INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

Several enterococcal species belong to the normal intestinal flora of humans and animals. Enterococcus faecalis, Enterococcus faecium, and more rarely other species may be involved in human infections. Antibiotic therapy is problematic because of resistances. In particular, penicillin, aminoglycoside, and glycopeptide resistances cause problems (12, 15).

To date, in animals, enterococcal infections are not known to play an important role, possibly because geriatric medicine and intensive care are less well developed. Recently the susceptibility of these intestinal bacteria to food animal growth-promoting antimicrobial agents not used in therapy has become a subject of concern (1, 11, 18). However, few data on the in vitro susceptibility of intestinal bacteria to these agents are available (7). Enterococci from animals have been tested for susceptibility and resistance to growth promoters and therapeutic agents by the disc diffusion method (13), agar dilution tests in Mueller-Hinton agar (6), and the broth dilution method (18). These differences in test methods make test results difficult to compare.

Enterococci need rich media for rapid growth (3). Most probably because the usual standardized Mueller-Hinton II medium was unsatisfactory in this respect, a modification of culture conditions, consisting of the use of brain heart infusion, was suggested by the National Committee for Clinical Laboratory Standards (NCCLS) (standard M100-S7 for use with M7-A4) (14) for testing resistance of E. faecalis and E. faecium to gentamicin, streptomycin, and vancomycin. This complex medium has not been standardized for antibiotic susceptibility testing, and differences may occur between different manufacturers. Furthermore, several species of enterococci belonging to the normal flora of animals require special growth conditions. A CO₂-enriched environment is necessary for the growth of Enterococcus columbae and Enterococcus cecorum (4). Other species show enhanced growth when incubated in a CO₂-enriched environment. The natural habitat of the intestinal enterococci is, however, largely anaerobic, and these facultative anaerobic bacteria have a preference for anaerobic conditions (3). These special requirements imply a need for specialized culture conditions.

The activity of several antibiotics is known to be influenced by the composition of the medium and/or by the incubation conditions. The possible effects of the composition of the medium and/or incubation conditions are much less well-known for the growth-promoting antibacterials. The only well-known phenomenon involving these substances is the influence of CO₂ on the pH of the medium, which affects the activity of tylosin, a macrolide antibiotic (9, 10, 16).

Therefore, we wished to investigate the effects of different modifications of the standardized Mueller-Hinton II agar dilution procedure on inhibitory activity of growth promoters. Such modifications have been proposed by the NCCLS for use with fastidious bacteria (standard M100-S7 for use with M7-A4) (14). The effects of sheep blood (5%), CO₂ incubation, and anaerobic incubation on the MICs of seven growth-promoting agents used in Europe for collection strains of animal-associated species of the genus Enterococcus were determined.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Strains. The tested collection strains of intestinal enterococci belonging to different species groups are shown in Table 1 (5, 21). The E. faecalis ATCC 29212 and Staphylococcus aureus ATCC 29213 control strains used for susceptibility tests were included in all tests.

Antibiotics. The following antibiotic preparations, manufactured for analytical purposes, were tested: avoparcin (American Cyanamid, Princeton, N.J.), bambermycin (Flavomycin; Hoechst, Frankfurt, Germany), virginiamycin (Pfizer, Rixensart, Belgium), bacitracin (67,000 U/mg) (Sigma, St. Louis, Mo.), tylosin (Sigma), avilamycin (Eli Lilly, Indianapolis, Ind.), and narasin (Eli Lilly). Antibiotics were dissolved in appropriate solvents to make stock solutions containing 1,000 µg/ml and then further diluted in sterile distilled water according to the methods recommended by the NCCLS (standard M100-S7 for use with M7-A4) (14).

Antimicrobial susceptibility tests. MIC tests were carried out on Mueller-Hinton II agar (Becton Dickinson, Cockeysville, Md.) with or without 5% sheep blood containing doubling dilutions of the antibiotics and used on the day of preparation. Appropriate antibiotic-free agar plates were included as controls. Inocula were prepared by diluting brain heart infusion (Oxoid, Basingstoke, United Kingdom) cultures in buffered saline overnight to a density of 0.5 on the McFarland turbidity scale and diluted 40-fold before inoculation. Plates were seeded with approximately 10⁵ CFU.

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

Effect of blood supplementation. When plates were incubated aerobically, major differences between Mueller-Hinton media with and without blood were noted in bambermycin MICs for several strains (44% of the strains) (Table 2). This increase in MIC on blood-containing media was not evident in the range of concentrations tested with strains for which MICs obtained on blood-free media were high. These all belonged to the E. faecium and E. gallinarum species groups. Narasin MICs were almost systematically (72% of the strains) 2 or 3 doubling dilutions higher on blood-supplemented agar. With bacitracin, differences of 1 or 2 doubling dilutions were seen with most strains (72%), and for some strains, a much higher MIC was seen when blood was added. With virginiamycin and avilamycin, the MIC was 1 or 2 doubling dilutions higher for 48 and 44% of the strains, respectively. With avoparcin and tylosin, no important differences were found. It should be noted that none of the strains of the E. cecorum group grew in these aerobically incubated tests. The same was seen with certain of the E. avium group strains but only on blood-free media.

Anaerobic versus aerobic incubation. When blood-supplemented plates were used (Table 3), the antibiotics bambermycin, avilamycin, narasin, and avoparcin showed slightly higher MICs upon anaerobic incubation than upon aerobic incubation for 27, 30, 65, and 34% of the strains, respectively. With tylosin, the opposite was seen in 46% of the strains tested. For virginiamycin and bacitracin, there was an almost equal distribution between higher and lower MICs when the aerobic test results were compared with the anaerobic ones.

Aerobic incubation versus incubation in a CO₂-enriched atmosphere. On Mueller-Hinton plates supplemented with blood, no major differences were noted (Table 4). For avilamycin, narasin, avoparcin, and to a lesser extent bacitracin, there was a tendency towards slightly lower MICs on plates incubated in CO₂ with 46, 77, 58, and 31% of the strains, respectively. In tests with tylosin, the opposite was seen with 50% of the strains. Bambermycin and virginiamycin results did not differ when strains were incubated aerobically or in a CO₂-enriched atmosphere.

Anaerobic incubation versus incubation in a CO₂-enriched atmosphere. Between anaerobic incubation or incubation in a CO₂-enriched atmosphere, there were no substantial differences in MIC results on plates with and without blood (Table 5).

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

Blood supplementation of Mueller-Hinton plates has been shown to have an effect on the MICs of sulfonamide, trimethoprim, cephalosporins, novobiocin, and nafcillin for enterococci (22). The addition of blood reduces MICs of cephalosporins, but not all cephalosporins react in the same way: some remain unaffected, and for others the effect does not exceed 1 or 2 doubling dilutions. This is dependent on the chemical structure of the cephalosporins (2, 17). In the present study, major detrimental effects of blood on antibacterial activity were seen with bambermycin. In the original description of this antibiotic (20), which at that time was called moenomycin, it was indicated that serum inactivates its in vitro activity by 10 to 0.1% compared to protein-free medium. The effects of blood seen in the present tests were much more dramatic, at least on the inhibitory activity on species and strains which were susceptible to relatively low concentrations of this antibiotic. A similar effect of blood may possibly be found in higher concentration ranges for the other strains.

Few reports deal with the effects of anaerobic incubation on MIC results. The activity of aminoglycosides is significantly reduced with Escherichia coli, Klebsiella pneumoniae and Proteus mirabilis incubated anaerobically without CO₂ (19). The reason for the negative effect on tylosin activity is found in the presence of CO₂, which has an effect on the pH of the medium (9, 10, 16). The causes of the slightly positive effects of CO₂ and anaerobic incubation on the inhibitory activity of avilamycin, narasin, and avoparcin observed in the present investigation are unknown, though they are most probably found in the effect of CO₂ on the activity of the antibiotics. To our knowledge, nothing is known about a pH effect on the MICs of these growth promoters. The activity of bambermycin has been reported to be affected by pH increases (20). This was not evident in our tests on blood-containing media, not even with strains such as the S. aureus control strain and the few other strains that were inhibited by relatively low concentrations of bambermycin in this medium (Table 6). The effects of this antibiotic on blood-free medium were strongly variable and strain dependent. It should be noted that the comparisons between aerobic incubation results and the results obtained in aerobic or aerobic CO₂-supplemented atmospheres are somewhat hindered by the presence of E. cecorum and of E. avium group strains which did not grow or which grew poorly aerobically. These strains were not included in the comparative tables.

Many changes in MICs corresponded to the species groups, or to several species of a species group, indicating that there may be species-specific growth requirements and that the species-specific metabolism may play a role in the differences noted. For the capnophilic bacteria, CO₂ dependency is clearly species group related; none of the strains in the E. cecorum group grow without CO₂, and some of the strains in the E. avium group have such poor growth without CO₂ that MICs could not be evaluated. However, for the other species-related changes, no obvious reason can be found for the differences.

It can be concluded that Mueller-Hinton II agar can be used for testing of the inhibitory activity of growth-promoting agents on enterococci. The addition of blood decreases the activity of these products to various degrees. The addition of CO₂ under both aerobic and anaerobic conditions enhances the inhibitory effects of narasin, avoparcin, and avilamycin and is detrimental to the activity of tylosin. When CO₂ supplementation is necessary because capnophilic strains are to be tested, the choice between anaerobic incubation with 4 to 10% CO₂ in the jars or incubation in a CO₂ incubator can be left open. The preferred conditions for testing the susceptibility when all enterococcal species are being compared is Mueller-Hinton II medium supplemented with blood and incubated in a CO₂-enriched atmosphere. When only E. faecium and E. faecalis are being tested, Mueller-Hinton II medium incubated aerobically gives satisfactory results. It should be recognized, however, that the omission of blood strongly influences results obtained with certain antibiotics, especially bambermycin and narasin.