Plasmid diversity in Escherichia coli isolated from processed poultry and poultry processors.

Plasmids of bacteria selected from different bacterial populations because they shared a distinctive antimicrobial resistance phenotype have sometimes had identical restriction fragments. Such identical plasmids are thought to belong to small and thus epidemic clones because the plasmid content of unselected resistant isolates has seemed diverse. To survey this presumed diversity and its implications for the lineage of resistance plasmids we examined the transferability, sizes and EcoR1 restriction fragment sizes of plasmids in both Escherichia coli isolated randomly from poultry raised by 16 growers as they were being processed through two plants and in isolates from the urine of women processing poultry in those plants. Forty two (24%) of 175 resistant isolates from poultry of 16 growers and 9 (26%) of 34 resistant isolates from the poultry processors transferred resistance conjugatively to varied combinations of antimicrobials. No poultry isolate had both the same expressed and the same transferred combination as any processor's isolate. The DNA bands which could be discerned in electrophoresis gels of restricted or unrestricted plasmid extracts of isolates or their transconjugants from 156 of the poultry and 24 of the poultry processors appeared diverse. Pairs of related-appearing plasmids were seen in consecutive isolates of poultry from each of two growers and in one pair from different growers. One set of identical-appearing plasmids was seen in 3 consecutive isolates from poultry of one grower, others in 2 consecutive isolates from a second grower's poultry, in 2 non-consecutive isolates of a third grower's, and in single isolates from poultry of 2 different growers. None of the plasmids from any of the human isolates appeared related to those from any other human isolate or to those of any poultry isolate. These results indicate that resistance plasmids are highly diverse and that all but two of the exceptions to complete diversity in the isolates surveyed here could be ascribed to cross colonization within flocks of individual poultry growers. Also, while none of the plasmids in the poultry isolates appeared ancestral to any of plasmids in the poultry processors' isolates, their diversity indicates that those sampled plasmids would be only a very small fraction of the total number of different plasmids in bacteria colonizing poultry processed at that time or earlier.