Macrolide-Lincosamide-Streptogramin Resistance Patterns in

ANTIMICROBIAL AGENTS AND CHEMOTHERAPY, Feb. 1981, p. 274-278 0066-4804/81/020274-05$02.00/0 Vol. 19, No. 2 Macrolide-Lincosamide-Streptogramin Resistance Patterns in Clostridium perfringens from Animals GUNINDRA N. DUTTA AND LUC A. DEVRIESE* Faculty of Veterinary Medicine, State University of Ghent, B-9000 Ghent, Belgium Different patterns of resistance against commonly used macrolide, lincosamide, and streptogramin antibiotics were found in Clostridium perfringens of animal origin. The patterns were designated as (i) macrolide-lincosamide-streptogramin group B generalized resistance, (ii) macrolide-lincosamide generalized resistance, (iii) macrolide-lincosamide inducible resistance, and (iv) macrolide-lincosamidestreptogramin low-level generalized resistance. The strains of the fourth pattern were able to inactivate pristinamycin and virginiamycin. The macrolide-susceptible strains showed a bimodal distribution of lincomycin and clindamycin susceptibility levels. The susceptible strains were inhibited by 0.25 Mg of lincomycin per ml and 0.03 Mig of clindamycin per ml. The low-level resistant strains were inhibited at concentrations of 2 to 4 Mg of lincomycin per ml and 0.5 to 2 Mg of cindamycin per ml. Two patterns of macrolide resistance, constitutive resistance and inducible or dissociated resistance, are known to occur in staphylococci (3, 7) and streptococci (9). These resistances are also involved with the lincosamide and streptogramin group B antibiotics (1, 2, 21). Another pattern of streptogramin resistance in Staphylococcus aureus, in which both A and B groups of the compound are involved, has been reported (4). The resistant strain was able to inactivate pristinamycin, a member of the streptogramin class of antibiotics. Resistance of Clostridium perfringens to erythromycin, lincomycin, or clindamycin has been reported (13, 15, 24). The patterns of macrolide-lincosamide-streptogramin (MLS) resistance in C. perfringens have not yet been described. MATERLALS AND METHODS Cultures. Ninety-four strains of C. perfringens, isolated during the period April 1979 to December 1979 from ceca of pigs (58 strains) and cattle (36 strains) brought for autopsy in the Veterinary Faculty, State University of Ghent, were used. Antibacterial compounds. The antibacterial compounds, as laboratory standard powder, were obtained as follows: clindamycin hydrochloride and lincomycin hydrochloride from The Upjohn Co., Kalamazoo, Mich.; erythromycin base from Abbott Laboratories, Brussels, Belgium; pristinamycin and spiramycin adipate from SPECIA, Paris, France; oleandomycin phosphate from Pfizer Co., Brussels, Belgium; tylosin base from Eli Lilly & Co., Indianapolis, Ind.; virginiamycin and the virginiamycin components M and S from Smith Kline & French Laboratories, Genval, Belgium. Clindamycin, lincomycin, oleandomycin, and spira- 274 mycin were dissolved and diluted in distilled water. Tylosin was first dissolved in 1 ml of ethanol, and then distilled water was added to give a stock solution of 1,000,Ag/ml. Further dilutions were made in distilled water. Erythromycin, pristinamycin, virginiamycin, and virginiamycin components M and S were dissolved in ethanol to give a solution of 2,000,ug/ml, and then an equal amount of distilled water was added to make a stock solution of 1,000,ug/ml. Further dilutions were made in distilled water. Antibiotic dilutions were incorporated to agar as suggested by Ericsson and Sherris (6). Susceptibility testing paper disks containing clindamycin (10 Mg) and lincomycin (10 Mg) were obtained from The Upjohn Co.; erythromycin (15 MAg) was obtained from Oxoid Ltd., Basingstoke, England. Paper disks containing oleandomycin (15 jig), spiramycin (30,Lg), virginiamycin component M (15 Lg), and virginiamycin component S (15,g) were prepared in the laboratory in sterile 6-mm antibiotic assay disks (Whatman Ltd., England) according to McDonald and Biberstein (12). Testing media. Minimal inhibitory concentration (MIC) tests, agar diffusion tests, and Gots tests were carried out in Wilkins-Chalgren agar (23) with hemin and vitamin K omitted. The medium was prepared from individual ingredients. The antibiotic-containing agar plates for MIC tests except those containing pristinamycin, virginiamycin, and virginiamycin components M and S were prepared in one lot and kept in a refrigerator for 1 to 10 days. The plates were dried at room temperature for 25 to 30 min in a laminar air flow after pouring. Because the activity of pristinamycin, virginiamycin, and virginiamycin components M and S deteriorated rapidly on storage, plates containing these antibiotics were prepared on the same day the tests were performed. Susceptibility tests. Twenty strains were tested in one batch. Three to four colonies from cultured VL blood agar plates (Gelose Viande-Levure medium

VOL. 19, 1981 [Pasteur Institute, Lille, France] supplemented with 5% sheep blood) were inoculated into Wilkins-Chalgren broth (hemin and vitamin K omitted) and incubated overnight anaerobically in GasPak jars at 37 C. The turbidity of the inocula was adjusted in 1 ml of clear Wilkins-Chalgren broth by adding the required drops of cultured broth to match the turbidity of onehalf the no. 1 McFarland standard. Inocula were applied to the antibiotic-containing plates with an inoculum replicator (18) and incubated anaerobically in an H2 and CO2 atmosphere at 37 C in GasPak jars. At the beginning and end of each series of tests, two plates of the same medium without antibiotics were inoculated. The plate inoculated at the beginning was incubated with the inoculated plates to serve as growth control; the other was incubated aerobically to detect possible aerobic contamination. No longer than 1 h elapsed between the preparation of inoculum and the placement of the inoculated plates in anaerobic jars. When testing the subsequent batches, one strain tested in the first batch was included in each batch as a control strain to see if there was loss of antibiotic activity in the refrigerated plates. After 24 h of incubation, the MIC of each strain was recorded as the lowest concentration of antibiotic yielding no growth, one discrete colony, or a fine barely visible haze as determined with the unaided eye (19). Agar diffusion tests. Seventeen strains (12 from pigs and 5 from cattle), apparently resistant to one or another of the tested compounds, were further tested with the agar diffusion tests described by Weisblum and Demohn (21) to investigate the nature of the resistance. The inocula were prepared in 2 ml of phosphate-buffered saline by adding the required drops of cultured Wilkins-Chalgren broth to match the turbidity of the no. 1 McFarland standard. Inocula were applied with cotton swabs to the agar plates containing Wilkins-Chalgren agar. Susceptibility disks were placed on the inoculated plates about 1 cm apart and incubated anaerobically in GasPak jars at 37 C for 24 h. Resistance was noted as inducible when a decreased radius of inhibition zone was observed on the side of the disk proximal to another disk (21). Gots tests. The resistant strains were also tested with the Gots tests (8) to determine whether they were able to inactivate the antibiotics. S. aureus ATCC 25923, with MICs of 0.12 yg/ml, 0.25,Lg/ml, 0.12 Ag/ml, 4,ug/ml, and 10,ug/ml against erythromycin, lincomycin, pristinamycin, virginiamycin component M, and virginiamycin component S, respectively, was used to seed the antibiotic-containing agar. Wilkins-Chalgren agar was melted and then cooled to about 45 C. Suspension of a 20-h agar culture of S. aureus was made in phosphate-buffered saline to match the turbidity of the no. 2 McFarland standard. One milliliter of the suspension was added to 100 ml of melted agar, mixed thoroughly, and poured in petri plates containing measured quantities of antibiotic dilutions to make a final concentration of 0.25, 0.5, and 1.tg of erythromycin per ml, 0.5, 1, and 2 ug of lincomycin per ml, 0.25, 0.5, and 1,ug of pristinamycin per ml, 5, 7.5, and 10 jig of virginiamycin component M per ml, and 15, 20, and 30,ug of virginiamycin component S per ml in each series of plates. The plates were dried at room temperature in a laminar air flow MLS RESISTANCE PATTERNS IN C. PERFRINGENS 275 for 20 min and were marked off into quadrants. Each strain from a freshly grown 20-h culture in VL blood agar plate was streaked onto one of the quadrants in such a manner as to form a circle or ellipse. One strain, susceptible to all antibiotics, was inoculated in the same way to serve as control. After 48 h of incubation anaerobically in GasPak jars at 37 C, the results were noted. A zone of satellite colonies of S. aureus in the agar under the surface growth extending both toward the center of the bacterial surface growth and around its periphery was recorded as positive, indicating inactivation of the antibiotics by the tested strains. Absence of growth of S. aureus was recorded as negative. Tests to assess degradation of antibiotics in broth. The standard curve for virginiamycin component M was obtained by carrying out the agar diffusion assay method as described by Sabath and Matsen (14), using Micrococcus luteus ATCC 9341. Two strains out of the three that showed the evidence of inactivation of antibiotics in the Gots tests were tested. The strains were inoculated in 10 ml of Wilkins-Chalgren broth containing 30,ug of virginiamycin component M per ml and incubated anaerobically. An uninoculated control tube containing the same amounts of media and antibiotics was also incubated with the inoculated tubes. After 16 and 24 h of incubation, 1 ml of medium from each tube was pipetted out, mixed with 2 ml of acetone, and kept at -20 C in airtight containers. After overnight preservation, the potency of antibiotic in the media was measured in M. luteus-seeded DST agar, using 0.02-ml-loaded 6-mm paper disks. RESULTS MIC tests. Results of the MIC tests showed that the activity of pristinamycin and virginiamycin in plates deteriorated rapidly on storage. The susceptibility of the control strain decreased at least 32 times when tested in plates containing pristinamycin and virginiamycin refrigerated for 2 days. Therefore, the tests with pristinamycin and virginiamycin were repeated and carried out along with the tests with virginiamycin components M and S in plates prepared on the same day the tests were performed. No loss of antimicrobial activity was detected when macrolide and lincosamide antibiotics were incorporated into the medium and stored in refrigerator for up to 10 days. The MICs of the tested antibiotics against 17 resistant strains and the susceptible strains are shown in Table 1. The strains were classified as susceptible, as low-level resistant or intermediate, or as resistant to the tested antibacterial agents according to MIC by the agar dilution method. The standard of classification was the relationship of the susceptibility of a strain to that of others of the same species (6). The MICs of lincomycin and clindamycin against the macrolide-susceptible strains of C. perfringens are shown in Table 2. Agar diffusion tests. Eight strains from pigs revealed the inducible nature of resistance. Anti-

276 DUTTA AND DEVRIESE TABLE 1. Patterns of resistance and MICs of C. perfringens strains of animal origin" MIC (jig/ml) Pattern of resistance Strain Macrolide Lincosamide Streptogramin ERY OLN SPI TLN LNC CLN PTN VIR VIR-M VIR-S MLS (B) generalized P-1-512 -512-512 -512 256 32 1 2 4 64 resistance P-2-512 '512-512 -512 256 32 1 2 4 64 C-1-512 -512-512 -512 256 32 1 2 4 64 ML generalized P-3 256-512 128 16-512 128 0.12 0.25 2 8 resistance P-4 256-512 128 16 '512 128 0.12 0.25 2 8 C-2 256-512 256 16.512 128 0.12 0.5 2 8 ML inducible resistance P-5 128h.512 64" 8h 256 32b 0.12 0.25 2 8 P-6 256-512 128b 8b 128h 32" 0.12 0.5 2 8 P-7 256-512 128h 41 128b 32" 0.12 0.5 2 8 P-8 256.512 128h 8h 256 64" 0.12 0.5 4 8 P-9 256-512 64b 4h 256 64" 0.12 0.5 4 8 P-10 256-512 64h 4b 16" 8h 0.12 0.5 4 8 P-11 256-512 64" 4" 256 32" 0.12 0.25 2 8 P-12 256-512 128" 4b 256 64" 0.12 0.25 2 8 MLS low-level generalized C-3 16 32 32 4 16 8 4 8 64 32 resistance C-4 16 32 32 2 16 8 4 8 64 16 C-5 16 32 16 2 16 8 4 8 64 16 Susceptible strains (46 from 0.5-2 4-16 4-8 0.12- -' -' 0.06-0.25-0.5-4 2-8 pigs and 31 from cattle) 0.5 0.12 0.5 " Abbreviations: ERY, erythromycin; OLN, oleandomycin; SPI, spiramycin; TLN, tylosin; LNC, lincomycin; CLN, clindamycin; PTN, pristinamycin; VIR, virginiamycin; VIR-M, virginiamycin component M; VIR-S, virginiamycin component S; MLS(B), macrolide, lincosamide, and streptogramin (group B); P, originating from pigs; C, originating from cattle. " Zone of induction was observed in agar diffusion tests. 'See Table 2. TABLE 2. MICs of lincomycin and clindamycin against 77 (46 from pigs and 31 from cattle) macrolidesusceptible strains of C. perfringens Antibacterial Suc % of strains with MIC (tg/ml) of: Source agent <0.03 0.06 0.12 0.25 0.5 1 2 4 Lincomycin Pigs 23 2 50 25 Cattle 41 10 41 8 Clindamycin Pigs 25 21 54 Cattle 51 10 31 8 biotics against which induction zones were observed are shown in Table 1. In all strains, erythromycin and oleandomycin acted as inducers. The patterns of resistance on the basis of MIC tests and agar diffusion tests are also shown in Table 1. Gots tests. Three cattle strains, belonging to the MLS low-level generalized resistance pattern (Table 1), were able to inactivate virginiamycin components M and S and pristinamycin. The inactivation was visible at all virginiamycin component M concentrations tested. Growth of the three strains was not optimal on plates containing virginiamycin component S, but the strains could inactivate the compound up to a concentration of 30,ug/ml. With pristinamycin, inactivation was marked up to a concentration ANTIMICROB. AGENTS CHEMOTHER. of 0.5 ug/ml. Although the strains were able to grow, they could not inactivate pristinamycin at a concentration of 1,ug/ml. Results with erythromycin and lincomycin were negative. Degradation of virginiamycin component M in broth. The tested strains were able to degrade virginiamycin component M in broth. After 16 h of incubation, the antibiotic concentration in the medium diminished from 10,ug/ ml (after addition of acetone) to below the detection limit (1.2,g/ml). The antibiotic activity on the uninoculated control tube decreased to 5,ug/ml in the same time period. DISCUSSION Four patterns of resistance were found on the basis of MIC tests and agar diffusion tests. The

VOL. 19, 1981 first pattern is designated MLS(B) generalized resistance to indicate that the pattern involves resistance against macrolide, lincosamide, and streptogramin group B antibiotics. This resistance is phenotypically very similar to the constitutive MLS resistance in staphylococci. The MICs of pristinamycin and virginiamycin against these strains were slightly higher than against the susceptible strains. But they were susceptible to virginiamycin component M, which belongs to the streptogramin group A antibiotics. The second pattem is called ML generalized resistance. Strains of this pattern.were susceptible to streptogramins. The third pattem of resistance, ML inducible resistance, was found in strains from pigs only. Erythromycin and oleandomycin induced resistance to other tested macrolides and lincosamides. This inducible resistance in C. perfringens differs from inducible resistance in S. aureus, where erythromycin and oleandomycin induce resistance to streptogramin group B antibiotics also. We did not test whether erythromycin could induce resistance to itself as reported by Weaver and Pattee (20) in S. aureus. The fourth pattern of resistance, designated MLS low-level generalized resistance, was found in strains from cattle only. The MICs of the tested macrolides and lincosamides against these strains were not as high as those described earlier. The strains were resistant to both groups of streptogramin. They were able to inactivate pristinamycin and virginiamycin components M and S in Gots tests. Pristinamycin was inactivated up to a concentration of 0.5,ug/ml. Although the strains were able to grow at a concentration of 1,ug/ml, the inactivation was not sufficient to allow the seed strain, S. aureus ATCC 25923, with a pristinamycin MIC of 0.12,tg/ml, to grow. The two strains tested in broth were also able to inactivate virginiamycin component M in 16 h. This drug inactivation capability of the strains suggests an enzymatic mechanism of resistance. An acetyl-transferase and a hydrolase capable of inactivating pristinamycin 2A (11) and pristinamycin 1A (10), respectively, have been described in a strain of S. aureus. An unusual trimodal distribution of lincomycin and clindamycin susceptibility levels was found. The strains investigated in this study can be classified into three groups: (i) a susceptible group, (ii) a low-level resistant or intermediate group, and (iii) a resistant group which includes the macrolide and streptogramin cross-resistant strains. Strains of the first two groups were susceptible to the macrolide antibiotics. A similar trimodal distribution of lincomycin susceptibility levels has been reported in C. perfringens MLS RESISTANCE PATTERNS IN C. PERFRINGENS 277 strains from humans (15). It is remarkable that similar percentages of susceptible, low-level resistaint or intermediate, and resistant strains against lincosamides occur in host species as different as humans, pigs, and cattle. The three classes of antimicrobial agents investigated in this study are inhibitors of the 50S ribosomal subunit (22). Toxigenic types of C. perfringens have been shown to carry plasmid ribonucleic acid (5). Sebald and co-workers (16) reported the isolation of a resistant strain of C. perfringens where the resistance against erythromycin and clindamycin appeared to be coded for by a plasmid, pip402. This R plasmid was not transferable to other C. perfringens strains (17). The significance of the resistance patterns reported here is a inatter of speculation. Some of the observed resistance patterns appear to be similar to those found in other bacterial species, but their relationship is unknown. It is also not known whether animal C. perfringens frequently colonize humans. All strains studied here were susceptible to penicillins (G. N. Dutta and L. A. Devriese, J. Vet. Pharmacol. Ther., in press), the antibiotics of choice in the treatment of clostridial infections. When C. perfringens strains with the resistance patterns reported here cause disease in animals, it is unlikely that they will respond to treatment with the MLS antibiotics. LITERATURE CITED 1. Barber, M., and P. M. Waterworth. 1964. Antibacterial activity of lincomycin and pristinamycin: a comparison with erythromycin. Br. Med. J. 2:603-606. 2. Bourse, R., and J. Monier. 1967. Effect de l'erythromycine sur la croisance de Staphylococcus aureus "resistant dissocie" en bacteriostase par un autre macrolide ou un antibiotique apparente. Ann. Inst. Pasteur Paris 110:67-79. 3. Chabbert, Y. 1956. Antagonisme in vitro entre l'erythromycine et la spiramycine. Ann. Inst. Pasteur Paris 90:787-790. 4. Dublanchet, A., C. J. Soussy, F. Squinazi, and J. Duval. 1977. Resistance de Staphylococcus aureus aux streptogramines. Ann. Microbiol. (Paris) 128A:277-287. 5. Duncan, C. L., C. M. Christenson, and J. I. Rood. 1978. Multiple plasmids in different toxigenic types of Clostridium perfringens: possible control of beta-toxin production, p. 246-248. In D. Schlessinger (ed.), Microbiology-1978. American Society for Microbiology, Washington, D.C. 6. Ericsson, H. M., and J. C. Sherris. 1971. Antibiotic sensitivity testing. Report of an international collaborative study. Acta. Pathol. Microbiol. Scand. Sect. B (Suppl. 217). 7. Garrod, L. P. 1957. The erythromycin group of antibiotics. Br. Med. J. 2:57-63. 8. Gots, J. S. 1945. The detection of penicillinase-producing properties of microorganisms. Science 102:309. 9. Hyder, S. L., and M. M. Streitfeld. 1973. Inducible and

278 DUTTA AND DEVRIESE constitutive resistance to macrolide antibiotics and lincomycin in clinically isolated strains of Streptococcus pyrogenes. Antimicrob. Agents Chemother. 4:327-331. 10. Le Goffic, F., M. L. Capmau, J. Abbe, C. Cerceau, A. Dublanchet, and J. Duval. 1977. Plasmid mediated pristinamycin resistance: PH IA, a pristinamycin 1A hydrolase. Ann. Microbiol. (Paris) 128B:471-474. 11. Le Goffic, F., M. L. Capmau, D. Bonnet, C. Cerceau, C. J. Soussy, A. Dublanchet, and J. Duval. 1977. Plasmid-mediated pristinamycin resistance PAC IIA: a new enzyme which modifies pristinamycin IIA. J. Antibiot. 30:665-669. 12. McDonald, M. C., and E. L. Biberstein. 1974. Determination of bacterial susceptibility to tylosin by singledisk agar diffusion tests. Am. J. Vet. Res. 35:1563-1565. 13. Rood, J. I., E. A. Maher, E. B. Somers, E. Campos, and C. L. Duncan. 1978. Isolation and characterization of multiply antibiotic-resistant Clostridiumperfringens strains from porcine feces. Antimicrob. Agents Chemother. 13:871-880. 14. Sabath, L. D., and M. Matsen. 1974. Assay of antimicrobial agents, p. 428-430. In E. H. Lennette, E. H. Spaulding, and J. P. Truant (ed.), Manual of clinical microbiology, 2nd ed. 1974. American Society for Microbiology, Washington, D.C. 15. Sapico, F. L., Y. Kwok, V. L. Sutter, and S. M. Finegold. 1972. Standardized antimicrobial disc susceptibility testing of anaerobic bacteria: in vitro susceptibility of Clostridium perfringens to nine antibiotics. Antimicrob. Agents Chemother. 2:320-325. 16. Sebald, M., D. Bouanchaud, and G. Beith. 1975. Nature ANTIMICROB. AGENTS CHEMOTHER. plasmidique de la resistance a plusiers antibiotiques chez C. perfringens type A souche 659. C. R. Acad. Sci. Paris Ser. D 280:2401-2404. 17. Sebald, M., and G. Brefort. 1975. Transfert du plasmide tetracycline-chloramphenicol chez Clostridiumperfringens. C. R. Acad. Sci. Paris Ser. D 281:317-319. 18. Steers, E., E. L. Foltz, and B. S. Graves. 1959. An inocula replicating apparatus for routine testing of bacterial susceptibility to antibiotics. Antibiot. Chemother. 9:307-311. 19. Sutter, V. L., A. L. Barry, T. D. Wilkins, and R. J. Zabransky. 1979. Collaborative evaluation of a proposed reference dilution method of susceptibility testing of anaerobic bacteria. Antimicrob. Agents Chemother. 16:495-502. 20. Weaver, J. R., and P. A. Pattee. 1964. Inducible resistance to erythromycin in Staphylococcus aureus. J. Bacteriol. 88:574-580. 21. Weisblum, B., and V. Demohn. 1969. Erythromycininducible resistance in Staphylococcus aureus: survey of antibiotic classes involved. J. Bacteriol. 98:447-452. 22. Weisblum, B., C. Siddhikol, C. J. Lai, and V. Demohn. 1971. Erythromycin-inducible resistance in Staphylococcus aureus: requirements for induction. J. Bacteriol. 106:835-847. 23. Wilkins, T. D., and S. Chalgren. 1976. Medium for use in antibiotic susceptibility testing of anaerobic bacteria. Antimicrob. Agents Chemother. 10:926-928. 24. Wilkins, T. D., and T. Thiel. 1973. Resistance of some species of Clostridium to clindamycin. Antimicrob. Agents Chemother. 3:136-137. Downloaded from http://aac.asm.org/ on July 7, 2018 by guest