Revealed: bacteria's survival guide

The bacteria that infect human beings have disarmed almost the entire modern arsenal of antibiotics. Not long after new antibiotics are deployed, resistance appears almost as if by magic.

Some strains of bacteria in hospitals are now resistant to virtually all common antibiotics -- they are a potentially lethal hazard to surgical patients.

The proximate causes of the problem are well known: over-prescription and poor patient compliance with antibiotic therapy, and the routine use of antibiotics as growth promoters in livestock feed formulas.

Such practices apply intense selection pressure to bacterial populations, winnowing out susceptible cells, and leaving resistant survivors. But selection pressure doesn't actually create resistance genes -- evolution just doesn't work that fast, even in bacteria.

In trying to determine the ultimate source of those resistance genes, a research team at Sydney's Macquarie University, led by Dr Michael Gillings, has made an astonishing discovery: a sort of microbial Internet, that gives bacteria access to a prodigious library of survival genes.

The discovery throws new light on how bacteria evolve and adapt to different environmental niches, and Gillings believes it has profound implications for the development of new antibiotics -- whatever ingenious new strategies or drugs emerge from bacterial genomics projects, bacteria are likely to circumvent them.

But the upside of the discovery, according to Gillings, is that the biotechnology industry will gain access to an almost limitless repertoire of ready-made genes for medical and industrial applications.

In hospital bacteria, resistance genes are usually arrayed on tiny loops of DNA called plasmids. By exchanging plasmids, resistant strains can rapidly disseminate resistance genes to other bacteria, even of different species.

Gillings says that the resistance genes are usually packaged within a universal DNA "cassette" and daisy-chained on the plasmid.

The daisy-chained genes are packaged within a larger structure called the integron, which contains a gene for an enzyme that cuts and splices DNA, allowing the daisy-chains to pop apart. Each individual cassette carries a DNA sequence at either end that acts like a Velcro tab -- as the cassette detaches, the ends join up, forming a tiny loop of DNA.

Gillings says these circularised cassettes drift around in the cell, or may be released into the environment, and be picked up by other bacteria.

The same integrase enzyme that allows the cassettes to be reshuffled or donated to neighbours opens up specialised 'slots' in bacterial DNA to accommodate new cassettes. This 'plug-and-play' mechanism allows new resistance genes to be transferred between the main chromosome and plasmids -- or donated to other microbes.

Gillings and his colleagues wondered how many wild-type bacteria also possess integrons. They took a 400mg soil sample -- less than a 10th of a teaspoon -- and used the polymerase chain reaction to search for the distinctive DNA tags that identify integrons.

"We expected to get only one or two hits from the whole sample," Gillings said. "But we found integron signals in every sub-sample we tested, and multiple genes with every sample. We're still haven't exhausted the diversity in our original 400mg soil sample, and the discovery curve is still rising at an angle of 45 degrees.

"We've now found thousands of genes, and when we compare their DNA sequences with known genes in databases, none of them have anything to do with antibiotic resistance. What we have discovered is a generic system for acquiring and rearranging genes in a very diverse range of bacteria. It's not just present in specialists, it's a ubiquitous feature of bacterial genomes."

In contrast to the specialised hospital bacteria, the wild-type bacteria that the Macquarie team has studied carry their integrons on their main chromosome.

Genome projects have now delivered gene maps and complete gene catalogues for more than 30 species of pathogenic bacteria that infect humans. Gillings says all these genomes share a basic set of housekeeping genes, accounting for about 60 to 70 per cent of the total.

In those bacteria that possess integrons, these may account for up to 5 per cent of the total genome -- the integrons serve as a sort of utility belt carrying all the specialised genes that the particular species requires to adapt and survive in its chosen environmental niche.

The Macquarie team has found that the specific cassettes found in these integrons not only between species, but between strains of the same species -- and in some cases the integrons are absent.

This, Gillings says, is a clear indication that the integrons are acquired by horizontal gene transfer within and between species.

"About 95 per cent of the genes in the integron have never been seen before. They have no homology to any genes in the DNA databases," Gillings says.

He notes that the genome projects on bacteria have involved selected clonal lines of the subject species -- 'hospital' strains that are highly adapted to the hospital environment, and which therefore represent only a tiny fraction of the genetic diversity in wild-type bacteria of the same species.

"There is a huge resource out there that genome sequencing projects are missing," he says.

The bacterial Internet allows 'hospital' bacteria to tap into this vast library of wild-type genes -- that's how resistance genes appear almost magically whenever a new antibiotic is brought into clinical use.

"We're battling an enemy with almost unlimited resources," Gillings says. "There's no reason to suspect that any gene, anywhere, even in humans, couldn't find its way into a bacterial gene cassette and be shared around.

"But from another perspective, the enormous hit rate we have been getting for integrons in wild-type bacteria means there is obvious potential for the biotechnology to exploit this resource."

Various strains of the crop pathogen Xanthomonas infect different crops. The Macquarie team -- Gillings, Assoc Prof Hatch Stokes and Dr Andrew Holmes -- is working with Dr Ric Cother at the NSW Department of Agriculture to investigate whether the specificity of these so-called pathovars for different crops stem from specialised arrays of gene cassettes in their integrons.

The collaborative project this month received a $147,00 three-year linkage research grant from the Australian Research Council (ARC).

While Xanthomonas provides a model system for investigating the potential role of integrons in pathogenicity, Gillings says the project also has important practical outcomes. "It could lead to a rapid detection and an identification of Xanthomonas infections in crops," he says.

"The surface of a crop's leaves is a pretty hostile environment -- the plants produce lots of anti-bacterial compounds, and they're exposed to ultraviolet light, and it's also a very dry environment.

"I'm not sure that gene cassettes are responsible for the pathogenicity of different strains, but it wouldn't surprise me to find that it's important for survival on plant surfaces."

Gillings says that while the discovery may temper optimism that genome projects will lead to new antibiotics that won't succumb to resistance, it raises the possibility that the integron/cassette system could itself be targeted to counter the resistance problem.