Risk assessment studies of E. coli host-vector systems.

These studies have demonstrated the poor survival of E. coli K-12 in the human and mouse gut and the absence of detectable transfer of pBR322 to endogenous intestinal flora. Even in the presence of mobilizing plasmids, no mobilization of pBR322 was detected, and transfer of the transferable plasmids was greatly reduced in the gut. Colonization of the mouse intestine by E. coli was shown to be affected by spontaneous chromosomal mutations to Sm or Rif resistance; these derivatives were weak competitors for colonization in the presence of the original E. coli K-12 organism. Moreover, E. coli K-12 appeared to be at a selective disadvantage in the human and mouse gut since even antibiotic selection failed to increase its survival. A background of data has been accumulated on the frequency of lambda, and M13 phages, and amber-suppressible bacteria in natural human and animal populations. These data should be useful to present and future risk-assessment evaluations. The frequency of antibiotic resistant coliforms in fecal samples from different human and animal populations was found to high. In the present study, antibiotic resistance was very common in both hospitalized and non-hospitalized populations. The percentage of resistant coliforms within individual specimens, however, was higher in patient population.