at The Institute for Genomic Research, the conclusion was unfounded.

His statistics imply that the true number of genes we got directly from bacteria is not 223, but maybe 40, tops. When more data come in, he says, the number could drop to zero.

Alien genes
Normally, genes travel vertically -- from a bacterium to its daughter cells, or from parents to offspring. The transfer in question is horizontal. Bacteria, for example, transfer genes for antibiotic resistance to other bacteria, and viruses regularly insert DNA into plant and animal chromosomes.

The current debate concerns whether bacteria pulled off that trick with our ancestors.

It's not as if we don't have enough genetic similarity to crummy ol' bacteria. After all, life seems to have descended from a common ancestor, and that left our chromosomes stuffed with thousands of bacterial genes. Some of these genes, involved in metabolism, energy and reproducing DNA, were so valuable that they didn't change much over the billions of years since eukaryotes (organisms with a cell nucleus) diverged from prokaryotes (those without a nucleus).

Equally unquestioned is the notion that even before that divergence, our predecessors "trapped" microbes and enslaved them to serve as mitochondria -- the cell's energy-producing factories. Plants, likewise, trapped free-living organisms to serve as chloroplasts -- the organelles that make sugar from sunlight and carbon dioxide.

With mitochondria and chloroplasts, the evidence is in the separate set of genes.

But did bacteria stick their DNA into our genes in a way that can be inherited? Salzberg isn't buying this claim.

The argument rests on the structural similarity of proteins made by the various genes. Proteins are hideously complicated, so if two have identical shape, it's odds-on that the gene that patterned for them was transferred among all organisms that make the protein. Independent invention of the identical gene is incredibly improbable.

If the gene appears in bacteria, humans and intermediate organisms (life forms that are more developed than bacteria but less lofty than we) then it probably spread through normal, vertical inheritance.

But if the gene is absent in intermediate organisms, horizontal transport becomes a possibility, since how else did we get the gene?

Not such a great buy
To answer the question, Salzberg analyzed data on proteins made by bacteria, humans and four other species whose genomes (complete genetic instructions) are known.

His analysis, however, took into account two factors that previous authors did not. First, genes change over time, but the rates differ among various organisms. Thus genes that look dissimilar may still be related.

Second, genes can disappear if they're not needed. So the absence of a gene in an intermediate organism does not prove it never had them. Statistically, Salzberg says, if you started with 30,000 genes -- as he did -- and examined four intermediate organisms, about 100 would be lost by chance. That, he says, would make it appear that the genes had jumped horizontally from bacteria to humans.

But the appearance would be deceiving. Once you take genetic change and gene loss into account, Salzberg says, "The evidence we have to date do not allow us say that any genes went directly from bacteria to humans." When the genomes of other organisms are deciphered, he predicts more intermediates will be found. "We fully expect [the number of horizontal transfers] to drop to zero."

No sale
Aside from statistics, there's logic behind the assertion. For bacterial DNA to appear in the offspring of a many-celled organism, it must enter reproductive ("germ line") cells. Any old cell just won't do. That may explain why bacteria can get away with horizontal transfer among themselves. "With bacteria, every organism is a germ line," Salzberg says.

Why does this evolutionary argument matter? Because statistical manipulation of DNA is a key tool for unraveling the history of life. "We're trying to infer something that happened millions or billions of years ago," says Salzberg. "Unless we have a time machine, we will have only statistical evidence, so we have to construct the evidence very carefully."

Sure, it's a bit disappointing, debunking the notion that bacteria can commandeer our DNA for their disgusting, single-celled purposes. But here's the bright side: now you don't have to worry that getting an infection, as Salzberg says, "would mean that your descendents would get new genes."

-- David Tenenbaum

Bibliography:
Microbial Genes in the Human Genome: Lateral Transfer or Gene Loss? Steven Salzberg et al, Sciencexpress, 17 May 2001.

International Human Genome Sequencing Consortium, Nature, 409, 860, 2001.