The mecA Homolog mecC Confers Resistance against β-Lactams in Staphylococcus aureus Irrespective of the Genetic Strain Background

Abstract

In staphylococci, methicillin resistance is mediated by mecA-encoded penicillin-binding protein 2a (PBP2a), which has a low affinity for beta-lactams. Recently, a novel PBP2a homolog was described as being encoded by mecC, which shares only 70% similarity to mecA. To prove that mecC is the genetic determinant that confers methicillin resistance in Staphylococcus aureus, a mecC knockout strain was generated. The S. aureus ΔmecC strain showed considerably reduced oxacillin and cefoxitin MICs (0.25 and 4 μg/ml, respectively) compared to those of the corresponding wild-type methicillin-resistant S. aureus (MRSA) strain (8 and 16 μg/ml, respectively). Complementing the mutant in trans with wild-type mecC restored the resistance to oxacillin and cefoxitin. By expressing mecC and mecA in different S. aureus clonal lineages, we found that mecC mediates resistance irrespective of the genetic strain background, yielding oxacillin and cefoxitin MIC values comparable to those with mecA. In addition, we showed that mecC expression is inducible by oxacillin, which supports the assumption that a functional beta-lactam-dependent regulatory system is active in MRSA strains possessing staphylococcal cassette chromosome mec (SCCmec) type XI. In summary, we showed that mecC is inducible by oxacillin and mediates beta-lactam resistance in SCCmec type XI-carrying strains as well as in different S. aureus genetic backgrounds. Furthermore, our results could explain the comparatively low MICs for clinical mecC-harboring S. aureus isolates.

INTRODUCTION

Methicillin-resistant Staphylococcus aureus (MRSA) is a global, major cause of health care-, community-, and livestock-associated infections, creating enormous disease burdens and leading to substantial efforts in preventive measures (1, 2). MRSA strains have been characterized by their beta-lactam resistance phenotype, based on an additional, low-affinity penicillin-binding protein (PBP) of subclass B1, known as PBP2a or PBP2′ (3). In the case of beta-lactam treatment, PBP2a compensates for the function of blocked PBPs, thus giving the strain the ability to grow in the presence of the antibiotic (4). PBP2a is encoded by the mecA gene, which is located, together with its regulators, mecR1 and mecI, on a mobile genetic element called staphylococcal cassette chromosome mec (SCCmec) (5, 6).

Although different SCCmec elements with various genetic information and organization systems have been described and evolved over time, mecA has hitherto been found to be highly conserved. In 2011, García-Álvarez et al. (7) and Shore et al. (8) discovered the novel SCCmec element type XI in S. aureus. SCCmec XI has structural similarities to other known SCCmec elements and is integrated within the same chromosomal locus, orfX. In contrast to the previously known SCCmec elements carrying mecA, this element carries another mec homolog, originally designated mecA_LGA251 and then reclassified as mecC. However, the molecular proof that mecC confers resistance to beta-lactams in S. aureus is still lacking.

The mecC gene shares 70% identity with mecA at the DNA level (7). This significant difference led to the assumption that the proteins encoded by mecC and mecA may differ from each other in terms of their structure and function. It was recently demonstrated that PBP2a and the mecC-encoded homolog (designated PBP2c by EUCAST) differ in their binding characteristics toward beta-lactams and their temperature-dependent activities (9). PBP2c shares only 63% identity to PBP2a at the amino acid level (7). It was found that PBP2c, in contrast to PBP2a, is thermosensitive, with its activity decreased at 37°C (9). Moreover, PBP2c exhibits a 4-fold higher binding affinity for oxacillin than that of PBP2a. In accordance with this, comparatively low MICs for clinical mecC-harboring S. aureus isolates (MICs of 0.75 to 32 mg/liter for oxacillin and 4 to 64 mg/liter for cefoxitin) have been described in the literature (7, 10, 11). This raises the question of whether strains harboring either PBP2a or PBP2c differ in their resistance levels as a consequence of transcriptional, structural, and/or functional differences of their PBP2 homologs.

Besides first attempts (9), it has not been proven on the molecular level that mecC is able to function as the genetic basis for phenotypic beta-lactam resistance in SCCmec type XI-carrying S. aureus strains. In this study, we showed by cloning strategies that mecC confers beta-lactam resistance in S. aureus strains harboring SCCmec type XI. Furthermore, we showed that mecC transcription is inducible by oxacillin, indicating a functional beta-lactam-dependent regulatory system encoded within SCCmec type XI. Finally, we compared mecC and mecA with regard to their abilities to confer resistance in different S. aureus lineages, comprising MRSA as well as methicillin-sensitive S. aureus (MSSA) backgrounds, and characterized their transcriptional promoter activation under different conditions.

MATERIALS AND METHODS

Strain collection and cultivation conditions.

The strains and plasmids used in this study are shown in Tables 1 and 2, respectively. The clinical S. aureus isolate W44646 was characterized previously (10). This strain showed a beta-lactam resistance phenotype and the presence of mecC, and the presence of SCCmec type XI was confirmed by PCR (10, 12). In general, S. aureus strains were cultivated on Columbia blood agar or tryptic soy agar (TSA), and Escherichia coli strains were grown on LB. Where indicated, media were supplemented with 100 μg/ml ampicillin, 2.5 μg/ml erythromycin, or 10 μg/ml chloramphenicol. Liquid cultures of S. aureus were incubated in tryptic soy broth (TSB) at 37°C with shaking at 160 rpm. Antimicrobial susceptibility testing was performed in Mueller-Hinton (MH) medium.

TABLE 1.

Bacterial strains used in this study

Strain	Description	Source or reference
S. aureus strains
W44646	Clinical human MRSA isolate carrying mecC; CC130, ST130, SCCmec XI, spa t843	Kriegeskorte et al. (10)
RN4220	R/M system-deficient strain; derivative from NCTC8325; CC8, ST8, spa t211	Kreiswirth et al. (36)
COLn	COL derivative; CC8, ST250, SCCmec I, spa t008	Katayama et al. (37)
ME131	COLn ΔSCCmec I	McCallum et al. (38)
W44646 ΔmecC	mecC insertion knockout mutant from W44646; Eryr	This study
JE2	USA300 derivative	NARSA strain collection
NE1868	Transposon insertion in mecA in USA300 derivative JE2; Eryr	NARSA strain collection
ATCC 43300	MRSA	ATCC strain collection
ATCC 25923	MSSA	ATCC strain collection
E. coli strain
Tg1	Wild-type E. coli K-12; supE thi-1 Δ(lac-proAB) Δ(mcrB-hsdSM)5 (rK− mK−) [F′ traD36 proAB lacIqZΔM15]	Gibson (39)

TABLE 2.

Plasmids used for cloning

Plasmid	Description	Source or reference
pQE30-Xa	Cloning vector for E. coli; Ampr	Qiagen
pBT2	E. coli-S. aureus shuttle vector; Ampr in E. coli; Cnr and temperature sensitive in S. aureus	40
pExpr	E. coli-S. aureus shuttle vector from pCN-47 (14); Ampr in E. coli; Eryr in S. aureus	NARSA strain collection
pExpr1	E. coli-S. aureus shuttle vector from pCN-47 (14); Ampr in E. coli; Erys Cnr in S. aureus	This study
pExpr2	E. coli-S. aureus shuttle vector from pCN-68 (14); Ampr in E. coli; Erys Cnr in S. aureus	This study
pQE30-Xa-mecC	pQE30-Xa with mecC	This study
pBT2-mecC-40/42	pBT2 with mecC	This study
pBT2-mecC-del-Ery	pBT2-mecC-40/42 with ermB insertion in mecC	This study
pExpr1-mecA	pExpr1 with mecA (and its native promoter)	This study
pExpr1-mecC	pExpr1 with mecC (and its native promoter)	This study
pExpr2-mecA	pExpr2 with mecA	This study
pExpr2-mecC	pExpr2 with mecC	This study
pRep-PmecA	mecA promoter-GFP fusion in pCN-54 vector (14); Ampr in E. coli; Eryr in S. aureus	This study
pRep-PmecC	mecC promoter-GFP fusion in pCN-54 vector (14); Ampr in E. coli; Eryr in S. aureus	This study

mecC gene knockout.

The plasmids and primers that were used to generate the mecC knockout mutant are listed in Tables 2 and 3, respectively. The mecC gene was amplified from genomic DNA of the clinical isolate W44646 by using primers mecA-AL-40 and mecA-AL-42. The PCR product was digested with EcoRI before being subcloned into the EcoRI site of the vector pQE30-Xa. From pQE30-Xa, the mecC gene was digested with EcoRI and ligated into the EcoRI site of the E. coli-S. aureus shuttle vector pBT2, thus creating pBT2-mecC-40/42.

TABLE 3.

Oligonucleotides for PCR, sequencing, and real-time PCR

Oligonucleotide	Sequence (5′–3′)	Application	Source or reference
mecA-AL-40	GCGCGAATTCTTACTGATCTATATCAAATTGAGTTTTTCC	mecC knockout	This study
mecA-AL-42	GCGCGAATTCATGAAAAAAATTTATATTAGTGTGCTAGTTC	mecC knockout	This study
ermB-F-(BglII)	GGAGATCTATCGATACAAATTCCCCGTAGGC	mecC knockout	This study
ermB-R-(BglII)	GGAGATCTATCGATAAGAAATAGATTTAAAAATTTCGCTGTTA	mecC knockout	This study
mecA-LGA-25-seq	TTGACAAACAGACTACAAATG	Sequencing	This study
mecA-LGA-26-seq	TTTGTGGAAACGCTTCTCAC	Sequencing	This study
mecA-AL-50	TATAGGCGCGCCTTACTGATCTATATCAAATTGAGTTTTTCC	mecC in pExpr2	This study
mecA-AL-44	GCGCGGTACCGTAGTACAAAAGGAGGAAGAGATGAA	mecC in pExpr2	This study
mecA-USA-45	GCGCGGTACCTGTAGTCTTATATAAGGAGTATATTGATG	mecA in pExpr2	This study
mecA-USA-51	CGTAGGCGCGCCTTATTCATCTATATCGTATTTTTTATTACC	mecA in pExpr2	This study
mecA-AL-40	GCGCGAATTCTTACTGATCTATATCAAATTGAGTTTTTCC	mecC in pExpr1	This study
mecA-AL-48	GCGCGGTACCGTTCACACCTCACTTCTTAAC	mecC in pExpr1	This study
mecA-USA-49	GCGCGGTACCTCTACACCTCCATATCACAA	mecA in pExpr1	This study
mecA-USA-42	GCGCGAATTCTTATTCATCTATATCGTATTTTTTATTACC	mecA in pExpr1	This study
pCN-Seq-for	GTTCTTTCCTGCGTTATCCC	Sequencing	This study
pCN-Seq-rev	GCTTTTTCGATTGATGAACACC	Sequencing	This study
mec-uni-1	CCTGAATCAGCTAATAATATTTCATT	Sequencing	This study
mecA-AL-seq 38	ATTAAAATCAGAGCGAGGCAAA	Sequencing	This study
mecA-CC398-seq 39	TGCAGAAAGACCAAAGCATAC	Sequencing	This study
mecA-CC398-seq 40	TTTCTACTTCACCATTATCG	Sequencing	This study
mecA-AL-52	GCGCGCATGCGTTCACACCTCACTTCTTAAC	mecC promoter	This study
mecA-AL-53	GCGCGAATTCCTCTTCCTCCTTTTGTACTA	mecC promoter	This study
mecA-USA-54	GCGCTGCATGCTCTACACCTCCATATCACAA	mecA promoter	This study
mecA-USA-55	CGCGCCGAATTCCAATATACTCCTTATATAAGAC	mecA promoter	This study
mecA_LGA251 for	ACTAGTATCTCGCCTTGG	RT-PCR (mecC)	This study
mecA_LGA251 ref	ATCCCGAGTGATTATCCC	RT-PCR (mecC)	This study
mecI_LGA251 for	ATGGGAAATAATGAATACGATTTGG	RT-PCR (mecISCCmecXI)	This study
mecI_LGA251 ref	AGACGATTGATTAGTGTTCTTATTG	RT-PCR (mecISCCmecXI)	This study
mecR1_LGA251 for	TAAACGGCAAGGAAACGAATGG	RT-PCR (mecR1SCCmecXI)	This study
mecR1_LGA251 ref	CCACTGGCATTATTCTTACC	RT-PCR (mecR1SCCmecXI)	This study
16S rRNA 1	CGGTCTTGCTGTCACTTATAG	RT-PCR (16S rRNA gene)	This study
16S rRNA 2	CTCTCAGGTCGGCTATGC	RT-PCR (16S rRNA gene)	This study
blaI for	GCAAGTTGAAATATCTATGG	blaI PCR	34
blaI rev	CATTTTCTGTACACTCTCATC	blaI PCR	Modified from reference 34
blaR1-for	CATGACAATGAAGTAGAAGC	blaR1 PCR	34
blaR1-rev	CTTATGATTCCATGACATACG	blaR1 PCR	34

To generate the mecC knockout vector, the erythromycin resistance cassette was cloned into the mecC gene. The ermB gene was amplified from the vector pCE8 by use of the primers ermB-F-(BglII) and ermB-R-(BglII). The PCR product was digested with BglII and cloned into the mecC gene within the vector pBT2-mecC-40/42, also digested with BglII. The resulting knockout vector, pBT2-mecC-del-Ery, was subcloned into S. aureus RN4220 before being transformed into the clinical isolate S. aureus W44646 and cultivated on TSA supplemented with 2.5 μg/ml erythromycin and 10 μg/ml chloramphenicol at 30°C. Insertional knockout of mecC was performed as described before (13), and mutants were checked by PCR, using primer combinations mecA-LGA-26-seq plus mecA-AL-42, mecA-LGA-25-seq plus mecA-AL-40, and mecA-LGA-25-seq plus mecA-LGA-26-seq.

mecA and mecC expression plasmids.

The plasmids and primers that were used for the cloning approaches are listed in Tables 2 and 3, respectively. First, the vectors pExpr1 and pExpr2 were generated from the vectors pCN47 (pExpr) and pCN68 (14) by replacing the erythromycin resistance cassette (ermC) with a chloramphenicol resistance cassette (cat194). The plasmids pExpr1 and pExpr2, both mediating chloramphenicol resistance in S. aureus, were applied for expression of mecA and mecC.

To express mecC and mecA under the control of their native promoters, the genes were cloned into the vector pExpr1. mecC was amplified from S. aureus W44646 by use of primers mecA-AL-40 and mecA-AL-48, and S. aureus USA300 was used to amplify mecA (by use of primers mecA-USA-49 and mecA-USA-42). Subsequently, both PCR products and the pExpr1 vector were cleaved with EcoRI and KpnI. pExpr1 was ligated with mecA or mecC and transformed into E. coli Tg1. After selection of clones, the clones were tested by PCR and confirmed by sequencing (using primers pCN-Seq-for, pCN-Seq-rev, mec-uni1, mecA-AL-seq 38, and mecA-CC398-seq 39). The appropriate plasmids were transformed into different S. aureus strains.

In a second approach, the mecA and mecC genes were expressed independently of their native promoters, instead being controlled by the blaZ promoter of the vector pExpr2. The mecC gene was amplified from S. aureus W44646 by use of primers mecA-AL-44 and mecA-AL-50, and mecA was amplified from S. aureus USA300 by use of primers mecA-USA-45 and mecA-USA-51. The PCR products and the pExpr2 vector were each restriction digested with KpnI and AscI. After ligating the PCR product and the vector, the resulting plasmids were transformed into E. coli Tg1 and selected. The clones were tested by PCR and confirmed by sequencing (using primers pCN-Seq-for, pCN-Seq-rev, mec-uni1, mecA-AL-seq 38, and mecA-CC398-seq 39). The appropriate plasmids were transformed into S. aureus.

GFP reporter plasmids.

To generate the mecA promoter- and mecC promoter-GFP fusion constructs, the vector pRep (14) with the gfpmut2 fragment, encoding green fluorescent protein (GFP), was used. The mecA promoter and mecC promoter sequences were amplified from strains USA300 and W44646 by using the primers mecA-USA-55 plus mecA-USA-54 and mecA-AL-52 plus mecA-AL-53. The pRep vector and PCR products were restriction digested with SphI and EcoRI, ligated, transformed into E. coli Tg1, and selected on erythromycin. After confirmation of the clones by PCR and sequencing (primer pCN-Seq-for), the plasmids were transferred into S. aureus RN4220, W44646, JE2, and COLn.

GFP reporter assay.

Bacterial cultures were grown in TSB medium with 2.5 μg/ml erythromycin and either 2 μg/ml oxacillin (Fig. 1), 0.2 μg/ml oxacillin (Fig. 2), or no oxacillin. The optical density at 578 nm (OD₅₇₈) was measured at hourly intervals. In parallel, the GFP fluorescence intensity of a 100-μl bacterial suspension was monitored with a SpectraMax Gemini XS spectrofluorometer (Molecular Devices, Sunnyvale, CA) at a 485-nm excitation wavelength and a 538-nm emission wavelength. The relative fluorescence unit (RFU) values were finally normalized to the OD (RFU/OD). As a control, the respective S. aureus strains carrying the pExpr empty vector (grown in TSB with 2.5 μg/ml erythromycin) were tested.

FIG 1.

Transcriptional activation of mecC, mecI_SCCmecXI, and mecR1_SCCmecXI in response to oxacillin in the SCCmec XI-carrying strain W44646. (a) n-Fold expression of mecC, mecI_SCCmecXI, and mecR1_SCCmecXI during early exponential growth in cultures incubated with 2 μg/ml oxacillin (white bars) or without oxacillin (black bars), as measured by quantitative real-time reverse transcription-PCR. (b) The promoter activity (RFU/OD) of mecC was measured over time in the presence of 2 μg/ml oxacillin (white bars) or the absence of oxacillin (black bars), using a GFP reporter assay.

FIG 2.

Reporter measurements of GFP fused to the mecA or mecC promoter. S. aureus strains JE2, W44646, and COLn, expressing plasmid-encoded GFP fused to the promoter of mecA (striped bars indicate RFU/OD, and triangles [▲] indicate OD₅₇₈ values) or mecC (black bars indicate RFU/OD, and diamonds [♦] indicate OD₅₇₈ values), were grown without oxacillin (a, c, and e) and with 0.2 μg/ml oxacillin (b, d, and f). As a control, the respective S. aureus strains carrying the pExpr empty vector (squares [■] indicate OD₅₇₈ values) were tested (a, c, and e).

Susceptibility testing.

Susceptibility testing was performed by the microdilution method according to CLSI guidelines. Antibiotic solutions (oxacillin and cefoxitin) were prepared in H₂O and diluted gradually from 1.024 μg/ml to 0.03125 μg/ml in 96-well plates. The plates were incubated for 24 h (cefoxitin) and 48 h (oxacillin) at 37°C before MIC values were determined.

Gene expression analysis via RNA isolation, reverse transcription, and real-time PCR.

The clinical S. aureus isolate W44646, carrying mecC within SCCmec XI, was used for gene expression measurements by real-time PCR. Colonies were inoculated from blood agar into 50 ml TSB with (2 μg/ml) or without oxacillin at an OD₅₇₈ of 0.05. For RNA isolation, samples were taken from the early exponential growth phase. Two milliliters of bacterial suspension was incubated for 5 min with 4 ml RNAprotect (Qiagen) and then centrifuged (10 min, 4°C, 4,000 × g). The pellet was resuspended in RNApro solution (MP Biomedicals), followed by mechanical disruption of cells. Subsequently, RNA was isolated by use of an RNeasy minikit (Qiagen) and an on-column DNase treatment (RNase-free DNase set; Qiagen). One microgram of RNA was used for reverse transcription by use of a QuantiTect reverse transcription kit (Qiagen) according to the manufacturer's protocol.

After reverse transcription, the cDNA was applied to real-time PCR. A CFX96 system (Bio-Rad Laboratories) was used for the measurements. Reaction mixtures were prepared with an EvaGreen kit (Segentic, Borken, Germany), and PCR was performed under the following conditions: 95°C for 15 min followed by 50 cycles of 10 s at 95°C, 10 s at 55°C, and 30 s at 72°C. Subsequent melting curve analyses were performed to evaluate the PCR specificity. 16S rRNA gene expression was used for normalization of expression levels. Gene expression rates were calculated using the software CFX Manager v 2.0 (Bio-Rad Laboratories).

RESULTS

mecC mediates resistance to cefoxitin and oxacillin in S. aureus strains harboring SCCmec XI.

To prove whether the mecC gene confers beta-lactam resistance in S. aureus, a mecC knockout was generated in the SCCmec type XI-harboring S. aureus strain W44646. The mecC knockout strain, W44646 ΔmecC, showed considerably reduced cefoxitin and oxacillin MICs compared to those for the wild-type strain, as tested by microdilution (Table 4). Cefoxitin and oxacillin MICs decreased from 16 to 4 μg/ml and from 8 to 0.25 μg/ml, respectively. Complementing W44646 ΔmecC in trans by using mecC (pExpr1-mecC) restored the resistance to oxacillin and cefoxitin (MIC of 64 μg/ml for cefoxitin and oxacillin). In accordance with these findings, the W44646 ΔmecC knockout mutant showed no growth on selective MRSA agar (ChromID MRSA; bioMérieux, Nürtingen, Germany), whereas the wild type and the W44646 ΔmecC/pExpr1-mecC complemented mutant were able to grow on this medium.

TABLE 4.

Susceptibility to cefoxitin and oxacillin of S. aureus wild-type and ΔmecC knockout mutant strains

Strain	MIC (μg/ml)
	Cefoxitin	Oxacillin
W44646	16	8
W44646 ΔmecC	4	0.25
W44646 ΔmecC/pExpr1-mecC	64	64
W44646 ΔmecC/pExpr1a	4	0.25

^aΔmecC knockout mutant carrying the vector without an insert.

Oxacillin induces mecC gene expression.

The expression levels of the mecC, mecR1_SCCmecXI, and mecI_SCCmecXI genes were analyzed in the SCCmec XI-harboring strain W44646. The transcription profiles of mecC, mecR1_SCCmecXI, and mecI_SCCmecXI were determined during growth of S. aureus W44646 by quantifying the respective RNA levels by use of real-time PCR. We analyzed changes in the transcriptional activation of mecC, mecR1_SCCmecXI, and mecI_SCCmecXI in response to a subinhibitory concentration (2 μg/ml) of the beta-lactam oxacillin. Interestingly, the transcription of mecC was strongly enhanced (Fig. 1a): it was found to be induced approximately 100-fold by oxacillin. Like that of mecC, the transcription of mecR1_SCCmecXI and mecI_SCCmecXI was found to be activated in the presence of oxacillin. Oxacillin led to 9- and 10-fold increases in mecR1_SCCmecXI and mecI_SCCmecXI transcription, respectively.

In addition to real-time PCR analysis, mecC-promoter-GFP fusion assays were conducted. In line with the real-time PCR results (Fig. 1a), we found a strong activation of the mecC promoter over time, in an oxacillin-dependent manner (applying 2 μg/ml oxacillin) (Fig. 1b). The promoter-GFP fusion assay showed that mecC transcription is inducible by oxacillin.

Like mecA, mecC mediates resistance in various strain backgrounds.

After introduction of the mecA and mecC genes in trans into methicillin-susceptible S. aureus strains (RN4220, W44646 ΔmecC, and NE1868), the MIC values of these strains increased considerably, and thus the strains were categorized as oxacillin and cefoxitin resistant. The plasmid-carried mecC gene—like mecA—conferred resistance to oxacillin and cefoxitin in all S. aureus strains. In the pExpr1 vector, mecC and mecA were expressed under the control of their native promoters (Table 5). Introducing pExpr1-mecC into S. aureus RN4220, ME131, NE1868, and W44646 ΔmecC led to increases in cefoxitin and oxacillin MICs ranging from 32 to 256 μg/ml and from 32 to 128 μg/ml, respectively. For mecA (pExpr1-mecA), we noted MICs of cefoxitin and oxacillin between 16 and ≥1,024 μg/ml and between 32 and ≥1,024 μg/ml, respectively.

TABLE 5.

Susceptibility to cefoxitin and oxacillin of S. aureus wild-type strains and mutants expressing mecA or mecC in trans, in vectors pExpr1 and pExpr2

Strain	MIC (μg/ml)
	pExpr1		pExpr2
	Cefoxitin	Oxacillin	Cefoxitin	Oxacillin
RN4220a	2	0.125	2	0.125
RN4220-mecA	16	64	16	4
RN4220-mecC	32	32	16	8–16
COLnb	256	512	256	512
ME131a	4	0.5	4	1
ME131-mecA	≥1,024	≥1,024	128	128
ME131-mecC	256	128	64	32
JE2b	32	64	32	64
NE1868a	4	0.25	2	0.25
NE1868-mecA	32	32	8	16
NE1868-mecC	32–64	32–64	8	2–4
W44646b	16	8	16	8
W44646 ΔmecCa	4	0.25	4	0.25
W44646 ΔmecC-mecA	128	256	16	16
W44646 ΔmecC-mecC	64	64	8	4–8

^aStrain carrying the respective vector without an insert (pExpr1 or pExpr2).

^bWild-type MRSA isolate.

In ME131, NE1868, and W44646 ΔmecC, the resistance to oxacillin and cefoxitin was restored and comparable to that of the respective parental strains (COLn, JE2, and W44646) after complementation with pExpr1-mecA or pExpr1-mecC in trans. To compare the impacts of mecC and mecA directly within the same genetic background and under the control of a standardized promoter, we cloned both mecC and mecA into the vector pExpr2, which harbors the blaZ promoter (14) (Table 5).

Expression of mecC (within pExpr2) in S. aureus RN4220, ME131, NE1868, and W44646 ΔmecC yielded cefoxitin MICs between 8 and 64 μg/ml and oxacillin MICs between 2 and 32 μg/ml. For pExpr2-mecA, we found MICs for cefoxitin and oxacillin of 8 to 128 μg/ml and 4 to 128 μg/ml, respectively.

Overall, plasmids in the ME131 background exhibited higher cefoxitin and oxacillin MICs than those in RN4220, NE1868, and W44646 ΔmecC, for both mecA and mecC. Comparing mecA and mecC directly, mecA yielded higher MICs than those with mecC in strains ME131 and W44646 ΔmecC. Furthermore, mecC constructs exhibited slightly higher MICs for cefoxitin than for oxacillin in the ME131 background and in all transformants carrying pExpr2-mecC. Apart from that, the absolute MIC values varied in a wide range depending on the different strain backgrounds. Although strains RN4220, COLn, and JE2 all belong to the clonal lineage CC8, they behave differently in terms of beta-lactam resistance and exhibited different oxacillin and cefoxitin MIC values when expressing mecA or mecC in trans.

mecC shows lower transcriptional promoter activity than that of mecA.

The promoter activities of mecA and mecC were compared using a GFP-promoter fusion assay. To figure out whether the promoter activity is strain dependent, we again compared different genetic strain backgrounds. The promoter fusion constructs were generated in identical manners, making it possible to compare the expression levels directly. S. aureus reporter strains (carrying the GFP-promoter fusions) were cultivated in TSB medium (with and without 0.2 μg/ml oxacillin). The reporter activities were measured via fluorescence intensities during growth of the culture (Fig. 2).

In all strains, JE2, COLn, and W44646, the mecA promoter led to more GFP expression than that with the mecC promoter. In the presence of oxacillin, a similar pattern was observed. The mecA promoter activity was shown to be higher than that of the mecC promoter with oxacillin induction in the different S. aureus backgrounds.

DISCUSSION

Since its first description in 1961, shortly after the introduction of methicillin, MRSA has become one of the most significant multiresistant pathogens worldwide, first within hospitals and, more recently, also in the community and including livestock-associated strains (15–20). So far, mecA has been the sole known genetic resistance determinant in MRSA, and it encodes the low-affinity penicillin-binding protein PBP2a (5, 21).

Recently, MRSA strains with a novel SCCmec XI element and a novel penicillin-binding protein gene homolog (mecC) were described in the United Kingdom and Denmark (7, 8). After their first description, SCCmec XI-carrying strains were identified in animal as well as human specimens in many European countries (10, 11, 22–25). Because mecC exhibits 70% identity to mecA on the nucleotide level and PBP2c shows 63% identity to PBP2a on the protein level, and because the majority of mecC-positive isolates were shown to be beta-lactam resistant, it was assumed that mecC is a novel genetic determinant mediating beta-lactam resistance in S. aureus. However, mecC has not been proven genetically to be the resistance determinant conferring non-beta-lactamase-caused beta-lactam resistance in S. aureus. In this study, we generated, for the first time, a mecC knockout mutant in an SCCmec XI background.

Using the mecC knockout mutant, we could clearly verify mecC as the genetic determinant conferring beta-lactam resistance. Insertional disruption of mecC led to decreased MICs for oxacillin and cefoxitin that could be categorized as susceptible. In line with these findings, we demonstrated that mecC gene transcription is activated in S. aureus strain W44646 (with the SCCmec XI background) in response to oxacillin. In accordance with the transcriptional real-time PCR experiments, promoter-GFP fusion assays showed that mecC transcription is inducible by oxacillin. However, the findings support the assumption that beta-lactam-dependent sensor and regulatory mechanisms encoded within SCCmec XI are functional. For mecA, it has been shown that MecI and MecR1, both encoded within SCCmec (class A mec gene complex), are involved in the transcriptional regulation of mecA (26, 27). In SCCmec type XI, the corresponding regulatory elements, mecI_SCCmecXI and mecR1_SCCmecXI, were found directly upstream of mecC. Proteins encoded by mecI_SCCmecXI and mecR1_SCCmecXI showed 66% and 45% amino acid identities to MecI and MecR1 proteins, respectively, encoded by previously described SCCmec elements (8). MecR1 has been identified as a transmembrane protein acting as a beta-lactam sensor and signal transducer, and MecI is a repressor that binds to the mecA promoter region. mecA transcription is induced after the repressor dimer is cleaved by MecR1. In the absence of the mecR1-mecI system, mec is transcribed constantly. Within our gene expression studies, we observed that the transcription of mecR1_SCCmecXI and mecI_SCCmecXI was enhanced in response to oxacillin.

In addition to mecR1-mecI, the blaR1-blaI system has been shown to be involved in the control of mecA transcription (28, 29). The mecR1-mecI and blaR1-blaI regulation systems are structurally and functionally related and show similar signaling pathways (30, 31). BlaR1 and BlaI are described as sensor-transducer and repressor domains, respectively, and play a role in regulating the transcription of mecA.

To identify differences in the regulation of mecA and mecC in different genetic backgrounds, we compared the transcriptional activations of mecA and mecC in S. aureus strains COLn, W44646, and JE2 by use of a promoter-GFP reporter assay. We found a higher mecA promoter activity than that of the mecC promoter on the transcriptional level, independent of the strain background. Since mec transcription is directly regulated by MecI or BlaI, one can suppose that binding of BlaI and/or MecI to the operator region results in decreased expression of mecC.

Among the strains tested in this study, only S. aureus W44646 harbors an intact mec regulon within its SCCmec type XI element (class E mec gene complex). The MRSA strains COLn and JE2 carry the SCCmec types I and IV, respectively, both with a class B mec gene complex, in which the mecI gene is absent and mecR1 is truncated (32). The BlaR1-BlaI regulatory system, which is known to affect PBP2a synthesis and beta-lactam resistance (33), was found to be absent in S. aureus JE2, COLn, and W44646 (the genes could not be detected by PCR using previously described primers [34]). Thus, among the strains tested, only S. aureus W44646 harbors a mec repressor: MecI_SCCmecXI. Since mecA transcription seems not to be repressed (in the absence of oxacillin) in S. aureus W44646 (Fig. 2c), we assume that mecA expression is probably not regulated by the MecI_SCCmecXI repressor. Comparing promoter and operator sequences of mecA (in S. aureus COL and USA300/JE2) and mecC (in S. aureus LGA251), we found differences in the nucleotide sequences, in particular in the MecI binding site (Fig. 3). The 4-bp inverted repeats separated by two nucleotides (TACA/TGTA DNA binding motif) in the MecI-protected regions are identical within the mecA and mecC operators. In contrast to the mecA operator, which consists of a 30-bp palindrome (with two connected 15-bp sites), the mecC operator contains palindromic regions (14 bp each) that are separated from each other by an 8-bp linker. We concluded that the repressor MecI_SCCmecXI might interact more specifically with the mecC operator than with the mecA operator. This assumption is supported by the fact that the MecI repressor protein encoded within SCCmec XI shares only 66% amino acid identity to MecI proteins encoded by previously described SCCmec elements (8).

FIG 3.

Operator-promoter sequences of mecA and mecC. The nucleotide sequences upstream of mecA and mecC are shown, and the predicted −10 and −35 promoter regions (gray), the ribosomal binding site (RBS), and the MecI-protected palindromic region (arrows), including 4-bp inverted repeats separated by two nucleotides (TACA/TGTA DNA binding motif), are highlighted. The nucleotide sequence upstream of mecA is from S. aureus USA300 and COL (accession no. CP000730 and CP000046), and that upstream of mecC is from S. aureus LGA251 (accession no. NC_017349).

In comparing the nucleotide sequences upstream of the mecA and mecC start codons, we detected further differences, in particular in the predicted −10 region of the promoter (Fig. 3). The −10 promoter sequence of mecA is TATACT (in S. aureus COL and USA300/JE2), and that for mecC was found to be TATTAT (in S. aureus LGA251). For mecA, it was observed before that mutations in the −10 promoter region affect mecA transcription (35). We assume that the differences in the mecA and mecC promoters influence transcription and result in a decreased transcriptional activation of mecC, as detected within GFP reporter assays.

However, a higher mecA/mecC expression level cannot generally be correlated with more resistance to beta-lactams, as stated before (28, 35).

In order to determine to what extent the different PBP2 homologs affect the beta-lactam resistance levels, we introduced plasmid-carried mecA and mecC into methicillin-susceptible S. aureus. With this approach, we could directly compare the abilities of mecA and mecC to confer beta-lactam resistance within the same genetic background. Notably, mecC mediated resistance independently of the tested S. aureus strain backgrounds, comprising MSSA (RN4220) as well as MRSA (W44646 ΔmecC, NE1868, and ME131) strains. In general, in all S. aureus strains included, we observed increases of oxacillin and cefoxitin MICs. Thus, both mecC and mecA conferred cefoxitin and oxacillin resistance in all S. aureus lineages tested.

In a direct comparison of mecA and mecC, mecA-carrying strains revealed slightly higher MICs of cefoxitin and oxacillin than those of mecC-carrying strains, at least for ME131 and W44646 ΔmecC. This correlates with the results of the promoter-GFP reporter assay and with observations recently made by Kim et al. (9). They demonstrated that PBP2c exhibits a 4-fold higher binding affinity for oxacillin than that of PBP2a, suggesting that strains carrying PBP2c exhibit less resistance against oxacillin than strains with PBP2a. Altogether, the results could explain the comparatively low MICs of oxacillin for mecC-harboring clinical S. aureus isolates, as described in the literature (7, 10).

Interestingly, cefoxitin and oxacillin MICs varied strongly between the different S. aureus strains into which either mecA or mecC was introduced. The mecA- and mecC-mediated resistance phenotypes were found to be more strain background dependent than gene dependent.

Further studies investigating the mechanisms influencing the regulation of PBP2a/PBP2c, its processing, and its activity, thereby affecting oxacillin and cefoxitin resistance levels, are warranted.

In conclusion, this is the first study proving mecC to be the genetic determinant conferring beta-lactam resistance in SCCmec type XI-carrying S. aureus. Furthermore, we found mecC to mediate oxacillin and cefoxitin resistance, like mecA, irrespective of the genetic strain background. Transcription of mecC was observed to be oxacillin inducible, indicating the presence of a functional beta-lactam-dependent regulatory system in S. aureus strains harboring the SCCmec XI element.