Role of murE in the Expression of β-Lactam Antibiotic Resistance in Staphylococcus aureus

Abstract

It was shown earlier that Tn551 inserted into the C-terminal region of murE of parental methicillin-resistant Staphylococcus aureus strain COL causes a drastic reduction in methicillin resistance, accompanied by accumulation of UDP-MurNAc dipeptide in the cell wall precursor pool and incorporation of these abnormal muropeptides into the peptidoglycan of the mutant. Methicillin resistance was recovered in a suppressor mutant. The murE gene of the same strain was then put under the control of the isopropyl-β-d-thiogalactopyranoside (IPTG)-inducible promoter P_spac. Bacteria grown in the presence of suboptimal concentrations of IPTG accumulated UDP-MurNAc dipeptide in the cell wall precursor pool. Both growth rates and methicillin resistance levels (but not resistance to other antibiotics) were a function of the IPTG concentration. Northern analysis showed a gradual increase in the transcription of murE and also in the transcription of pbpB and mecA, parallel with the increasing concentrations of IPTG in the medium. A similar increase in the transcription of pbpB and mecA, the structural genes of penicillin-binding protein 2 (PBP2) and PBP2A, was also detected in the suppressor mutant. The expression of these two proteins, which are known to play critical roles in the mechanism of staphylococcal methicillin resistance, appears to be—directly or indirectly—under the control of the murE gene. Our data suggest that the drastic reduction of the methicillin MIC seen in the murE mutant may be caused by the insufficient cellular amounts of these two PBPs.

Staphylococcus aureus mutant strain RUSA235 was originally isolated as a member of the large library of Tn551 insertional mutants of methicillin-resistant S. aureus (MRSA) strain COL in which the high and homogeneous level of methicillin resistance was reduced (6). The number of these determinants with the Tn551 inserts, initially termed fem (factors essential for methicillin resistance) or aux (auxiliary) genes, has now increased to more than 20 (1, 2, 6, 7). Several of the auxiliary genes are involved in peptidoglycan biosynthesis or cell wall turnover, some appear to have putative regulatory functions, and others encode proteins with functions as yet unidentified (7). With the possible exception of pbpB, the structural gene of penicillin-binding protein 2 (PBP2) (20), it is not clear how these genes cooperate with the methicillin resistance gene mecA (9, 30) in promoting high-level and homogeneous resistance to β-lactam antibiotics.

Genetic analysis of mutant RUSA235 identified the target of Tn551 as murE (15), an essential gene of S. aureus (12), encoding the UDP-N-acetylmuramyl tripeptide synthetase that catalyzes the addition of the l-lysine residue to the UDP-linked muramyl dipeptide cell wall precursor. In mutant RUSA235, the insert was 3 bp upstream of the termination codon, allowing production of a modified MurE protein with reduced specific activity (15). Biochemical analysis of RUSA235 demonstrated the accumulation of UDP-MurNAc dipeptide in the cytoplasmic cell wall precursor pool and incorporation of the dipeptide into the peptidoglycan of the mutant (16).

The initial purpose of the studies described in this communication was to examine the possibility that the hypersensitivity of RUSA235 to β-lactam antibiotics is related to some structural or functional defect in the cell wall containing the abnormal dipeptide components. Subsequently, the construction of a conditional murE mutant has allowed us to probe in more detail the role of MurE in cell wall synthesis and drug resistance.

MATERIALS AND METHODS

Bacterial strains, plasmids, and growth conditions.

The bacterial strains and plasmids used in this study are described in Table 1. S. aureus strains were grown in tryptic soy broth (TSB; Difco Laboratories, Detroit, Mich.) with aeration at 37°C or on tryptic soy agar (TSA; Difco Laboratories) plates at 37°C. Escherichia coli strains were grown in Luria-Bertani broth (Difco Laboratories) with aeration at 37°C. Erythromycin (10 μg/ml), chloramphenicol (10 μg/ml), and ampicillin (100 μg/ml) were used as recommended by the manufacturer (Sigma, St. Louis, Mo.) for selection and maintenance of S. aureus and E. coli transformants, respectively.

TABLE 1.

Strains and plasmids used in this study^a

Strain or plasmid	Relevant characteristicsa	Source or reference
S. aureus strains
RN4220	Mcs, restriction negative	R. Novick
COL	Homogeneous Mcr Ems; MIC, 800 μg/ml	RUb collection
RUSA235	COL::Tn551-Mcr Emr; MIC, 6.25 μg/ml	6
RUSA235-Homo*	COL::Tn551-Mcr Emr Homo*; MIC, 400 μg/ml	This study
RUSA235 Backcross	COL::Tn551-Mcr Emr; MIC, 6.25 μg/ml	16
RUSA235-Homo*	COL::Tn551-Mcr Emr; MIC, 6.25 μg/ml	This study
Backcross
RN4220spac::murE	RN4220 with Pspac-murE fusion in the chromosome	This study
COLspac::murE	COL with Pspac-murE fusion in the chromosome transformed with pMGPII	This study
COLpMGPII	COL transformed with pMGPII	22
E. coli DH5α	recA endA1 gyrA96 thi-1 hsdR17 supE44 relA1 φ80 ΔlacZΔM15	Bethesda Research Laboratories
Plasmids
pMGPI	S. aureus integrative vector with IPTG-inducible Pspac promoter and lacI gene; Apr Emr	21
pMGPII	S. aureus replicative plasmid containing lacI gene; Apr Cmr	21
pSGII	pMGPI with murE ribosome binding site and first 311 codons fused to Pspac promoter, Apr Emr	This study

^aOxacillin MICs are given. Ap^r, ampicillin resistant; Em^r, erythromycin resistant; Cm^r, chloramphenicol resistant.

^bRU, Rockefeller University.

To measure the growth rate of COL_spac::murE, an overnight culture was grown in TSB containing erythromycin (10 μg/ml), chloramphenicol (10 μg/ml), and 100 μM isopropyl-β-d-thiogalactopyranoside (IPTG). This starting culture was washed and resuspended in TSB without IPTG to remove traces of the inducer before the culture was diluted to an optical density at 620 nm (OD₆₂₀) of 0.02 in TSB containing 10 μg of chloramphenicol per ml and supplemented with increasing concentrations of IPTG (0, 25, 37.5, 75, and 200 μM). Each culture was incubated with agitation at 37°C, and the OD₆₂₀ was monitored.

DNA methods.

DNA manipulations were performed by standard methods (24). Restriction enzymes were used as recommended by the manufacturer (New England Biolabs, Beverly, Mass.). Routine PCR amplification was performed with Tth DNA polymerase (HT Biotechnology, Cambridge, United Kingdom). Wizard Plus Minipreps and Midipreps (Promega, Madison, Wis.) purification systems were used for plasmid extraction. PCR and digestion products were purified with Wizard PCR Preps and Wizard DNA Clean-up systems (Promega). Ligation reactions were performed with T4 ligase (New England Biolabs). DNA sequencing was done at the Rockefeller University Protein/DNA Technology Center by the BigDye terminator cycle sequencing method with either a 3700 DNA analyzer for capillary electrophoresis or ABI Prism 377 DNA sequencers for slab gel electrophoresis.

Determination of antibiotic resistance by population analysis.

Overnight cultures were plated at various dilutions on TSA plates containing increasing concentrations of the various antibiotics, and bacterial colonies were counted after incubation of the plates at 37°C for 48 h as previously described (5). Oxacillin, bacitracin, and cefotaxime were purchased from Sigma. Moenomycin was obtained through the courtesy of Aventis Pharma D, DI&A Natural Products (Bridgewater, N.J.).

Analysis of the UDP-linked precursor pool.

The UDP-linked cytoplasmic peptidoglycan precursor pool was extracted by a previously described procedure (16), except that the precursors were separated on a Hypersyl (Runcor Cheshire, United Kingdom) reverse-phase high-performance liquid chromatography (HPLC) octyldecyl silane column (3-μm particle size; 250 by 4.6 mm; 120-Å pore size) that was eluted with a linear 5 to 30% methanol gradient in 100 mM sodium phosphate buffer, pH 2.5, at a flow rate of 0.5 ml/min and assayed for absorbance at 254 nm.

Cell wall analysis.

Cell walls were isolated, the peptidoglycan was purified and hydrolyzed with the M1 muramidase, and the resulting muropeptides were reduced with borohydride and separated by reverse-phase HPLC as previously described (3).

Autolytic enzyme extract.

Crude autolytic extract was prepared by a method similar to that described previously (29). Strain COL was grown to mid-exponential phase in 250 ml of TSB at 37°C with aeration, chilled rapidly, harvested by centrifugation, washed once in ice-cold 50 mM Tris-HCl (pH 7.5), and extracted with 250 ml of 4% sodium dodecyl sulfate at 4°C for 30 min with stirring. The supernatant was used as the source of autolytic enzymes.

Cell wall hydrolysis in vitro.

Purified cell walls were suspended in buffer (50 mM Tris-HCl, pH 7.5) to an initial OD₆₂₀ of 0.5. Lysis was measured as a decrease in OD₆₂₀ during incubation of wall samples at 37°C with crude lytic enzyme extract (10 mg of protein/ml).

Construction of plasmid pSGII.

A DNA fragment containing the ribosome-binding site and the first 311 codons of the murE gene was amplified by PCR with PfuTurbo DNA polymerase (Stratagene, Heidelberg, Germany) and primers murEspacIA (5′-TAAGATCTACACCGCAATCATTGCCGCC-3′) and murEspacII (5′-ATTCCCGGGTTGTAGAAAAAGGAGCGGTTCAG-3′). The primers were engineered to carry BglII (murEspacIA) and SmaI (murEspacII) restriction sites (underlined in the primer sequences). The following PCR conditions were used: 94°C for 4 min; 40 cycles of 94°C for 45s, 55°C for 45 s, and 72°C for 1 min; and one final extension step of 72°C for 10 min. The purified PCR product was cleaned with a Wizard PCR Preps DNA purification system (Promega), digested with BglII and SmaI, and fused with the inducible spac promoter present in pMGPI (21), which was also digested with BglII and SmaI. The mixture was used to transform E. coli DH5α (Invitrogen, Carlsbad, Calif.) competent cells to obtain plasmid pSGII.

Construction of S. aureus strains with the murE gene under the control of an inducible promoter.

Plasmid pSGII was introduced into S. aureus RN4220 electrocompetent cells by electroporation with a Gene Pulser apparatus (Bio-Rad, Hercules, Calif.) essentially as previously described (13). The transformation mixture was plated on TSA containing erythromycin (10 μg/ml) and IPTG (300 μM). Plasmids pSGII and pMGPII (21) were sequentially transduced to MRSA strain COL by using phage 80α as previously described (17), except that 100 μM IPTG was added to the media used for preparation of the transducing lysate and for transduction. Transductants were selected with erythromycin (10 μg/ml) for pSGII and with chloramphenicol (10 μg/ml) for pMGPII. The correct sequence of pSGII insertion into COL was confirmed.

Experiments in liquid culture were performed by using the following protocol. Cultures were grown overnight at 37°C in TSB supplemented with 100 μM IPTG and the appropriate antibiotic(s). To remove traces of IPTG, bacterial cultures were centrifuged and the pellet was washed twice with TSB. The cells were resuspended in the same volume of TSB supplemented with various concentrations of IPTG and then used to determine various properties, such as susceptibility to antibiotics, growth rates, composition of the UDP-linked precursor pool, and transcription of several genetic determinants.

Determination of susceptibility to different types of antibiotics.

Cultures of COL_spac::murE were spread on the surface of TSA plates supplemented with different concentrations of IPTG (25, 37.5, 75, 150, and 500 μM), and antibiotic susceptibilities were tested with paper disks containing the antibiotics oxacillin (1 mg), ciprofloxacin (50 μg), vancomycin (30 μg), d-cycloserine (200 μg), and tetracycline (1 mg). Strain COL was used as a control. The sizes of inhibition halos were evaluated after incubation at 37°C for 9.5 h.

Determination of autolysis rates.

Triton X-100-stimulated autolysis in glycine buffer (pH 8.0) was measured as previously described (27). Cells were grown exponentially to an OD₆₂₀ of about 0.3. The cultures were rapidly chilled on ice, and the cells were washed once with ice-cold distilled water and suspended to an OD₆₂₀ of 1.0 in 50 mM glycine-0.01% Triton X-100 buffer. Autolysis was measured during incubation at 37°C by monitoring the OD₆₂₀.

Northern blotting analysis.

Cells were grown in TSB or TSB supplemented with increasing concentrations of IPTG at 37°C to an OD₆₂₀ of 0.7 to 0.8 (log-phase growth), and the RNA was extracted as previously described (28). Next, 7 μg of each RNA sample was analyzed by electrophoresis in a 1.2% agarose gel containing 0.66 M formaldehyde and morpholinepropanesulfonic acid (MOPS; Sigma). The RNA was blotted onto Hybond N⁺ membranes (Amersham, Buckinghamshire, United Kingdom) with a turbo blotter alkaline transfer system (Schleicher & Schuell, Inc., Keene, N.H.) with SSC20X. The PCR-amplified DNA probes were labeled with [α-³²P]dCTP (Amersham Life Sciences, Piscataway, N.J.) with a Ready to Go labeling kit (Amersham) and hybridized under high-stringency conditions. The blots were subsequently washed and autoradiographed.

Membrane purification.

Membrane proteins were prepared from bacterial cultures as previously described (25). Proteins were quantified with a DcProtein assay kit (Bio-Rad Laboratories).

Western blotting analysis.

For detection of PBP2A in the membrane protein fraction, 10 and 30 μg of each membrane protein preparation was resolved on 8% acrylamide gels and transferred to nitrocellulose membranes by Western blotting as previously described (31). Incubation with a monoclonal antibody against PBP2A of an MRSA strain (Eli Lilly & Co., Indianapolis, Ind.) was carried out with the ECL Western blot analysis system (Amersham) (31).

RESULTS

Selective suppression of β-lactam antibiotic resistance in Tn551 murE mutant RUSA235.

The susceptibilities of parental strain COL and mutant RUSA235 to bacitracin (MIC, 50 μg/ml), moenomycin (MIC, 0.25 μg/ml), and vancomycin (MIC, 1.5 μg/ml) were the same, while resistance to β-lactam antibiotics was drastically reduced. The oxacillin MIC went from 800 to 6 μg/ml, and that of cefotaxime went from 400 to 25 μg/ml. These findings indicate that the inhibitory effect of the murE mutation was selective to β-lactam antibiotics and did not affect inhibitors of earlier steps in cell wall biosynthesis.

Recovery of high-level β-lactam resistance in Homo* derivatives of mutant RUSA235.

The population analysis profile of methicillin resistance showed a heterogeneous phenotype. The methicillin MIC was reduced in 99.9% of the cells of strain RUSA235 cultures—from a MIC of 800 μg/ml in the parental strain to a MIC of 6 μg/ml in the mutant. Nevertheless, such cultures also contained subpopulations of bacteria that retained the parental level of oxacillin resistance. Such so-called Homo* subpopulations (8) were present with a frequency of 10⁻⁵ to 10⁻⁷ in mutant cultures. Homo* colonies picked from the agar plate gave rise to virtually homogeneous cultures with high-level methicillin resistance. When Homo* cultures were backcrossed into parental strain COL (by selection for the Tn551 marker, i.e., erythromycin resistance), the transductants showed the phenotype of the original RUSA235 mutant, indicating that the recovery of high-level antibiotic resistance in the Homo* cells was due to a compensatory (suppressor) mutation(s) distant from the Tn551 insertion (Fig. 1).

FIG. 1.

Heterogeneous expression of oxacillin resistance in S. aureus mutant RUSA235 carrying a Tn551 insert in murE. Expression of oxacillin resistance was determined by population analysis as described in Materials and Methods. Overnight cultures were plated on TSA or on TSA containing increasing concentrations of oxacillin. Plates were incubated for 48 h at 37°C, and the CFU were counted. Symbols: parental strain COL, ▴; mutant RUSA235, ○; RUSA235 backcross, □; RUSA235-Homo*, ▪; RUSA235-Homo* backcross, •.

Accumulation of dipeptide cell wall precursors in Homo* cultures.

The relative amounts of UDP-N-acetylmuramyl dipeptides and UDP-N-acetylmuramyl pentapeptide were compared in the cell wall precursor pools of parental strain COL, mutant 235, and Homo* cultures (Fig. 2). The relative amount (expressed as a percentage of all UDP-linked cell wall precursors) of muramyl dipeptides decreased from 18.5% in RUSA235 to about 8% in the Homo* extract, while muramyl pentapeptides increased from 55% in the mutant to about 67% in the Homo* extract. The corresponding values in parental strain COL were 5% dipeptides and 71% pentapeptide.

FIG. 2.

Alterations in the composition of the cytoplasmic cell wall precursor pool in murE mutant RUSA235. Cytoplasmic peptidoglycan precursors isolated from parental strain COL, mutant RUSA235, and derivative RUSA235-Homo* were prepared and analyzed by HPLC as described in Materials and Methods. The elution profiles of extracts from the parental strain (top), mutant RUSA235 (middle), and the Homo* derivative of the mutant (bottom) are shown. Cell wall precursors were identified by mass spectrometry. 1, UDP-MurNAc; 2, UDP-MurNAc-l-Ala; 3, UDP-MurNAc-l-Ala-d-Glu; 4, UDP-MurNAc-l-Ala-d-Glu-d-Ala-d-Ala pentapeptides. The component with a retention time of 80 min is vancomycin (26), which was used to induce accumulation of wall precursors. Abs, absorbance.

Cell wall peptidoglycan composition of cultures grown in the presence of sub-MICs of oxacillin.

Previous work has already demonstrated the basic abnormality of the peptidoglycan of RUSA235, namely, the appearance of two abnormal disaccharide dipeptide derivatives that differed from one another in the presence of isoglutamine or free glutamic acid residues (16). Figure 3 compares the muropeptide compositions of the peptidoglycans prepared from parental strain COL, mutant strain RUSA235, and its Homo* derivative, each grown either in TSB or in TSB supplemented with oxacillin at 5 μg/ml. This antibiotic concentration was previously shown to inhibit all of the native PBPs of S. aureus, leaving the mecA product PBP2A as the only transpeptidase to catalyze the cross-linking of muropeptides (4). The HPLC profiles of the three peptidoglycans were very similar; there was no evidence of the relative enrichment of the peptidoglycan in the abnormal dipeptide components when bacteria were grown with oxacillin in the medium.

FIG. 3.

Changes in the muropeptide composition of the peptidoglycan of murE mutant RUSA235. Muropeptide composition was determined in peptidoglycans isolated from parental strain COL, mutant RUSA235, and derivative RUSA235-Homo* grown in either TSB (A) or TSB containing oxacillin at 5 μg/ml (B). Peptidoglycan was isolated and hydrolyzed with muramidase, and the resulting muropeptides were separated by HPLC as described in Materials and Methods. Top, muropeptide pattern of parental strain COL; middle, muropeptide profile of mutant RUSA235; bottom, muropeptide profile of derivative RUSA235-Homo*. Abs, absorbance.

Purified cell walls prepared from strain COL, mutant RUSA235, and its Homo* derivative grown in the presence or absence of oxacillin showed comparable rates of cell wall degradation when treated with crude autolysin extract. The same strains also showed similar relative rates of autolysis when suspensions of the bacteria were exposed to the detergent Triton X-100 (Fig. 4).

FIG. 4.

Susceptibility of murE mutant cells and cell walls to autolytic degradation in vitro and in vivo. Cell walls prepared from strains COL (open columns), RUSA235 (dark gray columns), and RUSA235-Homo* (light gray columns) grown in the absence (A) or presence (B) of oxacillin at 5 μg/ml were tested for susceptibility to in vitro degradation by autolysins extracted from strain COL (see Materials and Methods). (C) Cultures of strains COL (open columns), RUSA235 (dark gray columns), and RUSA235-Homo* (light gray columns) grown in TSB were suspended in a Triton X-100 lysis buffer to an initial OD of 1.0, and the rates of autolysis were monitored as described in Materials and Methods. Data represent the means of two or three independent experiments.

Placing murE under the control of the inducible P_spac promoter.

Since murE is an essential gene, the impact of reduced levels of MurE on β-lactam resistance was tested by putting the transcription of the murE gene under the control of a promoter that can be induced with IPTG. The method of construction and the structure of the P_spac-controlled murE gene are shown in Fig. 5. For tighter repression of the P_spac promoter, multiple-copy plasmid pMGPII carrying the lacI repressor gene was also introduced into strain COL.

FIG. 5.

Construction of strain COL_spac::murE with the murE gene under the control of an inducible promoter. Only the uninterrupted copy of murE under the control of the P_spac promoter produces a functional protein.

Cultures of strain COL in which the expression of murE was inducible with IPTG were grown at different concentrations of the inducer. Overnight cultures of the bacteria grown in the presence of optimal concentrations of IPTG were centrifuged, washed twice with TSB, and resuspended in IPTG-free medium before being diluted to an OD₆₂₀ of 0.02 in TSB containing 10 μg of chloramphenicol per ml (in order to ensure the presence of plasmid pMGPII) and supplemented with increasing concentrations of IPTG. The growth rate of the cultures was followed by monitoring the rate of increase in OD. Figure 6 shows that the rate of bacterial growth was proportional to the concentration of IPTG in the medium.

FIG. 6.

Control of the growth rate of S. aureus strain COL by the rate of transcription of murE. A culture of strain COL_spac::murE grown overnight at 37°C in TSB supplemented with 100 μM IPTG and the appropriate antibiotics was centrifuged, washed twice with TSB, resuspended in the same volume of TSB, and distributed into test tubes containing TSB supplemented with various concentrations of IPTG and chloramphenicol (10 μg/ml). Growth of the cultures was monitored as described in Materials and Methods.

The same overnight cultures were also used to determine the effects of different IPTG concentrations on the expression of oxacillin resistance. The cultures were spread on agar supplemented with different concentrations of IPTG. A paper disk containing 1 mg of oxacillin was placed in the middle of the agar plates, and the diameter of the inhibition zone was determined after incubation of the plates at 37°C. Figure 7 shows that the diameter of the inhibition zone decreased in parallel with the increasing IPTG concentrations in the agar medium. The effect of IPTG was specific for β-lactam resistance; no changes in the diameters of zones of inhibition due to tetracycline, ciprofloxacin, vancomycin, or d-cycloserine were detected as a function of the concentration of IPTG in the agar medium (data not shown).

FIG. 7.

Effect of murE transcription on oxacillin resistance. Strain COL_spac::murE was plated on TSA supplemented with increasing concentrations of IPTG (25 [A], 37.5 [B], 75 [C], 150 [D], and 500 [E] μM). The sizes of inhibition halos around paper disks containing 1 mg of oxacillin were measured after incubation at 37°C for 9.5 h. The values under the panels are the diameters of inhibition zones.

Accumulation of UDP-N-acetylmuramyl dipeptides in cultures grown in suboptimal concentrations of IPTG.

Bacterial cultures were grown in TSB supplemented with different concentrations of IPTG. When the OD of the cultures reached 0.5, bacterial cultures were centrifuged and the composition of the cell wall precursor pool was determined as described in Materials and Methods. Cultures growing in the presence of increasing concentrations of IPTG showed a gradual increase in the amount of UDP-muramyl pentapeptide and a gradual decrease in the amounts of UDP-N-acetylmuramyl dipeptides in the cell wall precursor pool. For instance, the relative amount of the dipeptides decreased from 42 to 8% as the concentration of IPTG increased from 37.5 to 200 μM while the relative amount of muramyl pentapeptides increased from 27 to about 66% under the same growth conditions (Fig. 8).

FIG. 8.

Effect of murE transcription on the composition of the cytoplasmic peptidoglycan precursor pool. Cytoplasmic cell wall precursors were extracted from strain COL_spac::murE grown with different concentrations of IPTG and were analyzed as described in Materials and Methods. Numbers 1 through 4 identify the cell wall precursors listed in the legend to Fig. 2. Abs, absorbance.

Transcription of murE and methicillin resistance genes mecA and pbpB in cultures growing with various concentrations of IPTG.

Figure 9A through C show the Northern analysis of the transcription of murE and two genetic determinants known to be associated with β-lactam antibiotic resistance in S. aureus: mecA and pbpB, the structural determinants of PBP2A and PBP2, respectively. Cultures of COL_spac::murE were grown at four different concentrations of IPTG, and transcription levels were determined. In each one of the cases, the levels of transcription appeared to be a function of the IPTG concentration in the medium.

FIG. 9.

Transcription analysis of selected S. aureus genes in cultures of promoter-controlled murE and Tn551 murE mutants. Northern analysis was used to determine the levels of transcription of mecA (A), pbpB (B), and murE (C) in cultures of COL and COL_spac::murE grown in TSB supplemented with different concentrations of IPTG. Lanes: 1, COL; 2 through 5, COL_spac::murE grown in the presence of 25 (lane 2), 37.5 (lane 3), 75 (lane 4), and 200 (lane 5) μM IPTG. (D to G) Northern blotting analysis of the mecA (D), pbpB (E), pta (F), and murF (G) genes from parental strain COL (lane 1), the RUSA235 murE mutant (lane 2), and the RUSA235-Homo* derivative (lane 3). (H) Western blotting analysis of the amounts of PBP2A in parental strain COL (lane 1), the RUSA235 murE mutant (lane 2), and the RUSA235-Homo* derivative (lane 3). The analysis was done for 10 and 30 μg of total protein. The reasons for the smaller molecular size of the reactive band in panel C, line 1, are not clear. A similar change in molecular size has been observed before in spac control promoters (21). The change may involve some rearrangement around the initiation site of the promoter.

Led by these observations, we also determined the transcription of mecA and pbpB in strain COL, mutant RUSA235, and its Homo* derivative. Reduced transcription in the mutant and a return to a higher degree of expression in the Homo* derivative were noted (Fig. 9). The transcription levels of murF and pta (phosphate acetyltransferase)—two genes selected as controls—did not change in these three strains (Fig. 9).

DISCUSSION

Earlier data on the phenotype of murE insertional mutant RUSA235 pointed to abnormalities of cell wall peptidoglycan structure as the possible cause of the drastic reduction in the level of oxacillin resistance. In the cytoplasmic cell wall precursor pool of the murE mutant, there was an accumulation of UDP-N-acetylmuramyl dipeptides that then found their way into the polymerized cell wall of the mutant (16). The disaccharide dipeptides must be incorporated into the peptidoglycan through transglycosylation and will remain unavailable for cross-linking since they are missing both the critical diamino acid component and the C-terminal d-Ala-d-Ala residue.

We hypothesized that the drastic and selective reduction of resistance to β-lactam antibiotics may be related to the incorporation of abnormal muropeptides causing some sort of structural defect in the mutant cell wall. One may envision several specific scenarios. For instance, growth in the presence of β-lactam antibiotics may further increase the relative proportion of the abnormal dipeptide component to such a degree that it begins to jeopardize the structural stability of the cell wall. Mutant cell walls in which cross-linking is further inhibited by oxacillin may be hypersensitive to autolytic degradation. Increased sensitivity of poorly cross-linked staphylococcal cell walls to enzymatic degradation in vitro has been described (22). The abnormal dipeptide components may be recognized in β-lactam-treated bacteria as signals for a suicidal activation of enzymes that catalyze cell wall turnover.

Experimental tests of these hypotheses (described in this paper) gave negative results. Growth of the mutant bacteria in the presence of sub-MICs of oxacillin did not increase the incorporation of dipeptides into the peptidoglycan. Mutant and parental cell walls prepared from normal or oxacillin-treated bacteria were shown to have virtually identical sensitivities to degradation by autolytic extracts in vitro. Actually, the rate of autolysis induced by the detergent Triton X-100 was slightly slower in the mutant compared to that in the parental cells. Furthermore, there was no difference in cell wall lysis sensitivity between mutant RUSA235 and its Homo* derivative, in which high-level methicillin resistance was restored.

These data do not support our initial hypothesis that the mechanism of reduced β-lactam resistance was related to the structurally defective cell wall produced in the murE mutant.

In mutant RUSA235, the transposon is inserted 3 bp from the C terminus of the gene (15), resulting in the production of an abnormal protein that has reduced specific catalytic activity compared to the enzyme from parental cells (data not shown). In order to test the impact of inhibition of MurE on β-lactam resistance, we put the transcription of the gene under the control of an inducible promoter.

The introduction of this new experimental system provided a striking confirmation of the importance of MurE for the expression of oxacillin resistance. Not only was the growth rate of the cultures with the controllable murE gene a function of the IPTG concentration in the medium, but the level of oxacillin resistance also showed clear dependence on the transcription levels of murE and cultures grown at suboptimal concentrations of IPTG showed accumulation of UDP-N-acetylmuramyl dipeptides in the cell wall precursor pool. Thus, the basic microbiological and biochemical observations obtained in the Tn551 mutant RUSA235 were reproduced in parental strain COL by experimental modulation of the transcription of murE.

Unexpectedly, and most interestingly, the controlled rate of transcription of murE also brought along parallel changes in the transcription levels of mecA and pbpB, two genes that play a central role in methicillin resistance.

It has been well established that in MRSA cultures even low concentrations of β-lactam antibiotics can fully acylate and inactivate the normal complement of PBPs. In strain COL, with an oxacillin MIC of more than 400 μg/ml, as little as 2 to 5 μg of oxacillin per ml was shown to block the formation of most oligomeric muropeptides, resulting in the production of a peptidoglycan that was primarily composed of muropeptide monomers and dimers plus a small amount of trimeric muropeptides (4). According to our current model, in staphylococci growing in the presence of oxacillin, cell wall biosynthesis is mainly catalyzed by two proteins, methicillin resistance protein PBP2A, providing transpeptidase activity, and PBP2, providing the essential transglycosylase activity (20). The observations described in this communication indicate that the transcription rates of the genetic determinants of these two critical resistance proteins—mecA and pbpB—depend on the rate of transcription of murE, the structural gene of an enzyme that is involved with a step in the synthesis of the UDP-linked cell wall precursors. Apparently, the reduced rate of murE transcription in the bacteria brought along a parallel decline in the transcription of the two PBP-encoding genes and loss of oxacillin resistance. Analysis by Western blotting demonstrated that suppressor mutants of RUSA235 in which normal levels of oxacillin resistance were recovered contained larger amounts of PBP2A than did the mutant bacteria (Fig. 9).

Together, these data suggest that the mechanism by which β-lactam resistance is suppressed in murE mutants is related to a deficit in the cellular amounts of PBP2A and PBP2—two proteins that form the key components of the β-lactam resistance mechanism in S. aureus. Dependence of the methicillin MIC on the cellular amounts of PBP2A is not without precedent. In clinical isolates of MRSA that carry the specific chromosomal or plasmid-borne regulatory genes, the oxacillin MIC depends on the rate of transcription of mecA and/or the cellular amounts of PBP2A (23).

The apparently coordinate regulation of the transcription of two PBPs, PBP2A and PBP2, by the rate of transcription of a gene (murE) involved with the biosynthesis of cell wall precursors has interesting implications beyond the context of β-lactam resistance. Our data imply that PBP2 and PBP2A may be unstable—either in the functional or in the topographic sense. The selective localization of PBP2 at the sites of staphylococcal cell division has recently been demonstrated (19). It is conceivable that deposition of this essential protein at the zone of a new cell division requires the production of new molecules of this protein, in coordination with the rate of cell wall biosynthesis, a control site of which may be at the transcription of murE. Whether the transcriptional control is directly exerted through the MurE protein or through the change in the concentration of cell wall precursors, e.g., through the change in the concentration of the UDP-MurNAc pentapeptide cell wall precursor, remains to be determined. The participation of cell wall precursors in the control of expression of penicillinase-based β-lactam resistance and also in the control of the process of recycling of cell wall components has already been demonstrated in E. coli (10, 11, 14, 18).