The host range of RK2 minimal replicon copy-up mutants is limited by species-specific differences in the maximum tolerable copy number.

The minimal replicon of the broad-host-range plasmid RK2 consists of a gene, trfA (trans-acting replication), encoding a protein required for initiation of plasmid replication. The TrfA protein binds to iterons in the cis-acting origin of vegetative replication (oriV), but the exact mechanism by which TrfA-mediated replication initiation takes place is not known. We report here the isolation and characterization of five mini RK2 trfA mutant plasmids with an elevated plasmid copy number, four in Pseudomonas aeruginosa and one in Azotobacter vinelandii. The mutations are localized between or downstream of previously reported Escherichia coli copy-up mutations in trfA, and one of the mutations has been described earlier as an independent copy-up isolate in E. coli. The five mutant plasmids were all moderately copy up in both E. coli and their host of origin, in spite of the use of isolation procedures which were expected to select efficiently in favor of plasmid mutants specifying high copy numbers. In contrast, previously described high copy-up mutants isolated in E. coli could not be established in P. aeruginosa and A. vinelandii. These high copy-up mutants were shown to induce cell killing in E. coli under conditions where the plasmid copy number was increased as a physiological response to reduced growth rate. We propose that the reason for this killing effect is that the copy number under these conditions exceeds an upper tolerance level specific for E. coli. By assuming that the corresponding tolerance level is lower in P. aeruginosa and A. vinelandii than in E. coli, and that the mechanism of copy number regulation is similar, the model can explain the phenotypes of all tested copy up mutants in these two hosts. Analogous studies were also performed in Salmonella typhimurium and Acetobacter xylinum. The data obtained in these studies indicate that the above model is probably generally true for gram-negative bacteria, and the results also indicate that the maximum tolerable copy number is surprisingly low in some hosts.