LONDON ¿ Even if antibiotic consumption is slashed to a fraction of current levels, antibiotic-resistant strains of bacteria are not likely to disappear. This is the discouraging conclusion of a study into how emergence of resistance varies in different environments.

Dan Andersson, of the Swedish Institute for Infectious Disease Control in Solna, together with colleagues from Uppsala University in Sweden, examined antibiotic resistance in bacteria grown either in mice or in a laboratory medium. They found that the bacteria evolved different kinds of mutations according to the environment in which they grew.

Andersson, who is associate professor in the Department of Bacteriology at the institute, told BioWorld International: ¿What we found was that you can have antibiotic-sensitive strains, and antibiotic-resistant strains which have compensatory mutations to restore fitness, and that these have equal fitness even in the absence of antibiotic.¿

This result is discouraging, he added: ¿What it tells us is that we have to be extremely careful about how we use antibiotics. Our consumption of these drugs has to go down to prevent the appearance of fit, resistant bacteria. Once such compensated resistant strains are established, then even if consumption of antibiotics is reduced, these strains will not go away because the resistant strains with compensatory mutations are just as fit to reproduce as the antibiotic-sensitive strains.¿

The results of the study are reported in the Feb. 25, 2000, issue of Science in a paper titled, ¿Effects of Environment on Compensatory Mutations to Ameliorate Costs of Antibiotic Resistance.¿ The work was partly supported by Leo Pharmaceuticals Inc., of Copenhagen, Denmark.

The aim of the research, Andersson said, was to be able to understand and predict how rapidly resistance can develop, and how stable resistant strains would be once they had evolved. He explained, ¿Usually, antibiotic resistance confers some biological cost on the bacteria ¿ they are less fit as a result. But the cost of the resistance can be reduced by additional compensatory mutations. We decided to study the nature of these compensatory mutations in two different environments ¿ culture medium and mice.¿

Using Salmonella typhimurium, the team looked at mechanisms of resistance to two antibiotics, streptomycin and fusidic acid. The target of streptomycin in bacteria is the ribosome, thus preventing protein synthesis from taking place. Mutations that bring about resistance affect a gene that encodes a ribosomal protein, and these have the effect of preventing streptomycin from binding to the ribosome.

Andersson and his colleagues found that S. typhimurium strains that were resistant to streptomycin grew more slowly than wild-type strains in both mice and laboratory medium. The researchers selected fast-growing mutants in order to identify strains that had evolved compensatory mutations which restored (either partially or totally) their fitness to reproduce.

¿We found that, in mice, we got only one specific type of compensatory mutation, which was in the same gene ¿ in fact, in the same codon ¿ as that causing the resistance. But in the strains found in laboratory medium, we found that compensatory mutations had occurred in other genes, encoding other ribosomal proteins.¿

It was a similar story for the strains resistant to fusidic acid. Those mutants that had evolved in mice which had regained normal fitness had mutated back to the wild-type genotype, and turned out to be sensitive to fusidic acid once again. But those that had evolved in laboratory medium had compensatory mutations within the same gene as that responsible for resistance, and had retained their resistance to the antibiotic.

¿So we found two completely different types of compensation for resistance,¿ Andersson said. ¿This suggests that fundamental processes such as formation of mutations and selection of mutants can differ drastically according to environment.¿

A further observation was that the frequency with which the mutations occurred was much higher in mice than in laboratory medium. This finding could have implications for the pharmaceutical industry, Andersson said, because much work is carried out in culture media aimed at estimating how rapidly resistance might develop when a new antibiotic is used clinically. ¿What this study shows,¿ he said, ¿is that this approach may be completely irrelevant, and that the mutation rate is much higher when you are looking at an animal. It shows the importance of doing animal experiments to obtain relevant data.¿

Future experiments planned by the group will aim to determine why the mutation rate is different in different environments. ¿We want to find out what it is about the growth conditions in the mouse that lead to a faster mutation rate,¿ Andersson said. ¿It may be that this is because S. typhimurium normally grows only in cells called macrophages. These are important in defending the body against infection and produce compounds including oxygen radicals. It might be exposure to oxygen radicals which causes the different mutation spectrum in the mouse.¿

If this is the case, then it may be that this finding applies only to bacteria that grow inside macrophages. To check this, the team is carrying out similar experiments with bacteria that grow in other locations, to see if there is a similar difference in mutation rate and type.