Prevalence of SOS-mediated control of integron integrase expression as an adaptive trait of chromosomal and mobile integrons

BACKGROUND Integrons are found in hundreds of environmental bacterial species, but are mainly known as the agents responsible for the capture and spread of antibiotic-resistance determinants between Gram-negative pathogens. The SOS response is a regulatory network under control of the repressor protein LexA targeted at addressing DNA damage, thus promoting genetic variation in times of stress. We recently reported a direct link between the SOS response and the expression of integron integrases in Vibrio cholerae and a plasmid-borne class 1 mobile integron. SOS regulation enhances cassette swapping and capture in stressful conditions, while freezing the integron in steady environments. We conducted a systematic study of available integron integrase promoter sequences to analyze the extent of this relationship across the Bacteria domain. RESULTS Our results showed that LexA controls the expression of a large fraction of integron integrases by binding to Escherichia coli-like LexA binding sites. In addition, the results provide experimental validation of LexA control of the integrase gene for another Vibrio chromosomal integron and for a multiresistance plasmid harboring two integrons. There was a significant correlation between lack of LexA control and predicted inactivation of integrase genes, even though experimental evidence also indicates that LexA regulation may be lost to enhance expression of integron cassettes. CONCLUSIONS Ancestral-state reconstruction on an integron integrase phylogeny led us to conclude that the ancestral integron was already regulated by LexA. The data also indicated that SOS regulation has been actively preserved in mobile integrons and large chromosomal integrons, suggesting that unregulated integrase activity is selected against. Nonetheless, additional adaptations have probably arisen to cope with unregulated integrase activity. Identifying them may be fundamental in deciphering the uneven distribution of integrons in the Bacteria domain.

Multiresistance in Pasteurella multocida is mediated by coexistence of small plasmids

In most gram-negative bacteria, acquired multiresistance is conferred by large plasmids compiling numerous antimicrobial resistance genes. Here, we show an evolutionary alternative strategy used by Pasteurella multocida to become resistant to multiple clinically relevant antibiotics. Thirteen beta-lactam-resistant clinical isolates, concomitantly resistant to tetracyclines and/or streptomycin as well as to sulfonamides, were studied. Pulsed-field gel electrophoresis analysis revealed different profiles among the isolates, showing that clonal dissemination was not the sole event responsible for the spread of multiresistance. Each P. multocida strain carried two or three small plasmids between 4 and 6 kb in size. A direct association between resistance profile and plasmid content was found. Complete nucleotide sequencing of all plasmids revealed seven different replicons, six of them belonging to the ColE1 superfamily. All plasmids carried one, or a maximum of two, antimicrobial resistance determinants. Plasmids pB1000 and pB1002 bore bla(ROB-1), pB1001 carried tet(B), pB1003 and pB1005 carried sul2 and strA, pB1006 harbored tet(O), and p9956 bore the tet(H) gene. All plasmids except pB1002 and pB1006 were successfully transformed into Escherichia coli. pB1000, also involved in beta-lactam resistance in Haemophilus parasuis (A. San Millan et al., Antimicrob. Agents Chemother. 51:2260-2264, 2007), was mobilized in E. coli using the conjugation machinery of an IncP plasmid. Stability experiments proved that pB1000 was stable in P. multocida but highly unstable in E. coli. In conclusion, bla(ROB-1) is responsible for beta-lactam resistance in P. multocida in Spain. Coexistence and the spread of small plasmids are used by P. multocida to become multiresistant.