Skip to main content
NCBI home page
Antimicrobial Agents and Chemotherapy logoLink to Antimicrobial Agents and Chemotherapy
. 2004 Dec;48(12):4800–4807. doi: 10.1128/AAC.48.12.4800-4807.2004

Reduced Content of Lysyl-Phosphatidylglycerol in the Cytoplasmic Membrane Affects Susceptibility to Moenomycin, as Well as Vancomycin, Gentamicin, and Antimicrobial Peptides, in Staphylococcus aureus

Hiromi Nishi 1, Hitoshi Komatsuzawa 1,*, Tamaki Fujiwara 1, Nadine McCallum 2, Motoyuki Sugai 1
PMCID: PMC529239  PMID: 15561859

Abstract

An association between moenomycin resistance and vancomycin intermediate resistance in Staphylococcus aureus was demonstrated previously. Thus, to elucidate the mechanism of vancomycin intermediate resistance, we searched for factors contributing to moenomycin resistance. Random Tn551 insertional mutagenesis of methicillin-resistant S. aureus strain COL yielded three mutants with decreased susceptibilities to moenomycin. Correspondingly, these mutants also exhibited slightly decreased susceptibilities to vancomycin. Genetic analysis revealed that two of the mutants had Tn551 insertions in the fmtC (mprF) gene, which is associated with the synthesis of lysyl-phosphatidylglycerol. The third Tn551 insertion was located in the lysC gene, which is involved in the biosynthesis of lysine from aspartic acid. Consequently, mutations in both of these loci reduced the lysyl-phosphatidylglycerol content in the cell membrane, giving it a more negative net charge. The positively charged antibiotic gentamicin and cationic antimicrobial peptides such as β-defensins and CAP18 were more effective against the mutants. The levels of moenomycin and vancomycin binding to intact cells was also greater in the mutants than in the wild type, while the binding affinity was not altered when cells boiled in sodium dodecyl sulfate were used, indicating that both agents had higher affinities for the negatively charged membranes of the mutants. Therefore, the membrane charge of S. aureus appears to influence the efficacies of moenomycin, vancomycin, and other cationic antimicrobial agents.

Staphylococcus aureus is one of the major infectious disease-causing pathogens in humans. Antibiotics such as β-lactams, aminoglycosides, and quinolones are generally used for chemotherapy for these diseases; however, the emergence of methicillin-resistant S. aureus (MRSA), especially multidrug-resistant MRSA, has posed serious problems. Glycopeptides, such as vancomycin and teicoplanin, are usually effective against MRSA infections; however, vancomycin intermediate-resistant S. aureus (VISA) or glycopeptide intermediate-resistant S. aureus strains have now been isolated in several countries (3, 16, 29, 30). Highly vancomycin resistant strains, which contain the enterococcal vanA gene, have also emerged very recently (6), but the mechanism of resistance is clearly distinct from that in VISA strains. The mechanism used to acquire intermediate susceptibility is considered to be mediated by cell wall thickening, which facilitates the trapping of more vancomycin within the layers of cell wall peptidoglycan, which prevents it from reaching its target for lethality, namely, the peptidoglycan precursor attached to the cell membrane (7, 11). Although some factors affecting vancomycin susceptibility have been reported (2, 4, 5, 18, 19, 27, 28), the exact mechanism and the entire complement of factors associated with cell wall thickening and intermediate susceptibility remain unknown.

Spontaneous VISA mutants were previously isolated from an MRSA strain by exposure to vancomycin (11). Characterization of the mutants revealed that they contained a longer glycan chain length within their peptidoglycan and had decreased moenomycin susceptibility. Moenomycin is a cell wall synthesis inhibitor that acts on transglycosylases, which are thought to mediate the formation of the glycan chain of the peptidoglycan and the incorporation of the peptidoglycan precursor into cell wall peptidoglycan (9). Vancomycin also inhibits cell wall synthesis, although its target is different. Vancomycin binds to terminal d-Ala-d-Ala residues of the peptidoglycan precursor, which inhibits peptidoglycan chain elongation (1). Moenomycin-resistant mutants were then selected by exposure of the same parent strain, S. aureus COL, to moenomycin (21). These mutants also had a longer glycan chain length and a vancomycin-intermediate phenotype, thereby strongly suggesting that moenomycin resistance is associated with vancomycin intermediate resistance. Therefore, identification of the factor(s) that affects moenomycin susceptibility may contribute to the resolution of the mechanism of vancomycin intermediate susceptibility. However, when spontaneous mutants are used, it is difficult to identify the genetic factors facilitating phenotypic alterations, such as moenomycin or vancomycin resistance.

In the study described here, we isolated moenomycin-resistant mutants using transposon Tn551 mutagenesis and identified two genes involved in the synthesis of lysyl-phosphatidylglycerol.

MATERIALS AND METHODS

Bacterial strains and growth conditions.

The bacterial strains used in this study are listed in Table 1. S. aureus strains were grown in Trypticase soy broth (TSB; Becton Dickinson Microbiology Systems, Cockeysville, Md.). Erythromycin (30 μg/ml), chloramphenicol (50 μg/ml), or ampicillin (100 μg/ml) was added when necessary.

TABLE 1.

Strains used in this study

Strain Relevant characteristic(s) Reference or source
S. aureus
    COL Wild type 15
    COL with pRN3208 pRN3208; a thermosensitive plasmid harboring Tn551, Emr Cdr 15
    HN001 fmtC::Tn551 in COL, Emr This study
    HN002 lysC::Tn551 in COL, Emr This study
    HN003 fmtC::Tn551 in COL, Emr This study
    HN004 fmtC::Tn551 in COL, Emr; backcross from HN001 This study
    HN005 lysC::Tn551 in COL, Emr; backcross from HN002 This study
    HN006 fmtC::Tn551 in COL, Emr; backcross from HN003 This study
    HN007 lysC::Tn551 in KSA8, Emr; genetic outcross from HN002 This study
    COL-TS4 fmtC::Tn551 in COL, Emr 12
    KSA8 Wild type, methicillin resistant 12
    KSA8-TS4 fmtC::Tn551 in KSA8, Em-r 12
    BB270 Laboratory strain, methicilin resistant 12
    BB270-TS4 fmtC::Tn551 in BB270, Emr 12
    COL-VR1 Laboratory mutant from COL, vancomycin intermediate susceptible 11
    COL-VR1-TS4 fmtC::Tn551 in COL-VR1, Emr; genetic outcross from HN002 This study
E. coli
    pGEM-Teasy PCR fragment cloning vector, Ampr Promega
    HN101 PCR fragment (pTNR6 and pTNR8 from HN001 chromosome)/pGEM-Teasy This study
    HN102 PCR fragment (pTNR6 and pTNR8 from HN002 chromosome)/pGEM-Teasy This study
    HN103 PCR fragment (pTNR6 and pTNR8 from HN003 chromosome)/pGEM-Teasy This study
Open in a new tab

Isolation of moenomycin-resistant mutants by transposon mutagenesis.

Transposon mutagenesis was performed as described previously (12). Dilutions of an overnight culture of strain COL harboring the temperature-sensitive plasmid pRN3208, which carries transposon Tn551 (ermB), were grown at 30°C, plated on erythromycin-containing plates, and incubated at 42°C for 48 h. Colonies growing on these plates were screened for the insertion of Tn551 into the chromosome and loss of the plasmid moiety by replica plating onto plates containing erythromycin and cadmium (50 μg/ml), respectively. The curing efficiency of the plasmid was approximately 10−5. Mutants were then screened for decreased susceptibility to moenomycin. An overnight culture containing the Tn551 insertional mutants was spotted onto Trypticase soy agar (TSA) containing moenomycin at eight times the MIC, a concentration in the presence of which the wild type was unable to grow. After 24 h of incubation at 37°C, the growing cells were collected and the screening procedure was repeated. Mutants that were still able to grow on TSA containing eight times the original MIC of moenomycin were isolated.

Antibiotic susceptibility tests.

The MICs of various antibiotics were determined by microdilution, as described previously (14). The following concentrations of vancomycin were used for the determination of vancomycin MICs: 0.0625, 0.125, 0.25, 0.5, 1, 2, 3, 4, 5, 6, 8, 16, and 32 μg/ml. Population analysis profiles were determined by plating appropriate dilutions of an overnight culture on plates containing various concentrations of methicillin, vancomycin, or moenomycin. Colonies were counted after 48 h of incubation at 37°C (13). Vancomycin gradient plates were inoculated with a suspension of 107 bacterial cells, applied with a sterile cotton swab, along a gradient of 0 to 4 μg/ml. Gradient plates were incubated for 24 h at 37°C. All susceptibility tests were repeated at least three times to check the reproducibility of the results.

DNA manipulations.

Routine DNA manipulations, Southern blots and hybridizations, and DNA sequencing were performed essentially as described earlier (25). Restriction enzymes and shrimp alkaline phosphatase were purchased from Boehringer Mannheim Biochemica, Tokyo, Japan; and T4 DNA ligase was purchased from New England BioLabs, Beverly, Mass. DNA sequences were determined by the dideoxy chain termination method with an Autoread sequencing kit (Pharmacia Biotech, Tokyo, Japan). PCR reagents were from Boehringer Mannheim, and PCR was performed with the GeneAmp PCR System 2400 of Perkin-Elmer. Transductions were performed with phage 80alpha, as described previously (12).

Identification of Tn551 insertion sites.

Chromosomal DNA was isolated from the Tn551 mutants and digested with XbaI. After the DNA was heated at 75°C for 10 min, the digested DNA was self-ligated. By using this ligation mixture as a template, PCR amplification was performed with primers (5′-ACG GCG AAG GAT CAC TCA TGG-3′ and 5′-ATT TCT GAT GCG AGG TTC-3′) whose sequences were derived from the known Tn551 sequence (26). Primers were designed to anneal to the ends of Tn551, in opposite directions, to amplify the regions flanking the Tn551 insertions. PCR products were cloned into the pGEM-Teasy vector, and DNA sequencing was performed. On the basis of the DNA sequence of the regions flanking Tn551, the site of insertion was determined from The Institute for Genome Research database (http://www.tigr.org). To confirm the site of insertion, primers were then designed to amplify the entire gene into which Tn551 had been inserted.

Analysis of membrane composition.

Overnight cultures were diluted in fresh TSB and grown at 37°C with shaking until they reached an optical density at 660 nm of 0.3. Then, [2-3H]glycerol (37 kBq/ml) was added, and the cultures were incubated for 2 h at 37°C with shaking. The cells were then harvested by centrifugation and washed three times with phosphate-buffered saline (PBS). The cells were resuspended in chloroform-methanol (2:1 [vol/vol]) and sonicated for 15 min. Distilled water was then added to this suspension to separate the cell debris from the water-soluble fraction. The chloroform-methanol (2:1) phase was then extracted and dried in an evaporator. The samples were solubilized with carrier lipids containing a high concentration of S. aureus membrane lipids in order to visualize each lipid with an iodide stain. The samples were then spotted onto the bottom of a silica gel plate (Advantec), and two-dimensional thin-layer chromatography was performed as described by Kariyama (10). The solvents used for the first and second rounds of chromatography were chloroform-methanol-water (65:25:4) and chloroform-methanol-7 N NH3Cl (60:35:5), respectively. The silica plates were then exposed to iodine vapor in order to visualize the different lipids, including lysyl-phosphatidylglycerol, phosphatidylglycerol, cardiolipin, and glycolipid. Each spot stained by iodine was scraped off the plate and mixed with scintillation cocktail (Clearsol III; Nacalai Tesque), and the radioactivity was then measured with a scintillation counter. The percentage of each radioactive lipid present was calculated relative to the total amount of radioactivity.

Moenomycin and vancomycin binding assay.

Overnight cultures of S. aureus were harvested and washed with PBS. The cell suspensions were then divided into two fractions; one was used directly in the binding assay and the other was suspended in 4% sodium dodecyl sulfate (SDS) and boiled for 30 min, washed five times with distilled water, and then used in the binding assay. Several different concentrations of moenomycin and vancomycin (2, 10, 50, and 100 μg/ml) were prepared, with TSB used as the diluent. Two milligrams (wet weight) of S. aureus cells was added to 1 ml of the moenomycin or vancomycin solutions, and the mixtures were gently agitated for 30 min at 37°C. After centrifugation at 8,000 × g for 15 min, the supernatant was filtered through a 0.22-μm-pore-size membrane filter. Sterilized paper disks (diameter, 5 mm) were immersed in the supernatants and were then placed on TSA plates that had previously been swabbed with an overnight culture of S. aureus strain COL. After 24 h of incubation at 37°C, the diameters of the inhibitory zones were measured. For standardization, inhibitory zone diameters were measured for several different concentrations of antibiotic stock solutions that had had no bacterial contact. These standards were then used to calculate the concentration of the drug bound to S. aureus cells. The binding experiments were each repeated three times.

HPLC analysis of peptidoglycan. (i) Muropeptide analysis.

The preparation of murein and the reduction of muropeptide were carried out as described previously (11, 23). Mutanolysin-digested muropeptides were then fractionated by reverse-phase high-pressure liquid chromatography (RP-HPLC) (11, 23).

(ii) Glycan chain analysis.

Glycan chain analysis was performed by the same procedure used for muropeptide analysis, except that the cell wall was digested with lysostaphin (Sigma Chemical Co.) instead of mutanolysin.

Antibacterial assays of hBD3 and CAP18.

Overnight cultures of S. aureus were harvested, washed with PBS, and resuspended in 10 mM sodium phosphate buffer (NaPi) (pH 6.8). The bacterial suspensions were then diluted to 107 cells/ml with NaPi (pH 6.8), and 10-μl aliquots (105 cells) were added to 200 μl of NaPi with or without antibacterial peptides at various concentrations. Samples were then incubated anaerobically for 2 h at 37°C. The antibacterial peptides β-defensin-3 (hBD3) and CAP18 were synthesized and purified as described elsewhere (17). Appropriate dilutions of the reaction mixtures (100 μl) were then plated and incubated at 37°C overnight, and the numbers of CFU were determined. The antibacterial effects of the peptides were estimated according to the ratio of the number of surviving cells to the total number of cells added. The experiment was repeated at least three times to check the reproducibility of the results.

Growth curve analyses.

Growth rate experiments were performed with chemically defined medium (CDM), as described previously (8, 17). CDM depleted of lysine or aspartic acid, or both, was also prepared. Overnight cultures of S. aureus grown in CDM were harvested and then washed with the appropriate CDM. A small amount of bacterial suspension was then added to fresh CDM, CDM depleted of lysine, CDM depleted of aspartic acid, or CDM depleted of both lysine and aspartic acid. Cultures were incubated at 37°C with shaking, and growth was monitored by measuring the optical density at 660 nm.

RESULTS

Isolation of moenomycin-resistant mutants.

Three moenomycin-resistant mutants, mutants HN001 to HN003, were isolated from a library of 3,000 COL::Tn551 insertional mutants (Table 1). The moenomycin MICs for these mutants were four times higher than those for the wild type. Vancomycin MICs were also slightly higher for the mutants than for the parents, while the β-lactam and gentamicin MICs decreased (Table 2). Backcross mutants (mutants HN004 to HN006) all had the same phenotypes as the respective original mutants (data not shown), indicating that the phenotypes were caused by the Tn551 insertions. The population analysis profiles also confirmed the MIC results, showing that the mutant strain populations had become more resistant to vancomycin and moenomycin and less resistant to methicillin (Fig. 1). The Tn551 insertion from HN002 was also transduced to another MRSA strain, strain KSA8, to generate HN007. This mutant (mutant HN007) was also more resistant to moenomycin (data not shown). Subtle differences in vancomycin MICs were also confirmed by gradient plate analysis. Figure 2 shows the slight increase in the vancomycin resistance of mutant HN001 compared to that of the wild type.

TABLE 2.

MICs of various antibiotics for the mutants

Strain (genotype) MIC (μg/ml)
Oxacillin Methicillin Vancomycin Fosfomycin Bacitracin Moenomycin Gentamicin
COL (Wild type) 512 1,024 2 512 128 0.063 1
HN001 [COL(fmtC::Tn)] 128 256 3 512 64 0.25 0.125
HN002 [COL(lysC::Tn)] 128 256 3 256 128 0.25 0.25
HN003 [COL(fmtC::Tn)] 128 256 3 512 128 0.25 0.125
Open in a new tab

FIG. 1.

FIG. 1.

Open in a new tab

Population analysis of mutants grown in the presence of increasing concentrations of methicillin, moenomycin, and vancomycin. Overnight cultures of S. aureus wild-type strain COL (squares), HN001 (circles), and HN002 (triangles) were diluted and plated on TSA containing various concentrations of antibiotics. Colonies were counted after 48 h of incubation at 37°C.

FIG. 2.

FIG. 2.

Open in a new tab

Gradient plate analysis of the mutants. Vancomycin gradient plates (containing a vancomycin concentration gradient from 0 to 4 μg/ml) were inoculated with a suspension of 107 bacterial cells, applied with a sterile cotton swab along the gradient, and grown for 24 h at 37°C.

Identification of the genes disrupted by Tn551.

Inverse PCR was performed to obtain the DNA sequences of regions flanking the Tn551 insertions. PCR fragments amplified from the chromosomal DNA of mutants HN001, HN002, and HN003 were cloned into the pGEM-Teasy vector to generate pHN001, pHN002, and pHN003, respectively. As expected, the plasmid inserts all contained part of Tn551, along with a flanking region from the S. aureus chromosome. This sequence was then identified from the S. aureus genome database. In HN001 and HN003, Tn551 had inserted into the same gene, fmtC (mprF), and in HN002, Tn551 had inserted into the lysC gene (Fig. 3). Figure 3 shows a map of the fmtC and lysC loci, indicating the locations of the Tn551 insertions. fmtC encodes a protein of 840 amino acid residues with a molecular mass of 96.9 kDa. FmtC (MprF) was previously identified as a factor that affects susceptibility to methicillin (12) and to several antibacterial peptides (22). FmtC (MprF) is associated with the addition of lysine to phophatidylglycerol, a major component of the bacterial membrane. lysC is one of the genes constituting an operon required for lysine biosynthesis (31).

FIG. 3.

FIG. 3.

Open in a new tab

Restriction map of the fmtC and lysC loci in S. aureus. Arrows indicate the positions of Tn551 insertions.

Strains COL-TS4, KSA8-TS4, and BB270-TS4 are fmtC mutants (12) which were previously isolated as mutants exhibiting decreased levels of resistance to methicillin. COL-TS4 and KSA8-TS4 also had slightly decreased susceptibilities to vancomycin, while BB270-TS4 showed the same susceptibility as the wild type. We also checked the susceptibilities of VISA strain COL-VR1 and its corresponding fmtC mutant, COL-VR1-FmtC, to methicillin and vancomycin. The vancomycin MICs for COL-VR1 and COL-VR1-FmtC were 8 and 4 μg/ml, respectively. Inactivation of fmtC in COL-VR1 increased its vancomycin susceptibility, although the MIC for the mutant (4 μg/ml) was still higher than that for wild-type COL (2 μg/ml). The methicillin MIC for COL-VR1-FmtC (128 μg/ml) was also decreased compared to that for COL-VR1 (1,024 μg/ml).

Effects of lysine and asparatic acid on growth of the lysC mutant.

When parent strain COL was grown in the absence of lysine or aspartic acid, or both, its growth rate was decreased compared to its growth rate in normal CDM (Fig. 4). The growth rate of the lysC mutant, however, was decreased in normal CDM compared to that of COL. Depletion of aspartic acid or lysine from CDM decreased the growth rate of the mutant further, and the mutant could not grow in CDM when both amino acids were depleted.

FIG. 4.

FIG. 4.

Open in a new tab

Growth curves of S. aureus COL (closed symbols) and lysC mutant HN002 (open symbols) grown in different types of CDM. Growth was determined by measuring the optical density (OD) at 660 nm in CDM (squares), CDM without lysine (circles), CDM without aspartic acid (triangles), and CDM without both lysine and aspartic acid (diamonds).

Membrane compositions of the mutants.

Since fmtC (mprF) is associated with the addition of lysine into phosphatidylglycerol (22), we characterized the composition of the bacterial membrane. Table 3 shows the compositions of the bacterial mutant and wild-type membranes. The FmtC mutant (mutant HN001) contained a lower level of lysyl-phosphatidylglycerol and increased levels glycolipid and cardiolipin. Similarly, the lysC mutant (mutant HN002) had less lysyl-phosphatidylglycerol and more glycolipid and cardiolipin.

TABLE 3.

Membrane compositions of S. aureus cells

Strain (genotype) % of total radioactivitya
L-PG PG GL CL
COL (wild type) 3.4 90.0 5.6 1.0
HN001 (fmtC::Tn551) 0.2 89.7 7.3 2.8
HN002 (lysC::Tn551) 1.3 89.0 7.0 2.7
Open in a new tab
a

L-PG, lysyl-phosphatidylglycerol; PG, phosphatidylglycerol; GL, glycolipid; CL, cardiolipin.

Susceptibilities to antimicrobial peptides hBD3 and CAP18.

fmtC (mprF) inactivation was previously shown to lead to increased susceptibilities to antimicrobial peptides (22). Therefore, we investigated the susceptibilities of our mutants to the synthetic peptides hBD3 and CAP18 (Fig. 5). The fmtC mutant had dramatically increased susceptibilities to hBD3 and CAP18. The lysC mutant was also more susceptible, although the degree of increased susceptibility was relatively low compared to that of the fmtC mutant.

FIG. 5.

FIG. 5.

Open in a new tab

Antibacterial activities of hBD3 and CAP18. Different concentrations of the peptides were added to 200 μl of 10 mM sodium phosphate buffer (pH 6.8) containing 105 bacterial cells. After incubation at 37°C for 2 h, serial dilutions were plated onto TSA. The numbers of CFU were counted after 24 h of incubation at 37°C. Bacterial survival was determined from the number of bacteria that survived in the presence of the peptide as a percentage of the total number of CFU growing in the absence of peptide. The values shown correspond to the means of three independent experiments. Symbols: squares, wild-type strain COL; triangles, mutant HN001; circles, mutant HN002.

RP-HPLC peptidoglycan analyses.

The muropeptide profile of the peptidoglycan and the glycan chain distribution of both mutants were similar to those of the wild type (data not shown).

Moenomycin and vancomycin binding assay.

Binding assays were performed to test for differences in affinities of binding to moenomycin and vancomycin between the wild type and the mutants. For these assays two types of bacterial cell preparations were used: native bacterial cells and cells boiled in SDS (Fig. 6). The amount of binding and the kinetics of binding of moenomycin to cells boiled in SDS were similar for the mutants and the parent. However, moenomycin bound to native cells at a rate 10-fold higher than it bound to cells boiled in SDS, and the mutant showed a higher binding capacity than the parent. Vancomycin, on the other hand, bound to native cells and cells boiled in SDS at similar rates. The mutants appeared to have a 1.5-fold higher capacity to bind to native cells than the wild type, while the mutants and the wild type showed indistinguishable capacities to bind to boiled cells. We also tested the binding of moenomycin and vancomycin to other fmtC mutants (KSA8 and BB270) and observed similar results (data not shown).

FIG. 6.

FIG. 6.

Open in a new tab

Moenomycin and vancomycin binding to S. aureus cells. Different concentrations of moenomycin or vancomycin were mixed with intact S. aureus cells (closed symbols) or S. aureus cells boiled in SDS (open symbols). The amount of antibiotic binding to S. aureus cells was determined as described in Materials and Methods. Symbols: squares, wild-type strain COL; triangles, HN001; circles, HN002.

DISCUSSION

It was previously demonstrated (11, 21) that there is a correlation between moenomycin and vancomycin susceptibilities, although it was unclear whether these two resistance phenotypes shared parts of the same molecular mechanism. It was, however, shown that all moenomycin-resistant mutants, isolated from seven different S. aureus strains, exhibited intermediate resistance to vancomycin. Correspondingly, a vancomycin intermediate-resistant mutant, COL-VR1, derived from MRSA strain COL, also had an increased level of resistance to moenomycin, suggesting that the two mechanisms are related. Investigation of the factors responsible for increased moenomycin resistance is therefore likely to shed light on the mechanism of vancomycin intermediate resistance. Moenomycin resistance also proved easier to select for than vancomycin resistance, as it was very difficult to isolate mutants with a two- to fourfold increase in vancomycin resistance, while it was relatively easy to select moenomycin-resistant mutants.

In this study, we identified two genes (fmtC and lysC) which, when they were disrupted, increased the levels of resistance to both moenomycin and vancomycin. The fmtC gene was first identified as a factor affecting methicillin resistance (12) and was later identified as a factor affecting susceptibility to antimicrobial peptides (22). Peschel et al. (22) demonstrated that FmtC (MprF) is responsible for the addition of lysine to phosphatidylglycerol, which produces lysyl-phosphatidylglycerol, a major component of the cell membrane. The lysC gene is involved in the biosynthesis of lysine from aspartic acid (31). Lysine is one of the components of peptidoglycan; therefore, lysine synthesis is essential for cell viability. However, the ability of the lysC mutant to grow, albeit at a severely reduced rate, in CDM in the absence of lysine suggests that there is an additional, lysC-independent pathway(s) for lysine biosynthesis. Depletion of both lysine and aspartic acid from CDM reduced the growth rate of wild-type strain COL, while the lysC mutant could not grow at all without lysine and aspartic acid (Fig. 4). These results suggest that the biosynthetic pathway incorporating lysC is the major pathway for lysine biosynthesis. Therefore, the amount of lysine in the lysC mutant is hypothesized to be very low compared to that in the wild type. This decrease in the levels of lysine in the lysC mutant would also affect the levels of lysyl-phosphatidylglycerol (Table 3). Therefore, both lysC and fmtC mutations would result in the same phenotype, that is, a reduction in the lysyl-phosphatidylglycerol content of the membrane.

Due to the positive charge of lysyl-phosphatidylglycerol, the depletion of lysyl-phosphatidylglycerol is likely to result in membranes with a more negative net charge. Therefore, the mutants were tested for their susceptibilities to cationic antimicrobial peptides and gentamicin. Since antimicrobial peptides are thought to target the bacterial cell membrane (17, 22), the stronger the negative charge of the membrane is, the more susceptible to cationic peptides the bacterium is likely to be. Both the lysC and fmtC mutants were more susceptible to CAP18 and hBD3, both of which are positively charged antimicrobial peptides (Fig. 5). Gentamicin susceptibility was also increased in the fmtC and lysC mutants, implying that the more positively charged gentamicin molecules were being sequestered by the negatively charged membrane and entering the cytoplasm with a greater efficiency.

The mutants isolated in this study displayed decreased susceptibilities to moenomycin and vancomycin, although there was no cell wall thickening, no reduction in the degree of cross-linking, and no increase in the glycan chain length, phenotypes previously observed in several VISA strains, including strain COL-VR1 (data not shown). This suggests that there is more than one way for strains to become more resistant to both moenomycin and vancomycin. The nature of the mutations isolated here therefore raised the possibility that alterations in the membrane charge could affect susceptibilities to moenomycin and vancomycin. The results from the moenomycin and vancomycin binding assays support this hypothesis (Fig. 6). When intact cells were used in the binding assay, the amounts of moenomycin and vancomycin that bound to the mutants were higher than the amounts that bound to the wild type. However, when cells boiled in SDS were used in the assay, the amounts of binding were similar for the mutant and the wild type.

Interestingly, although more moenomycin and vancomycin bound to the mutants, it did not render them more susceptible; on the contrary, the mutants became more resistant. The increases in both vancomycin binding and vancomycin MICs were, however, both relatively small. It is possible that the nature in which moenomycin and vancomycin bind to the membranes of the mutants could, to some extent, preclude the drugs from acting on their lethal targets: peptidoglycan precursors in the case of vancomycin and transglycosylases in the case of moenomycin. The increased amount of positively charged vancomycin molecules bound to the cell surface could also affect the activities of autolysins, as autolysin activity has been shown to be dependent on the charge (20), which in turn could influence resistance levels.

Ruzin et al. (24) reported that mprF (fmtC) inactivation reduced the vancomycin resistance level of a VISA strain to that of a vancomycin-susceptible strain. They also showed that more vancomycin bound to the mutant cells than to the wild-type cells and suggested that the increased affinity for vancomycin binding led to the increased susceptibility of the fmtC mutant. However, our data indicate that fmtC inactivation in a vancomycin-susceptible background slightly decreases susceptibility to vancomycin (Table 2).

It was previously reported (12) that fmtC inactivation in S. aureus COL did not affect the vancomycin MIC. However, in that study a twofold broth microdilution method was used to measure the MICs, a method not sufficiently sensitive to measure the small variations in vancomycin susceptibility. When more sensitive methods were used to determine the vancomycin MICs (Table 2) and when the strains were analyzed on vancomycin gradient plates and by population analysis profiling, we found that vancomycin susceptibility was reduced in the COL fmtC mutant (Fig. 1 and Fig. 2).

We further investigated the effect of fmtC inactivation on two other vancomycin-sensitive strains and found that both of these strains also had slightly decreased susceptibilities to vancomycin (data not shown), implying that fmtC inactivation results in decreased susceptibility to vancomycin in most vancomycin-sensitive strains, although the degree to which fmtC inactivation affected vancomycin susceptibility did appear to be strain dependent (data not shown).

We then transduced the fmtC mutation into the VISA strain COL-VR1 and, like Ruzin et al. (24), found that fmtC inactivation in this VISA strain background increased vancomycin susceptibility, although the MIC for the strain was still higher than that for the susceptible parent. The COL-VR1 fmtC mutant also had increased susceptibilities to β-lactams. In previous studies (12), fmtC inactivation was shown to increase the susceptibilities of several MRSA strains to β-lactams. Therefore, fmtC inactivation increased the susceptibilities of all strains to β-lactams; however, its inactivation had different effects on vancomycin susceptibility, depending on the strains' initial resistance levels. These contradictory results suggest that fmtC inactivation may indirectly affect cell wall biosynthesis. We therefore hypothesized that fmtC inactivation could cause two different effects: in strains with a methicillin-resistant or vancomycin intermediate-resistant background, the influence of fmtC inactivation on cell wall biosynthesis could affect expression of the resistance phenotypes, making the strains more susceptible to methicillin and vancomycin, while in sensitive strains the changes in the membrane which potentially alter the way in which vancomycin and moenomycin bind could cause slight decreases in their susceptibilities. Therefore, once again, the genetic background of the strain is likely to have a significant effect on the influence of fmtC inactivation.

In conclusion, we identified two factors, fmtC and lysC, which, when they were inactivated, yielded mutants with altered, more negatively charged cell membranes. The findings obtained with these mutants demonstrate that membrane charge can have an affect on the susceptibilities of S. aureus to glycopeptides and moenomycin and to other, positively charged antibacterial agents, such as gentamicin and cationic antimicrobial peptides.

Acknowledgments

This work was supported by a grant-in-aid for scientific research (grant 14570238) from the Ministry of Education, Science, Sports and Culture of Japan.

Part of this study was carried out in the Research Center for Molecular Medicine, Faculty of Medicine, Hiroshima University.

REFERENCES

  • 1.Barna, J. C. J., and D. H. Williams. 1984. The structure and mode of action of glycopeptide antibiotics of the vancomycin group. Annu. Rev. Microbiol. 38:339-357. [DOI] [PubMed] [Google Scholar]
  • 2.Bischoff, M., M. Roos, J. Putnik, A. Wada, P. Glanzmann, P. Giachino, P. Vaudaux, and B. Berger-Bächi. 2001. Involvement of multiple genetic loci in Staphylococcus aureus teicoplanin resistance. FEMS Microbiol. Lett. 194:77-82. [DOI] [PubMed] [Google Scholar]
  • 3.Bobin-Dubreux, S., M.-E. Reverdy, C. Nervi, M. Rougier, A. Bolmstrom, F. Vandenesch, and J. Etienne. 2001. Clinical isolate of vancomycin-heterointermediate Staphylococcus aureus susceptible to methicillin and in vitro selection of a vancomycin-resistant derivative. Antimicrob. Agents Chemother. 45:349-352. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Boyle-Vavra, S., B. L. M. De Jonge, C. C. Ebert, and R. S. Daum. 1997. Cloning of the Staphylococcus aureus ddh gene encoding NAD+-dependent d-lactate dehydrogenase and insertional inactivation in a glycopeptide-resistant isolate. J. Bacteriol. 179:6756-6763. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Brandenberger, M., M. Tschierske, P. Giachino, A. Wada, and B. Berger-Bächi. 2000. Inactivation of a novel three-cistronic operon tcaR-tcaA-tcaB increases teicoplanin resistance in Staphylococcus aureus. Biochem. Biophysi. Acta 1523:135-139. [DOI] [PubMed] [Google Scholar]
  • 6.Chang, S., D. M. Sievert, J. C. Hageman, M. L. Boulton, F. C. Tenover, F P. Downes, S. Shah, J. T. Rudrik, G. R Pupp, W. J. Brown, D. Cardo, S. K. Fridkin, and the Vancomycin-Resistant Staphylococcus aureus Investigative Team. 2003. Infection with vancomycin-resistant Staphylococcus aureus containing the vanA resistance gene. N. Engl. J. Med. 348:1342-1347. [DOI] [PubMed] [Google Scholar]
  • 7.Cui, L., H. Murakami, K. Kuwahara-Arai, H. Hanaki, and K. Hiramatsu. 2000. Contribution of a thickened cell wall and its glutamine nonamidated component to the vancomycin resistance expressed by Staphylococcus aureus Mu50. Antimicrob. Agents Chemother. 44:2276-2285. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Hassain, M., M. H. Wilcox, P. J. White, M. K. Faulkner, and R. C. Spencer. 1992. Importance of medium and atmosphere type to both slime production and adherence by coagulase-negative staphylococci. J. Hosp. Infect. 20:173-184. [DOI] [PubMed] [Google Scholar]
  • 9.Huber, G., and G. Nesemann. 1968. Moenomycin, an inhibitor of cell wall synthesis. Biochem. Biophys. Res. Commun. 30:7-13. [DOI] [PubMed] [Google Scholar]
  • 10.Kariyama, R. 1982. Increase of cardiolipin content in Staphylococcus aureus by the use of antibiotics affecting the cell wall. J. Antibiot. 35:1700-1704. [DOI] [PubMed] [Google Scholar]
  • 11.Komatsuzawa, H., K. Ohta, Y. Yamada, K. Ehlert, H. Labischinski, J. Kajimura, T. Fujiwara, and M. Sugai. 2002. Increased glycan chain length distribution and decreased susceptibility to moenomycin in a vancomycin-resistant Staphylococcus aureus mutant. Antimicrob. Agents Chemother. 46:75-81. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.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. [DOI] [PubMed] [Google Scholar]
  • 13.Komatsuzawa, H., M. Sugai, C. Shirai, J. Suzuki, K. Hiramatsu, and H. Suginaka. 1995. Triton X-100 alters the resistance level of methicillin-resistant Staphylococcus aureus to oxacillin. FEMS Microbiol. Lett. 134:209-212. [DOI] [PubMed] [Google Scholar]
  • 14.Komatsuzawa, H., J. Suzuki, M. Sugai, Y. Miyake, and H. Suginaka. 1994. The effect of Triton X-100 on the in-vitro susceptibility of methicillin-resistant Staphylococcus aureus to oxacillin. J. Antimicrob. Chemother. 34:885-897. [DOI] [PubMed] [Google Scholar]
  • 15.Kornblum, J., B. J. Hartman, R. P. Novick, and A. Tomasz. 1986. Conversion of a homogeneously methicillin-resistant strain of Staphylococcus aureus to heterogeneous resistance by Tn551-mediated insertional inactivation. Eur. J. Clin. Microbiol. 5:714-718. [DOI] [PubMed] [Google Scholar]
  • 16.Marchese, A., G. Balistreri, E. Tonoli, E. A. Debbia, and G. C. Schito. 2000. Heterogeneous vancomycin resistance in methicillin-resistant Staphylococcus aureus strains isolated in a large Italian hospital. J. Clin. Microbiol. 38:866-869. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Midorikawa, K., K. Ouhara, H. Komatsuzawa, T. Kawai, S. Yamada, T. Fujiwara, K. Yamazaki, K. Sayama, M. A. Taubman, H. Kurihara, K. Hashimoto, and M. Sugai. 2003. Staphylococcus aureus susceptibility to innate antimicrobial peptides, β-defensins and CAP18, expressed by human keratinocytes. Infect. Immun. 71:3730-3739. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Milewski, W. M., S. Boyle-Vavra, B. Moreira, C. C. Ebert, and R. S. Daum. 1996. Overproduction of a 37-kilodalton cytoplasmic protein homologous to NAD+-linked d-lactate dehydrogenase associated with vancomycin resistance in Staphylococcus aureus. Antimicrob. Agents Chemother. 40:166-172. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Moreira, B., S. Boyle-Vavra, B. L. M. DeJonge, and R. S. Daum. 1997. Increased production of penicillin-binding protein 2, increased detection of other penicillin-binding proteins, and decreased coagulase activity associated with glycopeptide resistance in Staphylococcus aureus. Antimicrob. Agents Chemother. 41:1788-1793. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Neuhaus, F. C., and J. Baddiley. 2003. A continuum of anionic charge: structures and functions of d-alanyl-teichoic acids in gram-positive bacteria. Microbiol. Mol. Biol. Rev. 67:686-723. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.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. [DOI] [PubMed] [Google Scholar]
  • 22.Peschel, A., R. W. Jack, M. Otto, L. V. Collins, P. Staubitz, G. Nicholson, H. Kalbacher, W. F. Nieuwenhuizen, G. Jung, A. Tarkowski, K. P. M. van Kessel, and J. A. G. Strijp. 2001. Staphylococcus aureus resistance to human defensins and evasion of neutrophil killing via the novel virulence factor mprF is based on modification of membrane lipids with l-lysine. J. Exp. Med. 193:1067-1076. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Roos, M., E. Pittenauer, E. Schmid, M. Beyer, B. Reinike, G. Allmaier, and H. Labischinski. 1998. Improved high-performance liquid chromatographic separation of peptidoglycan isolated from various Staphylococcus aureus strains for mass spectrometric characterization. J. Chromatogr. B 705:183-192. [DOI] [PubMed] [Google Scholar]
  • 24.Ruzin, A., A. Severin, S. L. Moghazeh, J. Etienne, P. A. Bradford, S. J. Projan, and D. M. Shlaes. 2003. Inactivation of mprF affects vancomycin susceptibility in Staphylococcus aureus. Biochim. Biophys. Acta 1621:117-121. [DOI] [PubMed] [Google Scholar]
  • 25.Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory mannual, 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.
  • 26.Shaw, J. H., and D. B. Clewell. 1985. Complete nucleotide sequence of macrolide-lincosamide-streptogramin B-resistance transposon Tn917 in Streptococcus faecalis. J. Bacteriol. 164:782-796. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Shlaes, D. M., J. H. Shlaes, S. Vincent, L. Etter, P. D. Fey, and R. V. Goering. 1993. Teicoplanin-resistant Staphylococcus aureus expresses a novel membrane protein and increases expression of penicillin-binding protein 2 complex. Antimicrob. Agents Chemother. 37:2432-2437. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Sieradzki, K., M. G. Pinho, and A. Tomasz. 1999. Inactivated pbp4 in highly glycopeptide-resistant laboratory mutants of Staphylococcus aureus. J. Biol. Chem. 274:18942-18946. [DOI] [PubMed] [Google Scholar]
  • 29.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. [DOI] [PubMed] [Google Scholar]
  • 30.Smith, T. L., M. L. Pearson, K. R. Wilcox, C. Cruz, M. V. Lancaster, B. Robinson-Dunn, F. Tenover, M. J. Zervos, J. D. Band, E. White, and W. R. Jarvis. 1999. Emergence of vancomycin resistance in Staphylococcus aureus. N. Engl. J. Med. 340:493-501. [DOI] [PubMed] [Google Scholar]
  • 31.Wiltshire, M. D., and S. J. Foster. 2001. Identification and analysis of Staphylococcus aureus components expressed by a model system of growth in serum. Infect. Immun. 69:5198-5202. [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from Antimicrobial Agents and Chemotherapy are provided here courtesy of American Society for Microbiology (ASM)

RESOURCES