Previous Article | Next Article ![]()
Antimicrobial Agents and Chemotherapy, April 2002, p. 971-976, Vol. 46, No. 4
0066-4804/02/$04.00+0 DOI: 10.1128/AAC.46.4.971-976.2002
Copyright © 2002, American Society for Microbiology. All Rights Reserved.
Department of Microbiology, Bristol-Myers Squibb Co., Wallingford, Connecticut 06492
Received 23 March 2001/ Returned for modification 27 October 2001/ Accepted 2 January 2002
|
|
|---|
0.25 to 1 µg/ml. The BMS-247243 MIC90 for ß-lactamase-producing S. aureus strains was fourfold higher than that for ß-lactamase-nonproducing strains. BMS-247243 is hydrolyzed by staphylococccal ß-lactamases at 4.5 to 26.2% of the rates measured for cephaloridine. The affinity of BMS-247243 for PBP 2a was >100-fold better than that of methicillin or cefotaxime. BMS-247243 is bactericidal for MRSA, killing the bacteria twice as fast as vancomycin. These in vitro activities of BMS-247243 correlated with its in vivo efficacy against infections in animals, including the neutropenic murine thigh and rabbit endocarditis models involving MRSA strains. In conclusion, BMS-247243 has in vitro and in vivo activities against methicillin-resistant staphylococci and thus may prove to be useful in the treatment of infections caused by these multidrug-resistant organisms. |
|
|---|
30% among nosocomial S. aureus isolates to over 60% among nosocomial S. epidermidis isolates (16). Although methicillin-resistant (MR) S. aureus (MRSA) had largely been considered a hospital pathogen, its prevalence in the community is increasing (17). Cephalosporins, like all ß-lactams, inhibit bacteria by binding to enzymes involved in the biosynthesis of peptidoglycan. Binding of ß-lactams to these enzymes (also known as penicillin-binding proteins [PBPs]) leads to their inactivation and cell death. Unlike methicillin-susceptible strains of staphylococci, methicillin-resistant strains have acquired foreign genetic material which encodes a new PBP, known as PBP 2a (4, 7). PBP 2a has a low affinity for binding to ß-lactams (1), and in the presence of a ß-lactam antibiotic, it can compensate functionally for the high-affinity PBPs which are inactivated by the ß-lactam (7).
A program was initiated at Bristol-Myers Squibb with the goal of identifying a cephalosporin that has an increased affinity for PBP 2a of MRSA. BMS-247243 is the product of this effort (Fig. 1). BMS-247243 has three structural motifs: the cephalosporin nucleus and unique features on the C-7 and C-3 side chains (O. Kim, Y. Zhang, J. Wichtowski, D. Springer, B. Luh, J. Goodrich, R. Sterzycki, S. D'Andrea, P. Misco, Y. Ueda, and J. Bronson, Abstr. 40th Intersci. Conf. Antimicrob. Agents Chemother., abstr. 1062, 2000). At C-7, the dichloro group imparts lipophilicity, which is necessary for good anti-MRSA activity. The vinyl and amide groups at C-7 contribute to its favorable pharmocokinetics in animals. The C-7 acid is responsible for one of the four charges on this molecule. At neutral pH, BMS-247243 has a net charge of zero. The overall charge of the cephalosporin influences its potency against MRSA, its safety, and its solubility. At C-3, the morpholinium quaternary component is important for potency and pharmacokinetics, and the dimethyl groups contribute to improved solubility.
![]() View larger version (9K): [in a new window] |
FIG. 1. Chemical structure of BMS-247243.
|
(Part of this work was presented previously [J. Fung-Tomc, B. Minassian, M. Pucci, E. Gradelski, E. Huczko, T. Washo, and D. P. Bonner, Abstr. 40th Intersci. Conf. Antimicrob. Agents Chemother., abstr. 1063, 2000; E. Huczko, B. Minassian, E. Gradelski, J. Fung-Tomc, and D. Bonner, Abstr. 40th Intersci. Conf. Antimicrob. Agents Chemother., abstr. 1064, 2000; T. W. Hudyma, S. D'Andrea, O. Kim, B. Luh, J. Matiskella, P. Misco, D. Springer, Y. Zhang, J. Bronson, and Y. Ueda, Abstr. 40th Intersci. Conf. Antimicrob. Agents Chemother., abstr. 1060, 2000; Kim et al., 40th ICAAC; B. Kolek, E. Gradelski, D. Bonner, and J. Fung-Tomc, Abstr. 40th Intersci. Conf. Antimicrob. Agents Chemother., abstr. 1065, 2000; L. Lamb, I. Medina, C. Ferraro, S. Chaniewski, D. Taylor, Y. Tsai, and J. M. Clark, Abstr. 40th Intersci. Conf. Antimicrob. Agents Chemother., abstr. 1066, 2000; D. Springer, B. Luh, J. Goodrich, T. Hudyma, J. Bronson, and R. Miller, Abstr. 40th Intersci. Conf. Antimicrob. Agents Chemother., abstr. 1061, 2000].).
|
|
|---|
Bacterial strains. More than 250 clinical isolates of staphylococci collected within the past 10 years (half were collected within the past 5 years) were tested. The strains were categorized as methicillin susceptible (MS) or MR by PCR detection of the mecA gene (1). ß-Lactamase determination was performed by nitrocefin testing (11). ß-Lactamase extracts were prepared from strains RN107 and RN98 (kind gifts of R. Novick, Public Health Research, New York, N.Y.) and from strains 22260 and ST79/741 (type B) and strains FAR8 and FAR 10 (type D) (graciously provided by D. S. Kernodle, Veterans Affairs Medical Center, Nashville, Tenn.).
Isogenic ß-lactamase-nonproducing variants from four ß-lactamase-producing MRSA strains were obtained by curing the bacteria of the plasmid which encodes the enzyme. The plasmids were cured by incubating the strains, which were plated on Mueller-Hinton agar (MHA) plates, overnight at a high temperature (42°C). Colonies were picked, plated onto starch-iodine and MHA plates, and grown overnight at 35°C. The starch-iodine plates were then flooded with 30 mg of benzylpenicillin per ml and 0.32 M iodine in 0.12 M potassium iodide. Colonies that failed to produce a white halo on the plates flooded with starch-iodine were confirmed to be non-ß-lactamase producing by further testing and their inability to hydrolyze nitrocefin (3, 8).
MICs. MICs were determined by the NCCLS-recommended agar dilution method (11). The test medium was MHA for vancomycin and MHA supplemented with 2% NaCl for the ß-lactams. The bacterial inocula were 5 x 104 CFU/spot. The plates were incubated at 35°C for 24 h, and the MIC endpoint was defined as the lowest drug concentration that inhibited all visible growth.
Population analysis.
Efficiency of plating (EOP) is a measure of the degree of methicillin resistance expression (2). It was determined by quantitatively plating a suspension of each MRSA strain (containing
106 to 108 CFU/ml) onto MHA with or without 50 µg of methicillin per ml. The plates were incubated at 35°C for 48 h. EOP was defined as the number of CFU obtained on the plate with MHA and 50 µg of methicillin divided by the number of CFU obtained on the drug-free MHA plate. Heterogeneously MR strains have EOPs of <0.1, and homogeneously MR strains have EOPs of 0.1 to 1 to methicillin.
ß-Lactamase hydrolysis. The staphylococcal ß-lactamases used for evaluation were whole-culture extracts of S. aureus containing type A to D ß-lactamases grown in phosphate-CY medium (5). The stabilities of BMS-247243 and cephaloridine against ß-lactamases were evaluated with 100 µg of the cephalosporin per ml in 0.1 M sodium phosphate buffer (pH 6.0). The hydrolysis of BMS-247243 was expressed relative to that of cephaloridine, which was set at 100%.
PBP 2a binding assay. Cephalosporin binding was determined by an indirect method (13). The MRSA membrane proteins were prepared from ß-lactamase-negative MRSA strain A29225. Serial twofold dilutions of the cephalosporin were preincubated with the MRSA membrane proteins at 35°C for 20 min, after which [3H]benzylpenicillin (10 µg/ml, 26 Ci/nmol) was added and the mixture was incubated for another 20 min. The sample was boiled and electrophoresed on a sodium dodecyl sulfate-polyacrylamide gel for 20 h at 4°C with a constant current of 10 mA. The gel was exposed to X-ray film for up to 5 days, and a densitometer plot was generated for determination of the 50% inhibitory concentration (IC50), which was defined as the cephalosporin concentration that caused a 50% reduction in [3H]benzylpenicillin binding versus that for the control (which did not undergo preincubation with the cephalosporin).
Bactericidal activity.
The bactericidal activity of BMS-247243 was established by determination of the minimum bactericidal concentration (MBC) and time-kill analysis. The MBC was defined as the lowest drug concentration with CFU counts equal to n + 2
n, where n is equal to 0.1% of the initial inoculum (12, 15).
Time-kill analyses were performed in Mueller-Hinton broth (MHB) with 2% NaCl. The cells were grown to the logarithmic phase by preincubation of the inoculum (
5 x 105 CFU/ml) in fresh medium prior to the addition of drug. Aliquots were removed at 0, 2, 4, 6, 7, and 24 h after drug addition; and 50 µl of the undiluted aliquot and 10-fold serial dilutions of the aliquot were plated onto MHA plates for viable count determination. The plates were incubated at 35°C for 48 h.
Rabbit endocarditis. Rabbit endocarditis caused by MRSA strain A27223 was established by the procedure of Perlman and Freedman (14). A polyethylene catheter was inserted into the right carotid artery of female rabbits obtained from Convance Research Products, Denver, Pa. The catheter was advanced across the aortic valve until resistance was met. The catheter remained in place throughout the study. Five to 7 days later the rabbits were infected intravenously (i.v.) with 106 CFU of MRSA strain A27223. Therapy was started at 18 h postinfection and was continued for 3 consecutive days. BMS-247243 was administered i.v. at 10 mg/kg of body weight every 8 h for a daily dose of 30 mg/kg, while vancomycin was administered i.v. at 20 mg/kg every 12 h for a daily dose of 40 mg/kg. The pharmacokinetic parameters of BMS-247243 in rabbits dosed i.v. with 10 mg/kg were as follows: maximum concentration in plasma (Cmax), 94 µg/ml; plasma half-life (t1/2), 2.4 h; area under the concentration-time curve (AUC), 295 µg · h/ml; and level of binding to rabbit serum, 85%. For vancomycin, which was administered i.v. at 17.5 mg/kg, Cmax was 50.2 µg/ml and the plasma t1/2 was 1.3 h (9). Fifteen hours after the last treatment, the surviving animals were euthanatized and the cardiac vegetations were removed, weighed, and homogenized. Viable counts were determined by plating serial 10-fold dilutions of the homogenates onto sheep blood agar plates.
Thigh infection in immunocompetent and immunosuppressed mice. In mice with an intact immune system, the right thigh was infected on day 1 with 105 CFU of MRSA isolate A27218. The test compounds (five mice per group) were administered in 5% dextrose at 1, 4, and 7 h postinfection. The mice were killed on day 5. The infected muscle was removed and homogenized, and the viable counts were determined; these counts were compared to the numbers of CFU recovered from the thighs of nontreated animals.
Immunosuppression was induced with two 100-mg/kg intraperitoneal injections of cyclophosphamide monohydrate (Sigma) given 1 and 5 days before infection. Groups of five mice each were infected intramuscularly with 105 CFU of MRSA strain A27223 and were treated with a single i.v. dose of antibiotic at 2 h postinfection. At 24 h postinfection, the infected thighs were processed as described above.
MS S. aureus strain A15090 was also used in the neutropenic thigh infection model. However, the mice were treated at 1 h postinfection and the numbers of CFU in the infected muscle were determined at 1, 3, 5, and 24 h postinfection.
The pharmacokinetics of BMS-247243 in mice dosed i.v. with 20 mg/kg were as follows: Cmax, 61 µg/ml; plasma t1/2, 1.5 h; AUC, 107 µg · h/ml; and level of binding to mouse serum, 95%.
|
|
|---|
0.25 to 1 µg/ml (Tables 1 and 2). The BMS-247243 MICs at which 90% of strains are inhibited (MIC90s) for ß-lactamase-producing (BLA+) strains of S. aureus were fourfold higher than the MIC90s for ß-lactamase-negative (BLA-) strains. The MIC90s of BMS-247243 for MRSA and MR S. epidermidis strains were 4 and 2 µg/ml, respectively. The BMS-247243 MIC90 for MR S. haemolyticus was higher (8 µg/ml). |
View this table: [in a new window] |
TABLE 1. Activities of BMS-247243 against staphylococci
|
|
View this table: [in a new window] |
TABLE 2. MICs for other MS and MR Staphylococcus species of which five or fewer strains were tested
|
0.1. Since the susceptibility of an MRSA strain to ß-lactams is directly influenced by its level of methicillin resistance expression (10), the high proportion of homogeneously MR strains included in this evaluation substantiates the BMS-247243 MIC90 derived for MRSA strains. Because heterogeneous methicillin resistance is expressed by many clinical MRSA strains under routine susceptibility testing conditions, modifications of the MIC test are recommended for the testing of the susceptibilities of staphylococci to ß-lactams (6). These modifications include the use of a larger bacterial inoculum, supplementation of MHA with NaCl, and the use of a longer incubation period. When the bacterial inocula were increased by 100-fold, the BMS-247243 MICs increased two- and fourfold for ß-lactamase-nonproducing and -producing strains, respectively (data not shown). The BMS-247243 MIC results in Tables 1 and 2 were derived from tests with MHA and 2% NaCl; the BMS-247243 MICs obtained by tests with salt were twofold higher than those obtained in the absence of salt (data not shown). Likewise, by broth microdilution, the addition of salt to MHB also resulted in twofold higher BMS-247243 MICs compared to those obtained with unsupplemented MHB (data not shown). The BMS-247243 MICs obtained by broth microdilution were twofold lower than those obtained by agar dilution (data not shown). Prolonged incubation (48 versus 24 h) did not alter the BMS-247243 MICs (data not shown).
ß-Lactamase stability. As indicated in Table 1, the BMS-247243 MIC90 for BLA+, MS S. aureus strains was fourfold higher than that for BLA-, MS S. aureus strains, suggesting some influence of staphylococcal ß-lactamases on the activity of BMS-247243 against MS staphylococci. This influence was also examined by determining the effects of higher bacterial inocula or the addition of a ß-lactamase inhibitor, clavulanic acid. A 100-fold larger MS S. aureus inoculum size led to twofold increased BMS-247243 MICs for ß-lactamase-nonproducing strains and eightfold higher BMS-247243 MICs for ß-lactamase-positive strains (data not shown). Clavulanic acid, added in a proportion equal to that of BMS-247243, led to twofold decreases in the BMS-247243 MICs for ß-lactamase-positive MS S. aureus isolates and to no change in the BMS-247243 MICs for ß-lactamase-nonproducing strains (data not shown).
The influence of ß-lactamase on the BMS-247243 MICs for MRSA strains was minimal (Table 3). The BMS-247243 MICs were generally the same for isogenic MRSA pairs with and without the staphylococcal ß-lactamase. Unlike BLA+, MS S. aureus strains, the minimal effect of ß-lactamase on the BMS-247243 MICs for MRSA isolates might be explained by the higher levels of BMS-247243 needed to inactivate PBP 2a relative to the amount of BMS-247243 hydrolyzed by the staphylococcal ß-lactamases.
|
View this table: [in a new window] |
TABLE 3. MICs of BMS-247243 for isogenic pairs of MRSA strains with or without ß-lactamasea
|
PBP 2a binding. The binding affinity of BMS-247243 for PBP 2a was >100-fold better than those of methicillin and cefotaxime (Table 4). This increased binding affinity for PBP 2a correlates with the enhanced activity of BMS-247243 against MR staphylococci.
|
View this table: [in a new window] |
TABLE 4. PBP 2a binding for BMS-247243 and other ß-lactamsa
|
4 indicate that BMS-247243 is bactericidal. By time-kill studies, the maximum rate of killing of MRSA by BMS-247243 was obtained at four to eight times the MIC. The rates of killing of three MRSA strains by BMS-247243 and the comparator compounds are listed in Table 5. BMS-247243 decreased the viable counts by about 2 log10 at 6 h of incubation, twice the killing rate measured with vancomycin. The rates of bactericidal activity of BMS-247243 and vancomycin were halved in the presence of 2% NaCl. With methicillin, imipenem, or cefotaxime at 16 µg/ml, growth of MRSA strain A27223 was observed at 6 h of incubation (data not shown).
|
View this table: [in a new window] |
TABLE 5. Rate of MRSA killing by cell wall-active agentsa
|
![]() View larger version (14K): [in a new window] |
FIG. 2. Efficacy of BMS-247243 against rabbit endocarditis caused by MRSA strain A27223.
|
![]() View larger version (23K): [in a new window] |
FIG. 3. Comparative efficacies of BMS-247243 and vancomycin in a neutropenic murine thigh model caused by methicillin-susceptible S. aureus strain A15090.
|
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»