Overexpression of Genes of the Cell Wall Stimulon in Clinical Isolates of Staphylococcus aureus Exhibiting Vancomycin-Intermediate- S. aureu
http://www.100md.com
《细菌学杂志》
Wyeth Research, Pearl River, New York,Wyeth Research, Cambridge, Massachusetts,Wyeth Vaccines, Pearl River, New York,Laboratory of Microbiology, The Rockefeller University, New York, New York
ABSTRACT
Custom-designed gene chips (Affymetrix) were used to determine genetic relatedness and gene expression profiles in Staphylococcus aureus isolates with increasing MICs of vancomycin that were recovered over a period of several weeks from the blood and heart valve of a patient undergoing extensive vancomycin therapy. The isolates were found to be isogenic as determined by the GeneChip based genotyping approach and thus represented a unique opportunity to study changes in gene expression that may contribute to the vancomycin resistance phenotype. No differences in gene expression were detected between the parent strain, JH1, and JH15, isolated from the nares of a patient contact. Few expression changes were observed between blood and heart valve isolates with identical vancomycin MICs. A large number of genes had altered expression in the late stage JH9 isolate (MIC = 8 μg/ml) compared to JH1 (MIC = 1 μg/ml). Most genes with altered expression were involved in housekeeping functions or cell wall biosynthesis and regulation. The sortase-encoding genes, srtA and srtB, as well as several surface protein-encoding genes were downregulated in JH9. Two hypothetical protein-encoding genes, SAS016 and SA2343, were dramatically overexpressed in JH9. Interestingly, 27 of the genes with altered expression in JH9 grown in drug-free medium were found to be also overexpressed when the parental strain JH1 was briefly exposed to inhibitory concentrations of vancomycin, and more than half (17 of 27) of the genes with altered expression belonged to determinants that were proposed to form part of a general cell wall stress stimulon (S. Utaida et al., Microbiology 149:2719-2732, 2003).
INTRODUCTION
Staphylococcus aureus strains with decreased susceptibility to vancomycin (so-called vancomycin-intermediate S. aureus [VISA] strains) have been identified in clinical specimens in several countries during the past decade (10, 11, 25). Although several abnormal physiological properties have been described in VISA isolates, the mechanisms of resistance and whether or not it involves acquisition of genes has remained unclear. A major problem in mechanistic studies has been the lack of availability of isogenic vancomycin-susceptible strains that could be considered the parental strains of the VISA isolates. Even when a VISA isolate shared a common multilocus sequence type with a fully sequenced vancomycin-susceptible strain, the clinical origins and times of isolation were far apart, excluding the possibility of attributing genetic and phenotypic differences to the antibiotic susceptibility phenotype.
A recently described series of methicillin-resistant S. aureus (MRSA) isolates (JH1 through JH15) were recovered from a single patient during extensive chemotherapy with vancomycin (24). Bacterial isolates recovered at various times during therapy showed increasing vancomycin MICs which in isolates JH9 and JH14 eventually reached 8 μg/ml, a value typical of many clinical VISA isolates. The first isolate chronologically, JH1, and all of the subsequent isolates shared a common SCCmec element and multilocus sequence type and had identical PFGE patterns, indicating close genetic relatedness (24).
The purpose of the study described here was first to use a genome-scale DNA hybridization approach using custom designed S. aureus GeneChips (Affymetrix) to prove more rigorously the isogenic nature of the JH isolates. Once we established this, we planned to compare the gene expression profile of the vancomycin-susceptible "parental" isolate JH1 with those for the vancomycin-resistant isolates JH9 or JH14 in order to gain insights into the genetic and/or biochemical mechanism of VISA-type resistance.
MATERIALS AND METHODS
Strains, growth conditions, and determination of vancomycin susceptibility. The S. aureus isolates recovered from clinical material in Baltimore, Md., have been described previously (24). Bacteria were grown in tryptic soy broth (Difco, Detroit, MI) at 37°C with aeration. Growth was followed by monitoring the optical density at 620 nm (OD620) using an LKB spectrophotometer (Pharmacia LKB Biotechnology, Inc., Sweden). Viable titers were also determined routinely at sampling times by plating diluted cultures on tryptic soy agar (Difco). Vancomycin susceptibility was determined by the method of population analysis (24).
Treatment with vancomycin. JH1 and JH9 cultures were grown to exponential phase, corresponding to an OD620 of 0.4, at which time the cultures were treated with vancomycin at 1x and/or 8x of their respective MICs (in JH1 at 1 and 8 μg/ml and in JH9 at 8 μg/ml) for 30 min. Control cultures received no antibiotic. Portions of the cultures (200 μl) were removed at various intervals, serially diluted, and plated on tryptic soy agar. Colonies were counted after 48 h of incubation at 37°C.
Preparation of DNA and RNA. Preparation of chromosomal DNA was performed as described previously (1). During RNA preparation, cultures were mixed with two volumes of RNA Protect (QIAGEN, Inc., Valencia, Calif.) and, after incubation for 10 min at room temperature, bacterial pellets were collected by centrifugation and kept at –70°C. Extraction of RNA was carried out by using the RNeasy minikit (QIAGEN) according to the manufacturer's instructions.
GeneChip analysis of DNA and RNA samples. Custom-designed S. aureus GeneChips (Affymetrix, Inc., Santa Clara, Calif.) are described in detail by Dunman et al. (7). Chromosomal DNA was labeled as described previously (7). For RNA samples, reverse transcription, cDNA fragmentation, and terminal labeling of cDNA fragments with biotin were carried out according to the manufacturer's instructions for antisense prokaryotic arrays (Affymetrix). A total of 1.5 μg of labeled material was hybridized to each GeneChip for 16 h at 45°C. GeneChips were stained, washed, and scanned as described previously (2, 8). Signal intensities were normalized to account for loading errors and differences in labeling efficiencies. Gene expression results were calculated by using the average signal intensity from two biological replicates. All data analyzes were performed by using GeneSpring version 6.2 (Silicon Genetics) and Spotfire version 8.0. Dendrograms and Venn diagrams were generated by using GeneSpring software.
RESULTS
The JH strain series is isogenic. Sieradzki et al. have previously shown that the JH VISA isolates have identical pulsed-field gel electrophoresis patterns and share a common SCCmec type and multilocus sequence type (24). In order to determine whether the JH VISA isolates were isogenic, we performed a genome-scale DNA comparison for each of the JH strains using S. aureus GeneChips (Affymetrix). Based upon previously described methods, an absent, present, or marginal call was assigned for each of 7,792 genes and intergenic regions detected by this GeneChip. The scatter plots in Fig. 1 represent signal values for each of these GeneChip features for parent strain JH1 compared to JH9 or strain Mu50, a VISA isolate from Japan (11) with a similar pulse field type. Although small differences in the normalized signal intensities were detected between JH1 and JH9 (Fig. 1A), these did not reflect genetic differences between the two strains. Similarly, no differences in genetic content were detected for the other JH strains tested, including strains JH2, JH5, and JH14 or strain JH15 strain isolated from the nares of a patient contact (data not shown). By comparison, a large number of gene differences were detected between JH1 and Mu50 (Fig. 1B). These data suggest that the JH strain series are truly isogenic.
Genotypic characterization was subsequently extended to several VISA isolates from the United States and Japan, as well as the sequenced S. aureus strains, to assess the level of similarity between the JH isolates and previously characterized S. aureus strains. Cluster analysis confirmed that the JH strains are identical at the DNA level, as evident from the branch height grouping these strains (Fig. 2). The PC1 and PC3 VISA isolates (10, 11, 25) were also found to be identical. The JH series was most closely related to the VISA isolates from Michigan (VISA_MI), New Jersey (VISA_NJ), Pennsylvania (VISA_PA), and New York (VISA_PC). The Japanese VISA isolates, Mu50 and Mu3, were less closely related to the JH strain series.
Gene expression differences between JH1 and JH9. We have demonstrated that the JH series is isogenic, and as such it was possible to make valid gene expression comparisons between these strains. We compared the transcriptional profile of the JH1 parent strain and the vancomycin-resistant JH9 mutant in order to identify factors that may be contributing to vancomycin resistance. As many as 55 genes were found to be upregulated in JH9 and 169 genes were downregulated in JH9 compared to JH1 (Fig. 3). All genes with altered expression in JH9 are listed in Table 1 and are categorized according to functional group where possible.
The majority of the 169 genes with decreased expression in JH9 were involved in transport and metabolism of amino acids, nucleotides, carbohydrates, or inorganic ions and include the lac operon and the sirABC genes involved in iron transport. The tagB gene involved in teichoic acid biosynthesis had decreased expression in JH9. Interestingly, the mecA gene and its regulators mecRI were also found to be downregulated in JH9. A number of regulatory loci had decreased expression in JH9, including the sarS and sarU genes encoding MarR-like transcriptional regulators and the hld locus encoding RNAIII, the effector of the agr response. The icaR gene, involved in the regulation of the polysaccharide-producing ica locus was also downregulated in JH9, as was the lytR gene, part of the lytRS two-component regulatory system. The lytS gene showed a 1.9-fold decrease in expression in JH9 (data not shown) and therefore did not meet the 2-fold cutoff used to generate the gene list in Table 1. A number of surface protein-encoding genes were downregulated in JH9. These included spa, coa, sdrD, fnbAB, fib, sbi, clfB, sai-1, isdB, and the LPXTG protein-encoding genes SA0978 and SA2285. The srtA and srtB genes encoding the sortase enzyme involved in cross-linking of LPXTG surface proteins to the cell wall were also downregulated. The toxin-encoding genes, yent1, set15, lukS, and lukD and the cysteine protease sspB gene were downregulated in JH9. In addition, a large number of genes encoding hypothetical proteins of unknown function showed decreased expression in JH9.
A subset of the 55 genes that were upregulated in JH9 compared to JH1 (Fig. 3) are involved in cell envelope biogenesis and included genes belonging to the cap5 operon responsible for capsular polysaccharide synthesis and the icaD gene involved in surface polysaccharide synthesis. The fmtC gene involved in methicillin resistance also showed increased expression in JH9. However, the fmtB and fmhA genes with similar functions had decreased expression in JH9. The autolysin-encoding lytM and SA0620 genes were upregulated in JH9. Interestingly, the vraRS genes, encoding a two-component regulatory system and the ABC transporter encoding vraF and vraG genes that have previously been linked to glycopeptide resistance were upregulated by greater than twofold in JH9 (15). A large number of genes encoding hypothetical proteins were also upregulated in JH9. Two of these, SAS016 and SA2343, were particularly noteworthy with expression increases of 22.7- and 84.4-fold, respectively (Fig. 3).
Effect of vancomycin on gene expression in JH1 and JH9. The transcriptional profiles of JH1 and JH9 strains grown to exponential phase and treated with vancomycin were examined to identify genes whose expression is altered in response to this antibiotic. The significance of fold changes in gene expression listed was calculated by using P value cutoffs of 0.05 or 0.1. Seventy-two genes demonstrated expression changes of 2-fold upon treatment of JH1 with 8 μg of vancomycin/ml (8x MIC) (Table 2) and many of these also showed altered expression upon treatment with 1 μg of vancomycin/ml (1x MIC) (Table 2). In general, gene expression changes were greater after treatment with 8 μg of vancomycin/ml, suggesting a cumulative effect. The majority of genes whose expression was altered by vancomycin-encoded hypothetical proteins. The expression of a number of regulatory genes, including sarA, msrR, and vraRS, was also increased in the presence of vancomycin. In contrast, the sarS and rot regulatory genes were downregulated. Several cell wall-related genes were upregulated by vancomycin, and these included sgtB, encoding a transglycosylase domain protein similar to penicillin-binding protein 1A/1B in E. coli; murZ, which encodes a protein involved in cell wall biosynthesis; and the fmt gene encoding an autolysis and methicillin resistance-related protein. In addition, two autolysin-encoding genes were downregulated after vancomycin treatment. Most interestingly, 27 of the genes whose expression changed upon treatment of JH1 with 8 μg of vancomycin/ml also showed altered expression in the JH9 mutant grown in antibiotic-free medium. These include vraRS, sarS, sgtB, and the hypothetical protein-encoding genes SAS016 and SA2343. No changes in gene expression were detected when JH9 was treated with 8 μg of vancomycin/ml (1x MIC).
Gene expression differences between JH9 and JH14. JH9 and JH14 were isolated within 4 days of each other from the blood (JH9) and heart valve (JH14), respectively. Both have vancomycin MICs of 8 μg/ml (24). We have shown by using a genome-scale hybridization approach that these strains are identical (Fig. 1). The transcriptional profiles of strains JH9 and JH14 grown to exponential phase were compared in order to identify changes in gene expression that may correspond to environmental differences between their respective isolation sites (Table 3). Sixteen genes were identified with increased expression in JH14 compared to JH9, and five of these encoded ABC-type transport systems. The remaining genes encoded proteins involved in energy production, in conversion, transport and metabolism of amino acids or nucleotides, or in posttranslational modification or hypothetical proteins. No genes were downregulated in JH14 compared to JH9.
Identification of cell wall stress stimulon genes. Utaida et al. used transcriptional profiling to identify 105 genes in the S. aureus strain 8325-4 that were induced in response to the cell wall active antibiotics oxacillin, D-cycloserine, or bacitracin and proposed the idea of a cell wall stress stimulon in S. aureus (28). Similarly, Kuroda et al. examined the effects of vancomycin on gene expression in S. aureus N315 using microarrays (14) and found that 113 genes were upregulated in the presence of vancomycin. We compared the genes upregulated by vancomycin in the present study with those identified by the studies mentioned above and found 15 genes that are common to all three studies (Fig. 4). The genes encode proteins involved in cell wall biosynthesis (sgtB and murZ), teicoplanin resistance (tcaA), posttranslational modification (prsA and htrA), autolysis and methicillin resistance (fmt), and regulation (vraS), as well as several hypothetical proteins (Table 4). Ten of these genes were also found to be upregulated in JH9 compared to JH1. Only two genes, encoding autolysins, were downregulated in all three studies (Table 4).
DISCUSSION
The JH strains were isolated from a single patient during extensive vancomycin therapy. Consecutive isolates demonstrated a gradual reduction in vancomycin susceptibility and a decreased growth rate (24). We have shown here that the JH strains are isogenic by using a genome-scale genotyping approach. The JH strains are closely related to VISA isolates from the United States and Japan. The previously described PC1 and PC3 VISA strains were also isolated from a single patient and have vancomycin MICs of 2 and 8 μg/ml, respectively (25). We have shown that these two strains are genetically identical using the same methods.
Other studies that aimed to characterize the mechanisms responsible for the VISA phenotype were limited by the fact that no vancomycin-susceptible "parent" strain was available (15, 19). However, our findings clearly show that the JH strains are isogenic and thus represented a unique opportunity to study the changes in the expression of genes that are most likely associated with the vancomycin resistant phenotype. In the present study, we compared the transcriptional profile of the vancomycin-susceptible parent strain, JH1, with the vancomycin-resistant JH9 strain. JH1 was also compared to JH15, a vancomycin-susceptible strain isolated from the nares of a patient contact, and JH14, isolated from the patient's heart valve. No differences in gene expression were detected between JH1 and JH15 (data not shown), suggesting that this progenitor VISA isolate may exist in the community as a commensal isolate. Although both JH9 and JH14 had the same elevated MIC for vancomycin and were recovered at the same time, a small number of genes were overexpressed in JH14 compared to JH9. These include several ABC-type transporter genes and the clpC gene, which has previously been reported to play a role in stress tolerance and biofilm formation (9). JH14 was isolated from the heart valve, whereas JH9, like the rest of the JH isolates, was from the blood. It is possible that the altered expression of these genes in JH14 may play a role in the adaptation to and survival of the bacterium in the heart valve environment.
VISA strains often exhibit alterations in the cell wall that is believed to prevent access of vancomycin to its target (6, 22). It has previously been demonstrated that JH9 growing in antibiotic-free medium produces a thicker cell wall and poorly separated cells compared to JH1 (26). Furthermore, JH9 was found to grow slower and exhibited reduced peptidoglycan cross-linkage and cell wall turnover. In the present study, we identified a number of changes in gene expression that may be linked to this phenotype. Zymographic analysis revealed that the expression of a number of autolytic enzymes was increased in JH9 despite the observation that autolysis was suppressed in this strain (26). We have shown here that expression of a two autolysin genes, lytM and SA0620, are increased in JH9. The reduced autolysis phenotype observed in JH9 has been attributed to changes in cell wall teichoic acids (26). We observed a 3.7-fold decrease in expression of the tagB teichoic acid biosynthesis gene in JH9, which may alter the teichoic acid component of the cell wall. The cap5A-H genes were upregulated in JH9, and the overexpression of capsular polysaccharide may contribute to the amorphous extracellular material previously observed by using electron microscopy (26). Further differences in gene expression that alter the surface properties of JH9 include the downregulation of several surface protein genes, and in particular spa, with 50-fold-decreased expression. The srtA and srtB genes, responsible for cross-linking LPXTG motif-containing surface proteins to peptidoglycan, were also downregulated in JH9. Sieradzki and Tomasz attributed the reduced cross-linkage of peptidoglycan in JH9 with the suppressed expression of PBP4 (26). We only observed a 1.9-fold decrease in pbp4 gene expression (data not shown), but this may be enough to mediate these effects. Alternatively, posttranscriptional effects on PBP4 may be responsible for its reduced activity. We observed altered expression of the mecA, sgtB, fmtB, fmtC, and fmhA genes in the JH9 mutant, all of which encode proteins that play a characterized or putative role in cell wall biosynthesis (12, 13, 20, 21, 27). These changes likely contribute to the reduced rate of cell wall biosynthesis previously observed for JH9 (26). It is clear, however, that JH9 exists in a metabolically altered state, as evidenced by the reduced expression of a large number of housekeeping genes, and this may contribute to or be a consequence of the altered growth characteristics of JH9 compared to JH1 (24).
A number of regulatory genes were up- or downregulated in JH9 compared to JH1. The vraRS genes encoding a two-component regulatory system were overexpressed. This locus has previously been shown to play a role in vancomycin resistance and is induced in the presence of subinhibitory concentrations of this drug and other wall inhibitors (14, 15). The mecRI genes responsible for regulating the expression of mecA were also downregulated. It has been noted that a large percentage of VISA isolates belong to agr group II (18) and DNA hybridization studies have confirmed that the JH strains belong to this group (data not shown). The hld locus, encoding the RNAIII effector of the agr response was downregulated by 22-fold in the mutant. The agr response is growth phase dependent, and it is possible that, due to the growth defects observed for JH9, this strain may not have reached the same phase in the growth cycle as JH1 when the RNA was harvested. However, a previous study suggested that a loss of agr function caused by point mutations in the agr locus may contribute to the development of vancomycin resistance (23). It is possible that several of the gene expression changes observed for JH9 may be secondary effects due to altered expression of one of these regulatory loci and may not be directly related to the vancomycin resistance phenotype of JH9.
A large number of genes encoding hypothetical proteins had altered expression in JH9, but two open reading frame (ORFs), SAS016 and SA2343, were particularly noteworthy with expression increases of 23- and 84-fold, respectively. Bioinformatics analysis did not reveal many clues as to the function of the proteins encoded by these genes, but they are both conserved in all sequenced S. aureus and S. epidermidis strains. Interestingly, the region of DNA including and surrounding the SA2343 ORF (AB035451) was found to be overexpressed in two VISA isolates (15), and the SAS016 ORF has previously been annotated as vraX on the basis of its upregulation in the presence of imipenem (AB050664). Further characterization of these genes is under way.
We also examined the effect of vancomycin on gene expression in JH1 in order to test whether any of the genes which show altered expression in JH9 grown in antibiotic free medium, may also be involved in the vancomycin stress response. A total of 72 genes were identified as having expression changes of greater than twofold when JH1 was treated with 8 μg of vancomycin/ml (8x MIC) (P < 0.05 or 0.1). Many of these also showed expression changes, albeit smaller, upon treatment with 1 μg of vancomycin/ml, suggesting a dose-response effect. A large number of hypothetical protein-encoding genes had vancomycin-induced expression changes, and these included the SAS016 and SA2343 genes discussed above. In fact, the expression of the SAS016 ORF increases by >200-fold upon treatment with 8 μg of vancomycin/ml, suggesting that the SAS016 protein may play a significant role in the vancomycin stress response. A number of regulatory genes were up- or downregulated by vancomycin, including vraRS, as shown previously by Kuroda et al. (15). The VraRS two-component system has been shown to modulate the cell wall biosynthesis pathway in S. aureus (14), and 21 genes found to be under VraRS control were also identified here. These included the sgtB, murZ and fmt genes, which play a role in cell wall synthesis or turnover; the tcaA gene, which has been shown to play a role in glycopeptide resistance (3, 16); the yvqF gene, a homolog of which is upregulated by vancomycin and bacitracin in Bacillus subtilis (5, 6); and the SA2343 ORF discussed above. In addition, 27 of the 72 genes induced by vancomycin in JH1 were also found to have altered expression in the JH9 mutant grown in antibiotic-free medium, and these included 15 that were shown by Kuroda et al. to be under VraRS control (14). This suggests that the gene expression changes involved in the vancomycin stress response have become permanently registered in the JH9 genome even in the absence of vancomycin stress and may represent a secondary adaptation of the cell to a permanently stressed state.
Utaida et al. have proposed the idea of a cell wall stress stimulon representing a group of genes that are commonly induced upon treatment with cell wall active antibiotics (28). A similar cell wall stress stimulon has been identified in B. subtilis (5, 17). Utaida et al. examined the effects of oxacillin, D-cycloserine, and bacitracin on gene expression in S. aureus strain 8325-4. Similarly, Kuroda et al. described the effects of vancomycin on S. aureus strain N315. We identified 15 upregulated genes and 2 downregulated genes that are common in the studies described above, as well as the vancomycin-induced genes described here. The upregulated genes include vraS, sgtB, murZ, fmt, yvqF, tcaA, the SA2343 hypothetical protein-encoding gene and the htrA and prsA genes involved in posttranslational modification. A gene encoding a LytR-like regulator was also upregulated in all three studies. The LytRS two-component system has previously been shown to play a role in the control of autolysis (4). The two genes found to be downregulated in all three studies encode autolysins, which may be related to increased expression of the lytR-like gene. It is likely that the genes described above are key components of the cell wall stress stimulon, and it is not surprising, therefore, that many have functions related to cell wall metabolism.
The isogenic nature of the JH strain series provided a unique opportunity to study the gene expression changes involved in the development of the VISA phenotype. It is clear that this process is multifactorial. We have identified a number of genes that may play a role in the altered cell wall phenotype of the JH9 mutant and many additional housekeeping genes that are probably linked to the growth defects observed for this strain. The expression of numerous virulence genes was also downregulated in JH9 and implies a trade-off between bacterial resistance and fitness and/or virulence. In addition, we found that a subset of the genes with altered expression in JH9 are also involved in the vancomycin stress response. If the resistance phenotype observed for JH9 and other VISA isolates is related to a potentially unstable stress response, it may explain why VISA strains do not achieve vancomycin MICs higher than 16 μg/ml. We also performed a retrospective analysis to identify a subset of genes that are commonly up- or downregulated upon treatment with cell wall active antibiotics, and these are likely to play an important role in the cell wall stress stimulon.
The most provocative observation obtained in the present study was the striking similarity between the genetic determinants that were transiently overexpressed during vancomycin treatment of the susceptible "parental" isolate JH1 and were permanently overexpressed in the vancomycin-resistant isolates JH9 and JH14 grown in the absence of antibiotics. These findings strongly suggest that changes in the regulation of transcription of this particular set of genetic determinants are one of the fundamental features of the VISA-type resistance mechanism.
ACKNOWLEDGMENTS
Partial support for these investigations was provided by a grant from U.S. Public Health Service grant 2RO1 AI045738.
REFERENCES
Ausubel, F. M., R. Brent, R. E. Kingston, D. D. Moore, J. G. Seidman, J. A. Smith, and K. Struhl. 1995. Current protocols in molecular biology, p. 241-242. John Wiley & Sons, Inc., New York, N.Y.
Beenken, K. E., P. M. Dunman, F. McAleese, D. Macapagal, E. Murphy, S. J. Projan, J. S. Blevins, and M. S. Smeltzer. 2004. Global gene expression in Staphylococcus aureus biofilms. J. Bacteriol. 186:4665-4684.
Brandenberger, M., M. Tschierske, P. Giachino, A. Wada, and B. Berger-Bachi. 2000. Inactivation of a novel three-cistronic operon tcaR-tcaA-tcaB increases teicoplanin resistance in Staphylococcus aureus. Biochim. Biophys. Acta 1523:135-139.
Brunskill, E. W., and K. W. Bayles. 1996. Identification and molecular characterization of a putative regulatory locus that affects autolysis in Staphylococcus aureus. J. Bacteriol. 178:611-618.
Cao, M., T. Wang, R. Ye, and J. D. Helmann. 2002. Antibiotics that inhibit cell wall biosynthesis induce expression of the Bacillus subtilis sigma(W) and sigma(M) regulons. Mol. Microbiol. 45:1267-1276.
Cui, L., X. Ma, K. Sato, K. Okuma, F. C. Tenover, E. M. Mamizuka, C. G. Gemmell, M. N. Kim, M. C. Ploy, N. El-Solh, V. Ferraz, and K. Hiramatsu. 2003. Cell wall thickening is a common feature of vancomycin resistance in Staphylococcus aureus. J. Clin. Microbiol. 41:5-14.
Dunman, P. M., W. Mounts, F. McAleese, F. Immermann, D. Macapagal, E. Marsilio, L. McDougal, F. C. Tenover, P. A. Bradford, P. J. Petersen, S. J. Projan, and E. Murphy. 2004. Uses of Staphylococcus aureus GeneChips in genotyping and genetic composition analysis. J. Clin. Microbiol. 42:4275-4283.
Dunman, P. M., E. Murphy, S. Haney, D. Palacios, G. Tucker-Kellogg, S. Wu, E. L. Brown, R. J. Zagursky, D. Shlaes, and S. J. Projan. 2001. Transcription profiling-based identification of Staphylococcus aureus genes regulated by the agr and/or sarA loci. J. Bacteriol. 183:7341-7353.
Frees, D., A. Chastanet, S. Qazi, K. Sorensen, P. Hill, T. Msadek, and H. Ingmer. 2004. Clp ATPases are required for stress tolerance, intracellular replication and biofilm formation in Staphylococcus aureus. Mol. Microbiol. 54:1445-1462.
Hiramatsu, K. 2001. Vancomycin-resistant Staphylococcus aureus: a new model of antibiotic resistance. Lancet Infect. Dis. 1:147-155.
Hiramatsu, K., H. Hanaki, T. Ino, K. Yabuta, T. Oguri, and F. C. Tenover. 1997. Methicillin-resistant Staphylococcus aureus clinical strain with reduced vancomycin susceptibility. J. Antimicrob. Chemother. 40:135-136.
Komatsuzawa, H., K. Ohta, T. Fujiwara, G. H. Choi, H. Labischinski, and M. Sugai. 2001. Cloning and sequencing of the gene, fmtC, which affects oxacillin resistance in methicillin-resistant Staphylococcus aureus. FEMS Microbiol. Lett. 203:49-54.
Komatsuzawa, H., K. Ohta, M. Sugai, T. Fujiwara, P. Glanzmann, B. Berger Bachi, and H. Suginaka. 2000. Tn551-mediated insertional inactivation of the fmtB gene encoding a cell wall-associated protein abolishes methicillin resistance in Staphylococcus aureus. J. Antimicrob. Chemother. 45:421-431.
Kuroda, M., H. Kuroda, T. Oshima, F. Takeuchi, H. Mori, and K. Hiramatsu. 2003. Two-component system VraSR positively modulates the regulation of cell-wall biosynthesis pathway in Staphylococcus aureus. Mol. Microbiol. 49:807-821.
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.
Maki, H., N. McCallum, M. Bischoff, A. Wada, and B. Berger-Bachi. 2004. tcaA inactivation increases glycopeptide resistance in Staphylococcus aureus. Antimicrob. Agents Chemother. 48:1953-1959.
Mascher, T., N. G. Margulis, T. Wang, R. W. Ye, and J. D. Helmann. 2003. Cell wall stress responses in Bacillus subtilis: the regulatory network of the bacitracin stimulon. Mol. Microbiol. 50:1591-1604.
Moise-Broder, P. A., G. Sakoulas, G. M. Eliopoulos, J. J. Schentag, A. Forrest, and R. C. Moellering, Jr. 2004. Accessory gene regulator group II polymorphism in methicillin-resistant Staphylococcus aureus is predictive of failure of vancomycin therapy. Clin. Infect. Dis. 38:1700-1705.
Mongodin, E., J. Finan, M. W. Climo, A. Rosato, S. Gill, and G. L. Archer. 2003. Microarray transcription analysis of clinical Staphylococcus aureus isolates resistant to vancomycin. J. Bacteriol. 185:4638-4643.
Nishi, H., H. Komatsuzawa, S. Yamada, T. Fujiwara, M. Ohara, K. Ohta, M. Sugiyama, T. Ishikawa, and M. Sugai. 2003. Moenomycin resistance is associated with vancomycin-intermediate susceptibility in Staphylococcus aureus. Microbiol. Immunol. 47:927-935.
Pinho, M. G., S. R. Filipe, H. de Lencastre, and A. Tomasz. 2001. Complementation of the essential peptidoglycan transpeptidase function of penicillin-binding protein 2 (PBP2) by the drug resistance protein PBP2A in Staphylococcus aureus. J. Bacteriol. 183:6525-6531.
Reipert, A., K. Ehlert, T. Kast, and G. Bierbaum. 2003. Morphological and genetic differences in two isogenic Staphylococcus aureus strains with decreased susceptibilities to vancomycin. Antimicrob. Agents Chemother. 47:568-576.
Sakoulas, G., G. M. Eliopoulos, R. C. Moellering, Jr., C. Wennersten, L. Venkataraman, R. P. Novick, and H. S. Gold. 2002. Accessory gene regulator (agr) locus in geographically diverse Staphylococcus aureus isolates with reduced susceptibility to vancomycin. Antimicrob. Agents Chemother. 46:1492-1502.
Sieradzki, K., T. Leski, J. Dick, L. Borio, and A. Tomasz. 2003. Evolution of a vancomycin-intermediate Staphylococcus aureus strain in vivo: multiple changes in the antibiotic resistance phenotypes of a single lineage of methicillin-resistant S. aureus under the impact of antibiotics administered for chemotherapy. J. Clin. Microbiol. 41:1687-1693.
Sieradzki, K., R. B. Roberts, S. W. Haber, and A. Tomasz. 1999. The development of vancomycin resistance in a patient with methicillin-resistant Staphylococcus aureus infection. N. Engl. J. Med. 340:517-523.
Sieradzki, K., and A. Tomasz. 2003. Alterations of cell wall structure and metabolism accompany reduced susceptibility to vancomycin in an isogenic series of clinical isolates of Staphylococcus aureus. J. Bacteriol. 185:7103-7110.
Tschierske, M., C. Mori, S. Rohrer, K. Ehlert, K. J. Shaw, and B. Berger-Bachi. 1999. Identification of three additional femAB-like open reading frames in Staphylococcus aureus. FEMS Microbiol. Lett. 171:97-102.
Utaida, S., P. M. Dunman, D. Macapagal, E. Murphy, S. J. Projan, V. K. Singh, R. K. Jayaswal, and B. J. Wilkinson. 2003. Genome-wide transcriptional profiling of the response of Staphylococcus aureus to cell-wall-active antibiotics reveals a cell-wall-stress stimulon. Microbiology 149:2719-2732.(Fionnuala McAleese, Shang)
ABSTRACT
Custom-designed gene chips (Affymetrix) were used to determine genetic relatedness and gene expression profiles in Staphylococcus aureus isolates with increasing MICs of vancomycin that were recovered over a period of several weeks from the blood and heart valve of a patient undergoing extensive vancomycin therapy. The isolates were found to be isogenic as determined by the GeneChip based genotyping approach and thus represented a unique opportunity to study changes in gene expression that may contribute to the vancomycin resistance phenotype. No differences in gene expression were detected between the parent strain, JH1, and JH15, isolated from the nares of a patient contact. Few expression changes were observed between blood and heart valve isolates with identical vancomycin MICs. A large number of genes had altered expression in the late stage JH9 isolate (MIC = 8 μg/ml) compared to JH1 (MIC = 1 μg/ml). Most genes with altered expression were involved in housekeeping functions or cell wall biosynthesis and regulation. The sortase-encoding genes, srtA and srtB, as well as several surface protein-encoding genes were downregulated in JH9. Two hypothetical protein-encoding genes, SAS016 and SA2343, were dramatically overexpressed in JH9. Interestingly, 27 of the genes with altered expression in JH9 grown in drug-free medium were found to be also overexpressed when the parental strain JH1 was briefly exposed to inhibitory concentrations of vancomycin, and more than half (17 of 27) of the genes with altered expression belonged to determinants that were proposed to form part of a general cell wall stress stimulon (S. Utaida et al., Microbiology 149:2719-2732, 2003).
INTRODUCTION
Staphylococcus aureus strains with decreased susceptibility to vancomycin (so-called vancomycin-intermediate S. aureus [VISA] strains) have been identified in clinical specimens in several countries during the past decade (10, 11, 25). Although several abnormal physiological properties have been described in VISA isolates, the mechanisms of resistance and whether or not it involves acquisition of genes has remained unclear. A major problem in mechanistic studies has been the lack of availability of isogenic vancomycin-susceptible strains that could be considered the parental strains of the VISA isolates. Even when a VISA isolate shared a common multilocus sequence type with a fully sequenced vancomycin-susceptible strain, the clinical origins and times of isolation were far apart, excluding the possibility of attributing genetic and phenotypic differences to the antibiotic susceptibility phenotype.
A recently described series of methicillin-resistant S. aureus (MRSA) isolates (JH1 through JH15) were recovered from a single patient during extensive chemotherapy with vancomycin (24). Bacterial isolates recovered at various times during therapy showed increasing vancomycin MICs which in isolates JH9 and JH14 eventually reached 8 μg/ml, a value typical of many clinical VISA isolates. The first isolate chronologically, JH1, and all of the subsequent isolates shared a common SCCmec element and multilocus sequence type and had identical PFGE patterns, indicating close genetic relatedness (24).
The purpose of the study described here was first to use a genome-scale DNA hybridization approach using custom designed S. aureus GeneChips (Affymetrix) to prove more rigorously the isogenic nature of the JH isolates. Once we established this, we planned to compare the gene expression profile of the vancomycin-susceptible "parental" isolate JH1 with those for the vancomycin-resistant isolates JH9 or JH14 in order to gain insights into the genetic and/or biochemical mechanism of VISA-type resistance.
MATERIALS AND METHODS
Strains, growth conditions, and determination of vancomycin susceptibility. The S. aureus isolates recovered from clinical material in Baltimore, Md., have been described previously (24). Bacteria were grown in tryptic soy broth (Difco, Detroit, MI) at 37°C with aeration. Growth was followed by monitoring the optical density at 620 nm (OD620) using an LKB spectrophotometer (Pharmacia LKB Biotechnology, Inc., Sweden). Viable titers were also determined routinely at sampling times by plating diluted cultures on tryptic soy agar (Difco). Vancomycin susceptibility was determined by the method of population analysis (24).
Treatment with vancomycin. JH1 and JH9 cultures were grown to exponential phase, corresponding to an OD620 of 0.4, at which time the cultures were treated with vancomycin at 1x and/or 8x of their respective MICs (in JH1 at 1 and 8 μg/ml and in JH9 at 8 μg/ml) for 30 min. Control cultures received no antibiotic. Portions of the cultures (200 μl) were removed at various intervals, serially diluted, and plated on tryptic soy agar. Colonies were counted after 48 h of incubation at 37°C.
Preparation of DNA and RNA. Preparation of chromosomal DNA was performed as described previously (1). During RNA preparation, cultures were mixed with two volumes of RNA Protect (QIAGEN, Inc., Valencia, Calif.) and, after incubation for 10 min at room temperature, bacterial pellets were collected by centrifugation and kept at –70°C. Extraction of RNA was carried out by using the RNeasy minikit (QIAGEN) according to the manufacturer's instructions.
GeneChip analysis of DNA and RNA samples. Custom-designed S. aureus GeneChips (Affymetrix, Inc., Santa Clara, Calif.) are described in detail by Dunman et al. (7). Chromosomal DNA was labeled as described previously (7). For RNA samples, reverse transcription, cDNA fragmentation, and terminal labeling of cDNA fragments with biotin were carried out according to the manufacturer's instructions for antisense prokaryotic arrays (Affymetrix). A total of 1.5 μg of labeled material was hybridized to each GeneChip for 16 h at 45°C. GeneChips were stained, washed, and scanned as described previously (2, 8). Signal intensities were normalized to account for loading errors and differences in labeling efficiencies. Gene expression results were calculated by using the average signal intensity from two biological replicates. All data analyzes were performed by using GeneSpring version 6.2 (Silicon Genetics) and Spotfire version 8.0. Dendrograms and Venn diagrams were generated by using GeneSpring software.
RESULTS
The JH strain series is isogenic. Sieradzki et al. have previously shown that the JH VISA isolates have identical pulsed-field gel electrophoresis patterns and share a common SCCmec type and multilocus sequence type (24). In order to determine whether the JH VISA isolates were isogenic, we performed a genome-scale DNA comparison for each of the JH strains using S. aureus GeneChips (Affymetrix). Based upon previously described methods, an absent, present, or marginal call was assigned for each of 7,792 genes and intergenic regions detected by this GeneChip. The scatter plots in Fig. 1 represent signal values for each of these GeneChip features for parent strain JH1 compared to JH9 or strain Mu50, a VISA isolate from Japan (11) with a similar pulse field type. Although small differences in the normalized signal intensities were detected between JH1 and JH9 (Fig. 1A), these did not reflect genetic differences between the two strains. Similarly, no differences in genetic content were detected for the other JH strains tested, including strains JH2, JH5, and JH14 or strain JH15 strain isolated from the nares of a patient contact (data not shown). By comparison, a large number of gene differences were detected between JH1 and Mu50 (Fig. 1B). These data suggest that the JH strain series are truly isogenic.
Genotypic characterization was subsequently extended to several VISA isolates from the United States and Japan, as well as the sequenced S. aureus strains, to assess the level of similarity between the JH isolates and previously characterized S. aureus strains. Cluster analysis confirmed that the JH strains are identical at the DNA level, as evident from the branch height grouping these strains (Fig. 2). The PC1 and PC3 VISA isolates (10, 11, 25) were also found to be identical. The JH series was most closely related to the VISA isolates from Michigan (VISA_MI), New Jersey (VISA_NJ), Pennsylvania (VISA_PA), and New York (VISA_PC). The Japanese VISA isolates, Mu50 and Mu3, were less closely related to the JH strain series.
Gene expression differences between JH1 and JH9. We have demonstrated that the JH series is isogenic, and as such it was possible to make valid gene expression comparisons between these strains. We compared the transcriptional profile of the JH1 parent strain and the vancomycin-resistant JH9 mutant in order to identify factors that may be contributing to vancomycin resistance. As many as 55 genes were found to be upregulated in JH9 and 169 genes were downregulated in JH9 compared to JH1 (Fig. 3). All genes with altered expression in JH9 are listed in Table 1 and are categorized according to functional group where possible.
The majority of the 169 genes with decreased expression in JH9 were involved in transport and metabolism of amino acids, nucleotides, carbohydrates, or inorganic ions and include the lac operon and the sirABC genes involved in iron transport. The tagB gene involved in teichoic acid biosynthesis had decreased expression in JH9. Interestingly, the mecA gene and its regulators mecRI were also found to be downregulated in JH9. A number of regulatory loci had decreased expression in JH9, including the sarS and sarU genes encoding MarR-like transcriptional regulators and the hld locus encoding RNAIII, the effector of the agr response. The icaR gene, involved in the regulation of the polysaccharide-producing ica locus was also downregulated in JH9, as was the lytR gene, part of the lytRS two-component regulatory system. The lytS gene showed a 1.9-fold decrease in expression in JH9 (data not shown) and therefore did not meet the 2-fold cutoff used to generate the gene list in Table 1. A number of surface protein-encoding genes were downregulated in JH9. These included spa, coa, sdrD, fnbAB, fib, sbi, clfB, sai-1, isdB, and the LPXTG protein-encoding genes SA0978 and SA2285. The srtA and srtB genes encoding the sortase enzyme involved in cross-linking of LPXTG surface proteins to the cell wall were also downregulated. The toxin-encoding genes, yent1, set15, lukS, and lukD and the cysteine protease sspB gene were downregulated in JH9. In addition, a large number of genes encoding hypothetical proteins of unknown function showed decreased expression in JH9.
A subset of the 55 genes that were upregulated in JH9 compared to JH1 (Fig. 3) are involved in cell envelope biogenesis and included genes belonging to the cap5 operon responsible for capsular polysaccharide synthesis and the icaD gene involved in surface polysaccharide synthesis. The fmtC gene involved in methicillin resistance also showed increased expression in JH9. However, the fmtB and fmhA genes with similar functions had decreased expression in JH9. The autolysin-encoding lytM and SA0620 genes were upregulated in JH9. Interestingly, the vraRS genes, encoding a two-component regulatory system and the ABC transporter encoding vraF and vraG genes that have previously been linked to glycopeptide resistance were upregulated by greater than twofold in JH9 (15). A large number of genes encoding hypothetical proteins were also upregulated in JH9. Two of these, SAS016 and SA2343, were particularly noteworthy with expression increases of 22.7- and 84.4-fold, respectively (Fig. 3).
Effect of vancomycin on gene expression in JH1 and JH9. The transcriptional profiles of JH1 and JH9 strains grown to exponential phase and treated with vancomycin were examined to identify genes whose expression is altered in response to this antibiotic. The significance of fold changes in gene expression listed was calculated by using P value cutoffs of 0.05 or 0.1. Seventy-two genes demonstrated expression changes of 2-fold upon treatment of JH1 with 8 μg of vancomycin/ml (8x MIC) (Table 2) and many of these also showed altered expression upon treatment with 1 μg of vancomycin/ml (1x MIC) (Table 2). In general, gene expression changes were greater after treatment with 8 μg of vancomycin/ml, suggesting a cumulative effect. The majority of genes whose expression was altered by vancomycin-encoded hypothetical proteins. The expression of a number of regulatory genes, including sarA, msrR, and vraRS, was also increased in the presence of vancomycin. In contrast, the sarS and rot regulatory genes were downregulated. Several cell wall-related genes were upregulated by vancomycin, and these included sgtB, encoding a transglycosylase domain protein similar to penicillin-binding protein 1A/1B in E. coli; murZ, which encodes a protein involved in cell wall biosynthesis; and the fmt gene encoding an autolysis and methicillin resistance-related protein. In addition, two autolysin-encoding genes were downregulated after vancomycin treatment. Most interestingly, 27 of the genes whose expression changed upon treatment of JH1 with 8 μg of vancomycin/ml also showed altered expression in the JH9 mutant grown in antibiotic-free medium. These include vraRS, sarS, sgtB, and the hypothetical protein-encoding genes SAS016 and SA2343. No changes in gene expression were detected when JH9 was treated with 8 μg of vancomycin/ml (1x MIC).
Gene expression differences between JH9 and JH14. JH9 and JH14 were isolated within 4 days of each other from the blood (JH9) and heart valve (JH14), respectively. Both have vancomycin MICs of 8 μg/ml (24). We have shown by using a genome-scale hybridization approach that these strains are identical (Fig. 1). The transcriptional profiles of strains JH9 and JH14 grown to exponential phase were compared in order to identify changes in gene expression that may correspond to environmental differences between their respective isolation sites (Table 3). Sixteen genes were identified with increased expression in JH14 compared to JH9, and five of these encoded ABC-type transport systems. The remaining genes encoded proteins involved in energy production, in conversion, transport and metabolism of amino acids or nucleotides, or in posttranslational modification or hypothetical proteins. No genes were downregulated in JH14 compared to JH9.
Identification of cell wall stress stimulon genes. Utaida et al. used transcriptional profiling to identify 105 genes in the S. aureus strain 8325-4 that were induced in response to the cell wall active antibiotics oxacillin, D-cycloserine, or bacitracin and proposed the idea of a cell wall stress stimulon in S. aureus (28). Similarly, Kuroda et al. examined the effects of vancomycin on gene expression in S. aureus N315 using microarrays (14) and found that 113 genes were upregulated in the presence of vancomycin. We compared the genes upregulated by vancomycin in the present study with those identified by the studies mentioned above and found 15 genes that are common to all three studies (Fig. 4). The genes encode proteins involved in cell wall biosynthesis (sgtB and murZ), teicoplanin resistance (tcaA), posttranslational modification (prsA and htrA), autolysis and methicillin resistance (fmt), and regulation (vraS), as well as several hypothetical proteins (Table 4). Ten of these genes were also found to be upregulated in JH9 compared to JH1. Only two genes, encoding autolysins, were downregulated in all three studies (Table 4).
DISCUSSION
The JH strains were isolated from a single patient during extensive vancomycin therapy. Consecutive isolates demonstrated a gradual reduction in vancomycin susceptibility and a decreased growth rate (24). We have shown here that the JH strains are isogenic by using a genome-scale genotyping approach. The JH strains are closely related to VISA isolates from the United States and Japan. The previously described PC1 and PC3 VISA strains were also isolated from a single patient and have vancomycin MICs of 2 and 8 μg/ml, respectively (25). We have shown that these two strains are genetically identical using the same methods.
Other studies that aimed to characterize the mechanisms responsible for the VISA phenotype were limited by the fact that no vancomycin-susceptible "parent" strain was available (15, 19). However, our findings clearly show that the JH strains are isogenic and thus represented a unique opportunity to study the changes in the expression of genes that are most likely associated with the vancomycin resistant phenotype. In the present study, we compared the transcriptional profile of the vancomycin-susceptible parent strain, JH1, with the vancomycin-resistant JH9 strain. JH1 was also compared to JH15, a vancomycin-susceptible strain isolated from the nares of a patient contact, and JH14, isolated from the patient's heart valve. No differences in gene expression were detected between JH1 and JH15 (data not shown), suggesting that this progenitor VISA isolate may exist in the community as a commensal isolate. Although both JH9 and JH14 had the same elevated MIC for vancomycin and were recovered at the same time, a small number of genes were overexpressed in JH14 compared to JH9. These include several ABC-type transporter genes and the clpC gene, which has previously been reported to play a role in stress tolerance and biofilm formation (9). JH14 was isolated from the heart valve, whereas JH9, like the rest of the JH isolates, was from the blood. It is possible that the altered expression of these genes in JH14 may play a role in the adaptation to and survival of the bacterium in the heart valve environment.
VISA strains often exhibit alterations in the cell wall that is believed to prevent access of vancomycin to its target (6, 22). It has previously been demonstrated that JH9 growing in antibiotic-free medium produces a thicker cell wall and poorly separated cells compared to JH1 (26). Furthermore, JH9 was found to grow slower and exhibited reduced peptidoglycan cross-linkage and cell wall turnover. In the present study, we identified a number of changes in gene expression that may be linked to this phenotype. Zymographic analysis revealed that the expression of a number of autolytic enzymes was increased in JH9 despite the observation that autolysis was suppressed in this strain (26). We have shown here that expression of a two autolysin genes, lytM and SA0620, are increased in JH9. The reduced autolysis phenotype observed in JH9 has been attributed to changes in cell wall teichoic acids (26). We observed a 3.7-fold decrease in expression of the tagB teichoic acid biosynthesis gene in JH9, which may alter the teichoic acid component of the cell wall. The cap5A-H genes were upregulated in JH9, and the overexpression of capsular polysaccharide may contribute to the amorphous extracellular material previously observed by using electron microscopy (26). Further differences in gene expression that alter the surface properties of JH9 include the downregulation of several surface protein genes, and in particular spa, with 50-fold-decreased expression. The srtA and srtB genes, responsible for cross-linking LPXTG motif-containing surface proteins to peptidoglycan, were also downregulated in JH9. Sieradzki and Tomasz attributed the reduced cross-linkage of peptidoglycan in JH9 with the suppressed expression of PBP4 (26). We only observed a 1.9-fold decrease in pbp4 gene expression (data not shown), but this may be enough to mediate these effects. Alternatively, posttranscriptional effects on PBP4 may be responsible for its reduced activity. We observed altered expression of the mecA, sgtB, fmtB, fmtC, and fmhA genes in the JH9 mutant, all of which encode proteins that play a characterized or putative role in cell wall biosynthesis (12, 13, 20, 21, 27). These changes likely contribute to the reduced rate of cell wall biosynthesis previously observed for JH9 (26). It is clear, however, that JH9 exists in a metabolically altered state, as evidenced by the reduced expression of a large number of housekeeping genes, and this may contribute to or be a consequence of the altered growth characteristics of JH9 compared to JH1 (24).
A number of regulatory genes were up- or downregulated in JH9 compared to JH1. The vraRS genes encoding a two-component regulatory system were overexpressed. This locus has previously been shown to play a role in vancomycin resistance and is induced in the presence of subinhibitory concentrations of this drug and other wall inhibitors (14, 15). The mecRI genes responsible for regulating the expression of mecA were also downregulated. It has been noted that a large percentage of VISA isolates belong to agr group II (18) and DNA hybridization studies have confirmed that the JH strains belong to this group (data not shown). The hld locus, encoding the RNAIII effector of the agr response was downregulated by 22-fold in the mutant. The agr response is growth phase dependent, and it is possible that, due to the growth defects observed for JH9, this strain may not have reached the same phase in the growth cycle as JH1 when the RNA was harvested. However, a previous study suggested that a loss of agr function caused by point mutations in the agr locus may contribute to the development of vancomycin resistance (23). It is possible that several of the gene expression changes observed for JH9 may be secondary effects due to altered expression of one of these regulatory loci and may not be directly related to the vancomycin resistance phenotype of JH9.
A large number of genes encoding hypothetical proteins had altered expression in JH9, but two open reading frame (ORFs), SAS016 and SA2343, were particularly noteworthy with expression increases of 23- and 84-fold, respectively. Bioinformatics analysis did not reveal many clues as to the function of the proteins encoded by these genes, but they are both conserved in all sequenced S. aureus and S. epidermidis strains. Interestingly, the region of DNA including and surrounding the SA2343 ORF (AB035451) was found to be overexpressed in two VISA isolates (15), and the SAS016 ORF has previously been annotated as vraX on the basis of its upregulation in the presence of imipenem (AB050664). Further characterization of these genes is under way.
We also examined the effect of vancomycin on gene expression in JH1 in order to test whether any of the genes which show altered expression in JH9 grown in antibiotic free medium, may also be involved in the vancomycin stress response. A total of 72 genes were identified as having expression changes of greater than twofold when JH1 was treated with 8 μg of vancomycin/ml (8x MIC) (P < 0.05 or 0.1). Many of these also showed expression changes, albeit smaller, upon treatment with 1 μg of vancomycin/ml, suggesting a dose-response effect. A large number of hypothetical protein-encoding genes had vancomycin-induced expression changes, and these included the SAS016 and SA2343 genes discussed above. In fact, the expression of the SAS016 ORF increases by >200-fold upon treatment with 8 μg of vancomycin/ml, suggesting that the SAS016 protein may play a significant role in the vancomycin stress response. A number of regulatory genes were up- or downregulated by vancomycin, including vraRS, as shown previously by Kuroda et al. (15). The VraRS two-component system has been shown to modulate the cell wall biosynthesis pathway in S. aureus (14), and 21 genes found to be under VraRS control were also identified here. These included the sgtB, murZ and fmt genes, which play a role in cell wall synthesis or turnover; the tcaA gene, which has been shown to play a role in glycopeptide resistance (3, 16); the yvqF gene, a homolog of which is upregulated by vancomycin and bacitracin in Bacillus subtilis (5, 6); and the SA2343 ORF discussed above. In addition, 27 of the 72 genes induced by vancomycin in JH1 were also found to have altered expression in the JH9 mutant grown in antibiotic-free medium, and these included 15 that were shown by Kuroda et al. to be under VraRS control (14). This suggests that the gene expression changes involved in the vancomycin stress response have become permanently registered in the JH9 genome even in the absence of vancomycin stress and may represent a secondary adaptation of the cell to a permanently stressed state.
Utaida et al. have proposed the idea of a cell wall stress stimulon representing a group of genes that are commonly induced upon treatment with cell wall active antibiotics (28). A similar cell wall stress stimulon has been identified in B. subtilis (5, 17). Utaida et al. examined the effects of oxacillin, D-cycloserine, and bacitracin on gene expression in S. aureus strain 8325-4. Similarly, Kuroda et al. described the effects of vancomycin on S. aureus strain N315. We identified 15 upregulated genes and 2 downregulated genes that are common in the studies described above, as well as the vancomycin-induced genes described here. The upregulated genes include vraS, sgtB, murZ, fmt, yvqF, tcaA, the SA2343 hypothetical protein-encoding gene and the htrA and prsA genes involved in posttranslational modification. A gene encoding a LytR-like regulator was also upregulated in all three studies. The LytRS two-component system has previously been shown to play a role in the control of autolysis (4). The two genes found to be downregulated in all three studies encode autolysins, which may be related to increased expression of the lytR-like gene. It is likely that the genes described above are key components of the cell wall stress stimulon, and it is not surprising, therefore, that many have functions related to cell wall metabolism.
The isogenic nature of the JH strain series provided a unique opportunity to study the gene expression changes involved in the development of the VISA phenotype. It is clear that this process is multifactorial. We have identified a number of genes that may play a role in the altered cell wall phenotype of the JH9 mutant and many additional housekeeping genes that are probably linked to the growth defects observed for this strain. The expression of numerous virulence genes was also downregulated in JH9 and implies a trade-off between bacterial resistance and fitness and/or virulence. In addition, we found that a subset of the genes with altered expression in JH9 are also involved in the vancomycin stress response. If the resistance phenotype observed for JH9 and other VISA isolates is related to a potentially unstable stress response, it may explain why VISA strains do not achieve vancomycin MICs higher than 16 μg/ml. We also performed a retrospective analysis to identify a subset of genes that are commonly up- or downregulated upon treatment with cell wall active antibiotics, and these are likely to play an important role in the cell wall stress stimulon.
The most provocative observation obtained in the present study was the striking similarity between the genetic determinants that were transiently overexpressed during vancomycin treatment of the susceptible "parental" isolate JH1 and were permanently overexpressed in the vancomycin-resistant isolates JH9 and JH14 grown in the absence of antibiotics. These findings strongly suggest that changes in the regulation of transcription of this particular set of genetic determinants are one of the fundamental features of the VISA-type resistance mechanism.
ACKNOWLEDGMENTS
Partial support for these investigations was provided by a grant from U.S. Public Health Service grant 2RO1 AI045738.
REFERENCES
Ausubel, F. M., R. Brent, R. E. Kingston, D. D. Moore, J. G. Seidman, J. A. Smith, and K. Struhl. 1995. Current protocols in molecular biology, p. 241-242. John Wiley & Sons, Inc., New York, N.Y.
Beenken, K. E., P. M. Dunman, F. McAleese, D. Macapagal, E. Murphy, S. J. Projan, J. S. Blevins, and M. S. Smeltzer. 2004. Global gene expression in Staphylococcus aureus biofilms. J. Bacteriol. 186:4665-4684.
Brandenberger, M., M. Tschierske, P. Giachino, A. Wada, and B. Berger-Bachi. 2000. Inactivation of a novel three-cistronic operon tcaR-tcaA-tcaB increases teicoplanin resistance in Staphylococcus aureus. Biochim. Biophys. Acta 1523:135-139.
Brunskill, E. W., and K. W. Bayles. 1996. Identification and molecular characterization of a putative regulatory locus that affects autolysis in Staphylococcus aureus. J. Bacteriol. 178:611-618.
Cao, M., T. Wang, R. Ye, and J. D. Helmann. 2002. Antibiotics that inhibit cell wall biosynthesis induce expression of the Bacillus subtilis sigma(W) and sigma(M) regulons. Mol. Microbiol. 45:1267-1276.
Cui, L., X. Ma, K. Sato, K. Okuma, F. C. Tenover, E. M. Mamizuka, C. G. Gemmell, M. N. Kim, M. C. Ploy, N. El-Solh, V. Ferraz, and K. Hiramatsu. 2003. Cell wall thickening is a common feature of vancomycin resistance in Staphylococcus aureus. J. Clin. Microbiol. 41:5-14.
Dunman, P. M., W. Mounts, F. McAleese, F. Immermann, D. Macapagal, E. Marsilio, L. McDougal, F. C. Tenover, P. A. Bradford, P. J. Petersen, S. J. Projan, and E. Murphy. 2004. Uses of Staphylococcus aureus GeneChips in genotyping and genetic composition analysis. J. Clin. Microbiol. 42:4275-4283.
Dunman, P. M., E. Murphy, S. Haney, D. Palacios, G. Tucker-Kellogg, S. Wu, E. L. Brown, R. J. Zagursky, D. Shlaes, and S. J. Projan. 2001. Transcription profiling-based identification of Staphylococcus aureus genes regulated by the agr and/or sarA loci. J. Bacteriol. 183:7341-7353.
Frees, D., A. Chastanet, S. Qazi, K. Sorensen, P. Hill, T. Msadek, and H. Ingmer. 2004. Clp ATPases are required for stress tolerance, intracellular replication and biofilm formation in Staphylococcus aureus. Mol. Microbiol. 54:1445-1462.
Hiramatsu, K. 2001. Vancomycin-resistant Staphylococcus aureus: a new model of antibiotic resistance. Lancet Infect. Dis. 1:147-155.
Hiramatsu, K., H. Hanaki, T. Ino, K. Yabuta, T. Oguri, and F. C. Tenover. 1997. Methicillin-resistant Staphylococcus aureus clinical strain with reduced vancomycin susceptibility. J. Antimicrob. Chemother. 40:135-136.
Komatsuzawa, H., K. Ohta, T. Fujiwara, G. H. Choi, H. Labischinski, and M. Sugai. 2001. Cloning and sequencing of the gene, fmtC, which affects oxacillin resistance in methicillin-resistant Staphylococcus aureus. FEMS Microbiol. Lett. 203:49-54.
Komatsuzawa, H., K. Ohta, M. Sugai, T. Fujiwara, P. Glanzmann, B. Berger Bachi, and H. Suginaka. 2000. Tn551-mediated insertional inactivation of the fmtB gene encoding a cell wall-associated protein abolishes methicillin resistance in Staphylococcus aureus. J. Antimicrob. Chemother. 45:421-431.
Kuroda, M., H. Kuroda, T. Oshima, F. Takeuchi, H. Mori, and K. Hiramatsu. 2003. Two-component system VraSR positively modulates the regulation of cell-wall biosynthesis pathway in Staphylococcus aureus. Mol. Microbiol. 49:807-821.
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.
Maki, H., N. McCallum, M. Bischoff, A. Wada, and B. Berger-Bachi. 2004. tcaA inactivation increases glycopeptide resistance in Staphylococcus aureus. Antimicrob. Agents Chemother. 48:1953-1959.
Mascher, T., N. G. Margulis, T. Wang, R. W. Ye, and J. D. Helmann. 2003. Cell wall stress responses in Bacillus subtilis: the regulatory network of the bacitracin stimulon. Mol. Microbiol. 50:1591-1604.
Moise-Broder, P. A., G. Sakoulas, G. M. Eliopoulos, J. J. Schentag, A. Forrest, and R. C. Moellering, Jr. 2004. Accessory gene regulator group II polymorphism in methicillin-resistant Staphylococcus aureus is predictive of failure of vancomycin therapy. Clin. Infect. Dis. 38:1700-1705.
Mongodin, E., J. Finan, M. W. Climo, A. Rosato, S. Gill, and G. L. Archer. 2003. Microarray transcription analysis of clinical Staphylococcus aureus isolates resistant to vancomycin. J. Bacteriol. 185:4638-4643.
Nishi, H., H. Komatsuzawa, S. Yamada, T. Fujiwara, M. Ohara, K. Ohta, M. Sugiyama, T. Ishikawa, and M. Sugai. 2003. Moenomycin resistance is associated with vancomycin-intermediate susceptibility in Staphylococcus aureus. Microbiol. Immunol. 47:927-935.
Pinho, M. G., S. R. Filipe, H. de Lencastre, and A. Tomasz. 2001. Complementation of the essential peptidoglycan transpeptidase function of penicillin-binding protein 2 (PBP2) by the drug resistance protein PBP2A in Staphylococcus aureus. J. Bacteriol. 183:6525-6531.
Reipert, A., K. Ehlert, T. Kast, and G. Bierbaum. 2003. Morphological and genetic differences in two isogenic Staphylococcus aureus strains with decreased susceptibilities to vancomycin. Antimicrob. Agents Chemother. 47:568-576.
Sakoulas, G., G. M. Eliopoulos, R. C. Moellering, Jr., C. Wennersten, L. Venkataraman, R. P. Novick, and H. S. Gold. 2002. Accessory gene regulator (agr) locus in geographically diverse Staphylococcus aureus isolates with reduced susceptibility to vancomycin. Antimicrob. Agents Chemother. 46:1492-1502.
Sieradzki, K., T. Leski, J. Dick, L. Borio, and A. Tomasz. 2003. Evolution of a vancomycin-intermediate Staphylococcus aureus strain in vivo: multiple changes in the antibiotic resistance phenotypes of a single lineage of methicillin-resistant S. aureus under the impact of antibiotics administered for chemotherapy. J. Clin. Microbiol. 41:1687-1693.
Sieradzki, K., R. B. Roberts, S. W. Haber, and A. Tomasz. 1999. The development of vancomycin resistance in a patient with methicillin-resistant Staphylococcus aureus infection. N. Engl. J. Med. 340:517-523.
Sieradzki, K., and A. Tomasz. 2003. Alterations of cell wall structure and metabolism accompany reduced susceptibility to vancomycin in an isogenic series of clinical isolates of Staphylococcus aureus. J. Bacteriol. 185:7103-7110.
Tschierske, M., C. Mori, S. Rohrer, K. Ehlert, K. J. Shaw, and B. Berger-Bachi. 1999. Identification of three additional femAB-like open reading frames in Staphylococcus aureus. FEMS Microbiol. Lett. 171:97-102.
Utaida, S., P. M. Dunman, D. Macapagal, E. Murphy, S. J. Projan, V. K. Singh, R. K. Jayaswal, and B. J. Wilkinson. 2003. Genome-wide transcriptional profiling of the response of Staphylococcus aureus to cell-wall-active antibiotics reveals a cell-wall-stress stimulon. Microbiology 149:2719-2732.(Fionnuala McAleese, Shang)