In Vitro Activities of Daptomycin, Vancomycin, and Penicillin against Clostridium difficile, C. perfringens, Finegoldia magna, and Propionibacterium acnes

Daptomycin has in vitro activity against gram-positive anaerobicbacteria, although limited numbers of species have been tested.We studied the in vitro activities of daptomycin, vancomycin,and penicillin against more than 100 strains each of Clostridiumdifficile, C. perfringens, Finegoldia magna, and Propionibacteriumacnes. Daptomycin Etest MICs and results from time-kill studieswere determined for selected strains. For 392 of 421 strains(93%), daptomycin was inhibitory at $≤$ 1 µg/ml, including15 of 16 strains of C. difficile with elevated linezolid MICsof 8 and 16 µg/ml, all 32 strains with moxifloxacin MICsof $≥$ 4 µg/ml, and all 16 strains resistant to clindamycin.Daptomycin MICs were also $≤$ 1 µg/ml for all 16 F. magnastrains resistant to clindamycin and all 32 strains resistantto tetracycline. Only one strain, a C. perfringens strain, hada MIC of >2 µg/ml to daptomycin. Eighty-five and 92.5%of the Etest MICs were within 1 dilution of the agar dilutionmethod for all drugs at 24 and 48 h, respectively. In time-killstudies, a C. difficile strain was inhibited by both daptomycinand vancomycin at 1, 2, 4, 8, and 24 h; colony counts were decreasedby 2.3 to 2.9 log at 24 h. Vancomycin was not bactericidal forC. perfringens; however, daptomycin showed bactericidal activityas early as 1 h at four and eight times the MIC and at 2 and4 h at two and four times the MIC.

INTRODUCTION

Top

Introduction

Daptomycin is a cyclic lipopeptide antibiotic produced by Streptomycesroseosporus with bactericidal activity against resistant aerobicgram-positive organisms, such as methicillin-resistant Staphylococcusaureus, community-acquired methicillin-resistant Staphylococcusaureus, vancomycin-resistant enterococci, and penicillin-resistantpneumococci (3, 8), including multidrug-resistant strains (18).

It has been found to be safe and effective in complicated skinand skin structure infections (2, 21) and does not disrupt theanaerobic microflora or promote colonization by vancomycin-resistantenterococci or extended-spectrum-ß-lactamase-producingKlebsiella pneumoniae in mice (17). While antibiotic resistanceis an increasing worldwide problem, daptomycin has shown lowspontaneous resistance rates in vitro (19). Resistant strainshave rarely been reported thus far (13, 15, 16, 20); however,in one instance (13), daptomycin at a high concentration retainedbactericidal activity against resistant isolates of S. aureus.

Daptomycin has been shown to have in vitro activity againsta variety of anaerobic gram-positive bacterial pathogens, suchas Clostridium perfringens, Finegoldia magna, and Propionibacteriumacnes, that may be present in polymicrobial infections (4, 9-11),although limited numbers of each species have been tested previously.To explore its in vitro efficacy further, we tested more than100 strains each of Clostridium difficile, C. perfringens, Finegoldiamagna, and Propionibacterium acnes and compared the daptomycinMICs with those of vancomycin and penicillin. Additionally,daptomycin Etest MICs were determined for selected strains ineach group and time-kill studies with daptomycin and vancomycinwere performed on one strain of C. difficile and two strainsof C. perfringens.

MATERIALS AND METHODS

Top

Materials and Methods

The in vitro activities of daptomycin, vancomycin, and penicillinagainst more than 100 strains each of Clostridium difficile,C. perfringens, Finegoldia magna, and Propionibacterium acneswere determined. The strains selected included F. magna isolatesknown to be resistant to clindamycin and tetracycline and C.difficile isolates resistant to linezolid, clindamycin, andfluoroquinolones (5). Control strains included Staphylococcusaureus ATCC 29213, Eubacterium lentum ATCC 40335, and Bacteroidesfragilis ATCC 25285.

Susceptibility testing was performed according to the referenceagar dilution method previously described in the CLSI M11-A6document (7). Antimicrobial agents obtained were daptomycin(Cubist Pharmaceuticals, Inc., Lexington, MA), penicillin (SigmaChemical Co., St. Louis, MO), and vancomycin (Eli Lilly &Co., Indianapolis, IN). The antimicrobial agents were reconstitutedaccording to the manufacturers' instructions, serially diluted,and added to molten-supplemented brucella agar. For daptomycinplates, the calcium (Ca²⁺) concentration in brucella agar wasdetermined prior to testing using a Ca²⁺ electrode and preparedagar was supplemented to a final concentration of 50 mg/literof Ca²⁺. Plates were inoculated on the day of preparation. Drug-freeplates were included as growth controls.

The isolates were obtained from human clinical specimens, includingblood, skin and skin structure, intraabdominal, and feces, identifiedby standard methods (14) and maintained in skim milk at –70°C.The strains were taken from the freezer and transferred at leasttwice on brucella agar (Anaerobe Systems, Morgan Hill, CA) supplementedwith hemin, vitamin K₁, and 5% sheep blood (brucella blood agar[BBA]) to ensure purity and good growth. Colonies were suspendedin brucella broth to a density equal to a 0.5 McFarland standard.The suspensions were applied to the antibiotic plates with aSteers replicator that delivered a final inoculum of approximately10⁵ CFU/spot. The plates were incubated in an anaerobic chamberat 36°C for 44 to 48 h. The MIC was defined as the concentrationof the drug that completely inhibited growth or caused a markedreduction in the appearance of growth compared to a drug-freegrowth control.

The Etest was performed by suspending strains in brucella brothto a turbidity of a 0.5 to 1 McFarland standard, inoculatingthe suspension onto BBA according to the manufacturer's directions,and then applying the Etest strips containing a gradient ofdaptomycin plus calcium. The plates were incubated in the anaerobicchamber as described above and examined after 24 and 44 h. TheMIC was recorded as the point where the ellipse intersectedthe interpretive scale on the strip.

Time-kill studies were performed in reduced brucella broth supplementedwith vitamin K₁ and hemin. Drug concentrations used were two,four, and eight times the MIC as previously determined by agardilution; strains with higher MICs for daptomycin were selected.Drugs and test strains were added to tubes, which were thencapped with Hungate-type caps, in the anaerobic chamber andincubated in an ambient air incubator. Aliquots were removedwith a tuberculin syringe at 0, 1, 2, 4, 8, and 24 h and platedusing an automated spiral plater (Spiral Biotech, Inc., Norwood,MA) onto prereduced, anaerobically sterilized 150-mm-diameterBBA plates (Anaerobe Systems). Plates were incubated for 24h for Clostridia and 48 h for the other strains; colonies werequantitated according to the instrument's usage guidelines.Bactericidal activity was defined as a minimum 3-log reductionin bacterial counts at 24 h.

RESULTS AND DISCUSSION

Top

Results and Discussion

The agar dilution MICs are shown in Table 1. Although interpretivedaptomycin breakpoints for anaerobes are not yet available,most strains were inhibited by $≤$ 1 µg/ml and MICs were generallythe same or 1 dilution higher than vancomycin MICs. A few strains(14 of 102 C. difficile, 11 of 101 C. perfringens, 2 of 117P. acnes, and 3 of 101 F. magna strains) required 2 µg/mlfor inhibition. Only one strain, a C. perfringens strain, hada daptomycin MIC of >2 µg/ml, 8 µg/ml by agardilution (tested twice) and 6 µg/ml by Etest; however,this strain also had a linezolid MIC of 4 µg/ml and aquinupristin-dalfopristin MIC of 4 µg/ml (determined previously).The daptomycin MIC was $≤$ 1 µg/ml for 15 of 16 strains ofC. difficile that were previously determined to have elevatedlinezolid MICs of 8 and 16 µg/ml, all 32 strains thathad moxifloxacin MICs of $≥$ 4 µg/ml, and all 16 strains thatwere resistant to clindamycin. The daptomycin MIC was also $≤$ 1µg/ml for all 16 F. magna strains resistant to clindamycinand all 32 strains resistant to tetracycline.

View this table:
[in this window]
[in a new window]
TABLE 1. Agar dilution MICs of daptomycin, vancomycin, and penicillin against C. difficile, C. perfringens, P. acnes, and F. magna

The Etest and agar dilution comparisons for daptomycin are shownin Table 2. Eighty-five and 92.5% of the MICs were within 1dilution of the agar dilution reference method for all drugsat 24 and 48 h, respectively. C. perfringens was the only organismthat did not fall within 1 dilution of the agar dilution method,with three (30%) of the strains reading 2 dilutions higher byEtest than the reference method; this was ascribed to the spreadingnature of these strains.

View this table:
[in this window]
[in a new window]
TABLE 2. Comparison of Etest with agar dilution daptomycin MICs after 24 and 48 h of incubation

The time-kill assays for one strain of C. difficile and twostrains of C. perfringens are shown in Fig. 1. MICs for daptomycinwere 2 µg/ml for C. difficile RMA 9396 (Fig. 1, top) aswell as C. perfringens RMA 10436 (Fig. 1, middle) and 8 µg/mlfor C. perfringens RMA 10649 (Fig. 1, bottom); MICs for vancomycinwere 0.5 µg/ml for all three strains. Test concentrationsrepresent two, four, and eight times the MIC. The detectionlevel was $≥$ 20 CFU/ml. Although the C. difficile strain was inhibitedsimilarly by daptomycin and vancomycin at each time tested,neither drug was fully bactericidal at 24 h (colony counts weredecreased by 2.3 to 2.9 log). C. perfringens RMA 10436 was onlyslightly inhibited by vancomycin (a <1-log reduction fromthe control at 1, 2, 4, and 8 h) and was not bactericidal; however,daptomycin showed bactericidal activity as early as 1 h at fourand eight times the MIC and at 4 h for all concentrations. Vancomycinwas also not bactericidal for C. perfringens RMA 10649 (a <1-logreduction from the control at 1, 2, 4, 8, and 24 h), while daptomycinwas bactericidal at 1 h at eight times the MIC and at 2 h forall concentrations.

View larger version (26K):
[in this window]
[in a new window]
FIG. 1. Time-kill assays. MICs for daptomycin (D) were 2 µg/ml for C. difficile RMA 9396 (top) and C. perfringens RMA 10436 (middle) and 8 µg/ml for C. perfringens RMA 10649 (bottom). MICs for vancomycin (V) were 0.5 µg/ml for all three strains. Test concentrations for daptomycin and vancomycin were two, four, and eight times the MIC. The detection level is $≥$ 20 CFU/ml.

Conclusion. Daptomycin showed excellent in vitro activity, with most MICsat or below 1 µg/ml, against a large sampling of a varietyof anaerobic gram-positive bacteria that are frequently isolatedfrom clinical infections. Although there are no clinical datato support a breakpoint for anaerobes, $≤$ 1 µg/ml is at orbelow the CLSI breakpoints for Staphylococcus aureus and ß-streptococci( $≤$ 1 µg/ml) as well as vancomycin-sensitive Enterococcusfaecalis ( $≤$ 4 µg/ml) (6). Daptomycin is reversibly boundto human plasma proteins, primarily serum albumin protein, ina concentration-independent manner. The mean serum protein bindingof daptomycin is approximately 92% at a dosage of 4 or 6 mg/kgof body weight (manufacturer's data), but studies have shownthat serum increases the MIC by approximately twofold (12, 22).The Etest showed good correlation with the reference agar dilutionmethod when incubated for 48 versus 24 h (92.5 versus 85%).The time-kill studies demonstrate daptomycin activity that isboth concentration related and functional in an anaerobic environment.Although daptomycin was not fully bactericidal (colony countsshowed a 2.6- to 2.9-log reduction) against the one strain ofC. difficile tested, it showed good bactericidal activity (a>3-log reduction) against the two strains of C. perfringenstested. While a MIC of 8 µg/ml for the one strain of C.perfringens indicates in vitro resistance, it is difficult topredict clinical efficacy; however, the package insert providesmaximum concentration of drug in serum values of 57.8, 98.6,and 133 µg/ml at dosages of 4, 6, and 8 mg/kg, which areabove four and eight times the MIC for this strain. One studywith vancomycin-sensitive enterococci shows that MICs of 2,4, and 8 µg/ml are treatable in a mouse model (1); however,clinical data must be used to assess in vivo efficacies fororganisms with MICs greater than the established breakpoint.

ACKNOWLEDGMENTS

This study was supported by a grant from Cubist Pharmaceuticals,Inc., Lexington, MA.

FOOTNOTES

* Corresponding author: Mailing address: R. M. Alden Research Laboratory, 2001 Santa Monica Blvd., Suite 685W, Santa Monica, CA 90404. Phone: (310) 453-7820. Fax: (310) 453-7670. E-mail: k.l.tyrrell{at}verizon.net.

REFERENCES

Top