LONDON - High levels of antibiotic resistance have been found in gut bacteria from wild rodents that probably have never been exposed to antibiotics, prompting concerns that more careful use of antibiotics may not be enough to reduce antibiotic resistance.

Researchers from the University of Liverpool in the UK were so surprised by their findings that they immediately repeated the study. But their results were the same: Up to 90 percent of gut bacteria from the bank voles and wood mice tested were resistant to a range of common antibiotics.

C. Anthony Hart, professor of medical microbiology at the Centre for Comparative Infectious Diseases at Liverpool, told BioWorld International: "People are worried about antibiotic resistance at present and the idea is that if you use antibiotics more carefully, you will decrease the prevalence of resistance. I would not want to extrapolate from our study to say that that is wrong, but it means that we have to stop and think, 'Is this absolutely going to be so?'"

Kevin Kerr, senior lecturer in microbiology at the University of Leeds in the UK, said the study was an important piece of work. "It provides yet another worrying pointer as to how bad things are with respect to antibiotic resistance," he said. "It is certainly true that there is still no conclusive proof that more prudent use of antibiotics will have a significant effect on antibiotic resistance."

The study by Hart and his colleagues was part of a larger one into organisms such as cowpox virus, which are carried by wildlife but that can cause disease when transmitted to humans. The aim was to find out how these organisms - pathogenic in man - affect the health, fertility and life span of their animal hosts. The design of the study included trapping wild rodents such as the bank vole (Clethrionomys glareolus) and the wood mouse (Apodemus sylvaticus), taking blood from them, fitting a radio transponder to them and then releasing them, so that they could be recaught and tracked throughout their lives.

Hart and his colleagues decided to test the hypothesis that the prevalence of antibiotic resistance in wild rodents in two different woodland sites in an area of Northwest England would differ. One site was adjacent to farmland on which heifers were occasionally kept, while the other was very isolated, being surrounded by more woodland, gardens and an ornamental lake.

An undergraduate student, Moira Gilliver, was allocated the task as her degree course project of collecting feces from the voles and mice trapped in both sites, and testing them for antibiotic resistance.

Hart said, "We expected that there would not be much resistance to antibiotics in either population, because neither would have been exposed to antibiotics, although we thought that there might be more in the area adjacent to farmland."

Their results are reported in a letter to the Sept. 16, 1999, Nature, titled "Antibiotic resistance found in wild rodents." They wrote, "Overall, 90 percent of coliforms were resistant to amoxycillin, amoxycillin/clavulanic acid and cefuroxime. Depending on the bacterial species, 14 to 76 percent of coliforms were tetracycline resistant and 0 to 67 percent were trimethoprim resistant [and] 90 percent were ampicillin resistant."

In the latter case, expression of the enzyme beta-lactamase was the cause in more than half the isolates. The level of resistance to apramycin, an antibiotic used only in farming practice, was low, at 12 percent to 33 percent. Hart said: "This suggests that the high prevalence of resistance is not related to exposure to farm animals."

Quite why such high levels of resistance were present in these populations is a mystery, Hart said. "We can't say for absolutely certain that these animals have never had antibiotics, but it seems highly unlikely that they would. Given that antibiotics come from the natural world, and are produced by bacteria to kill off other bacteria, then it is possible that antibiotic resistance genes have developed in the same way. But I think the prevalence is so high that it could not be accounted for by that."

Kerr said that finding antibiotic resistance in this setting was not unexpected but that the high levels of resistance were surprising. He said, "Large quantities of antibiotics are given to animals reared for food production as growth promoters, sprayed on fruit bushes and in orchards, even on beehives and, worldwide, thousands of tons of antibiotics are discharged into the environment every year. When animals given these antibiotics defecate, the bacteria can contaminate land and then enter the water table. Scavenging birds may also feed in places where they can pick up resistant strains of bacteria and as a result these bacteria may be carried over long distances. But without further research, it is not possible to say whether this explains how these strains of resistant bacteria reached these rodents."

Hart and his colleagues are planning to study the genomes of the resistant isolates, to check whether the rodents carry the same isolates with the same antibiotic resistance throughout their lifetimes. "We are also looking at the nature of the antibiotic resistance genes, to try to see if that will give us a clue as to where they could have come from," Hart said.