Abstract
Coryneform bacteria are frequently isolated from bovine mastitis and are associated with economic losses. Generally, the MICs of the 15 antimicrobial agents tested at which 90% of the isolates tested are inhibited for 46 Corynebacterium bovis and 13 Corynebacterium amylocolatum strains were low. These are the first quantitative antimicrobial susceptibility data available for coryneforms from bovine mastitis. Data from this study suggest that comparable corynebacteria from humans have a much higher level of antimicrobial resistance to a variety of antimicrobial agents.
Coryneform bacteria are frequently isolated from bovine mastitis (2, 6, 7). In a previous study, the lipophilic species, Corynebacterium bovis and Corynebacterium amylocolatum were the most frequently isolated coryneform organisms (10). Antibiotic therapy is an essential component of mastitis control programs, and the accurate selection of the most effective antimicrobial agents depends upon the susceptibility of the organism to the agent. The purpose of this study was to determine the susceptibility of C. bovis and other coryneforms isolated from bovine mastitis to various antimicrobial agents.
Forty-six strains of C. bovis and 13 strains of C. amylocolatum from a previous study were used (10). Prior to the MIC determinations, all isolates were revived by subculture on Trypticase soy agar (Becton Dickinson Microbiology Systems, Cockeysville, Md.) supplemented with 5% sheep blood and 1.0% Tween 80 and incubated for 24 h at 35 to 37°C. MICs were determined using a broth microdilution method (Sensititre, Westlake, Ohio), except that the Mueller-Hinton broth was supplemented with 1% Tween 80 (6, 10). The MIC panel contained the following antimicrobial agents: ampicillin, oxacillin, cephalothin, ceftiofur, penicillin plus novobiocin, erythromycin, clindamycin, pirlimycin, tilmicosin, florfenicol, tetracycline, enrofloxacin, sarafloxacin, danofloxacin, and premafloxacin. The NCCLS-recommended quality control strains were included with each batch of organisms tested.
No information is available on the in vitro activity of antimicrobial agents commonly used to treat bovine mastitis caused by strains of C. bovis. The lack of in vitro antimicrobial susceptibility data may be due to the difficulty in cultivating C. bovis, as it fails to grow without lipid supplementation of the basal medium (3, 10). Previous studies (4, 8) determining the antimicrobial susceptibility of lipophilic corynebacteria isolated from humans, such as Corynebacterium jeikeium, have recommended the addition of rabbit serum or 0.1% Tween 80. However, the lipid requirements for C. jeikeium appear to be lower than those for C. bovis, as the former organism will grow on blood-supplemented media, while C. bovis grows very poorly, if at all. Results of a previous study (10) indicated that acceptable growth of C. bovis could be achieved by supplementing the basal medium with 1% Tween 80. Thus, we chose to use Mueller-Hinton agar supplemented with 1% Tween 80 for the test medium. Although this supplementation has not been recommended by NCCLS, the test results for the individual antimicrobial agents with the quality control organisms fell within guidelines published by NCCLS (6).
The MICs obtained with 46 strains of C. bovis are summarized in Table 1. Soriano et al. (8) determined the activity of various antimicrobial agents against 43 strains of C. jeikeium, a lipophilic corynebacterium isolated from humans. Phylogenetic studies (as reviewed in reference 3) have also determined that C. jeikeium is the organism most closely related to C. bovis. All of the antimicrobial agents tested except tilmicosin were active against the strains of C. bovis. For example, the MICs at which 90% of the isolates tested are inhibited (MIC90s) of ampicillin, oxacillin, cephalothin, and ceftiofur against C. bovis ranged from 0.25 to 4.0 μg/ml compared to 16.0 to >64.0 μg/ml for the type strain of C. jeikeium. In contrast, Soriano et al. (8) reported that the MIC90s of ampicillin, oxacillin, cephalothin, and cefuroxime for C. jeikeium were >256.0 μg/ml.
TABLE 1.
MICs for 46 strains of C. bovis isolated from bovine mammary glands
| Antimicrobial agent | MIC (μg/ml)
|
||
|---|---|---|---|
| 50% | 90% | Range | |
| Ampicillina | 0.25 | 0.25 | ≤0.06–0.25 |
| Oxacillina | 1.0 | 4.0 | 0.125–8.0 |
| Cephalothina | 0.5 | 0.5 | ≤0.06–64.0 |
| Ceftiofur | 0.125 | 0.5 | ≤0.06–64.0 |
| Penicillin plus novobiocina | 0.125 | 0.5 | ≤0.06–>64.0 |
| Erythromycina | ≤0.06 | ≤0.06 | ≤0.06–>64.0 |
| Tilmicosin | 1.0 | >32.0 | 0.5–>32.0 |
| Clindamycin | 0.125 | 0.25 | 0.125–0.5 |
| Pirlimycina | 0.125 | 0.25 | ≤0.06–>64.0 |
| Tetracycline | 0.25 | 0.25 | 0.125–>32.0 |
| Florfenicol | 1.0 | 2.0 | 1.0–>32.0 |
| Enrofloxacin | 0.125 | 0.25 | ≤0.03–>32.0 |
| Sarafloxacin | 0.25 | 0.5 | 0.25–>32.0 |
| Danofloxacin | 0.125 | 0.25 | 0.06–>32.0 |
| Premafloxacin | 0.015 | 0.015 | ≤0.0078–>8.0 |
Approved for use in treatment of bovine mastitis.
The macrolides (erythromycin and tilmicosin) and lincosamides (clindamycin and pirlimycin) inhibit protein synthesis in bacterial cells by binding to the same site on the ribosome (1, 5, 11). In the present study, the MIC90s for erythromycin, tilmicosin, clindamycin, and pirlimycin were ≤0.06, >32.0, 0.25, and 0.25 μg/ml, respectively. This is markedly different from the MIC90s obtained for C. jeikeium, which were >256.0 μg/ml for both erythromycin and clindamycin (8). Tilmicosin appears to be less potent than erythromycin, with a higher MIC50 (1.0 μg/ml) and MIC90 (>32.0 μg/ml). With the exception of one strain for which the MIC of erythromycin was high, the MICs for all strains were 0.5 μg/ml or below for this compound. In contrast, the tilmicosin MICs for all strains of C. bovis were 0.5 μg/ml or higher. The reason for the differences is unknown, but they are most likely due to differences in tilmicosin's and erythromycin's abilities to penetrate the cell.
Tetracycline and florfenicol are approved for use in veterinary medicine but are not available as intramammary infusion products. Both compounds were much more active against strains of C. bovis than against human strains of C. jeikeium (8). For example, the MIC90 of tetracycline was 0.25 μg/ml for C. bovis compared to 64.0 μg/ml for C. jeikeium (8).
The fluoroquinolones are widely used in human and veterinary medicine to treat a variety of diseases (11). Of these, the fluoroquinolones enrofloxacin, sarafloxacin, and danofloxacin demonstrated good activity, with MIC90s ranging from 0.25 to 0.5 μg/ml. In contrast, the expanded-spectrum fluoroquinolone (9) premafloxacin was much more active, with a MIC90 of 0.015 μg/ml.
C. amylocolatum is a normal resident of healthy human skin and is one of the corynebacteria most frequently isolated from this body site (3, 4). In the present study, we tested 13 strains isolated from bovine mastitis (Table 2). In contrast to the bovine strains, the human strains tend to be much more resistant to antimicrobial agents. For example, the MIC90s of ampicillin, oxacillin, cephalothin, and ceftriaxone for the human strains were >64.0 μg/ml, while the MIC90s for the bovine strains were 0.25, 2.0, 0.5, and 0.5 μg/ml (ceftiofur value), respectively (4).
TABLE 2.
MICs for 13 strains of C. amylocolatum isolated from bovine mammary glands
| Antimicrobial agent | MIC (μg/ml)
|
||
|---|---|---|---|
| 50% | 90% | Range | |
| Ampicillina | 0.125 | 0.25 | ≤0.06–0.25 |
| Oxacillina | 0.5 | 2.0 | 0.5–4.0 |
| Cephalothina | 0.25 | 0.5 | 0.13–0.5 |
| Ceftiofur | 0.125 | 0.5 | ≤0.06–64.0 |
| Penicillin plus novobiocina | ≤0.06 | 0.125 | ≤0.06–0.125 |
| Erythromycina | ≤0.06 | 0.13 | ≤0.06–>64.0 |
| Tilmicosin | 4.0 | 32.0 | 2.0–>32.0 |
| Clindamycin | 0.25 | 0.5 | 0.25–64.0 |
| Pirlimycina | 0.25 | 0.25 | 0.125–>64.0 |
| Tetracycline | 0.25 | 16.0 | 0.125–32.0 |
| Florfenicol | 32.0 | 32.0 | 1.0–>32.0 |
| Enrofloxacin | 0.125 | 0.25 | 0.06–0.25 |
| Sarafloxacin | 0.25 | 0.5 | 0.125–>32.0 |
| Danofloxacin | 0.125 | 0.25 | 0.06–0.5 |
| Premafloxacin | 0.015 | 0.015 | ≤0.0078–>8.0 |
Approved for use in treatment of bovine mastitis.
Similar results are also seen for C. amylocolatum with the macrolides and lincosamides, as the MIC90s for the human strains were >64.0 μg/ml for both erythromycin and clindamycin, compared to 0.13 and 0.5 μg/ml for these same agents with the bovine strains (4). As was the case with C. bovis, tilmicosin was much less active (MIC90 = 32.0 μg/ml) than erythromycin, clindamycin, or pirlimycin. Both tetracycline and florfenicol were much less active against the C. amylocolatum strains tested than against C. bovis. The MIC90 of ciprofloxacin was also >64.0 μg/ml for the human strains (4) compared to 0.25 μg/ml for enrofloxacin (a ciprofloxacin analogue) for the bovine strains.
This study represents the first quantitative assessment of the in vitro activity of various antimicrobial agents against bovine strains of C. bovis and C. amylocolatum. Generally, all the agents tested demonstrated good in vitro activity against these organisms. Data from this study suggest that comparable corynebacteria from humans have a much higher level of antimicrobial resistance to a variety of antimicrobial agents.
Acknowledgments
We thank Cheryl A. Case, Ellen S. Portis, and Sarah A. Salmon for technical assistance.
REFERENCES
- 1.Auckenthaler R W, Azwahlen A, Waldvogel F A. Macrolides. In: Peterson P K, Verhoef J, editors. The antimicrobial agents annual 1. New York, N.Y: Elsevier Scientific Publishers; 1986. p. 115. [Google Scholar]
- 2.Brooks B W, Barnum D A. The susceptibility of bovine udder quarters colonized with Corynebacterium bovis to experimental infection with Staphylococcus aureus or Streptococcus agalactiae. Can J Comp Med. 1984;48:146. [PMC free article] [PubMed] [Google Scholar]
- 3.Funke G, von Graevenitz A, Clarridge III J E, Bernard K A. Clinical microbiology of coryneform bacteria. Clin Microbiol Rev. 1997;10:125–159. doi: 10.1128/cmr.10.1.125. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Funke G, Pünter V, von Graevenitz A. Antimicrobial susceptibility patterns of some recently established coryneform bacteria. Antimicrob Agents Chemother. 1996;40:2874–2878. doi: 10.1128/aac.40.12.2874. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Hermans P E. Lincosaminides. In: Peterson P K, Verhoef J, editors. The antimicrobial agents annual 1. New York, N.Y: Elsevier Scientific Publishers; 1986. pp. 103–114. [Google Scholar]
- 6.National Committee for Clinical Laboratory Standards. Performance standards for antimicrobial disk and dilution susceptibility tests for bacteria isolated from animals. Tentative standard. NCCLS document M31-A. Wayne, Pa: National Committee for Clinical Laboratory Standards; 1999. [Google Scholar]
- 7.Pankey J W, Nickerson S C, Boddie R L, Hogan J S. Effects of Corynebacterium bovis infection on susceptibility to major pathogens. J Dairy Sci. 1985;68:2684–2693. doi: 10.3168/jds.S0022-0302(85)81153-X. [DOI] [PubMed] [Google Scholar]
- 8.Soriano F, Zapardiel J, Nieto E. Antimicrobial susceptibilities of Corynebacterium species and other non-spore-forming gram-positive bacilli to 18 antimicrobial agents. Antimicrob Agents Chemother. 1995;39:208–214. doi: 10.1128/aac.39.1.208. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Watts J L, Salmon S A, Sanchez M S, Yancey R J., Jr In vitro activity of premafloxacin, a new extended-spectrum fluoroquinolone, against pathogens of veterinary importance. Antimicrob Agents Chemother. 1997;41:1190–1192. doi: 10.1128/aac.41.5.1190. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Watts J L, Rossbach S. Identification of Corynebacterium bovis and other coryneforms isolated from bovine mammary glands. J Dairy Sci. 2000;83:2373–2379. doi: 10.3168/jds.S0022-0302(00)75126-5. [DOI] [PubMed] [Google Scholar]
- 11.Yao J D C, Moellering R C., Jr . Antibacterial agents. In: Murray P R, Baron E J, Pfaller M A, Tenover F C, Yolken R H, editors. Manual of clinical microbiology. 6th ed. Washington, D.C.: ASM Press; 1995. pp. 1281–1307. [Google Scholar]