Previous Article | Next Article ![]()
Antimicrobial Agents and Chemotherapy, July 2002, p. 2174-2178, Vol. 46, No. 7
0066-4804/02/$04.00+0 DOI: 10.1128/AAC.46.7.2174-2178.2002
Copyright © 2002, American Society for Microbiology. All Rights Reserved.
Laboratoire d'Antibiologie, Faculté de Médecine, Université de Nantes, Nantes,1 Laboratoire de Bactériologie, Centre Hospitalier Universitaire Côte-de-Nacre, Caen, France2
Received 28 September 2001/ Returned for modification 29 January 2002/ Accepted 10 April 2002
|
|
|---|
|
|
|---|
Quinupristin and dalfopristin are water-soluble injectable streptogramin B (SB) and SA antibiotics, respectively, whose combination in a 30:70 (wt/wt) ratio acts synergistically on gram-positive bacteria (5, 14). Streptogramins inhibit protein synthesis by binding to the ribosomal 50S subunit, and resistance is a concern mainly for the SB type (9, 20). The most frequent mechanism of quinupristin resistance encountered in recent years is target modification by methylation of an adenine residue in 23S rRNA, encoded in S. aureus by the ermA, ermB, or ermC gene. Constitutively expressed erm genes confer in vitro cross-resistance to macrolides, lincosamides, and SB. Some strains harbor inducible erm genes, whose expression is triggered only by 14- and 15-member macrolides (18). The msrA gene is believed to encode an efflux system, and vgb and vgbB genes encode a hydrolase that inactivates SB type antibiotics (22). Recently, mutations in the L22 ribosomal protein have been associated with resistance to quinupristin in S. aureus (B. Malbruny, A. Canu, B. Bozdogan, V. Zarrouk, B. Fantin, and R. Leclercq, Abstr. 40th Intersci. Conf. Antimicrob. Agents Chemother., abstr. 1928, 2000). Constitutive resistance to quinupristin has led to a relative loss of the bactericidal effect of quinupristin-dalfopristin (Q-D) in S. aureus rat and rabbit experimental endocarditis models (13, 15).
Previous experimental studies showed a potential benefit in combining Q-D with other antibiotics such as ß-lactams and rifampin against MRSA (26, 28). Gentamicin, like other aminoglycosides, is considered to exert its antibacterial effect via inhibition of protein synthesis after binding to the 30S ribosomal subunit (21). Its use, in association with other antibiotics, in the first days of treatment of serious S. aureus infections is usually recommended. However, the combination of gentamicin and Q-D was only studied in vitro, and conflicting results, including possible antagonism, were obtained (16, 17).
This study tested the combination of Q-D with gentamicin both in vitro and in vivo against dalfopristin- and gentamicin-susceptible S. aureus strains with various susceptibilities to quinupristin.
(This work was presented in part at the 40th Interscience Conference on Antimicrobial Agents and Chemotherapy, Toronto, Canada, 17 to 20 September 2000.)
|
|
|---|
Antibiotics and media. Gentamicin and Q-D were provided, respectively, by Schering-Plough (Hérouville-Saint-Clair, France), and Aventis (Paris, France). Unless otherwise stated, bacteria were grown overnight at 37°C either in cation-supplemented Mueller-Hinton broth (Sanofi Diagnostic Pasteur, Marne-la-Coquette, France) or on Mueller-Hinton agar (Difco, Becton-Dickinson France SA, Le Pont de Claix, France).
Search for quinupristin resistance determinants. Quinupristin resistance determinants were searched for by PCR. Genomic DNA was extracted from culture using the Qiagen DNeasy tissue kit. The sequences of primers for ermA, ermB, ermC, and msrAB are shown in Table 1. The 50-µl reaction mixture for PCR contained 10 ng of total bacterial DNA, 20 pmol (each) oligodeoxynucleotide, 200 µM deoxynucleoside triphosphates, 2 mM MgCl2, 1x PCR buffer, and 1 U of Taq DNA polymerase (Eurobio). PCR experiments were carried out in a Perkin-Elmer 4600 thermal cycler with a denaturation step (94°C for 5 min), followed by 35 amplification cycles (30 s of denaturation at 94°C, 45 s of annealing at 47°C, and 1 min of elongation at 72°C) and a final elongation step (72°C for 10 min). PCR products were separated by electrophoresis through a 2% agarose gel under a 100-V gradient.
|
View this table: [in a new window] |
TABLE 1. Sequences of primers for genes encoding resistance to quinupristin
|
Checkerboard method.
Interactions between Q-D and gentamicin were assessed by the checkerboard method in 96-well microtiter plates (Nalge Nunc International, Roskilde, Denmark), with an inoculum of 106 CFU/ml. Plates were incubated for 18 h at 37°C. Fractional inhibitory concentration indices were interpreted as follows:
0.5, synergy; >4, antagonism; >0.5 and
4, indifference (1).
Time-kill curves.
Time-kill curves were plotted on the basis of results obtained with an inoculum of 5 x 106 CFU/ml in glass tubes containing Mueller-Hinton broth and gentamicin and/or Q-D at concentrations equal to their MICs. Control tubes contained no antibiotics. After 0, 2, 6, and 24 h of incubation at 37°C, viable counts were determined by subculturing 50-µl serial dilutions of samples on agar plates with a spiral plater (Spiral Système, Saint-Nom-la-Bretèche, France). Colonies were counted after 48-h of incubation at 37°C. Antibiotic carryover was ruled out after the plating of a 100-CFU/ml suspension in the presence or absence of antibiotics. The lower limit of detection for the spiral plater was 2 log10 CFU/ml. Synergy was defined as a decrease of
2 log10 CFU/ml between the combination (Q-D and gentamicin) and its most active constituent after 24 h. Antagonism was defined as an increase of
2 log10 CFU/ml between the combination and its most active constituent. The interaction was otherwise considered indifferent.
Experimental staphylococcal endocarditis. New Zealand White rabbits (weight: 2.0 to 2.7 kg) were used in an aortic endocarditis model, as previously described (12). A polyethylene catheter was inserted into the left ventricle via the carotid artery and left in place throughout the experiment. One milliliter of an overnight bacterial culture (diluted in saline serum at 9 g/liter to obtain 106 CFU/ml) was inoculated into the ear vein 48 h after catheterization.
Twenty-four hours after inoculation, the animals were randomly assigned to groups receiving Q-D, gentamicin, Q-D and gentamicin, or no treatment (control group). Antibiotics were administered for 1 or 4 days in order to obtain serum concentrations close to those observed in humans after 1-h intravenous (i.v.) perfusions of 7.5 mg of Q-D/kg of body weight three times a day (t.i.d.) and 3 mg of gentamicin/kg once a day. For this purpose, Q-D was administered by intramuscular injection of 30 mg/kg t.i.d. (14, 15). The total daily dose of gentamicin (16.6 mg/kg) was infused i.v. at changing flow rates controlled by a computer dedicated to human pharmacokinetic simulations (6, 7).
Rabbits were sacrificed by i.v. injection of thiopental 24 h after inoculation in the control group and at the end of the last day of treatment in the other groups. Vegetations were excised, immediately placed on ice, weighed, homogenized in 0.5 ml of saline, and plated on Mueller-Hinton agar after serial dilutions. Viable counts after a 48-h incubation at 37°C were expressed as means ± standard deviations (SD) of log10 CFU per gram of vegetation. The lower limit of detection for this method is 1 CFU per 100 µl of undiluted vegetation homogenate.
To detect in vivo selection of resistant variants during treatment, undiluted vegetation homogenates were also plated on agar containing gentamicin or Q-D at a concentration corresponding to 16 times the MIC.
Determination of antibiotic concentrations in serum and pharmacokinetic analysis. Antibiotic concentrations in serum from three healthy rabbits were determined. Samples were drawn (i) 30 min (peak) and 23.5 h (trough) after the first injection of gentamicin and (ii) 15, 30, 45, 60, 120, 240, and 360 min after the first injection of Q-D. For streptogramin assays, 1-ml blood samples were immediately placed in tubes containing 0.25 ml of 0.25 N hydrochloric acid and then gently stirred and centrifuged (10 min at 1,500 x g). The upper phase was immediately stored at -80°C. Gentamicin concentrations were assessed by immunoenzymatic assays (EMIT; Behring). The lower limit of detection for gentamicin was 0.3 mg/liter.
Quinupristin, dalfopristin, and Q-D concentrations were determined by a microbiological assay based on the bactericidal synergy between quinupristin and dalfopristin. The bactericidal activities of pH-neutralized serum samples against Micrococcus luteus DRUG 320 were assessed and compared with a standard curve constructed on the basis of results for rabbit serum. Dalfopristin was added to sera at a final bacteriostatic concentration of 2 mg/liter for determination of quinupristin activity, and quinupristin was added at a final bacteriostatic concentration of 2 mg/liter for determination of dalfopristin activity. No antibiotic was added for determination of Q-D activity. After a 4-h incubation at 37°C, surviving bacteria were counted by plating 3 µl of diluted samples on Mueller-Hinton agar containing activated charcoal (10 g/liter) to prevent antibiotic carryover. Lower limits of detection were 0.09 mg/liter for quinupristin, 0.06 mg/liter for dalfopristin, and 0.23 mg/liter for Q-D. The coefficient of variation of the assay was 16%.
Pharmacokinetic data were analyzed by regression to determine the terminal slope and ß elimination half-lives. Areas under the curve of serum concentrations were determined by the trapezoidal method, with extrapolation from the last point toward infinity.
Statistics.
Statistical analysis was performed using StatView software (Abacus Concepts). Mean bacterial titers in vegetations were globally compared by analysis of variance (ANOVA). Scheffe's test was used to compare mean titers two by two. A P value
0.05 was considered significant.
|
|
|---|
|
View this table: [in a new window] |
TABLE 2. Agar dilution MICs of quinupristin, dalfopristin, Q-D, and gentamicin for S. aureus strains NA8 and BCB8
|
Checkerboard. Fractional inhibitory concentration indices of Q-D and gentamicin for strains NA8 and BCB8 were, respectively, 1.0 and 1.2. Therefore, the combination of the tested antibiotics was interpreted as indifferent for both strains.
Time-kill curves. Time-kill curves for strains NA8 and BCB8 are shown in Fig. 1. At a concentration of 0.5 mg/liter, Q-D proved more bactericidal against quinupristin-susceptible S. aureus NA8 than against BCB8: the initial inoculum was reduced by 3.0 log CFU/ml for NA8 and by 1.8 log CFU/ml for BCB8 after a 24-h incubation. No antagonism or synergism between Q-D and gentamicin was observed.
![]() View larger version (16K): [in a new window] |
FIG. 1. Time-kill curves for S. aureus NA8 and BCB8. Results are shown for the control group (diamonds) and groups of rabbits given gentamicin at 0.25 mg/liter (squares), Q-D at 0.5 mg/liter (triangles), or gentamicin at 0.25 mg/liter in combination with Q-D at 0.5 mg/liter (circles). Solid lines, strain NA8; dashed lines, strain BCB8.
|
|
View this table: [in a new window] |
TABLE 3. Pharmacokinetic serum parameters of quinupristin, dalfopristin, and Q-D in three healthy rabbits after a single intramuscular injection of Q-D (30 mg/kg)a
|
|
View this table: [in a new window] |
TABLE 4. Activity of drugs in 4-day regimensb
|
|
View this table: [in a new window] |
TABLE 5. Activity of drugs in 1-day regimensc
|
|
|
|---|
According to international guidelines, gentamicin was administered in our study by simulating a human i.v. regimen of a once-daily dose of 3 mg/kg (23). A previous study showed that this regimen was bactericidal against S. aureus NA8 after 48 h (3). Our data indicate that the in vivo activity of gentamicin administered as a once-daily dose was bactericidal against NA8 at the end of the first day of treatment, but required 4 days of treatment to be effective against BCB8. Sande and Courtney showed that gentamicin, in a t.i.d. regimen yielding a mean peak concentration of 5.0 mg/liter, had no bactericidal activity in a rabbit experimental staphylococcal endocarditis model after 3 and 5 days of treatment (24). Nevertheless we cannot exclude the possibility of synergy between Q-D and a multiple-dose regimen of gentamicin.
The combination of Q-D and gentamicin proved no more beneficial than Q-D or gentamicin alone against S. aureus in this study in terms of either in vitro and in vivo bactericidal activity or in vivo selection of a resistant variant. This lack of benefit may have been due to the high potency of the monotherapies, each of which reduced mean bacterial titers in vegetations by
3.5 log CFU/g after a 4-day regimen and selected no resistant variant. However, we have possibly underestimated the selection of such variant clones by using high antibiotic concentrations in solid media (i.e., 16 times the MIC). Furthermore we did not search variant clones showing resistance or increased resistance specifically to quinupristin or to dalfopristin. Such resistant variants were not selected in previous experimental endocarditis studies using S. aureus strains susceptible to dalfopristin but have been isolated when one dalfopristin-resistant strain was used (15, 27). Moreover, the combination of another antibiotic with Q-D might help prevent the emergence of constitutively resistant mutants during treatment of infections caused by S. aureus strains with inducible resistance to MLSB (MLSB-IR). This situation has been a matter of concern for Q-D and for other non-erm-inducing MLS antibiotics, by analogy with clindamycin, which is thought to select MLSB-CR mutants from MLSB-IR strains (18). Because of the low in vitro frequency of mutation from inducible to constitutive MLSB resistance (
10-7 to 10-8), this hypothesis needs to be tested by using experimental models yielding larger inocula than the rabbit endocarditis model (18).
Serum streptogramin concentrations were determined with a new bioassay based on the measurement of the bactericidal activity of the serum, but not its bacteriostatic activity, as in microbiological methods developed previously (13). This new assay uses one unique test strain without a streptogramin resistance mechanism, which prevents the instability due to formerly used strains (13). To measure the concentration of one compound, the other one was added at a high concentration, allowing for control of the synergy between quinupristin and dalfopristin. Conversely, for the global assay of Q-D, concentrations of separate compounds in test sera were not controlled, whereas control sera had a fixed 30:70 (wt/wt) ratio of quinupristin to dalfopristin (Q/D ratio). Although the bacteriostatic activities of Q-D against S. aureus, as assessed by determining MICs, are similar for Q/D ratios ranging from 90:10 to 10:90, no data about the influence of the Q/D ratio on Q-D in vitro bactericidal activity have been published (5). In our work, the sum of activities for separate compounds exceeded the activity from the global assay when the Q/D ratio was less than 0.6 (data not shown). The global Q-D assay should therefore be considered a functional one. Even if its result is expressed as a mass concentration, it yields nearly the amount of Q-D in serum with a 30:70 (wt/wt) Q/D ratio which has the same bactericidal activity as the test serum.
In conclusion, our experiments showed no additive benefit from the combination of Q-D and gentamicin against dalfopristin- and gentamicin-susceptible S. aureus, whether susceptible or resistant to quinupristin. Contrary to previous studies, Q-D showed in vivo bactericidal activity against the quinupristin-resistant strain. Further investigations are needed to determine the role of Q-D, with regard to vancomycin, in the management of MRSA infections.
Eric Batard was the recipient of a grant from the Direction Régionale des Affaires Sanitaires et Sociales des Pays de la Loire (France). This work was supported by Aventis Laboratories, France.
|
|
|---|
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»