Role of mutS and mutL genes in hypermutability and recombination in Staphylococcus aureus

Abstract

The mutator phenotype has been linked in several bacterial genera to a defect in the methyl-mismatch repair system, in which the major components are MutS and MutL. This system is involved both in mismatch repair and in prevention of recombination between homeologous fragments in Escherichia coli and has been shown to play an important role in the adaptation of bacterial populations in changing and stressful environments. In this report we describe the molecular analysis of the mutS and mutL genes of Staphylococcus aureus. A genetic analysis of the mutSL region was performed in S. aureus RN4220. Reverse transcriptase PCR experiments confirmed the operon structure already reported in other gram-positive organisms. Insertional inactivation of mutS and mutL genes and complementation showed the role of both genes in hypermutability in this species. We also designed an in vitro model to study the role of MutS and MutL in homeologous recombination in S. aureus. For this purpose, we constructed a bank of S. aureus RN4220 and mutS and mutL mutants containing the integrative thermosensitive vector pBT1 in which fragments with various levels of identity (74% to 100%) to the S. aureus sodA gene were cloned. MutS and MutL proteins seemed to have a limited effect on the control of homeologous recombination. Sequence of mutS and mutL genes was analyzed in 11 hypermutable S. aureus clinical isolates. In four of five isolates with mutated or deleted mutS or mutL genes, a relationship between alterations and mutator phenotypes could be established by negative complementation of the mutS or mutL mutants.

FIG. 1.

Genetic organization of the mutSL region in S. aureus and results of the transcriptional analysis by RT-PCR. (A) Schematic organization of the mutSL locus in S. aureus determined by in silico analysis. The hairpin symbol indicates a putative transcription terminator. The size of each intergenic region is indicated. Localization of the pBT1 insertion in the mutS gene of the S. aureus mutS1 strain is indicated. The black arrows flanking the dotted lines indicate the positions of the primers used in the RT-PCR analysis. Numbers in parentheses indicate the sizes of the amplified fragments. (B) Gel electrophoresis of the fragments obtained by RT-PCR. C, negative control without RNA; 4220, amplification obtained with S. aureus RN4220 total RNA extract; mutS, amplification obtained with S. aureus mutS1 total RNA extract. Primers used for each amplification are indicated.

FIG. 2.

Construction of the S. aureus mutS2 strain. Two fragments of mutS from S. aureus RN4220 (PCR 1, SNheIU5′/SSalIL5′; PCR 2, SKpnIU3′/SPvuIIL3′) and the totality of the Tn1545 kanamycin resistance cassette (PCR 3, KSalIU/KKpnIL) were amplified, restricted with the appropriate enzymes, and ligated. The construct obtained was amplified (PCR 4, SNheIU5′/SPvuIIL3′), and a NheI/PvuII restriction was performed on the resulting fragment and on the pBT1 thermosensitive vector (bla, β-lactamase gene; cat, chloramphenicol acetyltransferase gene; Ts, thermosensitivity; MCS, multiple cloning site). Both were ligated together, and the resulting construct (pBT1mutS2) was introduced into S. aureus RN4220. The double-recombination event was obtained at 42°C as previously described (4), using kanamycin (12.5 μg/ml) as the selective marker.

FIG. 3.

Estimation of the recombination frequencies in balance with sequence identity in S. aureus wt (RN4220) (A), mutS2 (B), and mutL (C) strains. Survival of the strains containing the different constructs after three subcultures at 42°C with chloramphenicol (and kanamycin for the mutS2 and mutL strains) was taken as an indicator of recombination frequencies. Points represent the mean ratio of the log of survival after three subcultures to the log of the corresponding initial inoculum. Bars indicate standard deviation. Regression lines and their equations are shown, as well as their correlation coefficients.