Journal of Antimicrobial Chemotherapy (1991) 27, 171-176
Antimicrobial activity of rifabutin in combination with two and three other antimicrobial agents against strains of Mycobacterium paratuberculosis Rodrick J. Chiodini
The inhibitory and bactericidal, synergistic, and antagonistic activities of rifabutin combined with ciprofloxacin, ethambutol, clofazimine, cefazolin, and amikacin in dual and triple combinations against various human and animal isolates of Mycobacterium paratuberculosis were determined. Synergism was observed when rifabutin was combined with either cefazolin or clofazimine in double combinations. The greatest amount of synergy occurred with therifabutin-cefazolincombination in which bactericidal synergism was present with all strains. Of the triple combinations examined, only rifabutin in combination with ethambutol and cefazolin or streptomycin and cefazolin showed bactericidal synergism against most of the strains. Although antagonism was not observed in any double combination with rifabutin, antagonism was shown with several of the triple combinations. The rifabutincefazolin and rifabutin-streptomycin-cefazolin combinations were found to have the greatest bactericidal synergism at concentrations well within achievable serum and tissue levels and may be appropriate choices for chemotherapeutic use. Introduction Mycobacterium paratuberculosis is the aetiological agent of paratuberculosis, a chronic granulomatous ileocolitis of domestic and wild ruminants (Chiodini, Van Kruiningen & Merkal, 1984). Under some circumstances, monogastric animals including horses, swine, and primates, may also become infected. The disease occurs throughout the world and in all of the United States. Prevalence rates in the United States have been estimated to range from 2-9% to as high as 17% of abattoir dairy cattle (Chiodini et al., 1984). There have been suggestions that M. paratuberculosis may be aetiologically involved in human disease (Chiodini, 1989). Treatment of infection with M. paratuberculosis in animals has not been successful. Although such treatment often results in clinical improvement, and in some cases remission, animals continue to shed M. paratuberculosis in faeces and, upon withdrawal of chemotherapy, clinical disease recurs (Chiodini el al., 1984). At most, therapeutic regimens control the infection as long as chemotherapy is continued. Even the administration of antimicrobial agents before experimental infection does not prevent infection (Rankin, 1955). Most of these studies, however, were conducted with monotherapeutic regimens which may be ineffective in other mycobacterial diseases. 171 0305-7453/91/020171+06 $02.00/0
© 1991 The British Society for Antimicrobial Chemotherapy
Downloaded from http://jac.oxfordjournals.org/ at University of Manchester on May 5, 2015
Mycobacteriology Unit SWP 526, Division of Gastroenterology, Department of Medicine, The Rhode Island Hospital, 593 Eddy Street, Providence, RI02903 USA and the Division of Biological and Medical Sciences, Department of Medicine, Brown University, Providence, USA
172
R. J. Chiodini
Regimens with multiple drugs have not been evaluated for the treatment of infection with M. paratuberculosis. The present study was undertaken to determine the bactericidal and inhibitory activity of rifabutin in combination with other antimicrobial agents in vitro, and to determine whether antimicrobial synergism occurred. Methods
Organisms
Media and antimicrobial agents Organisms were grown in 7H9 broth with Dubos oleic acid-albumin complex, 005% Tween 80, and 2mg/l mycobactinJ in 25 cm2 (30-ml) tissue culture flasks at 37°C without agitation, as previously described (Chiodini, 19906). The antibiotics evaluated are summarized in Table I. MIC, MBC and synergism determinations Broth assays were performed as previously described (Chiodini, 19906). Briefly, flasks containing broth were inoculated with 50 y\ of a mid-logarithmic phase culture Table I. Changes in MIC and MBC (mg/1) of antimicrobial agents in combination with rifabutin for strains of M. paratuberculosis
Antimicrobic Streptomycin Ciprofloxacin Ethambutol Clofazimine Cefazolin Amikacin
MIC range alone
(MIC) 0-25-10 (0-50) 012-0-25 (012) 5-0-10-0 (50) 0-03-0-12 (006) 1-2-> 100 (2-50) 1-25-2-50 (1-25)
MIC range with rifabutin (MIC^ 0-25-0-50 (0-25) 0-06-0-25 (012) 2-5-5-0 (2-50) 003-003 (003) 0-62-1-25 (0-62) 0-62-1-25 (0-62)
MBC range alone (MBC*) 0-25-10 (0-50) 0-25-10 (0-50) 50- > 100 (100)
0 12-10
(0-25) 100- > 10-0 (100) 1-25-50 (250)
MBC range with rifabutin (MBCso) 0-5-1-0 (0-50) 0-12-0-5 (0-25) 2-5-10-0 (50) 006-0-25 (012) 1-25-5-0 (2-5) 0-62-50 (2-50)
Downloaded from http://jac.oxfordjournals.org/ at University of Manchester on May 5, 2015
The human isolates of M. paratuberculosis examined included strains ATCC 43015, ATCC 43545, and ATCC 43544, isolated in the author's laboratory (Chiodini et al, 1986), strains ATCC 49164, ATCC 49166, isolated in The Netherlands by Dr J. Haagsma, and strain BB-410 (ATCC 49167), isolated in California by Dr B. Beaman. Additionally, the neotype strain of M. paratuberculosis (ATCC 19698) and a M. paratuberculosis isolate from primates, strain 3979 (ATCC 43546) (McClure et al., 1987) were also examined. The identification of all strains was confirmed by genetic analysis (Chiodini, 1990a).
Rifabutin synergism
173
Results Dual combinations
In combination with other antibiotics, the MICy, and MBCJO of rifabutin were reduced from 0-06 mg/1 and 1-0 mg/1, alone, to between 0-01 and 006 mg/1 and between 006 and 025 mg/1, respectively. The MIC and MBC of other antibiotics were likewise reduced when tested in combination with rifabutin, except for streptomycin and ciprofloxacin, of which the MICs were the same alone and in combination (Table I). The greatest reduction occurred when cefazolin or clofazimine were combined with rifabutin. With these two antibiotics, the MICjo and MIC*, of both were reduced four-fold, and the MBC^ and MBC,,, were reduced two-fold and 16-fold, respectively, in the case of clofazimine, and eight-fold in the case of cefazolin (Table I). Inhibitory synergism was not observed with rifabutin in combination with streptomycin, ciprofloxacin, ethambutol, or amikacin. Inhibitory synergism was detected against a single strain with rifabutin in combination with either clofazimine or cefazolin (Table II). Bactericidal synergism was observed with at least some strains when rifabutin was combined with ethambutol (five strains), clofazimine (six strains), and cefazolin (eight strains) (Table II). With the rifabutin-ethambutol combination, bactericidal synergism was seen against five of the six human strains (FBC, 0-25-0-50), but not with any of the strains isolated from other animals (FBC 0-75—1 0). With the rifabutin-clofazimine combination, bactericidal synergism was shown against all but two strains, one human isolate and the neotype strain (FBC, 1-0). With the rifabutin-cefazolin combination, a bactericidal synergistic effect was detected against all strains; the MBCg, of 0-12 mg/1 rifabutin and 2-5 mg/1 cefazolin had an FBCM value of 0-25. Bactericidal synergism was not detected with rifabutin in combination with either streptomycin, ciprofloxacin, or amikacin. Rifabutin appeared to play a critical role in the antimicrobial activity of some of the combinations. In four of the combinations examined (rifabutin with streptomycin, ethambutol, clofazimine, or cefazolin), a two-fold reduction in the concentration of rifabutin, without a change in the concentration of the other antibiotic, resulted in a complete loss of antimicrobial activity. The opposite effect, i.e., reduction in the concentration of the other antibiotics without a reduction in rifabutin, did not have as great an effect on the overall antimicrobial activity of the combination.
Downloaded from http://jac.oxfordjournals.org/ at University of Manchester on May 5, 2015
containing between 2-8 x 106 and 8-0 x 106cfu, as determined spectrophotometrically at 540 nm. Controls included a 1:100 dilution of the inoculum and a growth control. Similar tests were set up with combinations of antibiotics. After incubation at 36-38 c C until growth was observed in control cultures, the cultures were mixed, and then 100 //I samples were subjected to serial 10-fold dilution with inoculation of slants of Herrold's eggyolk medium with mycobactin J. Cultures were incubated at 37°C for 12-16 weeks, until individual colonies were discernible. The MIC was the lowest concentration of an antimicrobial agent that prevented growth of > 99% of the inoculum. The MBC was the lowest concentration of an antimicrobial agent which caused a loss in viability of $5 99-9% of the original inoculum. To determine whether the interaction of drugs in combination resulted in a synergistic or antagonistic effect, a fractional inhibitory (FIC) and bactericidal (FBC) concentration was calculated (Berenbaum, 1980). A FIC or FBC value of 0-5 or less was considered synergism and greater than 4-0 antagonism, for both double and triple combinations.
R. J. Chiodini
174
Table II. F1C and FBC values for rifabutin in combination with other antibiotics for strains of M. paratuberculosis
Antibiotic combination RF+ST RF + CP RF + ET RF + CL RF + CZ
FIC range
FBC F
RF + ST+CP RF + ST + ET RF + ST + CL RF + ST + CZ RF + CP + ET RF + CP + CL RF+CP+AM RF + ET + CL RF + ET + CZ RF+ET+AM
0-75-2-5 0-5-1-5 0-5-2-0 0-5-2-5 1-0-2-0 1-25-20 2-0-5-0 1-0-2-0 0-5-1-5 1-25-2-0
1-50 1-25
•0
•25 •5 •5
•5 2-25 •5 •0 •5
FBQo
10-2-5 0-75-1-5 0-25-10 0-25-10 0-25-0-5 1-0-2-5
20 10 0-5 0-5
0-75-3-5 0-5-3-0 1-0-3-0 0-5-1-5 1-5-2-5 1-0-4-0 2-0-100 0-75-2-5 0-25-1-25 0-5-5-0
30 1-5 2-5 0-5 20
0-25
2-5
2-25
3-5 1-5 0-5 1-5
RF, Rifabutin; ST, streptomycin; CP, ciprofloxacin; ET, ethambutol; CL, clofazimine; CZ, cefazolin; AM, amikacin.
Triple combinations
The amount of rifabutin required to achieve the MIC or MBC in combination with two other antibiotics was reduced two- to four-fold for the MICJO and 8-32-fold for the MBC,,). The greatest reduction in the MBC^, of rifabutin occurred in combination with streptomycin and cefazolin where the MBC^ of rifabutin was reduced 32-fold to 0-03 mg/1. A 16-fold reduction in the MBCj,, occurred with the rifabutin-ethambutolcefazolin combination. In general, streptomycin, ciprofloxacin, ethambutol, and amikacin could not be reduced below their MBC^ alone when used in combination with rifabutin and another antibiotic, and the MBCJO of clofazimine could only be reduced two-fold. Cefazolin was the only antibiotic that maintained its bactericidal activity with an eight-fold reduction in the MBC^ when combined with rifabutin and either streptomycin or ethambutol (Table II). There was no synergism or antagonism observed in either the inhibitory or bactericidal activity of the combinations rifabutinstreptomycin-ciprofloxacin, rifabutin-ciprofloxacin-ethambutol, or rifabutin-ethambutol-clofazimine. At least some synergism or antagonism was observed in all other combinations (Table II). In the combination rifabutin-streptomycin-ethambutol, the MICs were 0-013, 0-25 and 2-5 mg/1, respectively, for all strains, while the MBCs were variable with two- to four-fold differences in MBCs between strains. Synergism with this combination was observed in both inhibitory and bactericidal activity against only a single strain; in all others the effect was indifferent. Synergism was also observed against this single strain with the combination rifabutin-streptomycin-clofazimine (FIC, 0-5), but there was no synergistic activity against any other strains or in the bactericidal activity of the combination against this or other strains.
Downloaded from http://jac.oxfordjournals.org/ at University of Manchester on May 5, 2015
0 :M) 0 •0 0-75
RF + AK
0-75-2-0 0-75-2-0 0-75-2-0 0-5-1-5 0-5-1-5 0-75-2-0
range
Rifabutin synergism
175
Discussion Previous studies have determined the in-vitro activity of various antimicrobial agents against human and animal isolates of M'. paratuberculosis and suggested high inhibitory and bactericidal activity of these drugs (Chiodini, 19906). Since monotherapy has been shown to be ineffective in paratuberculosis, the in-vitro activity of multiple antibiotics combinations must be studied to define those most likely to be chemotherapeutically effective. The therapeutic use of synergistic combinations is particularly relevant to conditions of immunosuppression, such as the acquired immune deficiency syndrome (AIDS), and in those infections in which bacterial proliferation is not hindered by host defences, such as in paratuberculosis and leprosy. In the present investigation a stringent definition of synergism has been used ( ^ 0-5). Of the six double combinations examined, only two were found to act synergistically against strains of M. paratuberculosis (rifabutin with either clofazimine or cefazolin). Neither of these showed synergistic inhibitory activity against all strains, and only the rifabutin-cefazolin combination showed bactericidal synergism against all strains examined. Both combinations were bactericidal at concentrations well within therapeutic serum/tissue levels and may have clinical usefulness. Of the ten triple combinations, only two were found to be synergistic, presumably owing to the activity of cefazolin. This antibiotic was synergistic with rifabutin and, most probably, also with streptomycin and ethambutol. Although a direct assay of ethambutol and cefazolin was not performed, synergy was suggested by the observation that the MBC W of ethambutol alone was unchanged in all combinations except when combined with cefazolin in which the MBC was reduced four- to eight-fold. Likewise, the MBC of streptomycin was reduced to a concentration considered synergistic when combined with cefazolin. Although cefazolin has poor antimicrobial activity against
Downloaded from http://jac.oxfordjournals.org/ at University of Manchester on May 5, 2015
With the combination rifabutin-streptomycin-cefazolin, inhibitory synergism was also observed against only one strain, but bactericidal synergism was present against seven of the eight strains. The MBCy, and MBC,,, of this combination (both at 003, 0-25 and 2-5 mg/1, respectively) suggested synergy (FBC = 0-50) and were higher than the MBC of the strain in which synergy was not suggested. Four of the eight strains had FBC values of 0-25. Synergism was also observed with the combination rifabutin-ethambutol-cefazolin. Inhibitory synergism was present against only one strain, but there was a more generalized synergism in bactericidal activity. This combination had bactericidal synergism against five of the eight strains at or below the MBCJO (FBC, 0-50) of 012 mg/1 rifabutin with 2-5 mg/1 ethambutol and 2-5 mg/1 cefazolin. Owing almost exclusively on the concentration of ethambutol needed to achieve > 99-9% killing activity with this combination, activity against three of the strains at the MBC,,, was not considered synergistic. In the remaining combinations examined, synergy was not observed against any strain, and against at least one strain, antagonism was shown (Table II), e.g., rifabutinciprofloxacin-clofazimine (MBC antagonism against one strain), rifabutinciprofloxacin-amikacin (MIC and MBC antagonism against three strains), and rifabutin-ethambutol-amikacin (MBC antagonism against two strains). The greatest amount of antagonism was observed with the combination rifabutin, ciprofloxacin and amikacin in which FBC values of 50, 7-0, and 100 were obtained.
176
R. J. Chiodini
Acknowledgements
The support provided by research grants from Adria Laboratories, Columbus, Ohio is gratefully acknowledged. References Berenbaum, M. C. (1980). Correlations between methods for measurement of synergy. Journal of Infectious Diseases 142, 476-80. Chiodini, R. J. (1989). Crohn's disease and the mycobacterioses: a review and comparison of two disease entities. Clinical Microbiology Reviews 2, 90-117. Chiodini, R. J. (1990a). Characterization of Mycobacterium paratuberculosis and organisms of the Mycobacterium avium complex by restriction polymorphism of the rRNA gene region. Journal of Clinical Microbiology 28, 489-94. Chiodini, R. J. (1990ft). Bactericidal activities of various antimicrobial agents against human and animal isolates of Mycobacterium paratuberculosis. Antimicrobial Agents and Chemotherapy 34, 366-7. Chiodini, R. J., Van Kruiningen, H. J. & Merkal, R. S. (1984). Ruminant paratuberculosis (Johne's disease): the current status and future prospects. Cornell Veterinarian 74, 218-62. Chiodini, R. J., Van Kruiningen, H. J., Thayer, W. R. & Coutu, J. A. (1986). Spheroplastic phase of mycobacteria isolated from patients with Crohn's disease. Journal of Clinical Microbiology 24, 357-63. Hastings, R. C. & Franzblau, S. G. (1988). Chemotherapy of leprosy. Annual Review of Pharmacology and Toxicology 28, 231-45. Hornick, D. B., Dayton, C. S., Bedell, G. N. & Fick, R. B. (1988). Nontuberculous mycobacterial lung disease. Substantiation of a less aggressive approach. Chest 93, 550-5. McClure, H. M., Chiodini, R. J., Anderson, D. C , Swenson, R. B., Thayer, W. R. & Coutu, J. A. (1987). Mycobacterium paratuberculosis in a colony of stumptail macaques (Macaca arctoides). Journal of Infectious Diseases 155, 1011-9. Rankin, J. D. (1955). An attempt to prevent the establishment of Mycobacterium johnei in calves by means of isoniazid alone and in combination with streptomycin. Veterinary Record 67, 1105-7. (Received 1 May 1990; revised version accepted 2 October 1990)
Downloaded from http://jac.oxfordjournals.org/ at University of Manchester on May 5, 2015
M. paratuberculosis alone, data suggest that it is a potent modifier of rifabutin, ethambutol, and streptomycin activities and could play a major role in the therapeutic efficacy of combinations in the treatment of M. paratuberculosis infections and related mycobacterioses. The current recommendations for the treatment of mycobacterial infections is the implementation of at least triple drug therapy (Hastings & Franzblau, 1988). Such treatments offer greater efficacy than single- or dual drug therapy or the administration of different drugs during the course of a treatment regime. Since M. paratuberculosis is closely related to organisms of the M. avium complex (Chiodini, 1990a) treatment regimens are likely to require similar considerations. In one study employing a threedrug chemotherapeutic regimen in pulmonary M. avium infections, a mean of 3-6+/ —0-5 years of continuous treatment was required to obtain clinical improvement (Hornick et al., 1988). Since all antimicrobial agents function only during the replicative cycle of bacterial cells, and since M. paratuberculosis grows extremely slowly compared with other culturable mycobacteria, treatment regimens would undoubtedly need to be extended. Of all the combinations examined, the greatest amount of synergism was found with the rifabutin-cefazolin and rifabutin-streptomycin-cefazolin combinations. Alternate chemotherapeutic choices offering synergism include rifabutinclofazimine and rifabutin-ethambutol-cefazolin, but the effects were not as great.
Antimicrobial activity of rifabutin in combination with two and three other antimicrobial agents against strains of Mycobacterium paratuberculosis Rodrick J. Chiodini
The inhibitory and bactericidal, synergistic, and antagonistic activities of rifabutin combined with ciprofloxacin, ethambutol, clofazimine, cefazolin, and amikacin in dual and triple combinations against various human and animal isolates of Mycobacterium paratuberculosis were determined. Synergism was observed when rifabutin was combined with either cefazolin or clofazimine in double combinations. The greatest amount of synergy occurred with therifabutin-cefazolincombination in which bactericidal synergism was present with all strains. Of the triple combinations examined, only rifabutin in combination with ethambutol and cefazolin or streptomycin and cefazolin showed bactericidal synergism against most of the strains. Although antagonism was not observed in any double combination with rifabutin, antagonism was shown with several of the triple combinations. The rifabutincefazolin and rifabutin-streptomycin-cefazolin combinations were found to have the greatest bactericidal synergism at concentrations well within achievable serum and tissue levels and may be appropriate choices for chemotherapeutic use. Introduction Mycobacterium paratuberculosis is the aetiological agent of paratuberculosis, a chronic granulomatous ileocolitis of domestic and wild ruminants (Chiodini, Van Kruiningen & Merkal, 1984). Under some circumstances, monogastric animals including horses, swine, and primates, may also become infected. The disease occurs throughout the world and in all of the United States. Prevalence rates in the United States have been estimated to range from 2-9% to as high as 17% of abattoir dairy cattle (Chiodini et al., 1984). There have been suggestions that M. paratuberculosis may be aetiologically involved in human disease (Chiodini, 1989). Treatment of infection with M. paratuberculosis in animals has not been successful. Although such treatment often results in clinical improvement, and in some cases remission, animals continue to shed M. paratuberculosis in faeces and, upon withdrawal of chemotherapy, clinical disease recurs (Chiodini el al., 1984). At most, therapeutic regimens control the infection as long as chemotherapy is continued. Even the administration of antimicrobial agents before experimental infection does not prevent infection (Rankin, 1955). Most of these studies, however, were conducted with monotherapeutic regimens which may be ineffective in other mycobacterial diseases. 171 0305-7453/91/020171+06 $02.00/0
© 1991 The British Society for Antimicrobial Chemotherapy
Downloaded from http://jac.oxfordjournals.org/ at University of Manchester on May 5, 2015
Mycobacteriology Unit SWP 526, Division of Gastroenterology, Department of Medicine, The Rhode Island Hospital, 593 Eddy Street, Providence, RI02903 USA and the Division of Biological and Medical Sciences, Department of Medicine, Brown University, Providence, USA
172
R. J. Chiodini
Regimens with multiple drugs have not been evaluated for the treatment of infection with M. paratuberculosis. The present study was undertaken to determine the bactericidal and inhibitory activity of rifabutin in combination with other antimicrobial agents in vitro, and to determine whether antimicrobial synergism occurred. Methods
Organisms
Media and antimicrobial agents Organisms were grown in 7H9 broth with Dubos oleic acid-albumin complex, 005% Tween 80, and 2mg/l mycobactinJ in 25 cm2 (30-ml) tissue culture flasks at 37°C without agitation, as previously described (Chiodini, 19906). The antibiotics evaluated are summarized in Table I. MIC, MBC and synergism determinations Broth assays were performed as previously described (Chiodini, 19906). Briefly, flasks containing broth were inoculated with 50 y\ of a mid-logarithmic phase culture Table I. Changes in MIC and MBC (mg/1) of antimicrobial agents in combination with rifabutin for strains of M. paratuberculosis
Antimicrobic Streptomycin Ciprofloxacin Ethambutol Clofazimine Cefazolin Amikacin
MIC range alone
(MIC) 0-25-10 (0-50) 012-0-25 (012) 5-0-10-0 (50) 0-03-0-12 (006) 1-2-> 100 (2-50) 1-25-2-50 (1-25)
MIC range with rifabutin (MIC^ 0-25-0-50 (0-25) 0-06-0-25 (012) 2-5-5-0 (2-50) 003-003 (003) 0-62-1-25 (0-62) 0-62-1-25 (0-62)
MBC range alone (MBC*) 0-25-10 (0-50) 0-25-10 (0-50) 50- > 100 (100)
0 12-10
(0-25) 100- > 10-0 (100) 1-25-50 (250)
MBC range with rifabutin (MBCso) 0-5-1-0 (0-50) 0-12-0-5 (0-25) 2-5-10-0 (50) 006-0-25 (012) 1-25-5-0 (2-5) 0-62-50 (2-50)
Downloaded from http://jac.oxfordjournals.org/ at University of Manchester on May 5, 2015
The human isolates of M. paratuberculosis examined included strains ATCC 43015, ATCC 43545, and ATCC 43544, isolated in the author's laboratory (Chiodini et al, 1986), strains ATCC 49164, ATCC 49166, isolated in The Netherlands by Dr J. Haagsma, and strain BB-410 (ATCC 49167), isolated in California by Dr B. Beaman. Additionally, the neotype strain of M. paratuberculosis (ATCC 19698) and a M. paratuberculosis isolate from primates, strain 3979 (ATCC 43546) (McClure et al., 1987) were also examined. The identification of all strains was confirmed by genetic analysis (Chiodini, 1990a).
Rifabutin synergism
173
Results Dual combinations
In combination with other antibiotics, the MICy, and MBCJO of rifabutin were reduced from 0-06 mg/1 and 1-0 mg/1, alone, to between 0-01 and 006 mg/1 and between 006 and 025 mg/1, respectively. The MIC and MBC of other antibiotics were likewise reduced when tested in combination with rifabutin, except for streptomycin and ciprofloxacin, of which the MICs were the same alone and in combination (Table I). The greatest reduction occurred when cefazolin or clofazimine were combined with rifabutin. With these two antibiotics, the MICjo and MIC*, of both were reduced four-fold, and the MBC^ and MBC,,, were reduced two-fold and 16-fold, respectively, in the case of clofazimine, and eight-fold in the case of cefazolin (Table I). Inhibitory synergism was not observed with rifabutin in combination with streptomycin, ciprofloxacin, ethambutol, or amikacin. Inhibitory synergism was detected against a single strain with rifabutin in combination with either clofazimine or cefazolin (Table II). Bactericidal synergism was observed with at least some strains when rifabutin was combined with ethambutol (five strains), clofazimine (six strains), and cefazolin (eight strains) (Table II). With the rifabutin-ethambutol combination, bactericidal synergism was seen against five of the six human strains (FBC, 0-25-0-50), but not with any of the strains isolated from other animals (FBC 0-75—1 0). With the rifabutin-clofazimine combination, bactericidal synergism was shown against all but two strains, one human isolate and the neotype strain (FBC, 1-0). With the rifabutin-cefazolin combination, a bactericidal synergistic effect was detected against all strains; the MBCg, of 0-12 mg/1 rifabutin and 2-5 mg/1 cefazolin had an FBCM value of 0-25. Bactericidal synergism was not detected with rifabutin in combination with either streptomycin, ciprofloxacin, or amikacin. Rifabutin appeared to play a critical role in the antimicrobial activity of some of the combinations. In four of the combinations examined (rifabutin with streptomycin, ethambutol, clofazimine, or cefazolin), a two-fold reduction in the concentration of rifabutin, without a change in the concentration of the other antibiotic, resulted in a complete loss of antimicrobial activity. The opposite effect, i.e., reduction in the concentration of the other antibiotics without a reduction in rifabutin, did not have as great an effect on the overall antimicrobial activity of the combination.
Downloaded from http://jac.oxfordjournals.org/ at University of Manchester on May 5, 2015
containing between 2-8 x 106 and 8-0 x 106cfu, as determined spectrophotometrically at 540 nm. Controls included a 1:100 dilution of the inoculum and a growth control. Similar tests were set up with combinations of antibiotics. After incubation at 36-38 c C until growth was observed in control cultures, the cultures were mixed, and then 100 //I samples were subjected to serial 10-fold dilution with inoculation of slants of Herrold's eggyolk medium with mycobactin J. Cultures were incubated at 37°C for 12-16 weeks, until individual colonies were discernible. The MIC was the lowest concentration of an antimicrobial agent that prevented growth of > 99% of the inoculum. The MBC was the lowest concentration of an antimicrobial agent which caused a loss in viability of $5 99-9% of the original inoculum. To determine whether the interaction of drugs in combination resulted in a synergistic or antagonistic effect, a fractional inhibitory (FIC) and bactericidal (FBC) concentration was calculated (Berenbaum, 1980). A FIC or FBC value of 0-5 or less was considered synergism and greater than 4-0 antagonism, for both double and triple combinations.
R. J. Chiodini
174
Table II. F1C and FBC values for rifabutin in combination with other antibiotics for strains of M. paratuberculosis
Antibiotic combination RF+ST RF + CP RF + ET RF + CL RF + CZ
FIC range
FBC F
RF + ST+CP RF + ST + ET RF + ST + CL RF + ST + CZ RF + CP + ET RF + CP + CL RF+CP+AM RF + ET + CL RF + ET + CZ RF+ET+AM
0-75-2-5 0-5-1-5 0-5-2-0 0-5-2-5 1-0-2-0 1-25-20 2-0-5-0 1-0-2-0 0-5-1-5 1-25-2-0
1-50 1-25
•0
•25 •5 •5
•5 2-25 •5 •0 •5
FBQo
10-2-5 0-75-1-5 0-25-10 0-25-10 0-25-0-5 1-0-2-5
20 10 0-5 0-5
0-75-3-5 0-5-3-0 1-0-3-0 0-5-1-5 1-5-2-5 1-0-4-0 2-0-100 0-75-2-5 0-25-1-25 0-5-5-0
30 1-5 2-5 0-5 20
0-25
2-5
2-25
3-5 1-5 0-5 1-5
RF, Rifabutin; ST, streptomycin; CP, ciprofloxacin; ET, ethambutol; CL, clofazimine; CZ, cefazolin; AM, amikacin.
Triple combinations
The amount of rifabutin required to achieve the MIC or MBC in combination with two other antibiotics was reduced two- to four-fold for the MICJO and 8-32-fold for the MBC,,). The greatest reduction in the MBC^, of rifabutin occurred in combination with streptomycin and cefazolin where the MBC^ of rifabutin was reduced 32-fold to 0-03 mg/1. A 16-fold reduction in the MBCj,, occurred with the rifabutin-ethambutolcefazolin combination. In general, streptomycin, ciprofloxacin, ethambutol, and amikacin could not be reduced below their MBC^ alone when used in combination with rifabutin and another antibiotic, and the MBCJO of clofazimine could only be reduced two-fold. Cefazolin was the only antibiotic that maintained its bactericidal activity with an eight-fold reduction in the MBC^ when combined with rifabutin and either streptomycin or ethambutol (Table II). There was no synergism or antagonism observed in either the inhibitory or bactericidal activity of the combinations rifabutinstreptomycin-ciprofloxacin, rifabutin-ciprofloxacin-ethambutol, or rifabutin-ethambutol-clofazimine. At least some synergism or antagonism was observed in all other combinations (Table II). In the combination rifabutin-streptomycin-ethambutol, the MICs were 0-013, 0-25 and 2-5 mg/1, respectively, for all strains, while the MBCs were variable with two- to four-fold differences in MBCs between strains. Synergism with this combination was observed in both inhibitory and bactericidal activity against only a single strain; in all others the effect was indifferent. Synergism was also observed against this single strain with the combination rifabutin-streptomycin-clofazimine (FIC, 0-5), but there was no synergistic activity against any other strains or in the bactericidal activity of the combination against this or other strains.
Downloaded from http://jac.oxfordjournals.org/ at University of Manchester on May 5, 2015
0 :M) 0 •0 0-75
RF + AK
0-75-2-0 0-75-2-0 0-75-2-0 0-5-1-5 0-5-1-5 0-75-2-0
range
Rifabutin synergism
175
Discussion Previous studies have determined the in-vitro activity of various antimicrobial agents against human and animal isolates of M'. paratuberculosis and suggested high inhibitory and bactericidal activity of these drugs (Chiodini, 19906). Since monotherapy has been shown to be ineffective in paratuberculosis, the in-vitro activity of multiple antibiotics combinations must be studied to define those most likely to be chemotherapeutically effective. The therapeutic use of synergistic combinations is particularly relevant to conditions of immunosuppression, such as the acquired immune deficiency syndrome (AIDS), and in those infections in which bacterial proliferation is not hindered by host defences, such as in paratuberculosis and leprosy. In the present investigation a stringent definition of synergism has been used ( ^ 0-5). Of the six double combinations examined, only two were found to act synergistically against strains of M. paratuberculosis (rifabutin with either clofazimine or cefazolin). Neither of these showed synergistic inhibitory activity against all strains, and only the rifabutin-cefazolin combination showed bactericidal synergism against all strains examined. Both combinations were bactericidal at concentrations well within therapeutic serum/tissue levels and may have clinical usefulness. Of the ten triple combinations, only two were found to be synergistic, presumably owing to the activity of cefazolin. This antibiotic was synergistic with rifabutin and, most probably, also with streptomycin and ethambutol. Although a direct assay of ethambutol and cefazolin was not performed, synergy was suggested by the observation that the MBC W of ethambutol alone was unchanged in all combinations except when combined with cefazolin in which the MBC was reduced four- to eight-fold. Likewise, the MBC of streptomycin was reduced to a concentration considered synergistic when combined with cefazolin. Although cefazolin has poor antimicrobial activity against
Downloaded from http://jac.oxfordjournals.org/ at University of Manchester on May 5, 2015
With the combination rifabutin-streptomycin-cefazolin, inhibitory synergism was also observed against only one strain, but bactericidal synergism was present against seven of the eight strains. The MBCy, and MBC,,, of this combination (both at 003, 0-25 and 2-5 mg/1, respectively) suggested synergy (FBC = 0-50) and were higher than the MBC of the strain in which synergy was not suggested. Four of the eight strains had FBC values of 0-25. Synergism was also observed with the combination rifabutin-ethambutol-cefazolin. Inhibitory synergism was present against only one strain, but there was a more generalized synergism in bactericidal activity. This combination had bactericidal synergism against five of the eight strains at or below the MBCJO (FBC, 0-50) of 012 mg/1 rifabutin with 2-5 mg/1 ethambutol and 2-5 mg/1 cefazolin. Owing almost exclusively on the concentration of ethambutol needed to achieve > 99-9% killing activity with this combination, activity against three of the strains at the MBC,,, was not considered synergistic. In the remaining combinations examined, synergy was not observed against any strain, and against at least one strain, antagonism was shown (Table II), e.g., rifabutinciprofloxacin-clofazimine (MBC antagonism against one strain), rifabutinciprofloxacin-amikacin (MIC and MBC antagonism against three strains), and rifabutin-ethambutol-amikacin (MBC antagonism against two strains). The greatest amount of antagonism was observed with the combination rifabutin, ciprofloxacin and amikacin in which FBC values of 50, 7-0, and 100 were obtained.
176
R. J. Chiodini
Acknowledgements
The support provided by research grants from Adria Laboratories, Columbus, Ohio is gratefully acknowledged. References Berenbaum, M. C. (1980). Correlations between methods for measurement of synergy. Journal of Infectious Diseases 142, 476-80. Chiodini, R. J. (1989). Crohn's disease and the mycobacterioses: a review and comparison of two disease entities. Clinical Microbiology Reviews 2, 90-117. Chiodini, R. J. (1990a). Characterization of Mycobacterium paratuberculosis and organisms of the Mycobacterium avium complex by restriction polymorphism of the rRNA gene region. Journal of Clinical Microbiology 28, 489-94. Chiodini, R. J. (1990ft). Bactericidal activities of various antimicrobial agents against human and animal isolates of Mycobacterium paratuberculosis. Antimicrobial Agents and Chemotherapy 34, 366-7. Chiodini, R. J., Van Kruiningen, H. J. & Merkal, R. S. (1984). Ruminant paratuberculosis (Johne's disease): the current status and future prospects. Cornell Veterinarian 74, 218-62. Chiodini, R. J., Van Kruiningen, H. J., Thayer, W. R. & Coutu, J. A. (1986). Spheroplastic phase of mycobacteria isolated from patients with Crohn's disease. Journal of Clinical Microbiology 24, 357-63. Hastings, R. C. & Franzblau, S. G. (1988). Chemotherapy of leprosy. Annual Review of Pharmacology and Toxicology 28, 231-45. Hornick, D. B., Dayton, C. S., Bedell, G. N. & Fick, R. B. (1988). Nontuberculous mycobacterial lung disease. Substantiation of a less aggressive approach. Chest 93, 550-5. McClure, H. M., Chiodini, R. J., Anderson, D. C , Swenson, R. B., Thayer, W. R. & Coutu, J. A. (1987). Mycobacterium paratuberculosis in a colony of stumptail macaques (Macaca arctoides). Journal of Infectious Diseases 155, 1011-9. Rankin, J. D. (1955). An attempt to prevent the establishment of Mycobacterium johnei in calves by means of isoniazid alone and in combination with streptomycin. Veterinary Record 67, 1105-7. (Received 1 May 1990; revised version accepted 2 October 1990)
Downloaded from http://jac.oxfordjournals.org/ at University of Manchester on May 5, 2015
M. paratuberculosis alone, data suggest that it is a potent modifier of rifabutin, ethambutol, and streptomycin activities and could play a major role in the therapeutic efficacy of combinations in the treatment of M. paratuberculosis infections and related mycobacterioses. The current recommendations for the treatment of mycobacterial infections is the implementation of at least triple drug therapy (Hastings & Franzblau, 1988). Such treatments offer greater efficacy than single- or dual drug therapy or the administration of different drugs during the course of a treatment regime. Since M. paratuberculosis is closely related to organisms of the M. avium complex (Chiodini, 1990a) treatment regimens are likely to require similar considerations. In one study employing a threedrug chemotherapeutic regimen in pulmonary M. avium infections, a mean of 3-6+/ —0-5 years of continuous treatment was required to obtain clinical improvement (Hornick et al., 1988). Since all antimicrobial agents function only during the replicative cycle of bacterial cells, and since M. paratuberculosis grows extremely slowly compared with other culturable mycobacteria, treatment regimens would undoubtedly need to be extended. Of all the combinations examined, the greatest amount of synergism was found with the rifabutin-cefazolin and rifabutin-streptomycin-cefazolin combinations. Alternate chemotherapeutic choices offering synergism include rifabutinclofazimine and rifabutin-ethambutol-cefazolin, but the effects were not as great.
