ANTIMICROBIAL AGENTS AND CHEMOTHERAPY, Feb. 1992, 0066-4804/92/020387-07$02.00/0 Copyright © 1992, American Society for Microbiology
p. 387-393
Vol. 36, No. 2
Chemotherapeutic Efficacy of a Newly Synthesized Benzoxazinorifamycin, KRM-1648, against Mycobacterium avium Complex Infection Induced in Mice H. TOMIOKA,1 H. SAITO,'* K. SATO,1 T. YAMANE,2 K. YAMASHITA,2 K. HOSOE,2 K. FUJII,2 AND T. HIDAKA2 Department of Microbiology and Immunology, Shimane Medical University, Izumo 693,1 and Biochemical Research Laboratories, KANEKA Corporation, Takasago 676,2 Japan Received 27 September 1991/Accepted 3 December 1991
Newly synthesized benzoxazinorifamycin, KRM-1648, was studied for its in vivo anti-Mycobacterium avium complex (MAC) activities. When the MICs were determined by the agar dilution method with Middlebrook 7H11 agar medium, KRM-1648 exhibited similarly potent in vitro antimicrobial activities against the MAC isolated from AIDS and non-AIDS patients, indicating possible usefulness of KRM-1648 against AIDSassociated MAC infections. KRM-1648 exhibited potent therapeutic activity against experimental murine infections induced by M. intracellulare N-260 (virulent strain) and N-478, which has much weaker virulence. Similarly, KRM-1648 exhibited an excellent therapeutic efficacy against M. intracelulare infection induced in NK-cell-deficient beige mice (as a plausible model for AIDS-associated MAC infection), in which a much more progressed state of gross lesions and bacterial loads at the sites of infection were observed. When the infected beige mice were killed at weeks 4 and 8, obvious therapeutic efficacy was seen on the basis of reduction in the incidence and degree of lung lesions and bacterial loads in the lungs and spleen with infections due to M. intracelulare N-241, N-256, and N-260. In this case, the efficacy was the highest in N-260 infection, followed by strain N-241. When mice were observed until infection-induced death, survival time of the infected beige mice was found to be prolonged by KRM treatment. However, KRM-1648 was not efficacious in suppressing the progression of pulmonary lesions and the increase in bacterial loads at the sites of infection, including lungs and spleen, at the late phase of infection. This may imply some difficulty with chemotherapy for AIDSassociated MAC infection, even with KRM-1648 treatment, which has excellent in vitro and in vivo anti-MAC activities, as shown in the present study.
Rifampin (RMP) and other rifarmycins, including rifabutin (RBT), rifapentine, and so on, are highly active against slowly growing mycobacteria, particularly Mycobacterium tuberculosis and some nontuberculous mycobacteria (1, 3, 5, 7, 14, 18, 24, 28, 29, 36). However, they are not so excellent in their in vitro activity against M. avium complex (MAC) (5, 19, 28), supposedly because of the permeability barrier of the organisms (11, 25). Moreover, it is not known whether RMP and other rifamycins are really efficacious in the treatment of patients with MAC infections. Some investigators have reported that multidrug chemotherapy, including treatment with RMP, has been used in many cases of clinical management of AIDS- or non-AIDS-associated MAC patients with success (4, 6). In contrast, others have reported unavoidable difficulty in clinical control of MAC, even by using RMP combined with other drugs (36). For instance, Rosenzweig (27) reported that sustained conversion of sputum with such chemotherapy alone was achieved in only 41% of patients and, after a 6-month trial of drugs including RMP, 33% of the patients experienced relapse. Because MAC infections are rapidly increasing in immunocompromised hosts, especially in AIDS patients (38), new rifamycin derivatives having strong anti-MAC activity are desired. In relation, it is noteworthy that a substantial proportion of M. avium strains are considerably susceptible to RMP and RBT (5, 18), and there is a synergistic effect of these rifamycins with ethambutol (16, 17). Such a new *
rifamycin with a high in vitro anti-MAC activity and superior pharmacokinetics may be able to exhibit an excellent therapeutic efficacy against the MAC infections, when used in combination with other drugs, especially ethambutol. In a recent study (31), we found that newly synthesized rifamycin derivatives, benzoxazinorifamycins (KRMs), with chemical structures of 3'-hydroxy-5'-(4-alkylpiperazinyl)benzoxazinorifamycins (alkyl residues: isobutyl, propyl, sec-butyl, sec-butyl [R configuration], and sec-butyl [S configuration] for KRM-1648, -1657, -1668, -1686, and -1687, respectively), exhibited markedly stronger in vitro antimycobacterial activities, especially against the MAC. Their MICs for 50 and 90% of the strains against the MAC were about 64 times smaller than those of RMP. Moreover, KRMs, especially KRM-1648, exhibited higher antimicrobial activity against the MAC phagocytosed in macrophages than did RMP and RBT. Similar results were reported by Yamamoto et al. (37). In this study, KRM-1648 was studied for its in vivo activity against MAC infections induced in mice. MATERIALS AND METHODS
Organisms. For in vitro study, 77 strains of M. avium and 11 strains of M. intracellulare isolated from AIDS patients in the United States and 62 strains of M. avium and 49 strains of M. intracellulare isolated from non-AIDS patients in Japan were used. For in vivo studies, M. intracellulare N-241, N-256, N-260, and N-478 were used. All of the MAC strains were identified with the Rapid Diagnostic System for the MAC (Gen-Probe Inc., San Diego, Calif.), and they
Corresponding author. 387
388
ANTIMICROB. AGENTS CHEMOTHER.
TOMIOKA ET AL.
formed smooth, transparent, irregularly shaped colonies on 7H11 agar plates. Mice. Five-week-old female BALB/c (MAC-susceptible strain) (12), natural killer (NK)-cell-deficient beige (C57BL/6 bgibg) (26) or Std-ddY mice were used. Drugs. KRM-1648, -1657, -1668, -1686, and -1687, synthesized by Biochemical Research Laboratories, KANEKA Corp., Takasago, Japan, were used. The other rifamycins, RMP and RBT, were kind gifts from Daiichi Pharmaceutical Co., Tokyo, Japan, and Farmitalia Carlo Erba Research Laboratories, Milan, Italy, respectively. In vitro antimicrobial activity. Test organisms were cultured in 7H9 broth containing 0.05% Tween 80 at 37°C for 3 to 7 days until its optical density (540 nm, 1.5-cm light path) reached about 0.2. Test bacterial suspension was prepared by diluting the 7H9 broth culture with 0.1% Tween 80 in saline to give an optical density of 0.1 at 540 nm (approximately 107 CFU/ml) as measured by a spectrophotometer (Model 100-10; Hitachi Co. Ltd., Hitachi, Japan). The bacterial suspension was then diluted 10 times in the saline to prepare an inoculum. MIC was determined by the agar dilution method as previously described (35). Briefly, the inoculum (5 pAl) was spotted on 7H11 agar plates containing 100 -0.0125-,ug/ml doses (twofold dilution) of drugs. The MICs of the drugs were determined 14 days after cultivation at 37°C in a CO2 incubator (5% C02-95% humidified air). The MICs were read as minimum concentrations of drugs completely inhibiting the growth of organisms or allowing no more than five colonies to grow. In vivo activity. M. intracellulare was used to induce the MAC infection in mice, because its virulence to mice is significantly higher than that of M. avium (34a.). M. intracellulare N-241, N-256, N-260, or N-478 was cultured in 7H9 broth containing 0.05% Tween 80 at 37°C until its optical density (540 nm, 1.5-cm light path) reached about 0.15. The bacterial suspension was gently sonicated with Handy Sonic (Model 20P; Tomy Seiko, Co., Tokyo, Japan) and centrifuged at 1,000 rpm for 5 min to remove large bacterial clumps. The upper layer was then diluted with 7H9 medium to give an optical density of 0.1 at 540 nm and used as an inoculum for experimental infection. Mice were infected intravenously with the organisms at doses from 7.6 x 106 to 2.7 x 107 per mouse (see footnote a of each table for details). The animals were given either KRM, RMP, or RBT, finely emulsified in 2.5% gum arabic-0.2% Tween 80, by gavage once daily six times per week from day 1 to the end of the experiment. Therapeutic efficacy of a test drug was judged on the basis of the reduction of mortality of infected animals, incidence of gross lesions in the visceral organs, and bacterial load at the sites of infection, especially in the lungs and spleen. The number of CFUs in the visceral organs was counted by inoculating serial 10-fold dilutions of tissue homogenates with saline prepared by using a glass homogenizer onto 7H11 agar plates and cultivating them at 37°C for 2 weeks. The tissue homogenates were decontaminated with 0.2% NaOH for 20 s and then immediately neutralized with 0.5 N HCl, prior to the serial dilution. Distribution of drug in tissue. KRM-1648, RMP,
or
RBT
finely emulsified in 2.5% gum arabic-0.2% Tween 80 solution was given to male Std-ddY mice by gavage at the dose of 20 mg/kg of body weight. At intervals, the drug concentration in the plasma, lungs, and spleen was measured as follows. The visceral organs were homogenized in acetate buffer (pH 4.0) (for KRM-1648), phosphate-buffered saline (for RBT), or 2% sodium ascorbate in phosphate-buffered saline (for RMP), and then the homogenate or plasma was extracted with
acetone (for KRM-1648 and RBT) or methanol (for RMP). The resultant extract was condensed by subjecting Bond Elute C18 or C8 cartridge (Analytichem International Co., Harbor City, Calif.) and analyzed for concentration of each rifamycin by high-performance liquid chromatography. RESULTS In vitro activity of KRM-1648 against the MAC isolated from AIDS and non-AIDS patients. In our previous study, in vitro activity of KRM-1648 against MAC was examined by using only the strains isolated from non-AIDS patients. Therefore, it is unclear whether KRM-1648 is similarly active against MAC strains from AIDS patients and those from non-AIDS patients. Here, we determined the MICs of KRM-1648, RMP, and RBT against the MAC strains isolated from AIDS (M. avium, 77 strains; M. intracellulare, 11 strains) and non-AIDS (M. avium, 62 strains; M. intracellulare, 49 strains) patients. KRM-1648 was similarly efficacious against the MAC strains isolated from both AIDS and non-AIDS patients, and its MIC for 90% of strains tested (0.05 to 0.1 j,g/ml) was much smaller than those of RMP (3.13 to 25 ,ug/ml) and RBT (0.4 to 1.6 jig/ml). Therapeutic efficacy of KRM-1648 and other KRMs against the murine infections with M. intracellulare N-260 induced in BALB/c and beige mice. Table 1 shows the therapeutic activities of various KRMs against infection with M. intracellulare N-260 induced in BALB/c mice (experiment 1) and the efficacy of KRM-1648 against the N-260 infection induced in beige mice (experiment 2). Firstly, in experiment 1, when infected mice were given each of the five KRMs at doses of 0.2 and 0.4 mg per mouse, the incidence of gross lung lesions was completely inhibited, accompanied by a 1.9to 3.1-log decrease in the number of CFUs of the organisms in the lungs at week 8 compared with those for untreated animals, although such a marked reduction in the number of CFUs was not observed for the spleen. Moreover, infectioninduced splenomegaly (2.4-fold increase in spleen weight at week 8 for untreated mice) was also markedly suppressed; there was 70 to 85% reduction by the five KRMs at the dose of 0.4 mg per mouse compared with that for untreated control mice infected with M. intracellulare. In this experiment, KRM-1648, followed by KRM-1668, was most efficacious in controlling M. intracellulare infection in terms of inhibiting the increase in the number of CFUs in the lungs. It has been reported that NK-cell-deficient beige mice are much more susceptible to MAC infection (8, 9). As indicated in Table 1 (experiment 2), when the beige mice were infected with M. intracellulare N-260, much faster bacterial growth in the lungs and spleen was seen compared with those for BALB/c (experiment 1) and C57BL/6 (30) mice with a normal genotype (+/+). In this experiment, when KRM1648 was given at a dose of either 0.2 or 0.4 mg per mouse and the animals were killed at weeks 4 and 8, the agent efficiently reduced the incidence or degree of lung lesions or both and the growth of the organisms in the visceral organs. However, it was noted that KRM-1648 was unable to eliminate the organisms at the sites of infection, even when given at a dose of 0.4 mg per mouse, in the present administration protocol. Although a transient decrease in the number of CFUs in the spleen as well as lungs compared with the number of CFUs on day 1 was observed for mice given KRM-1648 at a dose of 0.4 mg per mouse in the relatively early phase of infection, such as at week 4, an increase in the number of organisms was seen at week 8, despite continuous KRM treatment. A similar tendency was also reported by
389
ANTI-MAC ACTIVITY OF BENZOXAZINORIFAMYCIN
VOL. 36, 1992
TABLE 1. Therapeutic efficacy of various KRMs against MAC infection with M. intracellulare N-260 induced in BALB/c (experiment 1) and beige (experiment 2) mice' No. of mice with indicated lung lesion score atb:
Experiment and drug
Expt 1 None KRM-1648
Dose (mg)
No. of None
1+
2+
3+
0 0 0 0 0
0 0 0 0 0
0 0 0 0 0
14 14 14 14 14 7 7 7
7 7 7 7 7
KRM-1668 KRM-1686 KRM-1687
0.2 0.4 0.2 0.4 0.4 0.4 0.4
Expt 2 None KRM-1648 KRM-1648
0.2 0.4
10 10 10
0 2 3
KRM-1657
0 1 2
5 2 0
Spleen
Lungs
8 weeks
4 weeks
mice
Log CFUC
None
1+
2+
3+
4+
4 weeks
8 weeks
4 weeks
8 weeks
0 7 7 7 7 7 7 7
0 0 0 0 0 0 0 0
7 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0
5.53 3.74 3.49
6.54 6.09 5.81
NCd NC NC
6.71 4.86 3.66 4.90 4.58 4.15 4.34 4.80
NC
7.00 6.77 6.44 6.78 6.66 6.50 6.56 6.51
0 0 0
0 0 0
0 1 4
0 4 1
5 0 0
7.09 5.21 4.48
8.02 6.34 5.47
7.14 6.12 5.69
8.36 6.90 6.48
0 0 0
3.96 3.53
6.17 6.08 NC NC
Mice were infected with 7.6 x 106 (experiment 1) or 1.5 x 107 (experiment 2) CFU of M. intracellulare N-260. Symbols: none, no macroscopic lesions; 1+, less than 20 small nodules; 2+, more than 20 small nodules; 3+, many small nodules with a few large nodules (diameter, _2 mm). c The mean values for seven (experiment 1) or five (experiment 2) mice are given. For experiment 1, the average standard error of the mean was 0.08 and did not exceed 0.22; for experiment 2, the average standard error of the mean was 0.12 and did not exceed 0.26. All the log CFU values for KRM-treated mice were significantly lower than those for untreated control mice (P < 0.01, Student's t test). At day 1, the values were 4.72 (lungs) and 6.01 log CFU (spleen) for experiment 1 and 4.43 (lungs) and 6.12 (spleen) log CFU for experiment 2 (n = 5). d NC, not counted. a
b
in mice given KRM was markedly lower than the degrees seen with animals given RMP or RBT. A statistically significant difference in the number of CFUs in the lungs and spleens for mice given KRM and those given either RMP or RBT was observed (P < 0.05, Student's t test). Thus, the therapeutic efficacy of KRM is thought to be much greater than those of RMP and RBT. Therapeutic activity of KRM-1648 against the murine infections with M. intracellulare N-478 induced in BALB/c mice. As shown in Table 3, infection induced in BALB/c mice by M. intracellulare N-478, which is less virulent to mice than N-260, was also efficiently controlled by KRM-1648, whose MIC against strain N-478 was 0.2 ,ug/ml (four times higher than that against strain N-260). No lung lesions were noted in any of the test mice, even at week 8. In this case, a 1.0- to 1.6-log decrease in the number of CFUs of organisms in the lungs of mice treated with KRM-1648 at doses of 0.2 to 0.8 mg per mouse was noted at week 8 after infection. In this case, a 2.3- and 1.8-fold increase in spleen weight for
Kuze et al. (23) in murine experimental infection induced by another MAC strain, M. intracellulare 31F093T. Table 2 compares the in vivo activity of KRM-1648 against MAC infection induced in beige mice with those of RMP and RBT. KRM completely inhibited the incidence of gross pulmonary lesions in MAC-infected beige mice at week 6 and caused a much lower degree of lung lesions at week 12 compared with those in untreated mice. Its efficacy to suppress the progression of lung lesions was much more obvious than those of RBT and RMP. MAC infectioninduced splenomegaly (11-fold increase in spleen weight for untreated mice) was also suppressed by treatment with these rifamycins. KRM-1648, RMP, and RBT caused, respectively, 93, 44, and 51% inhibition of the splenomegaly seen with untreated animals at week 12. KRM-1648 markedly suppressed the bacterial growth in the lungs and spleens of beige mice, and about 2- and 3-log reductions were seen at weeks 6 and 12, respectively, compared with those of untreated control mice. Moreover, the degree of lung lesions
Drug
None RMP RBT KRM
TABLE 2. Therapeutic efficacy of KRM-1648 against MAC infection with M. intracellulare N-260 induced in beige micea Log CFUC No. of mice with indicated lung lesion score atb: Dose No. of 12 weeks Spleen Lungs 6 weeks mice (mg)
0.4 0.4 0.4
13 13 13 13
None
1+
2+
3+
None
1+
2+
3+
4+
6 weeks
12 weeks
6 weeks
12 weeks
0 0 0 6
0 0 0 0
6 6 6 0
0 0 0 0
0 0 0 0
0 0 0 0
0 0 0
0 0 3
7 7 4
7.07
7
0
0
8.80 8.18d 8.46e 5.41d
7.75 7.40d 7.49 6.07d
9.17 8.39d 8.81 6.55d
6.84 6.68 4.84d
weizb infected with 8.7 x 106 CFU of M. intracellulare N-260. Lung lesions were scored as indicated in Table 1, footnote b. 4+, gross lesions with many large nodules. The mean values for six or seven mice are given. The average standard error of the mean was 0.10 and did not exceed 0.17. At day 1, the values were 4.87 (lungs) and 6.32 (spleen) log CFU. All the values (log CFU) for KRM-treated mice were significantly lower than those for either RMP- or RBT-treated mice (P < 0.01). d Significantly different from values obtained with untreated control mice (P < 0.01). e Significantly different from values obtained with untreated control mice (P < 0.05). a Beige mice b
c
390
ANTIMICROB. AGENTS CHEMOTHER.
TOMIOKA ET AL.-
TABLE 5. Therapeutic efficacy of KRM-1648 against MAC infection with M. intracellulare N-256 induced in beige mice: data for individual dead micea
TABLE 3. Therapeutic efficacy of various KRM-1648 doses against MAC infection with M. intracellulare N-478 induced in BALB/c micea Log CFUC
No. of mice with
KRM-1648 dose (mg)
score at 8b
None 1+
0 0.2 0.4 0.8
5 5 5 5
0 0 0 0
Spleen
Lungs
3+ 4 weeks 8 weeks 4 weeks 8 weeks
2+
0 0 0 0
0 0 0 0
4.41 3.48 3.34e 3.34
KRM-1648 dose (mg) dose (mg
Mouse Mouse no.
indicated lung lesion
5.03 4.07 4.04d 3.46e
6.81 6.41d 6.35d 6.07d
6.44
5.88d 5.84d 5.73d
a Mice were infected with 9.4 x 106 CFU of M. intracellulare N-478. lesions were scored as indicated in Table 1, footnote b. c The mean values for five mice are given. The average standard error of the mean was 0.18 and did not exceed 0.47. At day 1, values were 4.52 (lungs) and 5.83 (spleen) log CFU. d Significantly different from values obtained with untreated control mice (P < 0.01). e Significantly different from values obtained with untreated control mice (P < 0.05). b Lung
untreated mice at weeks 4 and 8 after infection, respectively, was seen. KRM given at doses of 0.2, 0.4, and 0.8 mg per mouse inhibited infection-induced splenomegaly by a 63 to 68% and a 50 to 53% ratio at weeks 4 and 8, respectively, compared with those for untreated control mice infected with the N-478 strain. Therapeutic activity of KRM-1648 against the murine infections with M. intracellulare N-256 and N-241 induced in beige mice. When therapeutic activity of KRM-1648 was examined for infection induced in beige mice by other virulent strains (Table 4), M. intracellulare N-256 (experiment 1) and N-241 (experiment 2), which have the same in vitro susceptibility to KRM-1648 (MIC, 0.05 p.g/ml), a comparable but somewhat weaker efficacy was observed compared with that of M. intracellulare N-260 infection, based on the incidence of pulmonary lesions and the number of CFUs of organisms recovered from the lungs and spleen at weeks 4 and 8. Table 5 shows the survival time, the degree of lung lesions, and the number of CFUs in the lungs and spleens at the deaths of individual animals due to N-256 infection. The mean survival time of infected animals was prolonged from
Log CFUc
Survival time (days)6
Degree of lung lesions
Lungs
Spleen
9.41
1 2 3 4 5
0 0 0 0 0
76 73 75 73 73
3+ 3+ 3+ 3+ 3+
9.62 9.17 9.48 9.38 9.33
9.60 9.53
6 7 8 9 10
0.2 0.2 0.2 0.2 0.2
73 73 75 73 75
3+ 4+ 3+ 3+ 4+
9.10 9.01
9.09
8.96 9.11 9.07
9.32 9.01
9.44 9.33
11 12 13 14 15
0.4 0.4 0.4 0.4 0.4
136 140 140 136 132
3+
10.40
10.04
4+ 3+ 4+ 3+
10.32 10.28 10.50 10.55
9.91 10.41 10.31 9.34
9.15 9.29
a The data for animals found dead due to strain N-256 infection at the dose of 1.3 x 107 CFU per mouse are indicated. I The mean survival times (days) were 74.0 for mice 1 to 5, 73.8 for mice 6 to 10, and 137 for mice 11 to 15. c Mean + standard error of the mean for lungs and spleen, respectively: 9.40 ± 0.08 and 9.40 ± 0.08 for mice 1 to 5, 9.11 + 0.06 and 9.18 ± 0.09 for mice 6 to 10, and 10.41 ± 0.05 and 10.00 ± 0.19 for mice 11 to 16.
74 to 137 days by KRM treatment at a dose of 0.4 mg per mouse, although KRM given at a dose of 0.2 mg per mouse failed to exhibit such an effect. Despite the fact that KRM given at a dose of 0.4 mg per mouse was able to prolong the survival time of infected beige mice, bacterial loads in the lungs and spleens of dead mice were larger for mice given KRM at this dose, compared with those for untreated control dead mice. However, it should be emphasized that KRM treatment was significantly efficacious in retarding the growth of the organisms in the visceral organs in the earlyto-middle phase of infection (Table 4). A similar situation was seen with infection due to strain
TABLE 4. Therapeutic efficacy of KRM-1648 against MAC infection with M. intracellulare N-256 (experiment 1) or N-241 (experiment 2) induced in beige micea No. of mice with indicated lung lesion score atb: KRM-1648
No. of
dose (mg)
mice
Expt 1 0
Log CFUc 4+
4 weeks
8 weeks
4 weeks
8 weeks
0
5
3
2
8.71 7.41d
8.13 7.14d
9.25 8.13d
3
0
7.68 6.02d 5.49
6.76d
6.58d
7.85d
0 3 2
5 2 0
9.10
8.17
9.54
8.62d 8.12d
6.95e
9.12d 8.12d
None
1+
2+
3+
None
1+
2+
3+
0 0 0
0 0 0
0 0 2
0 0 0
0 0 0
0 0 3
0.2 0.4
10 10 10
0 0 1
0 3 4
4 2
0
1 0 0
Expt 2 0 0.2 0.4
10 10 10
0 0 0
0 1 4
4 4 1
1 0 0
Spleen
Lungs
8 weeks
4 weeks
7.88 6.77
6.44d
6.77e
Beige mice were infected with 2.1 x 107 CFU of M. intracellulare N-256 (experiment 1) or 2.7 x 107 CFU of M. intracellulare N-241 (experiment 2). Lung lesions were scored as indicated in Table 1, footnote b. cThe mean values for five mice are given. The average standard error of the mean was 0.14 and did not exceed 0.24 for experiment 1 and was 0.14 and did not exceed 0.24 for experiment 2. At day 1, the values (log CFU) for lungs and spleen, respectively, were 4.81 and 6.48 (experiment 1) and 5.01 and 6.39 (experiment 2). d Significantly different from values obtained with untreated control mice (P < 0.01). e Significantly different from values obtained with untreated control mice (P < 0.05). a
b
ANTI-MAC ACTIVITY OF BENZOXAZINORIFAMYCIN
VOL. 36, 1992 TABLE 6. Therapeutic efficacy of KRM-1648 against MAC infection with M. intracellulare strain N-241 induced in beige mice: data for individual dead micea Mouse no.
KRM-1648 dose (mg)
Survival time (days)b
Degree of lung lesions
Lungs
Spleen
Log CFUC
1 2 3 4 5
0 0 0 0 0
111 111 109 69 70
3+ 3+ 4+ 3+ 3+
9.75 10.14 9.94 8.15 8.04
8.94 10.35 10.13 8.33 8.39
6 7 8 9 10
0.2 0.2 0.2 0.2 0.2
107 114 106 114 107
3+ 3+ 4+ 3+ 3+
9.49 10.32 8.26 9.76 10.41
9.70 10.27 8.47 10.39 9.84
11 12 13 14 15
0.4 0.4 0.4 0.4 0.4
107 125 119 114 126
3+ 3+ 4+ 4+ 4+
9.70 10.15 10.36 10.62 10.33
9.84 10.06 10.68 10.54 10.14
a The data for animals found dead due to strain N-241 infection at the dose of 1.2 x 107 CFU per mouse are indicated. b The mean survival times (days) were 94.0 for mice 1 to 5, 110 for mice 6 to 10, and 114 for mice 11 to 15. The mean values were not significantly different from those obtained with untreated control mice. c Mean ± standard error of the mean for lungs and spleen, respectively: 9.20 ± 0.46 and 9.23 ± 0.43 for mice 1 to 5, 9.65 ± 0.39 and 9.73 ± 0.34 for mice 6 to 10, and 10.23 ± 0.15 and 10.25 + 0.16 for mice 11 to 15.
N-241, as shown in Table 6. In this case, the prolongation of survival time by KRM treatment at a dose of 0.2 mg per mouse was more marked than that seen with N-256 infection, whereas the efficacy seen with mice given KRM at a dose of 0.4 mg per mouse was not as obvious as was seen with N-256 infection (Table 6 versus Table 5). In this case, the number of CFUs in the lungs and spleen at the death of the animals was also larger for mice given KRM-1648 at a dose of 0.4 mg per mouse than for untreated mice. Pharmacokinetics of KRM-1648. Table 7 compares the distribution of KRM-1648 in tissue with those of RMP and TABLE 7. Pharmacokinetics of KRM-1648: distribution of KRM-1648 in lungs, spleen, and plasma after a single oral administrationa Organ
Lungs
Spleen
(pug/g/ml) ±
Time after administration (h)
KRM-1648
RMP
RBT
1 5 8
1.10 ± 0.35b 5.27 ± 0.83 4.75 + 1.09
6.37 ± 0.64b 8.51 ± 2.10 6.87 + 0.69
9.86c
1 5 8
1.40 + 0.41 9.14 ± 0.56 7.65 ± 1.56
1.09 + 0.10 1.24 + 0.19 0.97 ± 0.09
NDd
Concn
SEM of:
3.22 1.39 ND ND
0.20 ± 0.14 9.51 ± 1.72 0.92 0.25 ± 0.05 12.73 ± 4.38 0.23 9.78 + 0.81 0.60 + 0.10 0.16 a Each drug was given to Std-ddY mice by gavage at a dose of 20 mg/kg of
Plasma
1 5 8
body weight. bn = 3. Cn = 2. d ND, not determined.
391
RBT. Although the concentration of KRM-1648 in plasma was not so sharply increased as in the case of RMP, its accumulation and retention in the spleen during 5 to 8 h after administration was much higher than that of RMP. The concentration of KRM-1648 in the lungs was somewhat lower but comparable to that of RMP and considerably higher than that of RBT. Therefore, the pharmacokinetics of KRM-1648 is comparable to that of RMP and superior to that of RBT. DISCUSSION
In this study, we found the following. First, KRM-1648 had a potent in vitro antimicrobial activity against MAC strains isolated from both AIDS and non-AIDS patients. There was no marked difference in the values against AIDS isolates and those against non-AIDS isolates. This is inconsistent with the finding by Horsburgh et al. (20) that there was a significant difference in MAC susceptibility to some antimicrobial agents, including RMP, with AIDS and nonAIDS isolates. They reported that the ratio of RMP (10 ,ug/ml)-susceptible strains in 75 non-AIDS isolates was 67%, which was significantly (P < 0.001) larger than the value (27%) for AIDS isolates (57 strains). The reason for this inconsistency is thought to be because they tested the drug susceptibility of all MAC isolates, without classifying them as M. avium or M. intracellulare. We found that the drug susceptibilities of the two species differed considerably from each other, in particular with RMP and new quinolones (32, 35). Moreover, the ratio of M. avium and M. intracellulare in AIDS isolates is markedly different from that in non-AIDS isolates (13, 20). For instance, Guthertz et al. (13) reported that 98% of the strains isolated from 44 patients with AIDS were identified as M. avium by Gen-Probe testing, while 40% of MAC isolates recovered from 109 patients without AIDS were M. intracellulare. These findings might be responsible for a certain degree of difference in the drug susceptibilities of MAC isolates from AIDS and non-AIDS patients, as they have observed. Secondly, this study revealed the superiority of the therapeutic efficacy of KRM-1648 against M. intracellulare infections induced in mice, in terms of suppression of splenomegaly, the incidence or degree of gross pulmonary lesions or both, and bacterial growth in the lungs and spleen. The pharmacokinetics of KRM-1648 was comparable to that of RMP and somewhat better than that of RBT (Table 7). Therefore, the excellent therapeutic efficacy of KRM-1648 against MAC infection seems to be due to its strong in vitro anti-MAC activity (31, 37). Thirdly, KRM-1648 exhibited therapeutic efficacy against N-260 strain infection induced in NK-cell-deficient beige mice (as a plausible model for AIDS-associated MAC infection) similar to that of a normal genotypic strain of mice. However, it is noted that in vivo activity of KRM-1648 in beige mice infected with M. intracellulare N-256 or N-241 was somewhat inferior to that seen with M. intracellulare N-260 infection, although it has the same MIC (0.05 ,ug/ml) against these three M. intracellulare strains. Although the precise reason for this is unclear, the following may be stated. First, a possible difference in the virulence of these M. intracellulare strains to mice does not seem to cause this phenomenon, because these three strains (N-260, N-256, and N-241) had similar levels of virulence to mice, showing comparable growth at the sites of infection (lungs and spleens) around weeks 8 to 12. Second, there may be more strict dependency of host defense mechanisms upon NK-cell
392
TOMIOKA ET AL.
functions with infections due to M. intracellulare strains, such as N-241 and N-256, compared with strain N-260 infection. In other words, KRM may require intact NK-cell functions to fully exhibit its in vivo antimicrobial activity, with N-241 or N-256 infection. Indeed, the possibility has been proposed that NK cells play some important role in host defense mechanisms not only in virus infections but also in other microbial infections (10, 15, 19). In relation to this, it is noteworthy that beige mice have an impaired cytotoxic T-cell activity in addition to NK-cell deficiency (2, 33). Therefore, it is also possible that KRM may need intact-cell functions of CD8+ T cells to efficiently express its in vivo activity against M. intracellulare infection. Indeed, CD8+ T cells have been proposed to play some important role in host resistance to mycobacterial infections (21), although CD4+ T cells mainly mediate the host immune response to mycobacterial infections (22). While NK cells are known to contribute to host resistance in the early phase of microbial infections (34), CD8+ T cells are generated in a relatively late phase of chronic infections, including mycobacterial infections (21). This may be a possible explanation for our finding that KRM-1648 failed to suppress the bacterial growth at the sites of infection in the progressed state of test M. intracellulare (N-241, N-256, and N-260)-induced infections, despite the fact that the agent showed an appreciable activity to inhibit bacterial growth in the early phase of infection. This can be easily explained if we assume that in MAC infections, CD8+ T cells rather than NK cells play an important role in the expression of host resistance. In any case, these situations may imply some difficulties with clinical treatment of AIDS-associated MAC infections even with KRMs, although they possess a superior in vitro activity against MAC strains isolated from both AIDS and non-AIDS patients and excellent in vivo anti-M. intracellulare activity in experimental infections in normal genotypic mice. In any case, KRM-1648 has an excellent in vivo activity against M. intracellulare infection induced in mice. Since rifamycins are known to exhibit synergistic in vitro activities against the MAC with other antimicrobial agents, especially ethambutol (16, 17), it is possible to improve the in vivo activity of KRM-1648 by combined use with other antimicrobial agents. On this point, further studies are under way. ACKNOWLEDGMENTS We thank Daiichi Pharmaceutical Co. and Farmitalia Carlo Erba Research Laboratories for the kind gifts of rifampin and rifabutin, respectively. REFERENCES 1. Arioli, V., M. Berti, G. Carniti, E. Randisi, and R. Scotti. 1981. Antimicrobial activity of DL 473, a new semisynthetic rifampicin derivative. J. Antibiot. 34:1026-1032. 2. Baca, M. E., A. M. Mowat, and D. M. V. Parrott. 1989. Immunological studies of NK cell-deficient beige mice. II. Analysis of T-lymphocyte functions in beige mice. Immunology 66:131-137. 3. Bruna, C. D., G. Schoppacassi, D. Ungheri, D. Jabes, E. Morvillo, and A. Sanfilippo. 1983. LM 427, a new spiropiperidylrifamaycin: in vitro and in vivo studies. J. Antibiot. 36:15021506. 4. Chiu, J., J. Nussbaum, S. Bozzette, J. G. Tilles, L. S. Young, J. Leedom, P. N. R. Heseltine, J. A. McCutchan, and California Collaborative Treatment Group. 1990. Treatment of disseminated Mycobacterium avium complex infection in AIDS with amikacin, ethambutol, rifampin, and ciprofloxacin. Ann. Intern. Med. 113:358-361.
ANTIMICROB. AGENTS CHEMOTHER. 5. Cynamon, M. H. 1985. Comparative in vitro activities of MDL 473, rifampin, and ansamycin against Mycobacterium intracellulare. Antimicrob. Agents Chemother. 28:440-441. 6. Davidson, P. T., V. Khanio, M. Goble, and T. S. Moulding. 1981. Treatment of disease due to Mycobacterium intracellulare. Rev. Infect. Dis. 3:1052-1059. 7. Dickinson, J. M., and D. A. Mitchison. 1990. In vitro activities against mycobacteria of two long-acting rifamycins, FCE22807 and CGP40/469A (SPA-S-565). Tubercle 71:109-115. 8. Gangadharam, P. R., C. K. Edwards III, P. S. Murthy, and P. F. Pratt. 1983. An acute infection model for Mycobacterium intracellulare disease using beige mice: preliminary results. Am. Rev. Respir. Dis. 127:648-649. 9. Gangadharam, P. R. J., V. K. Perumal, D. C. Harhi, and J. LaBrecque. 1989. The beige mouse model for Mycobacterium avium complex (MAC) disease: optimal conditions for the host and parasite. Tubercle 70:257-271. 10. Garcia-Penarrubia, P., F. T. Koster, R. 0. Kelley, T. D. McDowell, and A. D. Bankhurst. Antibacterial activity of human natural killer cells. J. Exp. Med. 169:99-113. 11. Gordon, J. H. N., and R. Kajioka. 1977. Permeability barrier to rifampin in mycobacteria. Antimicrob. Agents Chemother. 11: 773-779. 12. Goto, Y., R. Nakamura, H. Takahashi, and T. Tokunaga. 1984. Genetic control of resistance to Mycobacterium intracellulare infection in mice. Infect. Immun. 46:135-140. 13. Guthertz, L. S., B. Damsker, E. J. Bottone, E. G. Ford, T. F. Midura, and J. M. Janda. 1989. Mycobacterium avium and Mycobacterium intracellulare infections in patients with and without AIDS. J. Infect. Dis. 160:1037-1041. 14. Harris, G. D., W. G. Johanson, Jr., and D. P. Nicholson. 1975. Response to chemotherapy of pulmonary infection due to Mycobacterium kansasii. Am. Rev. Respir. Dis. 112:31-36. 15. Hatcher, F. M., and R. E. Kuhn. 1982. Destruction of Trypanosoma cruzi by natural killer cells. Science 218:295-296. 16. Heifets, L. B. 1982. Synergistic effect of rifampin, streptomycin, ethionamide, and ethambutol on Mycobacterium intracellulare. Am. Rev. Respir. Dis. 125:43-48. 17. Heifets, L. B., M. D. Iseman, and P. J. Lindholm-Levy. 1988. Combination of rifampin or rifabutin plus ethambutol against Mycobacterium avium complex. Am. Rev. Respir. Dis. 137: 711-715. 18. Heifets, L. B., P. J. Lindholm-Levy, and M. A. Flory. 1990. Bactericidal activity in vitro of various rifamycins against Mycobacterium avium and Mycobacterium tuberculosis. Am. Rev. Respir. Dis. 141:626-630. 19. Hidore, M. R., and J. W. Murphy. 1985. Correlation of natural killer cell activity and clearance of Cryptococcus neoformans from mice after adoptive transfer of splenic nylon wool-nonadherent cells. Infect. Immun. 51:547-555. 20. Horsburgh, C. R., Jr., D. L. Cohn, R. B. Roberts, H. Masur, R. A. Miller, A. Y. Tsang, and M. D. Iseman. 1986. Mycobacterium avium-M. intracellulare isolates from patients with or without acquired immunodeficiency syndrome. Antimicrob. Agents Chemother. 30:955-957. 21. Kaufmann, S. H. E. 1989. In vitro analysis of the cellular mechanisms involved in immunity to tuberculosis. Rev. Infect. Dis. ll:S448-S454. 22. Kaufmann, S. H. E. 1990. Interleukins, mycobacteria, and listeriae. Diagn. Microbiol. Infect. Dis. 13:429-433. 23. Kuze, F., T. Yamamoto, R. Amitani, and K. Suzuki. 1990. In vivo activities of new rifamycin derivatives against mycobacteria. Kekkaku 66:7-12. (In Japanese.) 24. Pezzia, W., J. W. Raleigh, M. C. Bailey, E. A. Toth, and J. Silverblatt. 1981. Treatment of pulmonary disease due to Mycobacterium kansasii: recent experience with rifampin. Rev. Infect. Dis. 3:1035-1039. 25. Rastogi, N., C. Frehel, A. Ryter, H. Ohayon, M. Lesourd, and H. L. David. 1981. Multiple drug resistance in Mycobacterium avium: is the wall architecture responsible for the exclusion of antimicrobial agents? Antimicrob. Agents Chemother. 20:666677. 26. Roder, J. C., and A. Duwe. 1979. The beige mutation in the
VOL. 36, 1992 selectively impairs natural killer cell function. Nature (London) 278:451-453. Rosenzweig, D. Y. 1976. Course and longterm follow-up of 100 cases of pulmonary infection due to Mycobacterium aviumintracellulare complex (M. a-i). Am. Rev. Respir. Dis. 113(Suppl.):55. Rynearson, T. K., J. S. Shronts, and W. Wolinsky. 1971. Rifampin: in vitro effect on atypical mycobacteria. Am. Rev. Respir. Dis. 104:272-274. Saito, H., K. Sato, and H. Tomioka. 1988. Comparative in vitro and in vivo activity of rifabutin and rifampicin against Mycobacterium avium complex. Tubercle 69:187-192. Saito, H., and H. Tomioka. 1990. The role of macrophages in host defence mechanisms against Mycobacterium avium complex infection induced in mice. Res. Microbiol. 141:206-212. Saito, H., H. Tomioka, K. Sato, M. Emori, T. Yamane, K. Yamashita, K. Hosoe, and T. Hidaka. 1991. In vitro antimycobacterial activities of newly synthesized benzoxazinorifamycins. Antimicrob. Agents Chemother. 35:542-547. Saito, H., H. Tomioka, K. Sato, H. Tasaka, M. Tsukamura, F. Kuze, and K. Asano. 1989. Identification and partial characterization of Mycobacterium avium and Mycobacterium intracelmouse
27.
28. 29. 30. 31.
32.
ANTI-MAC ACTIVITY OF BENZOXAZINORIFAMYCIN
393
lulare by using DNA probes. J. Clin. Microbiol. 27:994-997. 33. Saxena, R. K., Q. B. Saxena, and W. H. Adler. 1982. Defective T-cell response in beige mice. Nature (London) 295:240-241. 34. Sheliam, G. R., J. E. Allan, J. M. Papadimitriou, and G. J. Bancroft. 1981. Increased susceptibility to cytomegalovirus infection in beige mutant mice. Proc. Natl. Acad. Sci. USA 78:5104-5108. 34a.Tomioka, H., et al. Kekkaku, in press. 35. Tomioka, H., K. Sato, H. Saito, and Y. Yamada. 1989. Susceptibility of Mycobacterium avium and Mycobacterium intracellulare to various antibacterial drugs. Microbiol. Immunol. 33: 509-514. 36. Wolinsky, E. 1979. Nontuberculous mycobacteria and associated disease. Am. Rev. Respir. Dis. 119:107-159. 37. Yamamoto, T., R. Amitani, F. Kuze, and K. Suzuki. 1990. In vitro activities of new rifamycin derivatives against Mycobacterium tuberculosis and M. avium complex. Kekkaku 65:805810. (In Japanese.) 38. Young, L. S., C. Inderlied, 0. Berlin, and M. S. Gottlieb. 1986. Mycobacterial infections in AIDS patients, with an emphasis on the Mycobacterium avium complex. Rev. Infect. Dis. 8:10241033.
p. 387-393
Vol. 36, No. 2
Chemotherapeutic Efficacy of a Newly Synthesized Benzoxazinorifamycin, KRM-1648, against Mycobacterium avium Complex Infection Induced in Mice H. TOMIOKA,1 H. SAITO,'* K. SATO,1 T. YAMANE,2 K. YAMASHITA,2 K. HOSOE,2 K. FUJII,2 AND T. HIDAKA2 Department of Microbiology and Immunology, Shimane Medical University, Izumo 693,1 and Biochemical Research Laboratories, KANEKA Corporation, Takasago 676,2 Japan Received 27 September 1991/Accepted 3 December 1991
Newly synthesized benzoxazinorifamycin, KRM-1648, was studied for its in vivo anti-Mycobacterium avium complex (MAC) activities. When the MICs were determined by the agar dilution method with Middlebrook 7H11 agar medium, KRM-1648 exhibited similarly potent in vitro antimicrobial activities against the MAC isolated from AIDS and non-AIDS patients, indicating possible usefulness of KRM-1648 against AIDSassociated MAC infections. KRM-1648 exhibited potent therapeutic activity against experimental murine infections induced by M. intracellulare N-260 (virulent strain) and N-478, which has much weaker virulence. Similarly, KRM-1648 exhibited an excellent therapeutic efficacy against M. intracelulare infection induced in NK-cell-deficient beige mice (as a plausible model for AIDS-associated MAC infection), in which a much more progressed state of gross lesions and bacterial loads at the sites of infection were observed. When the infected beige mice were killed at weeks 4 and 8, obvious therapeutic efficacy was seen on the basis of reduction in the incidence and degree of lung lesions and bacterial loads in the lungs and spleen with infections due to M. intracelulare N-241, N-256, and N-260. In this case, the efficacy was the highest in N-260 infection, followed by strain N-241. When mice were observed until infection-induced death, survival time of the infected beige mice was found to be prolonged by KRM treatment. However, KRM-1648 was not efficacious in suppressing the progression of pulmonary lesions and the increase in bacterial loads at the sites of infection, including lungs and spleen, at the late phase of infection. This may imply some difficulty with chemotherapy for AIDSassociated MAC infection, even with KRM-1648 treatment, which has excellent in vitro and in vivo anti-MAC activities, as shown in the present study.
Rifampin (RMP) and other rifarmycins, including rifabutin (RBT), rifapentine, and so on, are highly active against slowly growing mycobacteria, particularly Mycobacterium tuberculosis and some nontuberculous mycobacteria (1, 3, 5, 7, 14, 18, 24, 28, 29, 36). However, they are not so excellent in their in vitro activity against M. avium complex (MAC) (5, 19, 28), supposedly because of the permeability barrier of the organisms (11, 25). Moreover, it is not known whether RMP and other rifamycins are really efficacious in the treatment of patients with MAC infections. Some investigators have reported that multidrug chemotherapy, including treatment with RMP, has been used in many cases of clinical management of AIDS- or non-AIDS-associated MAC patients with success (4, 6). In contrast, others have reported unavoidable difficulty in clinical control of MAC, even by using RMP combined with other drugs (36). For instance, Rosenzweig (27) reported that sustained conversion of sputum with such chemotherapy alone was achieved in only 41% of patients and, after a 6-month trial of drugs including RMP, 33% of the patients experienced relapse. Because MAC infections are rapidly increasing in immunocompromised hosts, especially in AIDS patients (38), new rifamycin derivatives having strong anti-MAC activity are desired. In relation, it is noteworthy that a substantial proportion of M. avium strains are considerably susceptible to RMP and RBT (5, 18), and there is a synergistic effect of these rifamycins with ethambutol (16, 17). Such a new *
rifamycin with a high in vitro anti-MAC activity and superior pharmacokinetics may be able to exhibit an excellent therapeutic efficacy against the MAC infections, when used in combination with other drugs, especially ethambutol. In a recent study (31), we found that newly synthesized rifamycin derivatives, benzoxazinorifamycins (KRMs), with chemical structures of 3'-hydroxy-5'-(4-alkylpiperazinyl)benzoxazinorifamycins (alkyl residues: isobutyl, propyl, sec-butyl, sec-butyl [R configuration], and sec-butyl [S configuration] for KRM-1648, -1657, -1668, -1686, and -1687, respectively), exhibited markedly stronger in vitro antimycobacterial activities, especially against the MAC. Their MICs for 50 and 90% of the strains against the MAC were about 64 times smaller than those of RMP. Moreover, KRMs, especially KRM-1648, exhibited higher antimicrobial activity against the MAC phagocytosed in macrophages than did RMP and RBT. Similar results were reported by Yamamoto et al. (37). In this study, KRM-1648 was studied for its in vivo activity against MAC infections induced in mice. MATERIALS AND METHODS
Organisms. For in vitro study, 77 strains of M. avium and 11 strains of M. intracellulare isolated from AIDS patients in the United States and 62 strains of M. avium and 49 strains of M. intracellulare isolated from non-AIDS patients in Japan were used. For in vivo studies, M. intracellulare N-241, N-256, N-260, and N-478 were used. All of the MAC strains were identified with the Rapid Diagnostic System for the MAC (Gen-Probe Inc., San Diego, Calif.), and they
Corresponding author. 387
388
ANTIMICROB. AGENTS CHEMOTHER.
TOMIOKA ET AL.
formed smooth, transparent, irregularly shaped colonies on 7H11 agar plates. Mice. Five-week-old female BALB/c (MAC-susceptible strain) (12), natural killer (NK)-cell-deficient beige (C57BL/6 bgibg) (26) or Std-ddY mice were used. Drugs. KRM-1648, -1657, -1668, -1686, and -1687, synthesized by Biochemical Research Laboratories, KANEKA Corp., Takasago, Japan, were used. The other rifamycins, RMP and RBT, were kind gifts from Daiichi Pharmaceutical Co., Tokyo, Japan, and Farmitalia Carlo Erba Research Laboratories, Milan, Italy, respectively. In vitro antimicrobial activity. Test organisms were cultured in 7H9 broth containing 0.05% Tween 80 at 37°C for 3 to 7 days until its optical density (540 nm, 1.5-cm light path) reached about 0.2. Test bacterial suspension was prepared by diluting the 7H9 broth culture with 0.1% Tween 80 in saline to give an optical density of 0.1 at 540 nm (approximately 107 CFU/ml) as measured by a spectrophotometer (Model 100-10; Hitachi Co. Ltd., Hitachi, Japan). The bacterial suspension was then diluted 10 times in the saline to prepare an inoculum. MIC was determined by the agar dilution method as previously described (35). Briefly, the inoculum (5 pAl) was spotted on 7H11 agar plates containing 100 -0.0125-,ug/ml doses (twofold dilution) of drugs. The MICs of the drugs were determined 14 days after cultivation at 37°C in a CO2 incubator (5% C02-95% humidified air). The MICs were read as minimum concentrations of drugs completely inhibiting the growth of organisms or allowing no more than five colonies to grow. In vivo activity. M. intracellulare was used to induce the MAC infection in mice, because its virulence to mice is significantly higher than that of M. avium (34a.). M. intracellulare N-241, N-256, N-260, or N-478 was cultured in 7H9 broth containing 0.05% Tween 80 at 37°C until its optical density (540 nm, 1.5-cm light path) reached about 0.15. The bacterial suspension was gently sonicated with Handy Sonic (Model 20P; Tomy Seiko, Co., Tokyo, Japan) and centrifuged at 1,000 rpm for 5 min to remove large bacterial clumps. The upper layer was then diluted with 7H9 medium to give an optical density of 0.1 at 540 nm and used as an inoculum for experimental infection. Mice were infected intravenously with the organisms at doses from 7.6 x 106 to 2.7 x 107 per mouse (see footnote a of each table for details). The animals were given either KRM, RMP, or RBT, finely emulsified in 2.5% gum arabic-0.2% Tween 80, by gavage once daily six times per week from day 1 to the end of the experiment. Therapeutic efficacy of a test drug was judged on the basis of the reduction of mortality of infected animals, incidence of gross lesions in the visceral organs, and bacterial load at the sites of infection, especially in the lungs and spleen. The number of CFUs in the visceral organs was counted by inoculating serial 10-fold dilutions of tissue homogenates with saline prepared by using a glass homogenizer onto 7H11 agar plates and cultivating them at 37°C for 2 weeks. The tissue homogenates were decontaminated with 0.2% NaOH for 20 s and then immediately neutralized with 0.5 N HCl, prior to the serial dilution. Distribution of drug in tissue. KRM-1648, RMP,
or
RBT
finely emulsified in 2.5% gum arabic-0.2% Tween 80 solution was given to male Std-ddY mice by gavage at the dose of 20 mg/kg of body weight. At intervals, the drug concentration in the plasma, lungs, and spleen was measured as follows. The visceral organs were homogenized in acetate buffer (pH 4.0) (for KRM-1648), phosphate-buffered saline (for RBT), or 2% sodium ascorbate in phosphate-buffered saline (for RMP), and then the homogenate or plasma was extracted with
acetone (for KRM-1648 and RBT) or methanol (for RMP). The resultant extract was condensed by subjecting Bond Elute C18 or C8 cartridge (Analytichem International Co., Harbor City, Calif.) and analyzed for concentration of each rifamycin by high-performance liquid chromatography. RESULTS In vitro activity of KRM-1648 against the MAC isolated from AIDS and non-AIDS patients. In our previous study, in vitro activity of KRM-1648 against MAC was examined by using only the strains isolated from non-AIDS patients. Therefore, it is unclear whether KRM-1648 is similarly active against MAC strains from AIDS patients and those from non-AIDS patients. Here, we determined the MICs of KRM-1648, RMP, and RBT against the MAC strains isolated from AIDS (M. avium, 77 strains; M. intracellulare, 11 strains) and non-AIDS (M. avium, 62 strains; M. intracellulare, 49 strains) patients. KRM-1648 was similarly efficacious against the MAC strains isolated from both AIDS and non-AIDS patients, and its MIC for 90% of strains tested (0.05 to 0.1 j,g/ml) was much smaller than those of RMP (3.13 to 25 ,ug/ml) and RBT (0.4 to 1.6 jig/ml). Therapeutic efficacy of KRM-1648 and other KRMs against the murine infections with M. intracellulare N-260 induced in BALB/c and beige mice. Table 1 shows the therapeutic activities of various KRMs against infection with M. intracellulare N-260 induced in BALB/c mice (experiment 1) and the efficacy of KRM-1648 against the N-260 infection induced in beige mice (experiment 2). Firstly, in experiment 1, when infected mice were given each of the five KRMs at doses of 0.2 and 0.4 mg per mouse, the incidence of gross lung lesions was completely inhibited, accompanied by a 1.9to 3.1-log decrease in the number of CFUs of the organisms in the lungs at week 8 compared with those for untreated animals, although such a marked reduction in the number of CFUs was not observed for the spleen. Moreover, infectioninduced splenomegaly (2.4-fold increase in spleen weight at week 8 for untreated mice) was also markedly suppressed; there was 70 to 85% reduction by the five KRMs at the dose of 0.4 mg per mouse compared with that for untreated control mice infected with M. intracellulare. In this experiment, KRM-1648, followed by KRM-1668, was most efficacious in controlling M. intracellulare infection in terms of inhibiting the increase in the number of CFUs in the lungs. It has been reported that NK-cell-deficient beige mice are much more susceptible to MAC infection (8, 9). As indicated in Table 1 (experiment 2), when the beige mice were infected with M. intracellulare N-260, much faster bacterial growth in the lungs and spleen was seen compared with those for BALB/c (experiment 1) and C57BL/6 (30) mice with a normal genotype (+/+). In this experiment, when KRM1648 was given at a dose of either 0.2 or 0.4 mg per mouse and the animals were killed at weeks 4 and 8, the agent efficiently reduced the incidence or degree of lung lesions or both and the growth of the organisms in the visceral organs. However, it was noted that KRM-1648 was unable to eliminate the organisms at the sites of infection, even when given at a dose of 0.4 mg per mouse, in the present administration protocol. Although a transient decrease in the number of CFUs in the spleen as well as lungs compared with the number of CFUs on day 1 was observed for mice given KRM-1648 at a dose of 0.4 mg per mouse in the relatively early phase of infection, such as at week 4, an increase in the number of organisms was seen at week 8, despite continuous KRM treatment. A similar tendency was also reported by
389
ANTI-MAC ACTIVITY OF BENZOXAZINORIFAMYCIN
VOL. 36, 1992
TABLE 1. Therapeutic efficacy of various KRMs against MAC infection with M. intracellulare N-260 induced in BALB/c (experiment 1) and beige (experiment 2) mice' No. of mice with indicated lung lesion score atb:
Experiment and drug
Expt 1 None KRM-1648
Dose (mg)
No. of None
1+
2+
3+
0 0 0 0 0
0 0 0 0 0
0 0 0 0 0
14 14 14 14 14 7 7 7
7 7 7 7 7
KRM-1668 KRM-1686 KRM-1687
0.2 0.4 0.2 0.4 0.4 0.4 0.4
Expt 2 None KRM-1648 KRM-1648
0.2 0.4
10 10 10
0 2 3
KRM-1657
0 1 2
5 2 0
Spleen
Lungs
8 weeks
4 weeks
mice
Log CFUC
None
1+
2+
3+
4+
4 weeks
8 weeks
4 weeks
8 weeks
0 7 7 7 7 7 7 7
0 0 0 0 0 0 0 0
7 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0
5.53 3.74 3.49
6.54 6.09 5.81
NCd NC NC
6.71 4.86 3.66 4.90 4.58 4.15 4.34 4.80
NC
7.00 6.77 6.44 6.78 6.66 6.50 6.56 6.51
0 0 0
0 0 0
0 1 4
0 4 1
5 0 0
7.09 5.21 4.48
8.02 6.34 5.47
7.14 6.12 5.69
8.36 6.90 6.48
0 0 0
3.96 3.53
6.17 6.08 NC NC
Mice were infected with 7.6 x 106 (experiment 1) or 1.5 x 107 (experiment 2) CFU of M. intracellulare N-260. Symbols: none, no macroscopic lesions; 1+, less than 20 small nodules; 2+, more than 20 small nodules; 3+, many small nodules with a few large nodules (diameter, _2 mm). c The mean values for seven (experiment 1) or five (experiment 2) mice are given. For experiment 1, the average standard error of the mean was 0.08 and did not exceed 0.22; for experiment 2, the average standard error of the mean was 0.12 and did not exceed 0.26. All the log CFU values for KRM-treated mice were significantly lower than those for untreated control mice (P < 0.01, Student's t test). At day 1, the values were 4.72 (lungs) and 6.01 log CFU (spleen) for experiment 1 and 4.43 (lungs) and 6.12 (spleen) log CFU for experiment 2 (n = 5). d NC, not counted. a
b
in mice given KRM was markedly lower than the degrees seen with animals given RMP or RBT. A statistically significant difference in the number of CFUs in the lungs and spleens for mice given KRM and those given either RMP or RBT was observed (P < 0.05, Student's t test). Thus, the therapeutic efficacy of KRM is thought to be much greater than those of RMP and RBT. Therapeutic activity of KRM-1648 against the murine infections with M. intracellulare N-478 induced in BALB/c mice. As shown in Table 3, infection induced in BALB/c mice by M. intracellulare N-478, which is less virulent to mice than N-260, was also efficiently controlled by KRM-1648, whose MIC against strain N-478 was 0.2 ,ug/ml (four times higher than that against strain N-260). No lung lesions were noted in any of the test mice, even at week 8. In this case, a 1.0- to 1.6-log decrease in the number of CFUs of organisms in the lungs of mice treated with KRM-1648 at doses of 0.2 to 0.8 mg per mouse was noted at week 8 after infection. In this case, a 2.3- and 1.8-fold increase in spleen weight for
Kuze et al. (23) in murine experimental infection induced by another MAC strain, M. intracellulare 31F093T. Table 2 compares the in vivo activity of KRM-1648 against MAC infection induced in beige mice with those of RMP and RBT. KRM completely inhibited the incidence of gross pulmonary lesions in MAC-infected beige mice at week 6 and caused a much lower degree of lung lesions at week 12 compared with those in untreated mice. Its efficacy to suppress the progression of lung lesions was much more obvious than those of RBT and RMP. MAC infectioninduced splenomegaly (11-fold increase in spleen weight for untreated mice) was also suppressed by treatment with these rifamycins. KRM-1648, RMP, and RBT caused, respectively, 93, 44, and 51% inhibition of the splenomegaly seen with untreated animals at week 12. KRM-1648 markedly suppressed the bacterial growth in the lungs and spleens of beige mice, and about 2- and 3-log reductions were seen at weeks 6 and 12, respectively, compared with those of untreated control mice. Moreover, the degree of lung lesions
Drug
None RMP RBT KRM
TABLE 2. Therapeutic efficacy of KRM-1648 against MAC infection with M. intracellulare N-260 induced in beige micea Log CFUC No. of mice with indicated lung lesion score atb: Dose No. of 12 weeks Spleen Lungs 6 weeks mice (mg)
0.4 0.4 0.4
13 13 13 13
None
1+
2+
3+
None
1+
2+
3+
4+
6 weeks
12 weeks
6 weeks
12 weeks
0 0 0 6
0 0 0 0
6 6 6 0
0 0 0 0
0 0 0 0
0 0 0 0
0 0 0
0 0 3
7 7 4
7.07
7
0
0
8.80 8.18d 8.46e 5.41d
7.75 7.40d 7.49 6.07d
9.17 8.39d 8.81 6.55d
6.84 6.68 4.84d
weizb infected with 8.7 x 106 CFU of M. intracellulare N-260. Lung lesions were scored as indicated in Table 1, footnote b. 4+, gross lesions with many large nodules. The mean values for six or seven mice are given. The average standard error of the mean was 0.10 and did not exceed 0.17. At day 1, the values were 4.87 (lungs) and 6.32 (spleen) log CFU. All the values (log CFU) for KRM-treated mice were significantly lower than those for either RMP- or RBT-treated mice (P < 0.01). d Significantly different from values obtained with untreated control mice (P < 0.01). e Significantly different from values obtained with untreated control mice (P < 0.05). a Beige mice b
c
390
ANTIMICROB. AGENTS CHEMOTHER.
TOMIOKA ET AL.-
TABLE 5. Therapeutic efficacy of KRM-1648 against MAC infection with M. intracellulare N-256 induced in beige mice: data for individual dead micea
TABLE 3. Therapeutic efficacy of various KRM-1648 doses against MAC infection with M. intracellulare N-478 induced in BALB/c micea Log CFUC
No. of mice with
KRM-1648 dose (mg)
score at 8b
None 1+
0 0.2 0.4 0.8
5 5 5 5
0 0 0 0
Spleen
Lungs
3+ 4 weeks 8 weeks 4 weeks 8 weeks
2+
0 0 0 0
0 0 0 0
4.41 3.48 3.34e 3.34
KRM-1648 dose (mg) dose (mg
Mouse Mouse no.
indicated lung lesion
5.03 4.07 4.04d 3.46e
6.81 6.41d 6.35d 6.07d
6.44
5.88d 5.84d 5.73d
a Mice were infected with 9.4 x 106 CFU of M. intracellulare N-478. lesions were scored as indicated in Table 1, footnote b. c The mean values for five mice are given. The average standard error of the mean was 0.18 and did not exceed 0.47. At day 1, values were 4.52 (lungs) and 5.83 (spleen) log CFU. d Significantly different from values obtained with untreated control mice (P < 0.01). e Significantly different from values obtained with untreated control mice (P < 0.05). b Lung
untreated mice at weeks 4 and 8 after infection, respectively, was seen. KRM given at doses of 0.2, 0.4, and 0.8 mg per mouse inhibited infection-induced splenomegaly by a 63 to 68% and a 50 to 53% ratio at weeks 4 and 8, respectively, compared with those for untreated control mice infected with the N-478 strain. Therapeutic activity of KRM-1648 against the murine infections with M. intracellulare N-256 and N-241 induced in beige mice. When therapeutic activity of KRM-1648 was examined for infection induced in beige mice by other virulent strains (Table 4), M. intracellulare N-256 (experiment 1) and N-241 (experiment 2), which have the same in vitro susceptibility to KRM-1648 (MIC, 0.05 p.g/ml), a comparable but somewhat weaker efficacy was observed compared with that of M. intracellulare N-260 infection, based on the incidence of pulmonary lesions and the number of CFUs of organisms recovered from the lungs and spleen at weeks 4 and 8. Table 5 shows the survival time, the degree of lung lesions, and the number of CFUs in the lungs and spleens at the deaths of individual animals due to N-256 infection. The mean survival time of infected animals was prolonged from
Log CFUc
Survival time (days)6
Degree of lung lesions
Lungs
Spleen
9.41
1 2 3 4 5
0 0 0 0 0
76 73 75 73 73
3+ 3+ 3+ 3+ 3+
9.62 9.17 9.48 9.38 9.33
9.60 9.53
6 7 8 9 10
0.2 0.2 0.2 0.2 0.2
73 73 75 73 75
3+ 4+ 3+ 3+ 4+
9.10 9.01
9.09
8.96 9.11 9.07
9.32 9.01
9.44 9.33
11 12 13 14 15
0.4 0.4 0.4 0.4 0.4
136 140 140 136 132
3+
10.40
10.04
4+ 3+ 4+ 3+
10.32 10.28 10.50 10.55
9.91 10.41 10.31 9.34
9.15 9.29
a The data for animals found dead due to strain N-256 infection at the dose of 1.3 x 107 CFU per mouse are indicated. I The mean survival times (days) were 74.0 for mice 1 to 5, 73.8 for mice 6 to 10, and 137 for mice 11 to 15. c Mean + standard error of the mean for lungs and spleen, respectively: 9.40 ± 0.08 and 9.40 ± 0.08 for mice 1 to 5, 9.11 + 0.06 and 9.18 ± 0.09 for mice 6 to 10, and 10.41 ± 0.05 and 10.00 ± 0.19 for mice 11 to 16.
74 to 137 days by KRM treatment at a dose of 0.4 mg per mouse, although KRM given at a dose of 0.2 mg per mouse failed to exhibit such an effect. Despite the fact that KRM given at a dose of 0.4 mg per mouse was able to prolong the survival time of infected beige mice, bacterial loads in the lungs and spleens of dead mice were larger for mice given KRM at this dose, compared with those for untreated control dead mice. However, it should be emphasized that KRM treatment was significantly efficacious in retarding the growth of the organisms in the visceral organs in the earlyto-middle phase of infection (Table 4). A similar situation was seen with infection due to strain
TABLE 4. Therapeutic efficacy of KRM-1648 against MAC infection with M. intracellulare N-256 (experiment 1) or N-241 (experiment 2) induced in beige micea No. of mice with indicated lung lesion score atb: KRM-1648
No. of
dose (mg)
mice
Expt 1 0
Log CFUc 4+
4 weeks
8 weeks
4 weeks
8 weeks
0
5
3
2
8.71 7.41d
8.13 7.14d
9.25 8.13d
3
0
7.68 6.02d 5.49
6.76d
6.58d
7.85d
0 3 2
5 2 0
9.10
8.17
9.54
8.62d 8.12d
6.95e
9.12d 8.12d
None
1+
2+
3+
None
1+
2+
3+
0 0 0
0 0 0
0 0 2
0 0 0
0 0 0
0 0 3
0.2 0.4
10 10 10
0 0 1
0 3 4
4 2
0
1 0 0
Expt 2 0 0.2 0.4
10 10 10
0 0 0
0 1 4
4 4 1
1 0 0
Spleen
Lungs
8 weeks
4 weeks
7.88 6.77
6.44d
6.77e
Beige mice were infected with 2.1 x 107 CFU of M. intracellulare N-256 (experiment 1) or 2.7 x 107 CFU of M. intracellulare N-241 (experiment 2). Lung lesions were scored as indicated in Table 1, footnote b. cThe mean values for five mice are given. The average standard error of the mean was 0.14 and did not exceed 0.24 for experiment 1 and was 0.14 and did not exceed 0.24 for experiment 2. At day 1, the values (log CFU) for lungs and spleen, respectively, were 4.81 and 6.48 (experiment 1) and 5.01 and 6.39 (experiment 2). d Significantly different from values obtained with untreated control mice (P < 0.01). e Significantly different from values obtained with untreated control mice (P < 0.05). a
b
ANTI-MAC ACTIVITY OF BENZOXAZINORIFAMYCIN
VOL. 36, 1992 TABLE 6. Therapeutic efficacy of KRM-1648 against MAC infection with M. intracellulare strain N-241 induced in beige mice: data for individual dead micea Mouse no.
KRM-1648 dose (mg)
Survival time (days)b
Degree of lung lesions
Lungs
Spleen
Log CFUC
1 2 3 4 5
0 0 0 0 0
111 111 109 69 70
3+ 3+ 4+ 3+ 3+
9.75 10.14 9.94 8.15 8.04
8.94 10.35 10.13 8.33 8.39
6 7 8 9 10
0.2 0.2 0.2 0.2 0.2
107 114 106 114 107
3+ 3+ 4+ 3+ 3+
9.49 10.32 8.26 9.76 10.41
9.70 10.27 8.47 10.39 9.84
11 12 13 14 15
0.4 0.4 0.4 0.4 0.4
107 125 119 114 126
3+ 3+ 4+ 4+ 4+
9.70 10.15 10.36 10.62 10.33
9.84 10.06 10.68 10.54 10.14
a The data for animals found dead due to strain N-241 infection at the dose of 1.2 x 107 CFU per mouse are indicated. b The mean survival times (days) were 94.0 for mice 1 to 5, 110 for mice 6 to 10, and 114 for mice 11 to 15. The mean values were not significantly different from those obtained with untreated control mice. c Mean ± standard error of the mean for lungs and spleen, respectively: 9.20 ± 0.46 and 9.23 ± 0.43 for mice 1 to 5, 9.65 ± 0.39 and 9.73 ± 0.34 for mice 6 to 10, and 10.23 ± 0.15 and 10.25 + 0.16 for mice 11 to 15.
N-241, as shown in Table 6. In this case, the prolongation of survival time by KRM treatment at a dose of 0.2 mg per mouse was more marked than that seen with N-256 infection, whereas the efficacy seen with mice given KRM at a dose of 0.4 mg per mouse was not as obvious as was seen with N-256 infection (Table 6 versus Table 5). In this case, the number of CFUs in the lungs and spleen at the death of the animals was also larger for mice given KRM-1648 at a dose of 0.4 mg per mouse than for untreated mice. Pharmacokinetics of KRM-1648. Table 7 compares the distribution of KRM-1648 in tissue with those of RMP and TABLE 7. Pharmacokinetics of KRM-1648: distribution of KRM-1648 in lungs, spleen, and plasma after a single oral administrationa Organ
Lungs
Spleen
(pug/g/ml) ±
Time after administration (h)
KRM-1648
RMP
RBT
1 5 8
1.10 ± 0.35b 5.27 ± 0.83 4.75 + 1.09
6.37 ± 0.64b 8.51 ± 2.10 6.87 + 0.69
9.86c
1 5 8
1.40 + 0.41 9.14 ± 0.56 7.65 ± 1.56
1.09 + 0.10 1.24 + 0.19 0.97 ± 0.09
NDd
Concn
SEM of:
3.22 1.39 ND ND
0.20 ± 0.14 9.51 ± 1.72 0.92 0.25 ± 0.05 12.73 ± 4.38 0.23 9.78 + 0.81 0.60 + 0.10 0.16 a Each drug was given to Std-ddY mice by gavage at a dose of 20 mg/kg of
Plasma
1 5 8
body weight. bn = 3. Cn = 2. d ND, not determined.
391
RBT. Although the concentration of KRM-1648 in plasma was not so sharply increased as in the case of RMP, its accumulation and retention in the spleen during 5 to 8 h after administration was much higher than that of RMP. The concentration of KRM-1648 in the lungs was somewhat lower but comparable to that of RMP and considerably higher than that of RBT. Therefore, the pharmacokinetics of KRM-1648 is comparable to that of RMP and superior to that of RBT. DISCUSSION
In this study, we found the following. First, KRM-1648 had a potent in vitro antimicrobial activity against MAC strains isolated from both AIDS and non-AIDS patients. There was no marked difference in the values against AIDS isolates and those against non-AIDS isolates. This is inconsistent with the finding by Horsburgh et al. (20) that there was a significant difference in MAC susceptibility to some antimicrobial agents, including RMP, with AIDS and nonAIDS isolates. They reported that the ratio of RMP (10 ,ug/ml)-susceptible strains in 75 non-AIDS isolates was 67%, which was significantly (P < 0.001) larger than the value (27%) for AIDS isolates (57 strains). The reason for this inconsistency is thought to be because they tested the drug susceptibility of all MAC isolates, without classifying them as M. avium or M. intracellulare. We found that the drug susceptibilities of the two species differed considerably from each other, in particular with RMP and new quinolones (32, 35). Moreover, the ratio of M. avium and M. intracellulare in AIDS isolates is markedly different from that in non-AIDS isolates (13, 20). For instance, Guthertz et al. (13) reported that 98% of the strains isolated from 44 patients with AIDS were identified as M. avium by Gen-Probe testing, while 40% of MAC isolates recovered from 109 patients without AIDS were M. intracellulare. These findings might be responsible for a certain degree of difference in the drug susceptibilities of MAC isolates from AIDS and non-AIDS patients, as they have observed. Secondly, this study revealed the superiority of the therapeutic efficacy of KRM-1648 against M. intracellulare infections induced in mice, in terms of suppression of splenomegaly, the incidence or degree of gross pulmonary lesions or both, and bacterial growth in the lungs and spleen. The pharmacokinetics of KRM-1648 was comparable to that of RMP and somewhat better than that of RBT (Table 7). Therefore, the excellent therapeutic efficacy of KRM-1648 against MAC infection seems to be due to its strong in vitro anti-MAC activity (31, 37). Thirdly, KRM-1648 exhibited therapeutic efficacy against N-260 strain infection induced in NK-cell-deficient beige mice (as a plausible model for AIDS-associated MAC infection) similar to that of a normal genotypic strain of mice. However, it is noted that in vivo activity of KRM-1648 in beige mice infected with M. intracellulare N-256 or N-241 was somewhat inferior to that seen with M. intracellulare N-260 infection, although it has the same MIC (0.05 ,ug/ml) against these three M. intracellulare strains. Although the precise reason for this is unclear, the following may be stated. First, a possible difference in the virulence of these M. intracellulare strains to mice does not seem to cause this phenomenon, because these three strains (N-260, N-256, and N-241) had similar levels of virulence to mice, showing comparable growth at the sites of infection (lungs and spleens) around weeks 8 to 12. Second, there may be more strict dependency of host defense mechanisms upon NK-cell
392
TOMIOKA ET AL.
functions with infections due to M. intracellulare strains, such as N-241 and N-256, compared with strain N-260 infection. In other words, KRM may require intact NK-cell functions to fully exhibit its in vivo antimicrobial activity, with N-241 or N-256 infection. Indeed, the possibility has been proposed that NK cells play some important role in host defense mechanisms not only in virus infections but also in other microbial infections (10, 15, 19). In relation to this, it is noteworthy that beige mice have an impaired cytotoxic T-cell activity in addition to NK-cell deficiency (2, 33). Therefore, it is also possible that KRM may need intact-cell functions of CD8+ T cells to efficiently express its in vivo activity against M. intracellulare infection. Indeed, CD8+ T cells have been proposed to play some important role in host resistance to mycobacterial infections (21), although CD4+ T cells mainly mediate the host immune response to mycobacterial infections (22). While NK cells are known to contribute to host resistance in the early phase of microbial infections (34), CD8+ T cells are generated in a relatively late phase of chronic infections, including mycobacterial infections (21). This may be a possible explanation for our finding that KRM-1648 failed to suppress the bacterial growth at the sites of infection in the progressed state of test M. intracellulare (N-241, N-256, and N-260)-induced infections, despite the fact that the agent showed an appreciable activity to inhibit bacterial growth in the early phase of infection. This can be easily explained if we assume that in MAC infections, CD8+ T cells rather than NK cells play an important role in the expression of host resistance. In any case, these situations may imply some difficulties with clinical treatment of AIDS-associated MAC infections even with KRMs, although they possess a superior in vitro activity against MAC strains isolated from both AIDS and non-AIDS patients and excellent in vivo anti-M. intracellulare activity in experimental infections in normal genotypic mice. In any case, KRM-1648 has an excellent in vivo activity against M. intracellulare infection induced in mice. Since rifamycins are known to exhibit synergistic in vitro activities against the MAC with other antimicrobial agents, especially ethambutol (16, 17), it is possible to improve the in vivo activity of KRM-1648 by combined use with other antimicrobial agents. On this point, further studies are under way. ACKNOWLEDGMENTS We thank Daiichi Pharmaceutical Co. and Farmitalia Carlo Erba Research Laboratories for the kind gifts of rifampin and rifabutin, respectively. REFERENCES 1. Arioli, V., M. Berti, G. Carniti, E. Randisi, and R. Scotti. 1981. Antimicrobial activity of DL 473, a new semisynthetic rifampicin derivative. J. Antibiot. 34:1026-1032. 2. Baca, M. E., A. M. Mowat, and D. M. V. Parrott. 1989. Immunological studies of NK cell-deficient beige mice. II. Analysis of T-lymphocyte functions in beige mice. Immunology 66:131-137. 3. Bruna, C. D., G. Schoppacassi, D. Ungheri, D. Jabes, E. Morvillo, and A. Sanfilippo. 1983. LM 427, a new spiropiperidylrifamaycin: in vitro and in vivo studies. J. Antibiot. 36:15021506. 4. Chiu, J., J. Nussbaum, S. Bozzette, J. G. Tilles, L. S. Young, J. Leedom, P. N. R. Heseltine, J. A. McCutchan, and California Collaborative Treatment Group. 1990. Treatment of disseminated Mycobacterium avium complex infection in AIDS with amikacin, ethambutol, rifampin, and ciprofloxacin. Ann. Intern. Med. 113:358-361.
ANTIMICROB. AGENTS CHEMOTHER. 5. Cynamon, M. H. 1985. Comparative in vitro activities of MDL 473, rifampin, and ansamycin against Mycobacterium intracellulare. Antimicrob. Agents Chemother. 28:440-441. 6. Davidson, P. T., V. Khanio, M. Goble, and T. S. Moulding. 1981. Treatment of disease due to Mycobacterium intracellulare. Rev. Infect. Dis. 3:1052-1059. 7. Dickinson, J. M., and D. A. Mitchison. 1990. In vitro activities against mycobacteria of two long-acting rifamycins, FCE22807 and CGP40/469A (SPA-S-565). Tubercle 71:109-115. 8. Gangadharam, P. R., C. K. Edwards III, P. S. Murthy, and P. F. Pratt. 1983. An acute infection model for Mycobacterium intracellulare disease using beige mice: preliminary results. Am. Rev. Respir. Dis. 127:648-649. 9. Gangadharam, P. R. J., V. K. Perumal, D. C. Harhi, and J. LaBrecque. 1989. The beige mouse model for Mycobacterium avium complex (MAC) disease: optimal conditions for the host and parasite. Tubercle 70:257-271. 10. Garcia-Penarrubia, P., F. T. Koster, R. 0. Kelley, T. D. McDowell, and A. D. Bankhurst. Antibacterial activity of human natural killer cells. J. Exp. Med. 169:99-113. 11. Gordon, J. H. N., and R. Kajioka. 1977. Permeability barrier to rifampin in mycobacteria. Antimicrob. Agents Chemother. 11: 773-779. 12. Goto, Y., R. Nakamura, H. Takahashi, and T. Tokunaga. 1984. Genetic control of resistance to Mycobacterium intracellulare infection in mice. Infect. Immun. 46:135-140. 13. Guthertz, L. S., B. Damsker, E. J. Bottone, E. G. Ford, T. F. Midura, and J. M. Janda. 1989. Mycobacterium avium and Mycobacterium intracellulare infections in patients with and without AIDS. J. Infect. Dis. 160:1037-1041. 14. Harris, G. D., W. G. Johanson, Jr., and D. P. Nicholson. 1975. Response to chemotherapy of pulmonary infection due to Mycobacterium kansasii. Am. Rev. Respir. Dis. 112:31-36. 15. Hatcher, F. M., and R. E. Kuhn. 1982. Destruction of Trypanosoma cruzi by natural killer cells. Science 218:295-296. 16. Heifets, L. B. 1982. Synergistic effect of rifampin, streptomycin, ethionamide, and ethambutol on Mycobacterium intracellulare. Am. Rev. Respir. Dis. 125:43-48. 17. Heifets, L. B., M. D. Iseman, and P. J. Lindholm-Levy. 1988. Combination of rifampin or rifabutin plus ethambutol against Mycobacterium avium complex. Am. Rev. Respir. Dis. 137: 711-715. 18. Heifets, L. B., P. J. Lindholm-Levy, and M. A. Flory. 1990. Bactericidal activity in vitro of various rifamycins against Mycobacterium avium and Mycobacterium tuberculosis. Am. Rev. Respir. Dis. 141:626-630. 19. Hidore, M. R., and J. W. Murphy. 1985. Correlation of natural killer cell activity and clearance of Cryptococcus neoformans from mice after adoptive transfer of splenic nylon wool-nonadherent cells. Infect. Immun. 51:547-555. 20. Horsburgh, C. R., Jr., D. L. Cohn, R. B. Roberts, H. Masur, R. A. Miller, A. Y. Tsang, and M. D. Iseman. 1986. Mycobacterium avium-M. intracellulare isolates from patients with or without acquired immunodeficiency syndrome. Antimicrob. Agents Chemother. 30:955-957. 21. Kaufmann, S. H. E. 1989. In vitro analysis of the cellular mechanisms involved in immunity to tuberculosis. Rev. Infect. Dis. ll:S448-S454. 22. Kaufmann, S. H. E. 1990. Interleukins, mycobacteria, and listeriae. Diagn. Microbiol. Infect. Dis. 13:429-433. 23. Kuze, F., T. Yamamoto, R. Amitani, and K. Suzuki. 1990. In vivo activities of new rifamycin derivatives against mycobacteria. Kekkaku 66:7-12. (In Japanese.) 24. Pezzia, W., J. W. Raleigh, M. C. Bailey, E. A. Toth, and J. Silverblatt. 1981. Treatment of pulmonary disease due to Mycobacterium kansasii: recent experience with rifampin. Rev. Infect. Dis. 3:1035-1039. 25. Rastogi, N., C. Frehel, A. Ryter, H. Ohayon, M. Lesourd, and H. L. David. 1981. Multiple drug resistance in Mycobacterium avium: is the wall architecture responsible for the exclusion of antimicrobial agents? Antimicrob. Agents Chemother. 20:666677. 26. Roder, J. C., and A. Duwe. 1979. The beige mutation in the
VOL. 36, 1992 selectively impairs natural killer cell function. Nature (London) 278:451-453. Rosenzweig, D. Y. 1976. Course and longterm follow-up of 100 cases of pulmonary infection due to Mycobacterium aviumintracellulare complex (M. a-i). Am. Rev. Respir. Dis. 113(Suppl.):55. Rynearson, T. K., J. S. Shronts, and W. Wolinsky. 1971. Rifampin: in vitro effect on atypical mycobacteria. Am. Rev. Respir. Dis. 104:272-274. Saito, H., K. Sato, and H. Tomioka. 1988. Comparative in vitro and in vivo activity of rifabutin and rifampicin against Mycobacterium avium complex. Tubercle 69:187-192. Saito, H., and H. Tomioka. 1990. The role of macrophages in host defence mechanisms against Mycobacterium avium complex infection induced in mice. Res. Microbiol. 141:206-212. Saito, H., H. Tomioka, K. Sato, M. Emori, T. Yamane, K. Yamashita, K. Hosoe, and T. Hidaka. 1991. In vitro antimycobacterial activities of newly synthesized benzoxazinorifamycins. Antimicrob. Agents Chemother. 35:542-547. Saito, H., H. Tomioka, K. Sato, H. Tasaka, M. Tsukamura, F. Kuze, and K. Asano. 1989. Identification and partial characterization of Mycobacterium avium and Mycobacterium intracelmouse
27.
28. 29. 30. 31.
32.
ANTI-MAC ACTIVITY OF BENZOXAZINORIFAMYCIN
393
lulare by using DNA probes. J. Clin. Microbiol. 27:994-997. 33. Saxena, R. K., Q. B. Saxena, and W. H. Adler. 1982. Defective T-cell response in beige mice. Nature (London) 295:240-241. 34. Sheliam, G. R., J. E. Allan, J. M. Papadimitriou, and G. J. Bancroft. 1981. Increased susceptibility to cytomegalovirus infection in beige mutant mice. Proc. Natl. Acad. Sci. USA 78:5104-5108. 34a.Tomioka, H., et al. Kekkaku, in press. 35. Tomioka, H., K. Sato, H. Saito, and Y. Yamada. 1989. Susceptibility of Mycobacterium avium and Mycobacterium intracellulare to various antibacterial drugs. Microbiol. Immunol. 33: 509-514. 36. Wolinsky, E. 1979. Nontuberculous mycobacteria and associated disease. Am. Rev. Respir. Dis. 119:107-159. 37. Yamamoto, T., R. Amitani, F. Kuze, and K. Suzuki. 1990. In vitro activities of new rifamycin derivatives against Mycobacterium tuberculosis and M. avium complex. Kekkaku 65:805810. (In Japanese.) 38. Young, L. S., C. Inderlied, 0. Berlin, and M. S. Gottlieb. 1986. Mycobacterial infections in AIDS patients, with an emphasis on the Mycobacterium avium complex. Rev. Infect. Dis. 8:10241033.
