Comparative study of the susceptibilities of major epidemic clones of methicillin-resistant Staphylococcus aureus to oxacillin and to the new broad-spectrum cephalosporin ceftobiprole

Abstract

Multidrug-resistant strains of Staphylococcus aureus continue to increase in frequency worldwide, both in hospitals and in the community, raising serious problems for the chemotherapy of staphylococcal disease. Ceftobiprole (BPR; BAL9141), the active constituent of the prodrug ceftobiprole medocaril (BAL5788), is a new cephalosporin which was already shown to have powerful activity against a number of bacterial pathogens, including S. aureus. In an effort to test possible limits to the antibacterial spectrum and efficacy of BPR, we examined the susceptibilities of the relatively few pandemic methicillin-resistant S. aureus (MRSA) clones that are responsible for the great majority of cases of staphylococcal disease worldwide. We also included in the tests the highly oxacillin-resistant subpopulations that are present with low frequencies in the cultures of these clones. Such subpopulations may represent a natural reservoir from which MRSA strains with decreased susceptibility to BPR may emerge in the future. We also tested the efficacy of BPR against MRSA strains with reduced susceptibility to vancomycin and against MRSA strains carrying the enterococcal vancomycin resistance gene complex. BPR was shown to be uniformly effective against all these resistant MRSA strains, and the mechanism of superb antimicrobial activity correlated with the strikingly increased affinity of the cephalosporin against penicillin-binding protein 2A, the protein product of the antibiotic resistance determinant mecA.

FIG. 1.

Oxacillin and BPR susceptibility profiles of major epidemic MRSA clones and their homogeneous oxacillin-resistant subpopulations. Aliquots of an overnight culture were plated after serial dilution on TSA containing increasing concentrations of oxacillin (circles) or BPR (squares). The numbers of CFU were counted after incubation for 48 h at 37°C. The antibiotic susceptibility profiles (closed symbols and solid lines) were first determined for E2125 (A), HPV107 (B), BK2464 (C), HDE288 (D), BK2529 (E), HAR24 (F), HAR22 (G), USA300 (H), COL (I), and HU25 (J). Colonies were then picked from the subpopulations that were able to grow in the presence of high concentrations of oxacillin (indicated by the arrowheads with asterisks), and the oxacillin and BPR susceptibility profiles were further determined (open symbols and dashed lines). Colonies were also picked from the subpopulations that were able to grow in the presence of elevated concentrations of BPR (arrowheads), and the BPR susceptibility profiles of the bacteria were determined (see Table 2).

FIG. 2.

Vancomycin, oxacillin, and BPR susceptibility profiles of VISA and VRSA strains. Aliquots of overnight cultures were plated on TSA containing increasing concentrations of vancomycin (closed squares), oxacillin (open circles), or BPR (closed circles). The numbers of CFU were counted after incubation at 37°C for 48 h. The antibiotic susceptibility profiles were determined for VISA strains Mu50 (A) and JH9 (B) and for VRSA strains VRS1 (C) and COLVA (D).

FIG. 3.

Inhibitory effects of BPR on the expression of the vancomycin resistance phenotype in VRSA strains VRS1 and COLVA. Aliquots of overnight cultures of strains VRS1 (A) and COLVA (B) were plated on TSA containing increasing concentrations of vancomycin in the absence of BPR (closed squares) or in the presence of a constant concentration (1 μg/ml) of BPR (open circles). Colonies were picked from the subpopulations of VRS1 or COLVA growing on plates containing 128 μg/ml of vancomycin and 1 μg/ml BPR (as indicated by arrows). Cultures of these colonies, named VRS1* and COLVA*, respectively, were tested for their vancomycin susceptibility profiles in the presence of a constant concentration (1 μg/ml) of BPR (open squares). (C) BPR susceptibility profiles of VRS1* (closed squares) and COLVA* (closed circles).

FIG. 4.

Synergistic effect of BPR on the vancomycin resistance of VRSA strains COLVA and VRS1. Strains VRS1 (A and B) and COLVA (C and D) were spread on TSA plates without (A and C) or with (B and D) 1 μg/ml of BPR added to the agar medium. The vancomycin susceptibilities were evaluated after overnight incubation at 37°C by the vancomycin Etest method.

FIG. 5.

Susceptibility of MRSA strain COL to the bactericidal effects of oxacillin and BPR. The viability of MRSA strain COL was determined during exposure to 10× the MIC of oxacillin (closed circles) or BPR (closed squares) and was compared to that of a culture grown in antibiotic-free medium (closed triangles). Exponentially growing cultures were exposed to the antibiotic at an OD₆₂₀ of 0.2 (zero time), and aliquots were removed at various intervals to determine the viable counts. Bacterial cells recovered from the first exposure to BPR for 24 h were again inoculated into fresh medium in the absence of BPR (empty triangles) or in the presence of the same concentration of BPR (empty squares), and the viable counts were determined.

FIG. 6.

Comparison of affinities of oxacillin and BPR for PBPs of S. aureus strain RU130. Membrane preparations (150 μg of proteins) were first incubated with increasing concentrations of oxacillin (A) or BPR (B) and then with a single saturating concentration of benzyl[¹⁴C]penicillin. After SDS-PAGE, the gel was exposed to a tritium storage phosphor screen for 2 weeks.

FIG. 7.

Comparison of the affinities of BPR and oxacillin for purified PBP 2A. Purified PBP 2A was initially incubated with different concentrations (from 0.5 μM to 20 μM) of BPR (squares) or oxacillin (circles), and then Bocillin FL (20 μM) was added as a reporter substrate. After SDS-PAGE, the gel was immediately scanned with a fluoroimager (Typhoon 9400).

FIG. 8.

Effects of β-lactam antibiotics on the circular dichroic spectra of PBP 2A. The conformational change of PBP 2A was determined by measuring the CD spectra in the absence of drug (circles) and in the presence of either BPR (squares) or oxacillin (triangles). Purified PBP 2A was incubated with β-lactam antibiotics for 30 min at 25°C, and then the CD spectra were recorded from 250 nm to 200 nm.

FIG. 9.

HPLC elution profiles of peptidoglycan purified from MRSA strain COL grown in the absence and in the presence of BPR. Peptidoglycans were purified from 1-liter cultures grown at 37°C to an OD₆₂₀ of 0.4 in the absence of BPR (A) or in the presence of a subinhibitory concentration (0.4 μg/ml) of BPR (B). Following digestion with mutanolysin, muropeptides were separated by HPLC and detected by measurement of the absorbance at 206 nm.