Bacteriocin production and resistance in a field population of Rhizobium leguminosarum biovar viciae

In 1993 isolates of Rhizobium leguminosarum bv. viciae were obtained from nodules formed by plants inoculated with soil samples from a variety of field sites in East Anglia (U.K.) and grouped into phenotypes (arbitrarily referred to as “strains”) on the basis of RAPD PCR and plasmid profiles ( Handley et al., 1997). At one of these sites, the 20 isolates obtained all appeared to be isolates of a single “strain”. This exclusive isolation of a single “strain” was not observed at any other site. In 1994 and 1995 the soil at this site was sampled again, and 153 isolates were characterised revealing at least 21 “strains”. None was the same as that found in 1993. We have screened each of the “strains” isolated from this site for bacteriocin production and resistance. The “strain” observed in 1993 was found to produce an extremely potent medium molecular weight bacteriocin to which 68% of the “strains” tested from other sites were sensitive to some degree. However, 62% of the “strains” isolated from this site in 1994 and 1995 were unaffected by the bacteriocin produced by the 1993 isolates, and the growth of all but one of the remaining “strains” was only slightly reduced. None was found to produce this bacteriocin. One “strain” which did appear to produce the same bacteriocin was subsequently found in a screen of 250 isolates obtained from a re-sample of stored soil collected in 1994. We believe that the bacteriocin produced by the “strain” obtained in 1993 was responsible for its dominance in that year, and that the population has responded with the subsequent proliferation of resistant strains.