Do bacteria swap genes in deadly game?

In 1982, contaminated hamburger meat triggered a rash of violent illness in the United States and signaled that a bacterial friend of people had turned foe. The culprit turned out to be a virulent strain of Escherichia coli, normally a helpful resident of the lower intestine (SN: 7/22/00, p.53).

Now, scientist from the University of Wisconsin-Madison have decoded the genome of the dangerous strain and compared it with the DNA sequence of its far more common, mild-mannered cousin (SN: 2/8/97, p. 84)

The two sequences provide hints as to how a bacterium so deadly could be the close relative of one so benign. The pathogen, the researchers assert in the Jan. 25 NATURE, had picked up chunks of DNA from unrelated, infective bacteria, acquiring unpleasant traits that can send people to the hospital.

James B. Kaper, who developers bacterial vaccines at the University of Maryland School of Medicine in Baltimore, asserts that the findings will lead to vaccines and new diagnostic techniques for harmful E. coli infection. The new data also "will provide tremendous insights into this fundamental question of how a pathogen becomes a pathogen."

Unlike most of the E. coli in the gut, the virulent strain O157:H7 has been responsible during recent decades for increasing numbers of deaths worldwide. Some of its genes enable it to better survive stomach acids and cling to instentinal walls. Others make it produce one toxin that causes bleeding lesions in the digestive tract and another, the Shiga toxin, that leads to kidney failure and even death.

To find out how the strains differ, the researchers compared the two genomes side-by-side. They expected the gene sequences to be nearly the same, says study coauthor Fred Blattner. They were looking for minor differences that would identify gene regions responsible for the pathogen's nasty affects.

Instead, they found that large swatches of DNA from the two cousins didn't match at all. In fact, about 20 percent of the two genomes were completely different, with the remainder of the DNA sequences being almost identical.

"It was the biggest surprise of my scientific career and one of the most interesting to come across," says Blattner. The genomes, he explains, are made up of a mosaic pattern. Section of a so-called DNA backbone seem unchanged since the two strains diverged 4.5 million years ago.

The pieces of dissimilar DNA, called islands, crop up throughout the genome. Some of these regions encode O157:H7's troublesome traits, such as toxins. The islands ends hint that bacteria-hopping viruses carried the pieces in from other species, says the researchers.

Despite calling this study "good work" and "an excellent starting point," Jonathan A. Eisen of The Institute for Genomic Research in Rockville, Md., argues for caution about the new report's conclusions. "Islands themselves cold have nothing to do with gene transfer," he says. "Personally, I'm not convinced."