Fate of Mutation Rate Depends on agr Locus Expression during ...

0 downloads 0 Views 922KB Size Report
Aug 6, 2009 - The Methodist Hospital Research Institute, Houston, Texas 77030. Received 6 ..... PCR) by using the SYBR green-based SensiMix one-step kit (Quantace/Bioline,. Tauton, MA) ...... Jeffress Memorial Trust (A.E.R.). Microarray ... Special thanks goes to Kathryn Stockbauer and Philip Randall, from the Office of ...
ANTIMICROBIAL AGENTS AND CHEMOTHERAPY, July 2011, p. 3176–3186 0066-4804/11/$12.00 doi:10.1128/AAC.01119-09 Copyright © 2011, American Society for Microbiology. All Rights Reserved.

Vol. 55, No. 7

Fate of Mutation Rate Depends on agr Locus Expression during Oxacillin-Mediated Heterogeneous-Homogeneous Selection in Methicillin-Resistant Staphylococcus aureus Clinical Strains䌤† Konrad B. Plata, Roberto R. Rosato, and Adriana E. Rosato* Center for Molecular and Translational Human Infectious Diseases Research, The Methodist Hospital Research Institute, Houston, Texas 77030 Received 6 August 2009/Returned for modification 2 October 2009/Accepted 14 April 2011

Methicillin-resistant Staphylococcus aureus (MRSA) strains are characterized by a heterogeneous expression of resistance. We have previously shown in clinical oxacillin-susceptible, mecA-positive MRSA strains that selection from a very heterogeneous (HeR) to highly homogeneous (HoR) resistant phenotype was mediated by acquisition of mutations through an oxacillin-induced SOS response. In the present study, we used a spotted DNA microarray to evaluate differential gene expression during HeR-HoR selection and found increased expression of the agr two-component regulatory system. We hypothesized that increased expression of agr represents a mechanistically relevant component of this process. We demonstrated that inactivation of agr during the HeR-HoR selection process results in a significant increase in mutation rate; these effects were reversed by complementing the agr mutant. Furthermore, we found that extemporal ectopic expression of agr and, more specifically, RNAII in agr-null mutant HeR cells suppressed mutation frequency and the capacity of these cells to undergo the HeR-HoR selection. These findings sustain the concept that increased expression of agr during HeR-HoR selection plays a critical role in regulating the ␤-lactam-induced increased mutation rate in very heterogeneous MRSA strains. Moreover, they indicate that a temporally controlled increase in agr expression is required to tightly modulate SOS-mediated mutation rates, which then allows for full expression of oxacillin homogeneous resistance in very heterogeneous clinical MRSA strains. quorum-sensing system (37) that coordinates the expression of many gene products that are required for staphylococcal virulence and infection and other accessory gene functions (17). It suppresses expression of surface proteins and upregulates expression of several exoproteins (34). The Agr system consists of RNAII and RNAIII, two divergent transcripts which are transcribed from P2 and P3 promoters, respectively (38). RNAII encodes the structural components of the quorumsensing system, including AgrBDCA (42). AgrC is a transmembrane protein functioning as a histidine kinase, which is the sensory component of the two-component regulatory system (27). AgrD encodes an autoinduction signal that is processed and transported by AgrB (52); when a threshold concentration of the autoinducing peptide (AIP) is detected in the environment, AgrC undergoes autophosphorylation (18). Phosphorylated AgrC transduces the information to the response regulator AgrA, which, once activated, transcriptionally activates P2 and P3 promoters, increasing RNAII and RNAIII levels (35). Although RNAIII encodes delta-toxin, it is the RNAIII molecule itself that is the regulatory molecule of the Agr system (38, 39). Moreover, according to current knowledge, modulation of target expression by agr is achieved via RNAIII (38), which represents a paradigmatic example of virulence factor control via a small regulatory RNA (sRNA) (43) and the central role of sRNAs in quorum sensing (48). RNAIII generally acts by an antisense base-pairing mechanism (2) and regulates many target genes via control of a repressor protein called rot, a member of the sarA family of transcriptional regulators (2, 31, 44). RNAII-dependent mechanisms have also been described, notably in the case of phenol-soluble modulins (PSMs). These proteins play a key role in S. aureus immune

Methicillin-resistant Staphylococcus aureus (MRSA) is one of the most important nosocomial pathogens, responsible for a wide range of severe infections such as endocarditis, osteomyelitis, and sepsis. MRSA is an increasing threat both in clinical settings and, more recently, in community populations (16). In S. aureus, methicillin resistance is due to the acquisition of a large DNA element, termed staphylococcal cassette chromosome mec (SCCmec), which integrates site and orientation specifically into the S. aureus chromosome (19, 21). The prerequisite for methicillin resistance located on SCCmec is mecA, which encodes a ␤-lactam-insensitive penicillin binding protein (PBP), PBP2a, that can continue to cross-link the cell wall once the native PBPs (i.e., PBP1 to ⫺4) have been inactivated (22). One of the main characteristics of MRSA is its heterogeneous expression of resistance. In some isolates with methicillin or oxacillin (Oxa) resistance, only a small portion (ⱕ0.1%) of the population expresses resistance to ⱖ10 ␮g of oxacillin per ml (heterotypic resistance [HeR]), while in other isolates, most of the population expresses resistance to a high level (homotypic resistance [HoR]) (11, 49). Besides antimicrobial resistance, S. aureus also has the potential to cause serious and diverse forms of infections. The accessory gene regulator (agr) is a typical Gram-positive

* Corresponding author. Mailing address: The Methodist Hospital Research Institute, 6670 Bertner Ave., RIB 6-113, Houston, TX 77030. Phone: (713) 441-4369. Fax: (713) 441-7295. E-mail: aerosato@tmhs .org. † Supplemental material for this article may be found at http://aac .asm.org/. 䌤 Published ahead of print on 2 May 2011. 3176

VOL. 55, 2011

agr IN MRSA WITH LOW-LEVEL RESISTANCE

evasion and are regulated by direct binding of the AgrA response regulator to the psm promoter region (41). We have demonstrated in previous studies that, in very heterogeneous MRSA clinical strains (mecA positive, oxacillin susceptible), selection from an HeR to HoR phenotype by subinhibitory concentrations of ␤-lactam antibiotics was associated with a mutational event through the triggering of a ␤-lactam-mediated SOS response that involved lexA and recA genes (9). In an effort to expand our previous work, we demonstrat in the present study that this process is also associated with increased transcription levels of agr and agr-regulated genes. Furthermore, our results indicate that the agr two-component regulator plays an important role in modulating the mutation rate generated through an oxacillin-mediated SOS response. The present findings provide new insights into the impact and role of agr in the clinically relevant oxacillin-resistant phenotype in heterogeneous MRSA strains. MATERIALS AND METHODS Materials and media. Mueller-Hinton agar (MHA; BBL Microbiology Systems, Cockeysville, MD) and tryptic soy agar with 5% sheep blood (TSA II; BD Diagnostics System, Franklin Lakes, NJ), tryptic soy agar (TSA; Difco Laboratoires, Detroit, MI) with and without additives (Sigma, St. Louis, MO; United States Biochemicals, Cleveland, OH), and Luria-Bertani (LB) broth (USB Corporation, Cleveland, OH) were used for subculture and maintenance of S. aureus strains. When selection was necessary, chloramphenicol (Cm; 15 ␮g/ml), tetracycline (5 ␮g/ml), rifampin (100 ␮g/ml), or oxacillin (0.5 ␮g/ml or 128 ␮g/ml) was added to the culture medium (Sigma-Aldrich, St. Louis, MO). Bacterial strains. S. aureus oxacillin-susceptible mecA-mediated strains were provided by Fred Tenover, Gordon Archer, and Janet Swenson of the Centers for Disease Control and Prevention (CDC) in Atlanta, GA. S. aureus 13011 (9), SA200643, and SA200644 are representative strains of a collection used in this study. S. aureus N315 (a standard reference strain for genome analysis preMRSA) was used as a control (24). Phenotypic methods to detect methicillin resistance. Disk diffusion susceptibility tests and MICs were performed following the method and guidelines of the Clinical and Laboratory Standards Institute (CLSI; NCCLS M2-A8) (7). Selection in broth from the SA13011 HeR-HoR phenotype. Selection of SA13011 from the heterotypic (HeR) to the homotypic (HoR) resistance phenotype was performed as follows. The bacteria were grown overnight in 5 ml LB broth without antibiotic. Cultures were then diluted to an optical density at 600 nm (OD600) of ⬃0.025 in 100 ml LB broth either with or without 0.5 ␮g/ml oxacillin (Sigma) and were grown at 37°C with shaking (180 rpm). The ODs were monitored every hour for 35 h. Growing cells were finally streaked onto an oxacillin gradient plate with a concentration gradient from 0 to 128 ␮g/ml. Selection from heterotypic to homotypic cells was verified by resistance population analysis as described by Chambers et al. (4), except that methicillin was substituted for oxacillin. Determination of mutation frequency. Mutation frequencies for resistance to rifampin were determined during the SA13011 HeR (without oxacillin [⫺Oxa])to-HoR (HeR with 0.5 ␮g/ml Oxa [⫹Oxa]) selection process. Inoculated flasks were incubated at 37°C with shaking at 145 rpm; aliquots of 100 ␮l were taken at different time intervals, including 3, 6, 9, 27, 30, and 33 h. All of the variants were selected on TSA plates containing 100 ␮g/ml rifampin and TSA plates containing serial dilutions to determine CFU/ml. Mutation frequencies were expressed as the number of antibiotic-resistant mutants recovered as a fraction of the viable count. Three independent cultures were sampled in triplicate to minimize error caused by inter- and intrasample variation. agr phenotype and genotype studies. ␦-Hemolysin (Hld) is encoded by the hld gene within the RNAIII region (20). ␦-Hemolysin production was measured by streaking SA13011-HeR and derivatives adjacent to a ␤-hemolysin disk (Remel, Lenexa, KS) on a tryptic soy agar plate with 5% sheep blood, incubating them at 37°C overnight, and evaluating for synergistic hemolysis within the ␤-hemolysin zone produced by the disk containing bacterial growth. The presence of synergistic hemolysis within the ␤-hemolysin zone would indicate the production of ␦-hemolysin by the test organism and therefore functional agr locus, while agr dysfunction was defined as complete absence of ␦-hemolysin, evidenced by the lack of synergistic hemolysis. The S. aureus 13011 agr group was determined by

3177

multiplex PCR with primers Pan, agr1, agr2, agr3, and agr4 as described by Gilot et al. (14). Strain COL was used as a reference strain for agr group I, S. aureus strain N315 was the reference strain for agr group II, RN8462 was the reference strain for agr group III, A980740 was the reference strain for agr group IV, and RN6911 was the reference strain for the agr-null group. Cloning, transformation, and DNA manipulation. Chromosomal DNA was prepared by using a genomic DNA preparation kit (Qiagen, Chatsworth, CA) according to the manufacturer’s directions. All restriction endonuclease digestions and ligations were performed in accordance with the manufacturer’s (New England BioLabs, Beverly, MA.) specifications. Sequence analysis was performed by using the automated laser fluorescence technique employing fluorescein-labeled oligonucleotides (Applied Biosystems, Foster City, CA). Consensus sequences were assembled from both orientations with Vector NTI Advance 10 software for Windows (InforMax, Bethesda, MD). S. aureus N315 (accession no. BA000018) was used as a positive control. Construction of SA13011 agr mutant (null) and complementation. To generate an agr-null mutant, the agr deletion in RN6911 (38) was transduced into the clinical strain SA13011-HeR by phage 80␣-mediated transduction as previously described (36). The SA13011 agr mutant was named LR3-HeRagr. trans-complementation of agr was performed by using a construct encompassing the complete agr operon. Transacting products that are required for complete agr activity, including an upstream region putative ribosomal binding site and agr promoters (P2 and P3), were amplified by using primers Agr-FL-F and AgrFL-R (Table 1). The 3.8-kb fragment PCR products were purified with the QIAquick gel extraction kit, ligated into the PCR 2.1-TOPO vector (Invitrogen, Carlsbad, CA), and transformed into chemically competent Escherichia coli TOP10 cells (Invitrogen). A staphylococcal origin of replication was introduced by cloning plasmid pSK265 into the unique BamHI site on PCR 2.1-TOPO, and the construct was moved into S. aureus RN4220 by electroporation (11). Phage 80␣ was used to transduce the constructs from RN4220 into LR3-HeRagr, generating the complemented strain LR4HeRagr::agr (Table 1). Identical approaches were used for generation of the agr-null mutant and complemented strains in MRSA heterogeneous strains SA2006044 and SA2006043 (obtained from CDC); these strains have the same phenotype as SA13011. The agr-null mutants were named SA43-HeRagr and SA44-HeRagr, and their respective complemented strains were named SA43-HeRagr::agr and SA44-HeRagr::agr. Plasmid curing. Strain LR7-HeR (SA13011-HeR expressing pSK265-fulllength agr) was cured of plasmid pSK265 by daily passaging in 10 ml of tryptic soy broth (TSB) lacking antibiotics for 7 days (1). Cultures were then streaked onto blood agar, and individual colonies were replica plated onto TSA plates with or without chloramphenicol (Cm). To confirm the loss of plasmid, PCR amplification of the cat gene was performed for those colonies that did not grow in chloramphenicol. Plasmids containing the cat gene from pSK265 plasmid were included in each PCR experiment as positive controls. The cured strain was named LR8 (LR7-HeR cured). Inducible expression of RNAII/III in agr-null mutant LR3-HeRagr. RNAII and RNAIII genes were individually cloned under the control of Pspac isopropylD-thiogalactopyranoside (IPTG)-inducible vector pCL15 (29); a 176-bp fragment containing the RNAIII gene without the P3 promoter was amplified by using primers RNAIII-w/P3-F and RNAIII-w/P3-R (Table 1). Similarly, a 2.8-kb fragment was amplified for the RNAII gene without P2 promoter by using primers AgrF-w/P2-F and AgrR-w/P2-R (Table 1); in both cases, SA13011genomic DNA was used as a template. PCR products were purified using the QIAquick gel extraction kit, ligated into the PCR 2.1-TOPO vector (Invitrogen, Carlsbad, CA), and transformed into chemically competent E. coli TOP10 cells (Invitrogen). A fragment containing the insert was excised from PCR 2.1-TOPO by HindIII and XbaI sites and subcloned into the same restriction sites of pCL15. Plasmids purified from E. coli were confirmed by sequencing and electroporated into S. aureus RN4220. Transformants were selected on TSA containing chloramphenicol, sequenced, and transduced into agr-null mutant LR3-HeRagr by phage 80␣, resulting in the LR9-HeR and LR10-HeR strains, respectively. Analysis of gene expression by real-time RT-PCR and microarray transcriptional profiling. RNA extraction was performed by using the RNeasy extraction kit (Qiagen) as previously described (15). Equal amounts of total RNA were used to determine gene expression by real-time reverse transcriptase PCR (RTPCR) by using the SYBR green-based SensiMix one-step kit (Quantace/Bioline, Tauton, MA) according to the manufacturer’s protocol. Results for the experimental genes were normalized to 16S rRNA levels and were compared according to the threshold cycle (CT) value; oligonucleotide primers are shown in Table 1. Microarray transcriptional profiles were carried out by using a spotted DNA microarray (TIGR version 6 S. aureus slides) containing 4,546 oligonucleotides (70-mer) covering the genomes of S. aureus strain COL (2,654 open reading frames [ORFs]), N315 (2,623 ORFs), Mu50 (2,748 ORFs), MRSA strain 252

3178

PLATA ET AL.

ANTIMICROB. AGENTS CHEMOTHER. TABLE 1. Bacterial strains, plasmids, and primers used in this study

Strain no., plasmid, or primer

Strain name

Relevant genotype, phenotype, or 5⬘33⬘ sequence

SA13011-HeR SA13011-HoR LR3-HeRagr LR3-HoRagr LR4-HeRagr::agr LR4-HoRagr::agr LR5-HeR (empty vector) LR6-HoR (empty vector) LR7-HeR LR7-HoR LR8-HeR LR8-HoR LR9-HeR LR9-HoR LR10-HeR LR10-HoR LR11-HeR LR11-HoR LR12-HeR LR12-HoR RN4220 RN6911

Oxas mecA (⫹) Selected from SA13011 HeR (⫹0.5 ␮g/ml Oxa) ⌬agr deletion mutant of parental strain SA13011 LR3-HeRagr (⫹0.5 ␮g/ml Oxa) LR3-HeRagr/pSK265-agr LR4-HeRagr::agr (⫹0.5 ␮g/ml Oxa) LR3-HeRagr/pSK625 LR5-HeR (⫹0.5 ␮g/ml Oxa) SA13011-HeR expressing pSK265-full-length agr LR7 (⫹0.5 ␮g/ml Oxa) LR7-HeR cured of pSK265-full-length agr LR8 (⫹0.5 ␮g/ml Oxa) LR3-HeRagr/pCL15-RNAIII without P3 promoter LR9-HeR (⫹0.5 ␮g/ml Oxa) LR3-HeRagr/pCL15-RNAII without P2 promoter LR10-HeR (⫹0.5 ␮g/ml Oxa) LR3-HeRagr/pSK265-RNAIII LR3-HeRagr/pSK265-RNAIII (⫹0.5 ␮g/ml Oxa) LR3-HeRagr/pSK265-RNAII LR3-HeRagr/pSK265-RNAII (⫹0.5 ␮g/ml Oxa) Restriction-deficient mutagenized RN450 agr-null mutant

This This This This This This This This This This This This This This This This This This This This 11 38

Plasmids E. coli TOPO PCR2.1 S. aureus pSK265 S. aureus pCL15

Ampr Kanr High-copy staphylococcal replicon IPTG-inducible S. aureus replicon

Invitrogen 30 29

Primers Agr-FL-F Agr-FL-R 16S-F 16S-R RNAIII-F RNAIII-R RNAIII-w/P3-F RNAIII-w/P3-R Agr-w/P2-F Agr-w/P2-R Agr-S1 Agr-S2 RecA-F RecA-R

GCGCCATAGGATTGTAGAGTG CTCAGTAAGAATCCATTTCGCCC TCCGGAATTATTGGGCGTAA CCACTTTCCTCTTCTGCACTCA GATGGCTTAATAACTCATAC AAGGAGTGATTTCAATGGCAC TTAAGGAAGGAGTGATTTCAATGGCAC GATGGCTTAATAACTCATAC GAGGAGAGTGGTGTAAAATTGAAT ACAATTGAATACGCCGTTAACTGACTT ATGGTTATTAAGTTGGGATGG CAGCGGGTACTTTAGGTT GAGAAATCTTTCGGTAAAGGT GTGAAGCGCTACTGTTGTCTTACC

This study

Strains 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22

(2,744 ORFs), methicillin-susceptible S. aureus (MSSA) strain 476 (2,619 ORFs), and pLW043 (62 ORFs), as previously described (15). TIFF images of the hybridized arrays were analyzed using TIGR-Spotfinder software (http://www .tigr.org/software/). The data set was normalized by applying the LOWESS algorithm (block mode; smooth parameter 0.33) and using TIGR-MIDAS (http: //www.tigr.org/software/) software, and significant changes were identified with SAM (significance analysis of microarrays; http://www-stat.stanford.edu/⬃tibs /SAM/index.html) software. Differential expression was defined as a change of more than 2-fold in transcript versus the comparator strain.

RESULTS Differential gene expression analysis between SA13011-HeR and SA13011-HoR. S. aureus 13011-HeR is a representative mecA-positive, oxacillin-susceptible strain with MICs of oxacillin and cefoxitin of 2 ␮g/ml and 8 ␮g/ml, respectively (9, 12). Selection from an HeR to HoR phenotype was obtained by growing the HeR cells in broth containing subinhibitory concentrations of oxacillin (0.5 ␮g/ml), as previously shown (9).

Source or reference

study study study study study study study study study study study study study study study study study study study study

This study This study This study This study 45 This study

With the purpose of continuing previous studies directed at identifying genes that may be involved in the process of HeRHoR selection in SA13011, differential gene expression analysis was performed by spotted DNA microarray. Pairwise comparisons were made in triplicates between isogenic SA13011-HeR and SA13011-HoR strains (Table 2). RNAs were extracted from cells grown in LB media and collected at a similar exponential log phase (i.e., OD600 of ⬃0.6). As we previously reported (9, 15), the gene encoding PBP2a, mecA (SA0038_N315), was upregulated. Interestingly, among upregulated genes, we found the two-component regulator agr— notably accessory gene regulator A (agrA; SA2026_COL), accessory gene regulator B (agrB; SA1842_N315), and accessory gene regulator C (agrC; SA1843_N315)—transcribed through RNAII. Moreover, several of the genes that were either up- or downregulated corresponded to genes known to be under the regulation of the agr system, including the agr effector molecule

VOL. 55, 2011

agr IN MRSA WITH LOW-LEVEL RESISTANCE

3179

TABLE 2. Differential gene expression between SA13011-HeR and SA13011-HoR as determined by spotted DNA microarraya ORF

Gene

Product of putative function

Fold change

SA0038_N315 SA2026_COL SA1842_N315 SA1843_N315 MW1959_MW2 SA1007_N315 SA0138_COL SA0148_COL SA0150_MRSA252 SA0147_N315 SA0152_N315 SA0901_N315 SA1629_N315 SA1865_COL MW0932_MW2 SA2428_N315 SA1206_N315 SA0909_N315 SAV1052_MU50 SA1758_N315 SA0265_N315 SA0107_N315 SA0594_N315 SA0596_N315

mecA agrA agrB agrC hld ␣-Hemolysin gene cap cap

Penicillin-binding protein 2⬘ Accessory gene regulator protein Accessory gene regulator B Accessory gene regulator C ␦-Hemolysin ␣-Hemolysin precursor Capsular polysaccharide biosynthesis protein Cap5C Capsular polysaccharide biosynthesis galactosyltransferase Cap5M Capsular polysaccharide synthesis enzyme Capsular polysaccharide synthesis enzyme Cap5D Capsular polysaccharide synthesis enzyme Cap5I Serine protease V8 protease; glutamyl endopeptidase Serine protease SplC Serine protease SplE Serine protease; V8 protease; glutamyl endopeptidase Arginine deaminase Factor essential for expression of methicillin resistance Fmt, autolysis and methicillin-resistant-related protein Autolysin Staphylokinase precursor Peptidoglycan hydrolase Inmunoglobulin G binding protein A precursor Teichoic acid translocation permease protein Teichoic acid byosynthesis

⫹3.5 ⫹2.3 ⫹4.1 ⫹2.2 ⫹4.9 ⫹3.7 ⫹4.0 ⫹4.6 ⫹3.9 ⫹4.3 ⫹4.2 ⫹5.8 ⫹5.8 ⫹4.2 ⫹5.4 ⫹12.8 ⫹2.8 ⫹3.6 ⫹2.6 ⫹3.8 ⫹2.3 ⫺17.8 ⫺1.5 ⫺2.8

splC splE arcA femA fmt atl lytM spa tagG

a The data set was normalized by applying the LOWESS algorithm (block mode; smooth parameter 0.33) and using TIGR MIDAS software (http://www.tigr.org /software/), and significant changes were identified with SAM (significance analysis of microarrays) software.

hld, which is transcribed by RNAIII and encodes the structural protein ␦-hemolysin (hld; MW2-1959). ␣-Hemolysin precursor (SA1007_N315); serine protease V8 glutamyl endopeptidase (SAN315-0901; staphylococcal nuclease [nuc; SA0746_N315]); capsular polysaccharide biosynthesis protein genes cap5C (SACOL_0138), cap5M (SACOL_0148), cap5I (SACOL_0152), cap5D (SACOL_0147), and cap8O (SACOL_0158); and the arc operon (arginine deaminase operon arcCDBA; SA2428_N315) were upregulated (Table 2). The immunoglobulin G binding protein A precursor gene, spa (SA0107_N315), which was previously shown to be negatively regulated by agr (10), appeared downregulated (Table 2). Other upregulated genes included cell-wall-associated genes fmt (SA0909_N315) and femA (SA1206_N315), the cell-wall-anchored surface protein gene clfA, and autolysis genes atl (SAV1052_Mu50) and lytM (SA0265_N315); teichoic acid translocation permease protein gene tagG (SA0594_N315) and teichoic acid biosynthesis protein (SA0596_N315) were downregulated. The studies were then focused on the analysis of agr in the context of ␤-lactam-induced HeR-HoR selection. Strains SA13011, SA2006044, and SA2006043 were determined to belong to agr group 2, identical to S. aureus N315. Furthermore, all of the HoR colonies displayed exactly the same phenotype vis a vis agr (data not shown). A detailed time course analysis of RNAIII was performed by real-time RT-PCR to address the possibility that increased agr expression observed in homotypic cells may constitute a cell-density-mediated effect (37): i.e., accumulation of agr-autoinducing peptide (AIP), rather than a HeR/HoR selection-mediated increase. SA13011-HeR/HoR cells were harvested from colonies (blood agar plates), grown in liquid LB media, and collected at OD600s of 0.25, 0.5, 1.0, and 1.7. RNA was extracted as described in Materials and Methods, and equal amounts (20 ng) of total RNA were used

to determine RNAIII mRNA expression; the results were then normalized to 16S rRNA levels. As displayed in Table 3, an increase in gene expression of RNAIII was gradually observed in SA13011-HeR following cell growth density, reaching a maximal value at stationary phase (i.e., OD600 of 1.7). However, homotypic cells displayed a very significant increase compared to the heterotypic counterparts collected under the same experimental conditions, ruling out the possibility that the increase in agr expression observed in cells growing and selected in the presence of oxacillin may merely reflect an increase in cell density or AIP. Increased agr expression during HeR-HoR selection regulates oxacillin-induced SOS response-mediated mutation rate. In previous studies, we reported that ␤-lactam-mediated SA13011 HeR-HoR selection was associated with an oxacillininduced SOS response and increased mutation rates; these events were determined by the preexistence of a hypermutable population that favored the HoR selection following exposure

TABLE 3. Expression analysis of RNAIII mRNA contents in SA13011-HeR and SA13011-HoR determined by real-time RT-PCR Fold increase in expressiona OD600

0.25 0.52 1.0 1.7

HeR

HoR

1.0 ⫾ 0.05 0.9 ⫾ 0.04 4.3 ⫾ 0.8 6.2 ⫾ 1.1

4.8 ⫾ 0.7* 12.2 ⫾ 1.1* 22.9 ⫾ 1.8* 24.7 ⫾ 2.1*

a The results are expressed as fold increase with respect to the value corresponding to the SA13011-HeR OD600 of 0.25. Data represent means ⫾ SD of three independent experiments performed in triplicate. ⴱ, significantly higher than the corresponding value in SA13011-HeR cells (P ⱕ 0.001).

3180

PLATA ET AL.

ANTIMICROB. AGENTS CHEMOTHER.

FIG. 1. Characterization of agr expression by RT-PCR (A) and real-time RT-PCR (B). RNA was isolated from SA13011-HeR; its highly resistant derivative, SA13011-HoR; and their agr-null counterparts, LR3-HeRagr and LR3-HoRagr. Cells were harvested at the exponential phase of growth. RT-PCR (A) was performed using S1 and S2 primers encompassing a product of 1.1 kb containing hld, the P2 and P3 promoters, and agrB. RT-PCR products from 16S rRNA were used as loading and RNA quality controls; S. aureus N315 was used as a reference control. (C) Analysis of RNAIII mRNA expression by real-time RT-PCR. Relative values of RNAIII mRNA over 16S rRNA are shown. RNA samples for a time course analysis were prepared from SA13011-HeR (⫹Oxa) (HoR) and LR3-HeRagr (⫹Oxa) (HoR) cells. (D) ␦-Hemolysis assay. Each panel shows results for the test strains (i.e., agr-positive SA13011-HeR and SA13011-HoR) and their respective agr-null mutants, LR3-HeRagr and its derivative, LR3-HoRagr, streaked horizontally. The inner circle denotes the enhanced zone of hemolysis created by the interaction of the ␤-hemolysis disk and ␦-hemolysis (SA13011-HoR).

to the antibiotic (9). To investigate the role that increased expression of agr may play during the HeR-HoR selection process, agr-null mutants were generated by phage-80-mediated transduction from the agr-null mutant RN6911 (38). Successful transduction resulted in strains LR3-HeRagr and LR3HoRagr (strains 3 and 4 in Table 1); agr expression was verified by PCR (DNA) (data not shown), RT-PCR (Fig. 1A), and real-time RT-PCR (Fig. 1B). Analysis of agr mRNA levels in both LR3-HeRagr and LR3-HoRagr strains showed an absence of agr expression in these cells (Fig. 1A and B), while increased levels of agr were clearly detected in SA13011-HoR compared to the SA13011-HeR counterpart. agr expression was further characterized in these strains by monitoring mRNA levels of the agr regulatory molecule RNAIII during the HeR-HoR selection (Fig. 1C). Real-time RT-PCR analysis of RNA samples prepared from cells collected at different time intervals showed a progressive increase in the expression of RNAIII in SA13011-HeR (⫹Oxa) (HoR) (Fig. 1C, data points B1 to B5). In contrast, no changes were observed in LR3HeRagr (⫹Oxa) (LR3-HoRagr) (Fig. 1C, data points D1 to D5), confirming lack of agr expression during the selection

(Fig. 1C). Functional analyses of agr involvement were performed in SA13011-HeR/HoR and their respective agr-null mutants by using ␦-hemolysin production, a surrogate marker of agr activity (45). Consistent with data on agr expression analysis, an enhanced zone of hemolysis created by the interaction of ␤-hemolysin (disk) and ␦-hemolysin was discernible with the wild-type strain, strain SA13011-HoR selected with oxacillin (strain 2) (Fig. 1D, arrow). In contrast, agr-null mutants LR3-HeRagr and LR3-HoRagr (strains 3 and 4, respectively) displayed marked attenuation of the hemolytic properties, clearly demonstrating the absence of agr expression. To investigate the role that increased expression of agr may play in relation to oxacillin-induced SOS response, both growth and mutation frequencies for resistance to rifampin were determined in relation to agr. The growth pattern determined by measuring the optical density at 600 nm was followed in the absence and presence of oxacillin (0.5 ␮g/ml) at the indicated time intervals (3, 6, 9, 27, 30, and 33 h) (Fig. 2). In the absence of oxacillin, similar growth rates were observed between both SA13011-HeR and LR3-HeRagr strains (Fig. 2, data points A1 and ⫺2 and C1 and ⫺2, respectively); however, marked dif-

VOL. 55, 2011

agr IN MRSA WITH LOW-LEVEL RESISTANCE

3181

agr-null mutants LR3-HeRagr and LR3-HoRagr (strains 3 and 4, respectively) displayed values corresponding to an ⬃5-log increase (i.e., 1.2 ⫻ 10⫺8/1.4 ⫻ 10⫺3, respectively, at time point 27 h) (Table 4). These observations were extended to other heterogeneous MRSA strains with phenotypes similar to SA13011. For example, increased mutation frequencies of ⬃3 logs were also observed in MRSA strains SA200643-HeR/HoR and SA200644-HeR/HoR during the process of selection (⫾0.5 ␮g/ml Oxa) at the indicated time intervals (3 to 33 h). As was the case with SA13011, the corresponding agr-null mutants SA44-HeRagr and SA44-HoRagr (see Table S1 in the supplemental material) and SA43-HeRagr and SA44-HoRagr (data not shown) displayed an increase of ⬃5 to 6 logs under the same growth conditions. To further assess agr modulatory effects, the mutant LR3HeRagr strain was complemented with a cloned full-length agr (LR4-HeRagr::agr; strain 5) (Table 1). Constitutive expression of agr was verified following complementation in RNA samples prepared from cells collected during HeR-HoR selection at the exponential phase of growth: i.e., OD600 of 0.6 at ranges of 3 to 6 h for cells growing without oxacillin (HeR), and 21 to 24 h for nascent cells growing with 0.5 ␮g/ml oxacillin (HoR). As shown in Fig. 3, agr trans-complementation (LR4-HeRagr::agr; strain 5) resulted in agr transcription levels similar to those corresponding to wild-type SA13011HoR (strain 2); as expected, these values were not affected by exposure to oxacillin (LR4-HoRagr::agr; strain 6). No changes in agr expression were observed in empty-vector-complemented strains LR5-HeR (strain 7) and LR6-HoR (strain 8) (⫾0.5 ␮g/ml Oxa), respectively. agr complementation restored to a pattern similar to the wild-type SA13011-HoR strain (strain 2) both the growth of the agr mutant strain LR4-HoRagr::agr (strain 6) in the presence of 0.5 ␮g/ml oxacillin (data not shown) and the frequency of mutation during the selection (i.e., 5.2 ⫻ 10⫺8 versus 5.4 ⫻ 10⫺5 for strains 5 and 6 and 2.3 ⫻ 10⫺8 versus 6.4 ⫻ 10⫺5 for strains 1 and 2, respectively), an ⬃3-log difference (time point 27 h) (Table 4). The agr mutants harboring the empty vector (i.e., LR5-HeR and LR6-HoR; strains 7 and 8, respectively) showed no differences compared to levels corresponding to the agr mutants

FIG. 2. Time course analysis of S. aureus 13011 selection from heterotypic to homotypic oxacillin resistant compared to the agr-null LR3-HeRagr strain and its derivative, LR3-HeRagr (⫹Oxa) (HoR), as determined by optical density at 600 nm. Points C1 and C2 correspond to SA13011-HeR grown in the absence of oxacillin; points B1 to B5 correspond to SA13011-HeR grown in the presence of 0.5 ␮g/ml oxacillin, leading to the selection of the homotypic counterpart derivative SA13011-HoR (points B3 to B5); points A1 and A2 correspond to LR3-HeRagr grown in the absence of oxacillin; and points D1 to D5 correspond to LR3-HeRagr (HoR) grown in the presence of 0.5 ␮g/ml oxacillin. Values represent means ⫾ standard deviations (SD) for three separate experiments. *, significantly lower than the corresponding SA13011-HeR (⫹Oxa) values for points D3 to D6 (P ⱕ 0.01).

ferences appeared when cells were exposed to 0.5 ␮g/ml oxacillin. For example, after 27 h, the growth rate of LR3-HeRagr (⫹Oxa) (HoR) cells was slow compared to that of SA13011HoR cells (OD600s of 0.42 and 0.72) (Fig. 2, data points D3 and B3, respectively). At 30 h, homotypic cells selected from LR3HeRagr (⫹Oxa) reached only an OD600 of 0.78 (data point D4), while their SA13011-HoR counterparts were at an OD600 of 1.6 (data point B4). This indicated that absence of agr significantly affected the process of HeR-HoR selection. In parallel, the number of mutants generated by LR3-HeRagr (strain 3) (Table 1) was followed in absence and presence of oxacillin (0.5 ␮g/ml) at 3-, 6-, 9-, 27-, 30-, and 33-h time intervals (Table 4). While in SA13011-HeR/HoR, the mutation rate represented an ⬃3-log increase (i.e., from 2.3 ⫻ 10⫺8 to 6.4 ⫻ 10⫺5 at time point 27 h for strains 1 and 2, respectively) and

TABLE 4. Mutation rate analysis during HeR-HoR selection in SA13011 strains 1 to 12 Strain no.a

Mutation frequency atb: Strain name 3h

6h

9h

27 h

30 h

33 h

1 2 3 4 5 6

SA13011-HeR SA13011-HoR LR3-HeRagr LR3-HoRagr LR4-HeRagr::agr LR4-HoRagr::agr

2.3 ⫻ 10 3.0 ⫻ 10⫺9 1.2 ⫻ 10⫺8 2.1 ⫻ 10⫺8 5.2 ⫻ 10⫺8 6.2 ⫻ 10⫺9

5.0 ⫻ 10 6.1 ⫻ 10⫺8 4.8 ⫻ 10⫺7 1.5 ⫻ 10⫺7 2.2 ⫻ 10⫺8 2.7 ⫻ 109

7.5 ⫻ 10 6.1 ⫻ 10⫺6 3.8 ⫻ 10⫺7 2.6 ⫻ 10⫺6 1.6 ⫻ 10⫺8 3.0 ⫻ 10⫺6

1.1 ⫻ 10 6.4 ⫻ 10⫺5 5.4 ⫻ 10⫺7 1.4 ⫻ 10⫺3 2.0 ⫻ 10⫺7 5.4 ⫻ 10⫺5

3.7 ⫻ 10 3.0 ⫻ 10⫺4 2.5 ⫻ 10⫺6 7.8 ⫻ 10⫺2 8.3 ⫻ 10⫺7 2.6 ⫻ 10⫺4

1.3 ⫻ 10⫺8 1.2 ⫻ 10⫺4 5.0 ⫻ 10⫺6 3.5 ⫻ 10⫺2 3.7 ⫻ 10⫺7 1.6 ⫻ 10⫺4

7 8 9 10 11 12

LR5-HeR (empty vector) LR6-HoR (empty vector) LR7-HeR LR7-HoR LR8-HeR LR8-HoR

2.9 ⫻ 10⫺8 8.1 ⫻ 10⫺8 1.3 ⫻ 10⫺9 2.8 ⫻ 10⫺7 3.3 ⫻ 10⫺8 2.5 ⫻ 10⫺8

4.0 ⫻ 10⫺7 3.5 ⫻ 10⫺8 9.0 ⫻ 10⫺8 1.3 ⫻ 10⫺7 1.2 ⫻ 10⫺7 8.9 ⫻ 10⫺7

1.2 ⫻ 10⫺7 1.0 ⫻ 10⫺5 5.1 ⫻ 10⫺8 5.3 ⫻ 10⫺7 8.0 ⫻ 10⫺8 4.2 ⫻ 10⫺7

9.3 ⫻ 10⫺6 2.3 ⫻ 10⫺3 3.8 ⫻ 10⫺7 1.0 ⫻ 10⫺6 2.5 ⫻ 10⫺8 1.6 ⫻ 10⫺4

5.6 ⫻ 10⫺6 4.3 ⫻ 10⫺2 6.7 ⫻ 10⫺7 1.2 ⫻ 10⫺7 1.3 ⫻ 10⫺7 7.6 ⫻ 10⫺4

1.3 ⫻ 10⫺6 2.8 ⫻ 10⫺2 5.2 ⫻ 10⫺7 2.4 ⫻ 10⫺8 2.0 ⫻ 10⫺7 5.3 ⫻ 10⫺4

a

⫺8

⫺8

⫺7

⫺8

⫺8

Described in Table 1. Mutation frequencies are expressed as the number of antibiotic-resistant mutants recovered as a fraction of the viable count. Three independent cultures were sampled in triplicate to minimize error caused by inter- and intrasample variation. b

3182

PLATA ET AL.

FIG. 3. Quantitation of agrD mRNA by RT-PCR in SA13011 strains and derivatives (strains 1 to 12) (Table 1) grown in the absence (⫺) and presence (⫹) of 0.5 ␮g/ml oxacillin. Values of specific agrD mRNA/16S rRNA are shown in the vertical axis. *, significantly higher than sample 1 (P ⬍ 0.001). Three independent cultures were sampled in triplicate to minimize error caused by inter- and intrasample variation.

LR3-HeRagr and LR3-HoRagr (strains 3 and 4, respectively) (Table 4). No significant changes were observed in mutation rates between strains growing in the absence of oxacillin (strains 1, 3, 5, and 7) (Table 4). Together, these results indicate that increased expression of agr during the HeR-HoR selection process plays an important functional role in modulating the mutation rate associated with the acquisition of the homotypic phenotype in the present clinical heterogeneous S. aureus strains. Ectopic expression of full-length agr blocks oxacillin-mediated HeR-HoR selection. Additional experiments were performed in order to reach a better understanding and characterization of the agr functional role during oxacillin-mediated HeR-HoR selection. Full-length agr was cloned into the pSK265 vector and expressed in SA13011-HeR (LR7-HeR; strain 9) (Table 1). Constitutive expression of ectopic agr was monitored by real-time RT-PCR, resulting in the following agrD fold changes: SA13011-HeR, 1 (reference value) (Fig. 3; strain 1); SA13011-HeR (⫹Oxa) (HoR), 7.3 ⫾ 0.51 (Fig. 3; strain 2); LR7-HeR, 8.5 ⫾ 1.4 (Fig. 3; strain 9); and LR7-HoR, 13.8 ⫾ 1.8 (Fig. 3; strain 10). No phenotypic differences were found between SA13011-HeR and LR7-HeR strains grown in the absence of oxacillin (MICs of oxacillin, 2 ␮g/ml), indicating that the sole agr-increased expression was not sufficient to confer the HoR phenotype. LR7-HeR was then tested for its capacity to induce HeR-HoR selection in the presence of subinhibitory concentrations of oxacillin (0.5 ␮g/ml). The growth pattern was determined by measuring the optical density at 600 nm, as shown in Fig. 4A (right panel). LR7-HeR cells growing with oxacillin did not display the characteristic drop of cell density that preceded the selection of SA13011-HoR cells (Fig. 4A, left graph). Comparable results were observed with other subinhibitory concentrations of oxacillin, including 0.25 and 0.75 ␮g/ml (data not shown). Importantly, MICs of oxacillin showed that the level of resistance remained unchanged after exposure of LR7-HeR to 0.5 ␮g/ml oxacillin (i.e., MIC, 2 ␮g/ml). Oxacillin resistance levels comparing LR7-HeR (⫾0.5

ANTIMICROB. AGENTS CHEMOTHER.

␮g/ml Oxa) to SA13011-HeR/HoR were confirmed by using 0to 128-␮g/ml oxacillin gradient plates with a 0.5 McFarland inoculum (Fig. 4B). While confluent growth up to 128 ␮g/ml was observed for SA13011-HoR (Fig. 4B, lane 4), no cells were detected in lanes corresponding to SA13011-HeR (lane 3) and LR7-HeR (⫾0.5 ␮g/ml Oxa) (lanes 1 and 2, respectively). In addition, oxacillin resistance population analysis demonstrated that LR7-HeR cells grown in the presence of oxacillin were inhibited at an oxacillin concentration of 1 ␮g/ml, while 5 ⫻ 103 SA13011-HeR cells (⫹Oxa) (HoR) were still growing at oxacillin concentrations of ⱖ64 ␮g/ml (Fig. 4C). Interestingly, agr ectopic overexpression inhibited completely the appearance of rifampin mutants in LR7-HeR (⫾0.5 ␮g/ml Oxa) (Table 4). To exclude the possibility that impaired HeR-HoR selection in the LR7-HeR strain containing the pSK265-agr plasmid may be due to a second-site mutation(s), the LR7HeR strain was cured of the pSK265-agr plasmid by daily passaging in antibiotic (chloramphenicol)-free media (1). Importantly, the LR8-cured strain displayed the same pattern of HeR-HoR selection as parental SA13011 both in terms of agr expression levels (Fig. 3; strains 11 and 12 versus strains 1 and 2, respectively) and in terms of mutation rates (Table 4), discarding the existence of second-site mutations that may be responsible for the observed phenotypes. Impact of agr on mutation rate is mediated by RNAII. As mentioned previously, the Agr system consists of two divergent transcripts, RNAII and RNAIII, which are transcribed from the P2 and P3 promoters, respectively (38). To determine the level of involvement of each of these components in the regulation of mutation rates and the oxacillin resistance phenotype during HeR-HoR selection, RNAII and RNAIII were individually cloned in either the constitutive pSK265 vector or in the IPTG-inducible pCL15 plasmid (with genes under the control of the Pspac promoter) (29). The agr-null mutant strain LR3-HeRagr was used as the recipient strain. For IPTG-inducible expression, optimal conditions of gene expression were determined by analyzing both RNAII (i.e., agrBDCA) and RNAIII mRNA levels by using increasing concentrations of IPTG ranging from 0.25 mM to 3 mM; a 1 mM IPTG concentration was selected for ulterior use as it did induce expression levels comparable to those determined in SA13011-HoR (data not shown). Mutation frequencies for resistance to rifampin were determined by using strains LR9 (LR3-HeRagr/pCL15-RNAIII), LR10 (LR3-HeRagr/pCL15-RNAII), LR11 (LR3-HeRagr/ pSK265-RNAIII), and LR12 (LR3-HeRagr/pSK265-RNAII). LR9 and LR10 were grown in the presence of 15 ␮g/ml chloramphenicol (the pCL15 resistance marker) plus 1 mM IPTG ⫾ 0.5 ␮g/ml oxacillin. LR11 and LR12 were analyzed in the presence of 15 ␮g/ml chloramphenicol ⫾ 0.5 ␮g/ml oxacillin (Table 5). RNA samples were extracted under the same conditions to verify gene expression levels (Fig. 5). No significant differences in mutation rates were observed between the SA13011-HeR strain (Table 4) and the corresponding LR9/10 (⫹Cm ⫹ 1 mM IPTG) and LR11/12 strains grown in the absence of oxacillin (Table 5). However, important differences were observed when they were grown in the presence of the ␤-lactam antibiotic. Selection of strains complemented with RNAIII, expressed either inducibly or constitutively [LR9 and LR11 (⫹Oxa), respectively], revealed an increased mutation

VOL. 55, 2011

agr IN MRSA WITH LOW-LEVEL RESISTANCE

3183

FIG. 4. (A) Time course analysis of selection from heterotypic to homotypic oxacillin resistance in the S. aureus 13011 (left graph) and LR7 (right graph) strains, as determined by optical density at 600 nm. Values represent means ⫾ SD for three separate experiments. (B) Oxacillin gradient plate (0 to 128 ␮g/ml) of LR7-HeR (lane 1), LR7-HoR (lane 2), SA13011-HeR (lane 3), and SA13011-HoR (lane 4). Absence of growth is observed consistently with a decrease in the number of resistant cells in LR7-HoR (lane 2) compared to SA13011-HoR (confluent growth; lane 4). A representative picture of three independent cultures is shown. (C) Oxacillin resistance population analysis of SA13011-HeR, SA13011-HoR, LR7-HeR, and LR7-HoR. Expression of oxacillin population (EOP) curves were generated by testing the indicated cells obtained directly from plates containing the corresponding colonies growing on various concentrations of oxacillin. The colonies to be tested were resuspended in 1 ml of LB broth. Shown on the y axis is the number of S. aureus cells (in log10 CFU per milliliter on oxacillin and CFU per milliliter on MHA) remaining on the plates containing various concentrations of oxacillin (shown on the x axis).

rate similar to patterns observed with mutant LR3-HoRagr (Table 4) (i.e., MICs of oxacillin of ⱖ256 ␮g/ml) (Table 5). In contrast, 0.5-␮g/ml oxacillin selection of strains expressing either inducible or constitutive RNAII [LR10 and LR12

(⫹Oxa)] displayed a marked suppression of the mutation rate (Table 5). Consistent with these observations, LR10 and LR12 plus oxacillin also displayed reduced MICs of oxacillin (MIC, 8 ␮g/ml). Taken together, these results indicate that, in the pres-

TABLE 5. Mutation rate analysis during HeR-HoR selection in mutant agr strains 13 to 20 overexpressing constitutive or inducible RNAII or RNAIII Strain no.a

1 2 3 4 5 6 7 8 9 10 a

Strain name (treatment)

SA13011-HeR SA13011-HoR LR9-HeR (Cm ⫹ 1 mM IPTG) LR9-HoR (Cm ⫹ 1 mM IPTG ⫹ 0.5 ␮g/ml Oxa) LR10-HeR (Cm ⫹ 1 mM IPTG) LR10-HoR (Cm ⫹ 1 mM IPTG ⫹ 0.5 ␮g/ml Oxa) LR11-HeR LR11-HoR (⫹0.5 ␮g/ml Oxa) LR12-HeR LR12-HoR (⫹0.5 ␮g/ml Oxa)

Oxa MIC (␮g/ml)

Mutation frequency atb: 3h

6h

9h

27 h

30 h

33 h

1 ⱖ256 1 ⱖ256

⫺8

2.3 ⫻ 10 3.0 ⫻ 10⫺9 1.4 ⫻ 10⫺8 7.3 ⫻ 10⫺8

⫺8

5.0 ⫻ 10 6.1 ⫻ 10⫺8 1.1 ⫻ 10⫺7 1.5 ⫻ 10⫺8

⫺7

7.5 ⫻ 10 6.1 ⫻ 10⫺6 2.5 ⫻ 10⫺7 6.9 ⫻ 10⫺7

⫺8

1.1 ⫻ 10 6.4 ⫻ 10⫺5 8.6 ⫻ 10⫺7 5.4 ⫻ 10⫺3

⫺8

3.7 ⫻ 10 3.0 ⫻ 10⫺4 7.5 ⫻ 10⫺6 8.1 ⫻ 10⫺3

1.3 ⫻ 10⫺8 1.2 ⫻ 10⫺4 2.0 ⫻ 10⫺7 3.68 ⫻ 10⫺3

1

1.4 ⫻ 10⫺7

6.8 ⫻ 10⫺7

1.6 ⫻ 10⫺7

4.5 ⫻ 10⫺7

3.7 ⫻ 10⫺7

2.0 ⫻ 10⫺7

8

4.7 ⫻ 10⫺8

2.2 ⫻ 10⫺8

3.1 ⫻ 10⫺7

1.6 ⫻ 10⫺7

1.4 ⫻ 10⫺7

3.7 ⫻ 10⫺7

1 ⱖ128 1 8

5.7 ⫻ 10⫺8 2.0 ⫻ 10⫺7 5.0 ⫻ 10⫺8 4.4 ⫻ 10⫺6

9.6 ⫻ 10⫺7 8.5 ⫻ 10⫺7 1.2 ⫻ 10⫺9 5.5 ⫻ 10⫺6

1.2 ⫻ 10⫺9 6.5 ⫻ 10⫺6 6.5 ⫻ 10⫺7 1.3 ⫻ 10⫺6

1.3 ⫻ 10⫺9 2.3 ⫻ 10⫺3 7.1 ⫻ 10⫺7 1.5 ⫻ 10⫺6

2.5 ⫻ 10⫺7 1.4 ⫻ 10⫺2 1.0 ⫻ 10⫺7 6.7 ⫻ 10⫺6

3.6 ⫻ 10⫺7 1.9 ⫻ 10⫺2 3.2 ⫻ 10⫺7 2.2 ⫻ 10⫺5

Described in Table 1. Mutation frequencies are expressed as the number of antibiotic-resistant mutants recovered as a fraction of the viable count. Three independent cultures were sampled in triplicate to minimize error caused by inter- and intrasample variation. b

3184

PLATA ET AL.

ANTIMICROB. AGENTS CHEMOTHER.

FIG. 6. Quantitation of recA mRNAs by real-time RT-PCR in the indicated SA13011 and LR3-HeRagr strains (Table 1). Values of specific recA mRNA/16S rRNA are shown in the vertical axis. #, significantly higher than SA13011 (⫺Oxa) (reference sample value of 1; P ⬍ 0.001). *, LR3-HeRagr significantly higher than the corresponding strains SA13011 (⫺Oxa) (reference sample value of 1; P ⬍ 0.05), LR3-HeRagr and LR3-HeRagr (⫹Oxa) (OD of 0.2; P ⬍ 0.001), and LR3-HeRagr (⫹Oxa) (ODs of 0.5 and 0.8). **, LR3-HeRagr significantly higher than corresponding strain SA13011 (⫹Oxa) (P ⬍ 0.001). Three independent cultures were sampled in triplicate to minimize error caused by inter- and intrasample variation. FIG. 5. Quantitation of agrD (A) and RNAIII (B) mRNAs by RTPCR in S. aureus strains LR3, LR9, and LR10 (Tables 1 to 5) grown with (⫹ox or ⫹I) or without 0.5 ␮g/ml oxacillin or 1 mM IPTG, respectively. Values of specific agrD mRNA/16S rRNA are shown in the vertical axis. *, significantly higher than the corresponding reference sample LR10 (A) or LR9 (B), which had a value of 1 (P ⬍ 0.001). LR9 ⫹ I and LR10 ⫹ I are used as negative controls for agrD and RNAIII, respectively. Three independent cultures were sampled in triplicate to minimize error caused by inter- and intrasample variation.

ent MRSA heterogeneous clinical strains, the process of oxacillin-mediated HeR-HoR selection—including increased mutation rate and oxacillin-resistant phenotype—are under the regulatory activity of the agr P2-RNAII regulatory component. Importantly, they also indicate that, as mentioned above, the sole agr-increased expression is not sufficient to confer the HoR phenotype. Furthermore, extemporal expression of agr suppressed mutation and reduced the ability of HeR cells to become highly resistant (HoR) to oxacillin. recA expression during oxacillin-mediated HeR-HoR selection: role of agr. The present results demonstrated that SA13011 HeR-HoR selection is associated with the triggering of the SOS response resulting in significantly increased mutation rates, a process regulated by changes in expression of the two-component regulator agr. In order to identify potential links between agr and the SOS response, we analyzed expression of recA, one of the key regulatory proteins involved in this system (28, 32, 40). Consistent with its role, expression of recA determined by real-time RT-PCR showed a significant increase during oxacillin-mediated SA13011 HeR-HoR selection (Fig. 6). Importantly, when these samples were compared against agr-null mutant cells (i.e., LR3-HeRagr), a further and significant increase in recA expression was observed, highlighting the existence of a link between SOS response recA and HeR/HoR mutation rate regulator agr.

DISCUSSION In the present study, we have investigated the role of increased expression of agr during ␤-lactam-induced HeR-HoR selection in heterogeneous MRSA clinical strains. Despite its low level of resistance, a highly homotypic resistant population was selected when these strains were exposed to subinhibitory concentrations of oxacillin. The present findings expand our previous observations showing that HeR-HoR selection in these MRSA clinical strains is associated with an increased mutation rate mediated by an oxacillin-induced SOS response (9). Transcriptional profile analyses performed between SA13011-HeR/HoR populations revealed that in addition to genes previously identified belonging to the SOS response pathway (9), there was increased expression of the agr global regulatory system. The interest in investigating the role of agr during the process of selection originated from the observation that an important number of genes that were either up- or downregulated were well-known targets of the two-component agr regulatory system (10). From a functional point of view, the present results show that inactivation of agr impaired the clinical SA13011 growth rate during HeR-HoR selection. Mechanistically, this phenomenon was associated with a very significant increase in the number of rifampin mutants. Paradoxically, overexpression of agr in SA13011-HeR cells not only abolished the capacity of these cells to undergo HeR-HoR selection in the presence of subinhibitory concentrations of oxacillin, but also their capacity to generate rifampin mutants. These observations strongly support the concept that increased expression of agr requires temporal regulation since an early increase in agr levels conspires against an appropriate mutation rate response. On the other hand, no increase at all results in excessive mutation rates, which also impairs the ability of cells to reach maximal resistance. These observations are in agreement with current models suggesting that an increase in mutation rates represents a highly geneti-

VOL. 55, 2011

cally regulated process (26, 47), reinforcing the concept that oxacillin-mediated HeR-HoR selection requires both increased expression of mecA (9) and strictly controlled SOS response-mediated mutation rates. Also of note, the current findings established that agr-modulated, oxacillin-mediated HeR-HoR selection depends on genes carried in the RNAII transcript (i.e., agrA, agrC, agrD, and agrB) rather than RNAIII. Consistent with these observations, recent reports have shown that regulatory roles of agr on mechanisms of virulence in community-associated MRSA (CA-MRSA) strains are mediated by RNAII (41). Together, these data support the notion that modulation of agr expression during HeR-HoR selection plays a key role as a regulator of mutations that are generated through an oxacillin-induced SOS response. An association between stress response and regulatory genes has been recently observed in CA-MRSA strains with heterogeneous resistance to nafcillin (46). Exposure of CA-MRSA to subinhibitory concentrations of levofloxacin or ciprofloxacin was shown to select for increased resistance to nafcillin. Interestingly, in this study, transcriptional expression profiling showed that agrA and agrB genes were overexpressed in fluoroquinolone-exposed mutants, suggesting the potential involvement of the agr regulatory pathway (46). In addition, the agr system has been linked to genetic changes in vancomycinintermediate S. aureus (VISA) isolates, notably to low-level vancomycin resistance in S. aureus. This emphasizes the importance of the agr locus in the expression of the resistant phenotype (8, 23, 33, 51). Recent studies involving VISA strains have also shown that the sarA locus, which plays a significant role in the regulation of the agr operon (3, 5), is required for the expression of both vancomycin intermediate and ciprofloxacin resistance. Also interesting, sarA inactivation was demonstrated to lead to a reduction of resistance levels to both antibiotics (25). Furthermore, ciprofloxacin has also been shown to induce the SOS response in S. aureus, a process in which a functional sarA response was required for the cell to respond to the agent (6). However, in all of these studies, the specific role of agr was not addressed and remained a matter of speculation. Additionally, as first shown in our previous study (9), mRNA expression of key SOS response regulatory gene recA appeared to be increased during HeR-HoR selection. Levels of recA mRNA were significantly higher in the absence of agr, providing evidence of potential links between SOSmediated mutation rates and agr-modulated target genes involved in this pathway. Further studies to address the mechanisms involved in these SOS-agr interactions are ongoing. In summary, the present results indicate that while selection of homotypic resistant cells by increasing the SOS-mediated mutation rate may represent an adaptative strategy enhancing the chances of surviving ␤-lactam antibiotics, it does require a very strict control from a genetic standpoint, as previously shown in other models (13, 50). In this context, our findings indicate that a temporally controlled increase in agr expression is required to tightly modulate SOS-mediated mutation rates, which allows for full expression of oxacillin homogeneous resistance in very heterogeneous clinical MRSA strains. These observations contribute to the understanding of the mechanisms associated with the development of resistance to ␤-lactams in S. aureus and demonstrate the critical role that the agr

agr IN MRSA WITH LOW-LEVEL RESISTANCE

3185

two-component regulatory system—in addition to expression of mecA and the SOS response—may play in this process. ACKNOWLEDGMENTS This work was supported by NIH grant 1 R01 AI080688-01A2 (A.E.R.) and by an award from the Thomas F. Jeffress and Kate Miller Jeffress Memorial Trust (A.E.R.). Microarray studies were supported through a PFGRC grant (A.E.R.) from the J. Craig Venter Institute. Special thanks goes to Kathryn Stockbauer and Philip Randall, from the Office of Academic Development, TMHRI, for assistance with manuscript editing. REFERENCES 1. Banerjee, R., M. Gretes, L. Basuino, N. Strynadka, and H. F. Chambers. 2008. In vitro selection and characterization of ceftobiprole-resistant methicillin-resistant Staphylococcus aureus. Antimicrob. Agents Chemother. 52: 2089–2096. 2. Boisset, S., et al. 2007. Staphylococcus aureus RNAIII coordinately represses the synthesis of virulence factors and the transcription regulator Rot by an antisense mechanism. Genes Dev. 21:1353–1366. 3. Chakrabarti, S. K., and T. K. Misra. 2000. SarA represses agr operon expression in a purified in vitro Staphylococcus aureus transcription system. J. Bacteriol. 182:5893–5897. 4. Chambers, H. F., G. Archer, and M. Matsuhashi. 1989. Low-level methicillin resistance in strains of Staphylococcus aureus. Antimicrob. Agents Chemother. 33:424–428. 5. Cheung, A. L., and S. J. Projan. 1994. Cloning and sequencing of sarA of Staphylococcus aureus, a gene required for the expression of agr. J. Bacteriol. 176:4168–4172. 6. Cirz, R. T., et al. 2007. Complete and SOS-mediated response of Staphylococcus aureus to the antibiotic ciprofloxacin. J. Bacteriol. 189:531–539. 7. Clinical and Laboratory Standards Institute. 2008. Performance standards for antimicrobial susceptibility testing. CLSI document M100-S19, 19th informational supplement. CLSI, Wayne, PA. 8. Cui, L., J. Q. Lian, H. M. Neoh, E. Reyes, and K. Hiramatsu. 2005. DNA microarray-based identification of genes associated with glycopeptide resistance in Staphylococcus aureus. Antimicrob. Agents Chemother. 49:3404– 3413. 9. Cuirolo, A., K. Plata, and A. E. Rosato. 2009. Development of homogeneous expression of resistance in methicillin-resistant Staphylococcus aureus clinical strains is functionally associated with a beta-lactam-mediated SOS response. J. Antimicrob. Chemother. 64:37–45. 10. Dunman, P. M., et al. 2001. Transcription profiling-based identification of Staphylococcus aureus genes regulated by the agr and/or sarA loci. J. Bacteriol. 183:7341–7353. 11. Finan, J. E., A. E. Rosato, T. M. Dickinson, D. Ko, and G. L. Archer. 2002. Conversion of oxacillin-resistant staphylococci from heterotypic to homotypic resistance expression. Antimicrob. Agents Chemother. 46:24–30. 12. Forbes, B. A., et al. 2008. Unusual form of oxacillin resistance in methicillinresistant Staphylococcus aureus clinical strains. Diagn. Microbiol. Infect. Dis. 61:387–395. 13. Frank, E. G., D. G. Ennis, M. Gonzalez, A. S. Levine, and R. Woodgate. 1996. Regulation of SOS mutagenesis by proteolysis. Proc. Natl. Acad. Sci.U. S. A. 93:10291–10296. 14. Gilot, P., G. Lina, T. Cochard, and B. Poutrel. 2002. Analysis of the genetic variability of genes encoding the RNA III-activating components Agr and TRAP in a population of Staphylococcus aureus strains isolated from cows with mastitis. J. Clin. Microbiol. 40:4060–4067. 15. Goldstein, F., et al. 2007. Identification and phenotypic characterization of a beta-lactam-dependent, methicillin-resistant Staphylococcus aureus strain. Antimicrob. Agents Chemother. 51:2514–2522. 16. Herold, B. C., et al. 1998. Community-acquired methicillin-resistant Staphylococcus aureus in children with no identified predisposing risk. JAMA 279:593–598. 17. Heyer, G., et al. 2002. Staphylococcus aureus agr and sarA functions are required for invasive infection but not inflammatory responses in the lung. Infect. Immun. 70:127–133. 18. Ingavale, S. S., W. van Wamel, and A. L. Cheung. 2003. Characterization of RAT, an autolysis regulator in Staphylococcus aureus. Mol. Microbiol. 48: 1451–1466. 19. Ito, T., et al. 2001. Structural comparison of three types of staphylococcal cassette chromosome mec integrated in the chromosome in methicillinresistant Staphylococcus aureus. Antimicrob. Agents Chemother. 45:1323– 1336. 20. Janzon, L., and S. Arvidson. 1990. The role of the delta-lysin gene (hld) in the regulation of virulence genes by the accessory gene regulator (agr) in Staphylococcus aureus. EMBO J. 9:1391–1399. 21. Katayama, Y., T. Ito, and K. Hiramatsu. 2000. A new class of genetic element, staphylococcus cassette chromosome mec, encodes methicillin re-

3186

22. 23.

24. 25.

26.

27.

28. 29.

30.

31.

32. 33.

34.

35.

PLATA ET AL.

sistance in Staphylococcus aureus. Antimicrob. Agents Chemother. 44:1549– 1555. Kirby, W. M. 1944. Extraction of a highly potent pencilling inactivator from penicillin resistant staphylococci. Science 99:452–453. Kuroda, M., K. Kuwahara-Arai, and K. Hiramatsu. 2000. Identification of the up- and down-regulated genes in vancomycin-resistant Staphylococcus aureus strains Mu3 and Mu50 by cDNA differential hybridization method. Biochem. Biophys. Res. Commun. 269:485–490. Kuroda, M., et al. 2001. Whole genome sequencing of meticillin-resistant Staphylococcus aureus. Lancet 357:1225–1240. Lamichhane-Khadka, R., et al. 2009. sarA inactivation reduces vancomycinintermediate and ciprofloxacin resistance expression by Staphylococcus aureus. Int. J. Antimicrob. Agents 34:136–141. Layton, J. C., and P. L. Foster. 2003. Error-prone DNA polymerase IV is controlled by the stress-response sigma factor, RpoS, in Escherichia coli. Mol. Microbiol. 50:549–561. Lina, G., et al. 1998. Transmembrane topology and histidine protein kinase activity of AgrC, the agr signal receptor in Staphylococcus aureus. Mol. Microbiol. 28:655–662. Little, J. W. 1993. LexA cleavage and other self-processing reactions. J. Bacteriol. 175:4943–4950. Luong, T. T., and C. Y. Lee. 2006. The arl locus positively regulates Staphylococcus aureus type 5 capsule via an mgrA-dependent pathway. Microbiology 152:3123–3131. McKinney, T. K., V. K. Sharma, W. A. Craig, and G. L. Archer. 2001. Transcription of the gene mediating methicillin resistance in Staphylococcus aureus (mecA) is corepressed but not coinduced by cognate mecA and beta-lactamase regulators. J. Bacteriol. 183:6862–6868. McNamara, P. J., K. C. Milligan-Monroe, S. Khalili, and R. A. Proctor. 2000. Identification, cloning, and initial characterization of rot, a locus encoding a regulator of virulence factor expression in Staphylococcus aureus. J. Bacteriol. 182:3197–3203. Miller, C., et al. 2004. SOS response induction by beta-lactams and bacterial defense against antibiotic lethality. Science 305:1629–1631. Mongodin, E., et al. 2003. Microarray transcription analysis of clinical Staphylococcus aureus isolates resistant to vancomycin. J. Bacteriol. 185:4638– 4643. Morfeldt, E., L. Janzon, S. Arvidson, and S. Lofdahl. 1988. Cloning of a chromosomal locus (exp) which regulates the expression of several exoprotein genes in Staphylococcus aureus. Mol. Gen. Genet. 211:435–440. Morfeldt, E., I. Panova-Sapundjieva, B. Gustafsson, and S. Arvidson. 1996. Detection of the response regulator AgrA in the cytosolic fraction of Staphylococcus aureus by monoclonal antibodies. FEMS Microbiol. Lett. 143:195–201.

ANTIMICROB. AGENTS CHEMOTHER. 36. Novick, R. 1967. Properties of a cryptic high-frequency transducing phage in Staphylococcus aureus. Virology 33:155–166. 37. Novick, R. P., and E. Geisinger. 2008. Quorum sensing in staphylococci. Annu. Rev. Genet. 42:541–564. 38. Novick, R. P., et al. 1993. Synthesis of staphylococcal virulence factors is controlled by a regulatory RNA molecule. EMBO J. 12:3967–3975. 39. Peng, H. L., R. P. Novick, B. Kreiswirth, J. Kornblum, and P. Schlievert. 1988. Cloning, characterization, and sequencing of an accessory gene regulator (agr) in Staphylococcus aureus. J. Bacteriol. 170:4365–4372. 40. Perez-Capilla, T., et al. 2005. SOS-independent induction of dinB transcription by beta-lactam-mediated inhibition of cell wall synthesis in Escherichia coli. J. Bacteriol. 187:1515–1518. 41. Queck, S. Y., et al. 2008. RNAIII-independent target gene control by the agr quorum-sensing system: insight into the evolution of virulence regulation in Staphylococcus aureus. Mol. Cell 32:150–158. 42. Recsei, P., et al. 1986. Regulation of exoprotein gene expression in Staphylococcus aureus by agar. Mol. Gen. Genet. 202:58–61. 43. Romby, P., F. Vandenesch, and E. G. Wagner. 2006. The role of RNAs in the regulation of virulence-gene expression. Curr. Opin. Microbiol. 9:229–236. 44. Said-Salim, B., et al. 2003. Global regulation of Staphylococcus aureus genes by Rot. J. Bacteriol. 185:610–619. 45. Sakoulas, G., et al. 2002. Accessory gene regulator (agr) locus in geographically diverse Staphylococcus aureus isolates with reduced susceptibility to vancomycin. Antimicrob. Agents Chemother. 46:1492–1502. 46. Tattevin, P., L. Basuino, and H. F. Chambers. 2009. Subinhibitory fluoroquinolone exposure selects for reduced beta-lactam susceptibility in methicillin-resistant Staphylococcus aureus and alterations in the SOS-mediated response. Res. Microbiol. 160:187–192. 47. Tenaillon, O., E. Denamur, and I. Matic. 2004. Evolutionary significance of stress-induced mutagenesis in bacteria. Trends Microbiol. 12:264–270. 48. Toledo-Arana, A., F. Repoila, and P. Cossart. 2007. Small noncoding RNAs controlling pathogenesis. Curr. Opin. Microbiol. 10:182–188. 49. Tomasz, A., S. Nachman, and H. Leaf. 1991. Stable classes of phenotypic expression in methicillin-resistant clinical isolates of staphylococci. Antimicrob. Agents Chemother. 35:124–129. 50. van den Broek, D., A. W. T. Chin, G. V. Bloemberg, and B. J. Lugtenberg. 2005. Role of RpoS and MutS in phase variation of Pseudomonas sp. PCL1171. Microbiology 151:1403–1408. 51. Walsh, T. R., and R. A. Howe. 2002. The prevalence and mechanisms of vancomycin resistance in Staphylococcus aureus. Annu. Rev. Microbiol. 56: 657–675. 52. Zhang, L., L. Gray, R. P. Novick, and G. Ji. 2002. Transmembrane topology of AgrB, the protein involved in the post-translational modification of AgrD in Staphylococcus aureus. J. Biol. Chem. 277:34736–34742.