|Home | About | Journals | Submit | Contact Us | Français|
Excision of staphylococcal cassette chromosome mec (SCCmec) is mediated through the ccrA- and -B-encoded recombinases. We investigated the effects of different antimicrobial agents on ccrA expression by using a ccrA::lacZ fusion and reverse transcription-PCR with methicillin (meticillin)-resistant Staphylococcus aureus strains MW2 (SCCmec IVa) and N315 (SCCmec II). Upregulation of ccrA was observed upon exposure to β-lactam antibiotics. Vancomycin increased ccrA expression in MW2 but had no effect on N315. Vancomycin may contribute to the transfer of SCCmec IVa but have no effect in SCCmec II.
Resistance of staphylococci to oxacillin and other beta-lactam antibiotics is mediated by penicillin binding protein 2a, encoded by the mecA gene. mecA is carried on staphylococcal cassette chromosome mec (SCCmec). Currently, there are five different types of SCCmec elements, with several subtypes, characterized by a unique combination of mec and ccr complexes (7). While SCCmec types I to III have primarily been observed in the hospital, SCCmec types IV and V were initially described for community-acquired methicillin (meticillin)-resistant Staphylococcus aureus (MRSA) isolates (3). It has been shown that the ccrAB gene products can spontaneously excise the SCCmec cassette (6, 9). Excision of SCCmec from its chromosomal background has also been documented to occur at high frequency when the ccrAB genes are resident on a high-copy-number plasmid (5). This suggests that an increase in expression of these genes may be associated with a high frequency of excision. This study was conducted to investigate the influence of external factors, such as antimicrobials and other stresses, on the expression of ccrAB.
(This work was presented in part at the 47th Interscience Conference on Antimicrobial Agents and Chemotherapy [ICAAC], Washington, DC, 2007.)
MRSA strains MW2 (SCCmec type IVa) and N315 (SCCmec type II) were used as templates for the amplification of their respective ccrA fragments. Plasmids were transformed into Escherichia coli DH5-alpha. Shuttle vectors were moved from E. coli to S. aureus RN4220 by electroporation as previously described (4).
A ccrA::lacZ fusion was created by amplifying the upstream promoter region, including the first 21 nucleotides of ccrA, from MW2 and N315 by using primers C1F and C1RT (Table (Table1).1). The PCR product was ligated to the SmaI-cut promoterless ß-galactosidase gene in plasmid pMC1871 and transformed into E. coli. A negative-control out-of-frame construct was produced as described above. Plasmid orientation was confirmed by PCR and sequencing. The ccrA::lacZ construct was excised with PstI, ligated into the same restriction site of shuttle plasmid pMK4, and transformed into E. coli (12). The resultant plasmid was isolated and used to transform RN4220 by electroporation (4). Plasmids were then transduced into MW2 and N315 by using phage 80-alpha (10).
β-Galactosidase activity was measured by using o-nitrophenyl-β-d-galactopyranoside (ONPG) as previously described (8). Briefly, cells were grown in brain heart infusion broth until an optical density at 600 nm of approximately 0.5 was reached. The commonly used antimicrobials ampicillin, sulbactam, ampicillin-sulbactam, oxacillin, cefoxitin, gentamicin, ciprofloxacin, linezolid, tetracycline, and vancomycin were added at CLSI breakpoint concentrations (2), and the cells were incubated for a further 15 min. Heat shock was induced by incubation at 42°C. Control samples were untreated. After cells were broken open by milling, the suspension was centrifuged and 1 ml of lysate added to 0.2 ml of ONPG (4 μg/ml) and incubated for 4 h at 37°C. To confirm the results, ccrA expression was quantified by reverse transcription-PCR (RT-PCR) under the same conditions as described above, using wild-type strains MW2 and N315. Briefly, 0.5 ml of cells, after challenge, was added to 1 ml RNAprotect (Qiagen, Hilden, Germany) and total RNA extracted with a Qiagen RNeasy kit. Contaminating DNA was removed with on-column DNase treatment (Qiagen). Total RNA was quantified by using a spectrophotometer, and 0.5 μg RNA was reverse transcribed with a QuantiTect reverse transcription kit (Qiagen). Real-time PCR was performed with a LightCycler (Roche, Mannheim, Germany) with QuantiTect SYBR green (Qiagen), using ccrA-specific primers (Table (Table1).1). 16S RNA was used as a housekeeping gene. A titration of vancomycin (doubling dilutions starting from 64 μg/ml) was performed using MW2 and N315 for RT-PCR and the constructs for ONPG hydrolysis.
We generated two ccrA::lacZ fusions: pMccrA::lacZ, from MW2 (pM), and pNccrA::lacZ, from N315 (pN). Both were transduced into MW2 and N315, resulting in MW2-pM, MW2-pN, N315-pM, and N315-pN. No β-galactosidase activity was observed in wild-type MW2 or N315 or the out-of-frame construct. Baseline levels of β-galactosidase activity were observed for transductants, consistent with ccrA expression values observed by RT-PCR.
There was no significant difference in β-galactosidase activity between the constructs pM and pN. Increased activity was observed in N315 transductants (compared to the level for MW2) and upon incubation with beta-lactam antibiotics in comparison to the control levels (Table (Table2).2). Ampicillin increased the expression of the ccrA::lacZ gene, while sulbactam alone had little or no effect; however, in combination, their effect was synergistic. Heat shock lowered lacZ expression. Vancomycin increased ccrA::lacZ expression in MW2 twofold but had the opposite effect in N315 (Table (Table1).1). Other antibiotics had little or no effect and were excluded from further study. Expression patterns were reproducible in all instances.
RT-PCR confirmed these results, with the exception of heat shock, resulting in an almost fourfold increase in expression of MW2-ccrA transcripts. To further investigate the polar effect of vancomycin, a titration was performed (Fig. 1A and B). A subinhibitory concentration had limited effect on ONPG hydrolysis and N315 ccrA transcripts. In contrast, MW2-pM ß-galactosidase activity and MW2 ccrA transcripts rose to a maximum at 16 μg/ml. Small increases of ß-galactosidase activity and ccrA expression were found with N315 at 8 μg/ml and 2 μg/ml, respectively.
In the present study, we sought to determine what effect antimicrobials have on expression of ccrA. We did not investigate the expression of ccrB; however, given that these genes are in an operon, it is highly likely that ccrB is similarly expressed. We constructed a reporter system, putting the regulation of the lacZ gene under the control of ccrA and -B transcription factors. Anthony et al., using an RNA microarray, have shown that mecA is constitutively expressed in an oxacillin-resistant MRSA isolate and that treatment with mupirocin had a downregulatory effect on mecA (1). In contrast, a heterogenous oxacillin phenotype showed a much lower level of mecA expression, which was raised upon exposure to oxacillin. Steidl et al. found that cell wall stress stimulon genes were induced in the presence of cell wall-active compounds (11). Among the antibiotics that we tested, only ß-lactams and vancomycin had a marked effect on ccrA expression. Interestingly, Wielders et al. report the possible in vivo transfer of the mecA cassette into a methicillin-susceptible S. aureus isolate after treatment with ß-lactams (13). Thus, treatment with ß-lactams may have promoted the transfer of SCCmec into the previously susceptible strain. In the present study, oxacillin had the greatest effect when the ß-galactosidase assay was used. However, when RT-PCR was used, cefoxitin and ampicillin were the more powerful inducers. The finding that heat shock gave an increase in ccrA transcripts but downregulated the ß-galactosidase activity was reproducible, and we interpret this as an initial response to a stress stimulus which is either short lived or not translated, because the heat shock mechanism overrides all other metabolic events. Alternatively, incubation at 42°C may have affected ß-galactosidase.
The therapeutic use of vancomycin to treat MRSA is well established. In our current study, we found that exposure to breakpoint concentrations of vancomycin had a downregulatory effect on both ß-galactosidase activity and ccrA transcripts in an SCCmec type II strain. In contrast, in an SCCmec type IVa strain, exposure to vancomycin had an upregulatory effect with the two independent methods. Both MW2 and N315 had vancomycin MICs that were <1 μg/ml. With MW2, we detected a >2-fold increase at 2 μg/ml vancomycin, which rose to a maximum at 16 μg/ml. This interesting finding leads one to hypothesize that the use of vancomycin may potentially aid the spread of SCCmec IVa.
This work was supported in part by Deutsche Forschungsgemeinschaft (DFG) project number WI2070/2-1.
Published ahead of print on 13 July 2009.