- Email: [email protected]
In vitro activity of new rifamycins aganst rifampicin-resistant M. tuberculosis and MAIS-complex mycobacteria
Tubercle (1987) 68, 177-182 0 Longman Group UK Ltd. 1987
IN VlTRO ACTIVITY OF NEW RiFAMYClNS AGAlRtST RiFAlUPiCWRESISTANT TUBERCULOSlS AND MAIS-COIRPLEX MYCOBACTEAA
M.
Jean M. Dickinson and D.A. Mitchison Department
of Bacteriology,
Royal Postgraduate
Medical
School, Ducane Road, London W12 OHS
Summary Comparisons were made of the in vitro activity of rifampicin, and the rifamycin derivatives, rifapentine, rifabutin, CGP 29861, CGP 7040 and CGP 27557, against rifampicin-sensitive and rifampicin-resistant strains of Mycobacterium tubercdosis and against the Mycobacferium avium/intrace//u/are/scrofu/aceum (MAIS) complex. The new rifamycins had MlCs four to eight times lower than those of rifampicin against sensitive M. tuberculosis strains. Of the 35 rifampicin-resistant strains of M. tubercu/osis, 31 % were sensitive to rifabutin but only 3-11 % to the other rifamycins. The proportions of the MAIS strains found to be sensitive were 35 % for rifampicin, 50-60 % for CGP 27557, rifapentine and rifabutin and 85-92 % for CGP 29861 and CGP 7040. R&urn6 L’auteur a comparit I’activite in vitro de la rifampicine et des d&iv& de la rifamycine-rifapentine, rifabutine, CGP 29 861, CGP 7040 et CGP 27 557-vis-&vis des souches de Mycobacterium tuberculosis sensibles ti la rifampicine et resistantes ti la rifampicine et vis-8-vis du complexe Mycobacterium avium/intrace//u/are/ scrofulaceum (MAIS). Les nouvelles rifamycines avient des CIM 4 B 8 fois plus basses que celle de la rifampicine vis-ti-vis des souches sensibles de M. tuberculosis. Parmi 35 souches de M. tuberculosis rksistantes ZIla rifampicine, 31 % Btaient sensibles B la rifabutine, mais seulement 3 6 11 % aux autres rifamycines. Les proportions des souches du complexe MAIS trouvdes sensibles ont &!tBde 35 % en ce qui concernait la rifampicine, 50-69 % pour CPG 27 557, rifapentine et rifabutine et 85-92 % pour CGP 29 861 et CGP 7040. Resumen Se cornpar la actividad in vitro de la rifampicina y de derivados de la rifamicina rifapentina, rifabutina, CGP 29861, CGP 7040 y CGP 27557- contra cepas de Mycobacferium tuberculosis resistentes a la rifampicina y sensibles a la rifampicina y contra el complejo Mycobacterium avium/intrace//u/are/scrofu/eceum (MAIS). Las nueuvas rifamicinas tenian MlCs 4 a 8 veces m6s bajas que la de la rifampicina contra cepas sensibles de M. tuberculosis. De un total de 35 cepas resistentes a la rifampicina de M. tuberculosis, el31 % eran sensibles a la rifabutina, pero ~610 3-l 1 % a las otras rifamicinas. Las proporciones de cepas de MAIS que resultaron sensibles fueron: 35 % para la rifampicina, 50-60 % para CGP 27557, rifapentina y rifabutina y 85-92 % para CGP 29861 y CGP 7040. Introduction Rifampicin
is a major antibiotic
in the treatment
of tuberculosis,
leprosy
and some other
178
Dickinson
and
Mitchison
mycobacterial infections. It is not surprising that attempts have been made to tailor the basic rifamycin molecules to produce semi-synthetic drugs with properties potentially advantageous over those of rifampicin. Information on six new semi-synthetic rifamycins is available. The in vitro characteristics, pharmacokinetics and activity in experimental murine tuberculosis of rifabutin, a spiropiperidyl rifamycin also known as LM 427 and ansamycin (a misnomer since other antibiotics belong to the broad class of ansamycins) have been described [I, 21, as have the corresponding properties of rifapentine, a cyclopentyl rifamycin also known as DL 473 and MDL [3,41, and the long-acting rifamycin derivatives CGP 29861, CGP 7040, CGP 27557 [5,6,7,8] and FCE 22250 [9, IO]. Claims have been made that rifabutin has in vitro activity against rifampicin-resistant strains of Mycobacterium tuberculosis because of a mode of action considered to be different to that of rifampicin [I 1, 12, 13, 141. Rifabutin has also been widely used on an individual patient basis in the United States for the treatment of patients with acquired immune deficiency syndrome (AIDS) who have become infected with organisms of the Mycobacterium avium/intrace/lulare/scrofu/aceum (MAIS) complex [9, 11, 141. We describe here in vitro studies carried out to compare five of these six new rifamycins and rifampicin in their activities against 1. rifampicin-resistant M. tuberculosis and 2. strains of the MAIS complex.
Materials
and methods
Rifamycins The six rifamycins,
the abbreviation
Rifamycin
Abbreviation
Rifampicin Rifabutin Rifapentine CGP 29861 CGP 7040 CGP 27557
RMP RBU RPE c29 c70 C27
for them
Alternative
used
here and their
alternative
names
are:
names
LM 427, ansamycin DL 473, MDL 473
Mycobacteria The strains of M. tuberculosis consisted of 10 drug-sensitive strains obtained pre-treatment (eight from Hong Kong patients, one from a British patient and strain H37Rv) and 48 strains used as controls for sensitivity tests in our laboratory [15], many of which were resistant to rifampicin. Most of these strains were obtained during unsuccessful treatment with a some strains being isolated about a month before the rifampicin-containing regimen, emergence of rifampicin resistance, some at the time that resistance was first detected and some a few months after the emergence of resistance. The 26 MAIS-complex strains consisted of a single strain from each of the serovars except 17 and 18, and were obtained from Dr P. A. Jenkins, Mycobacterium Reference Laboratory, Public Health Laboratory Service, Cardiff, South Wales. Minimal
inhibitory
concentrations
The minimal inhibitory concentration (MIC) of the six drugs against all strains of M. tuberculosis was estimated by growing the strains in 7HS medium with Tween and dropping 0.03 ml, containing about 500 colony forming units, onto sectors of plates of 7HlO
Activity of new rifamycins
179
oleic acid albumin agar made selective by the addition of carbenicillin 100 mg/l, polymyxin B 200 units/ml, trimethoprim 20 mg/l and amphotericin B 10 mg/l (161 and containing serial 2-fold dilutions of the rifamycins rangingsfrom 0.04-10 mg/l with a rifamycin-free control plate. The end-point was the lowest concentration showing no growth after incubation at 37 “C for 28 days. The strains of M. tuberculosis had also been tested during the preceding 6 months for sensitivity to rifampicin in standard tests on Lowenstein-Jensen medium 1171. The MAIS-complex strains were first grown in 7H9 medium with Tween and, when fully grown, 0.03 ml from a calibrated dropping pipette was added to 5 ml volumes of 7H9 liquid medium without Tween containing serial 2-fold dilutions of the rifamycins. The end-point was the lowest concentration inhibiting any growth after incubation for 14 days. The 7HlO and 7H9 media were obtained from Difco Laboratories, E Molesey, Surrey. Results
M. tuberculosis When tested in duplicate standard Lowenstein-Jensen medium sensitivity tests, 23 strains of M. tuberculosis were found to be sensitive to RMP. The MlCs of all six rifamycins obtained on 7HlO plates for these sensitive strains are shown in Table I. No differences were found between the 10 strains obtained pre-treatment and the remaining 13 strains obtained during treatment. Although the mean MIC could not be calculated because the concentrations used did not extend sufficiently low, it is evident that all five new rifamycins had lower MlCs than RMP, particularly RBU, C29 and C70 whose distributions suggested that they were about eight times more active. The lowest concentration of RMP that inhibited all strains was 1.25 mg/l, which was therefore taken as the cut-off concentration for sensitive strains. The MlCs for the remaining 35 RMP-resistant strains are set out in Table II. Using the definition obtained from Table I, a strain was considered as sensitive to a rifamycin if it had an MIC of less than 1.25 mg/l. Of the 11 strains classified as sensitive to RBU, one strain had low MlCs that fell within the distributions of MlCs for sensitive strains for each of the rifamycins (Table I) except that its MlCs for C29 and C70 were just too high. The remaining 10 strains fell into the lower portions of the distributions of each of the rifamycins. Thus it seems that there was a group of 11 rifampicin-resistant strains, comprising about one-third of those we tested, which had lower MlCs to any rifamycin, including RMP, than the remaining strains. These strains were appreciably more sensitive to RBU that to the other
Table I.
Minimal
inhibitory concentrations
of rifamycins against 23 sensitive strains of
M. tuberculosis. M/C
Rifam
ycin
(md,) >2.5 2.5 1.25 0.6 0.3 0.16 0.08 0.04
or less
RMP
RPE
RBU
C29
C70
C27
0 0 0 1 14 6 0 2
0 0 0 0 0 2 15 6
0 0 0 0 0 0 2 21
0 0 0 0 0 0 0 23
0 0 0 0 0 0 1 22
0 0 0 0 0 4 10 9
180
Dickinson
and Mitchison
Table II. Minimal inhibitory strains of M. tuberculosis.
concentrations
M/C
of rifamycins
against
35 rifampicin-resistant
Rifamycin
Img/ll
RMP >I0 10
5 2.5 1.25 0.6 0.3 0.16 0.08 0.04
19
6 2 2 5 0
or less
Sensitive (MIC cl.25 No.
(1) (2) (2) (5)
RPE
RBlJ
c29
c70
07
20 4 (1) 4 (3) 0 5 (5)
2 18 4 0 0 4 (41
18 3 5 2 4 2
10 5 2 8 (3) 6 (4) 3 (3)
19 6 (2) 5 (4) 0 4 14) 0
1 (1)
l(1’) 0 0 0
0 l(1’) 0 0
5 (5) 0 2 (2’) 0
1
2
11
3
3
6
31
9
(2) (2) (4) (2)
0 0 ill”) 0
0 1 (1’) 0 0
0 1 (I’) 0 0
mg/l)
Percent
4
1
11
3
The figures in parentheses are the distributions for the 11 strains most sensitive to RBU. *Indicates the MIC for the single most sensitive strain.
Table III. Minimal inhibitory MAIS-complex mycobacteria.
concentrations
M/C Img/l)
Rifam
110 10
5 2.5 1.25
0.6 0.3 0.16 0.08 0.4 0.02
or less
Sensitive (MIC (0.6 No. Percent
of
rifamycins
against
ycin
RMP
RPE
RBU
C29
C70
C27
3 1 3 3 3 4 0 1 2 3 3*
0 1 0 1 4 2 2 6 4 4 2
0 1 1 0 1 5 4 7 5 0 2
0 0 1 1 1 1 8 4 7 1 2
0 1 0 0 1 0 2 4 14 1 3
1 2 1 3 3 3 5 4 1 2 1
mgll) 9
18
18
22
24
13
35
69
69
85
92
50
“Two of the three strains had an MIC of 0.16 or less because the lowest concentration of RMP in the batch of medium used was 0.3 mgll.
26
strains
of
Activity of new rifamycins
181
rifamycins, the MlCs being about four times lower, on average. Using our definition of sensitivity, all 11 (31 %) were sensitive to RBU but only l-4 strains (3-I 1 %I were sensitive to the other rifamycins. However, 9 of the 11 strains had MlCs of 0.6 or 0.3 mg/l RBU, only just within our classification as sensitive. In summary, only RBU had appreciable in vitro activity against RMP-resistant M. tuberculosis, but only against about one-third of the strains and usually with MlCs sufficiently high to raise doubts about their likely response to RBU. MAE-complex The MlCs of the six rifamycins against MAIS-complex strains are set out in Table III. A strain was considered sensitive if the MIC was less than 0.6 mg/l, a lower concentration than in the solid medium tests on M. tuberculosis because the incubation period was shorter. The proportion of sensitive strains was 35 % with RMP, 50-60 % with RPE, RBU and C27, and 85-92 % with C29 and C70. Discussion
The new rifamycins RPE and C27 appeared to have MlCs against drug-sensitive strains of M. tuberculosis about four times lower than those of RMP. Similar results for RPE were obtained by Arioli et al. [3], although Grosset and his colleagues [41 found that the MlCs of RMP and RPE were similar. RBU, C29 and C70 appeared to be eight times more active than RMP. In experiments to be reported elsewhere, C70 was found to be appreciably more stable in vitro than RMP, so that the estimate of activity relative to RMP may have been overestimated for C70 though not for the other four rifamycins. The sixth new rifamycin, FCE 22250, was also included in some of the tests done, but it was found to be much less stable than RMP during incubation. Because its instability was probably responsible for the much higher MlCs obtained, the results are not presented here. Our results suggest that only RBU is likely to have any activity against RMP-resistant strains of M. tuberculosis. Even so, only 31 % of these strains appeared sensitive, and with MlCs slightly higher than the MlCs of RMP against sensitive strains. This proportion is similar to the 36 % estimated by Woodley and Kilburn [I I] and is consistent with the data of Heifets and lseman [141. The 11 strains that we found sensitive to RBU were also the least resistant to the other rifamycins, including RMP, the MlCs for RBU being about four times lower than for the other rifamycins. Considering activity against the MAIS-complex strains, the two rifamycins with the lowest MlCs appeared to be C29 and C70, although the activity of the latter may have been overestimated relative to the other rifamycins because of its exceptional stability. Both of these compounds appeared more active than RPE, RBU and C27, which in their turn were more active than RMP. One cannot assume that a finding of high in vitro activity implies similar activity in human disease, since other factors come into play. The extent of binding to plasma proteins could well influence their activity in animals and man. RPE is bound to a greater extent than RMP ]18,191, RBU is bound to a lesser extent [communication from Messrs Farmitalia Carlo Erba] while the binding of C29, C70 and C27 is similar [communication from Messrs Ciba-Geigy]. The distribution between extracellular and intracellular rifamycin concentrations could also be of great importance. After dosage with RBU, exceptionally high tissue concentrations but very low plasma concentrations are found [2, communication from Farmitalia Carlo Erba]. Although it is six to seven times more active weight for weight than RMP in experimental murine tuberculosis 121, this is a disease where the tubercle bacilli are predominantly
182
Dickinson
and Mitchison
intracellular. In contrast, the location of bacilli in human disease, whether in pulmonary tuberculosis or in the overwhelming infection with MAIS-complex organisms in AIDS, is predominantly extracellular, so that the activity of RBU in man must remain in doubt until it has
been
demonstrated
by
clinical
studies.
Acknowledgement This research was supported by the Medical Research Council. We are grateful to Messrs Merrell Dow Pharmaceuticals for supplying rifapentine, to Messrs Farmitalia Carlo Erba for rifabutin and FCE 22250 and to Messrs Ciba-Geigy for CGP 29861, CGP 7040 and CGP 27557.
References 1 Sanfilippo, A., Delia Bruna, C., Marsili, L., Morvillo, E., Pasqualucci, C. FL, Schioppacassi, G., Ungheri, D. (1980). Biological activity of a new class of rifamycins Spiro-piperidyl-rifamycins. Journal OfAntibiotics, 33, 1193. 2 Della Bruna, C., Schioppacassi, G., Ungheri, D., Jabes, D., Morvillo, E., Sanfilippo, A. (1983). LM 427, a new spiropiperidylrifamycin: in vitro and in vivo studies. Journal of Antibiotics, 36, 1502. 3 Arioli, V., Berti, M., Carniti, G., Randisi, E., Rossi, E., Scotti, R. (1981). Antibacterial activity of DL 473, a new semisynthetic rifamycin derivative. Journal of Antibiotics, 34, 1026. 4 Truffot-Pernot, C., Grosset, J., Bismuth, R., Lecoeur, H. (1983). Activite de la rifampicine administree de maniere intermittente et de la cyclopentyl rifamycine (ou DL 473) sur la tuberculose experimentale de la souris. Revue francais Maladies Respiratoire, 11, 875. 5 Vischer, W. A., Imhof, P., Hauffe, S., Degen, P. (1986). Pharmokinetics of new long-acting rifamycin-derivatives in man. Bulletin of the lntemational Union against Tuberculosis, 61, 8. 6 Vischer, W. A., Gowrishankar, R., Ashtekar, D. R., Costa-Pereira, R., Subrahmanyam, D., Kump, W., Traxler, P. (1986). Antitubercular activity in vitro and in vivo of new long-acting rifamycin derivatives. Bulletin of the lnrernational Union against Tuberculosis, 61. 8. 7 Tosch, W., Ban, E. (1988). Pharmacokinetics of new long-acting riafmycin-derivatives in animals. Bulletin of the lnremational Union against Tuberculosis, 61, 9. 8 Lecoeur, H., Truffot-Pernot, C., Grosset, J. (1986). Activity of CGP 29861. a new long lasting rifamycin-derivative against Mycobacterium tuberculosis in the mouse. Bulletin of the lntemational Union against Tuberculosis, 61, IO. 9 Ungheri, D., Morvillo, E., Sanfilippo, A. (1983). Attivita delle ansamicine LM 427 e FCE 22250 su micobatteri atipici in vitro. Giornale Italian0 Chemioterapia, 30. 97. IO Della Bruna, C., Ungheri, D., Sebben, G., Sanfilippo, G., Franceschi, G., Marsili, L. (1983). FCE 22250: Abstract: A long-acting antibiotic belonging to the novel class of 3-azinomethyl rifamycins. 73th international Congress of Chemotherapy 28th August-2nd September, Eds. K. H. Spitzy, K. Karrer. Part 111, p. 39. Vienna. 11 Woodley, C. L., Kilburn, J. 0. (1982). In vitro susceptibility of Mycobacrerium avium complex and Mycobacrerium ruberculosis strains to a Spiro-piperidyl rifamycin. American Review of Respiratory Disease, 126, 586. 12 Ungheri, D., Della Bruna, C., Sanfilippo, A. (1984). Activity of the spiropiperidyl rifamycin LM 427 on rifampicin resistant Mycobacterium tuberculosis. Giomale kaliano Chemioterapia, 31, 211. 13 Ungheri, D., Della Bruna, C., Sanfilippo, A. (1984). Studies on the mechanism of action of the spiropiperidylrifamycin LM 427 on rifampicin-resistant M. tuberculosis. Drugs Experimental Clinical Research, 10, 681. 14 Heifets, L. B., Iseman, M. D. (1985). Determination of in vitro susceptibility of mycobacteria to ansamycin. American Review of Respiratory Disease, 132, 710. 15 Canetti, G., Fox, W., Khomenko, A., Mahler, H. T., Menon, N. K., Mitchison, D.A., Rist, N., Smelev, N. A. (1969). Advances in techniques of testing mycobacterial drug sensitivity and the use of sensitivity tests in tuberculosis control programmes. Bulletin of the World Health Organization, 41, 21. 16 Mitchison, D. A., Allen, B. W., Carrol, L., Dickinson, J. M., Aber, V. R. (1972). A selective oleic acid albumin agar medium for tubercle bacilli. Journal of Medical Microbiology, 5, 165. 17 Medical Research Council (1973). Co-operative controlled trial of a standard regimen of streptomycin, PAS and isoniazid and three alternative regimens of chemotherapy in Britain. Tubercle, 64, 99. 18 Assandri, A., Perazzi, A., Berti, M. (1977). Studies of binding of C,-substituted rifamycins to human and bovine albumin. Journal of Antibiotics, 30, 409. 19 Assandri. A., Cristina, T., Moro, L (1978). Physiological disposition of a series of rifamycins in the rat: a comparative study. Journal of Antibiotics, 31, 894.
IN VlTRO ACTIVITY OF NEW RiFAMYClNS AGAlRtST RiFAlUPiCWRESISTANT TUBERCULOSlS AND MAIS-COIRPLEX MYCOBACTEAA
M.
Jean M. Dickinson and D.A. Mitchison Department
of Bacteriology,
Royal Postgraduate
Medical
School, Ducane Road, London W12 OHS
Summary Comparisons were made of the in vitro activity of rifampicin, and the rifamycin derivatives, rifapentine, rifabutin, CGP 29861, CGP 7040 and CGP 27557, against rifampicin-sensitive and rifampicin-resistant strains of Mycobacterium tubercdosis and against the Mycobacferium avium/intrace//u/are/scrofu/aceum (MAIS) complex. The new rifamycins had MlCs four to eight times lower than those of rifampicin against sensitive M. tuberculosis strains. Of the 35 rifampicin-resistant strains of M. tubercu/osis, 31 % were sensitive to rifabutin but only 3-11 % to the other rifamycins. The proportions of the MAIS strains found to be sensitive were 35 % for rifampicin, 50-60 % for CGP 27557, rifapentine and rifabutin and 85-92 % for CGP 29861 and CGP 7040. R&urn6 L’auteur a comparit I’activite in vitro de la rifampicine et des d&iv& de la rifamycine-rifapentine, rifabutine, CGP 29 861, CGP 7040 et CGP 27 557-vis-&vis des souches de Mycobacterium tuberculosis sensibles ti la rifampicine et resistantes ti la rifampicine et vis-8-vis du complexe Mycobacterium avium/intrace//u/are/ scrofulaceum (MAIS). Les nouvelles rifamycines avient des CIM 4 B 8 fois plus basses que celle de la rifampicine vis-ti-vis des souches sensibles de M. tuberculosis. Parmi 35 souches de M. tuberculosis rksistantes ZIla rifampicine, 31 % Btaient sensibles B la rifabutine, mais seulement 3 6 11 % aux autres rifamycines. Les proportions des souches du complexe MAIS trouvdes sensibles ont &!tBde 35 % en ce qui concernait la rifampicine, 50-69 % pour CPG 27 557, rifapentine et rifabutine et 85-92 % pour CGP 29 861 et CGP 7040. Resumen Se cornpar la actividad in vitro de la rifampicina y de derivados de la rifamicina rifapentina, rifabutina, CGP 29861, CGP 7040 y CGP 27557- contra cepas de Mycobacferium tuberculosis resistentes a la rifampicina y sensibles a la rifampicina y contra el complejo Mycobacterium avium/intrace//u/are/scrofu/eceum (MAIS). Las nueuvas rifamicinas tenian MlCs 4 a 8 veces m6s bajas que la de la rifampicina contra cepas sensibles de M. tuberculosis. De un total de 35 cepas resistentes a la rifampicina de M. tuberculosis, el31 % eran sensibles a la rifabutina, pero ~610 3-l 1 % a las otras rifamicinas. Las proporciones de cepas de MAIS que resultaron sensibles fueron: 35 % para la rifampicina, 50-60 % para CGP 27557, rifapentina y rifabutina y 85-92 % para CGP 29861 y CGP 7040. Introduction Rifampicin
is a major antibiotic
in the treatment
of tuberculosis,
leprosy
and some other
178
Dickinson
and
Mitchison
mycobacterial infections. It is not surprising that attempts have been made to tailor the basic rifamycin molecules to produce semi-synthetic drugs with properties potentially advantageous over those of rifampicin. Information on six new semi-synthetic rifamycins is available. The in vitro characteristics, pharmacokinetics and activity in experimental murine tuberculosis of rifabutin, a spiropiperidyl rifamycin also known as LM 427 and ansamycin (a misnomer since other antibiotics belong to the broad class of ansamycins) have been described [I, 21, as have the corresponding properties of rifapentine, a cyclopentyl rifamycin also known as DL 473 and MDL [3,41, and the long-acting rifamycin derivatives CGP 29861, CGP 7040, CGP 27557 [5,6,7,8] and FCE 22250 [9, IO]. Claims have been made that rifabutin has in vitro activity against rifampicin-resistant strains of Mycobacterium tuberculosis because of a mode of action considered to be different to that of rifampicin [I 1, 12, 13, 141. Rifabutin has also been widely used on an individual patient basis in the United States for the treatment of patients with acquired immune deficiency syndrome (AIDS) who have become infected with organisms of the Mycobacterium avium/intrace/lulare/scrofu/aceum (MAIS) complex [9, 11, 141. We describe here in vitro studies carried out to compare five of these six new rifamycins and rifampicin in their activities against 1. rifampicin-resistant M. tuberculosis and 2. strains of the MAIS complex.
Materials
and methods
Rifamycins The six rifamycins,
the abbreviation
Rifamycin
Abbreviation
Rifampicin Rifabutin Rifapentine CGP 29861 CGP 7040 CGP 27557
RMP RBU RPE c29 c70 C27
for them
Alternative
used
here and their
alternative
names
are:
names
LM 427, ansamycin DL 473, MDL 473
Mycobacteria The strains of M. tuberculosis consisted of 10 drug-sensitive strains obtained pre-treatment (eight from Hong Kong patients, one from a British patient and strain H37Rv) and 48 strains used as controls for sensitivity tests in our laboratory [15], many of which were resistant to rifampicin. Most of these strains were obtained during unsuccessful treatment with a some strains being isolated about a month before the rifampicin-containing regimen, emergence of rifampicin resistance, some at the time that resistance was first detected and some a few months after the emergence of resistance. The 26 MAIS-complex strains consisted of a single strain from each of the serovars except 17 and 18, and were obtained from Dr P. A. Jenkins, Mycobacterium Reference Laboratory, Public Health Laboratory Service, Cardiff, South Wales. Minimal
inhibitory
concentrations
The minimal inhibitory concentration (MIC) of the six drugs against all strains of M. tuberculosis was estimated by growing the strains in 7HS medium with Tween and dropping 0.03 ml, containing about 500 colony forming units, onto sectors of plates of 7HlO
Activity of new rifamycins
179
oleic acid albumin agar made selective by the addition of carbenicillin 100 mg/l, polymyxin B 200 units/ml, trimethoprim 20 mg/l and amphotericin B 10 mg/l (161 and containing serial 2-fold dilutions of the rifamycins rangingsfrom 0.04-10 mg/l with a rifamycin-free control plate. The end-point was the lowest concentration showing no growth after incubation at 37 “C for 28 days. The strains of M. tuberculosis had also been tested during the preceding 6 months for sensitivity to rifampicin in standard tests on Lowenstein-Jensen medium 1171. The MAIS-complex strains were first grown in 7H9 medium with Tween and, when fully grown, 0.03 ml from a calibrated dropping pipette was added to 5 ml volumes of 7H9 liquid medium without Tween containing serial 2-fold dilutions of the rifamycins. The end-point was the lowest concentration inhibiting any growth after incubation for 14 days. The 7HlO and 7H9 media were obtained from Difco Laboratories, E Molesey, Surrey. Results
M. tuberculosis When tested in duplicate standard Lowenstein-Jensen medium sensitivity tests, 23 strains of M. tuberculosis were found to be sensitive to RMP. The MlCs of all six rifamycins obtained on 7HlO plates for these sensitive strains are shown in Table I. No differences were found between the 10 strains obtained pre-treatment and the remaining 13 strains obtained during treatment. Although the mean MIC could not be calculated because the concentrations used did not extend sufficiently low, it is evident that all five new rifamycins had lower MlCs than RMP, particularly RBU, C29 and C70 whose distributions suggested that they were about eight times more active. The lowest concentration of RMP that inhibited all strains was 1.25 mg/l, which was therefore taken as the cut-off concentration for sensitive strains. The MlCs for the remaining 35 RMP-resistant strains are set out in Table II. Using the definition obtained from Table I, a strain was considered as sensitive to a rifamycin if it had an MIC of less than 1.25 mg/l. Of the 11 strains classified as sensitive to RBU, one strain had low MlCs that fell within the distributions of MlCs for sensitive strains for each of the rifamycins (Table I) except that its MlCs for C29 and C70 were just too high. The remaining 10 strains fell into the lower portions of the distributions of each of the rifamycins. Thus it seems that there was a group of 11 rifampicin-resistant strains, comprising about one-third of those we tested, which had lower MlCs to any rifamycin, including RMP, than the remaining strains. These strains were appreciably more sensitive to RBU that to the other
Table I.
Minimal
inhibitory concentrations
of rifamycins against 23 sensitive strains of
M. tuberculosis. M/C
Rifam
ycin
(md,) >2.5 2.5 1.25 0.6 0.3 0.16 0.08 0.04
or less
RMP
RPE
RBU
C29
C70
C27
0 0 0 1 14 6 0 2
0 0 0 0 0 2 15 6
0 0 0 0 0 0 2 21
0 0 0 0 0 0 0 23
0 0 0 0 0 0 1 22
0 0 0 0 0 4 10 9
180
Dickinson
and Mitchison
Table II. Minimal inhibitory strains of M. tuberculosis.
concentrations
M/C
of rifamycins
against
35 rifampicin-resistant
Rifamycin
Img/ll
RMP >I0 10
5 2.5 1.25 0.6 0.3 0.16 0.08 0.04
19
6 2 2 5 0
or less
Sensitive (MIC cl.25 No.
(1) (2) (2) (5)
RPE
RBlJ
c29
c70
07
20 4 (1) 4 (3) 0 5 (5)
2 18 4 0 0 4 (41
18 3 5 2 4 2
10 5 2 8 (3) 6 (4) 3 (3)
19 6 (2) 5 (4) 0 4 14) 0
1 (1)
l(1’) 0 0 0
0 l(1’) 0 0
5 (5) 0 2 (2’) 0
1
2
11
3
3
6
31
9
(2) (2) (4) (2)
0 0 ill”) 0
0 1 (1’) 0 0
0 1 (I’) 0 0
mg/l)
Percent
4
1
11
3
The figures in parentheses are the distributions for the 11 strains most sensitive to RBU. *Indicates the MIC for the single most sensitive strain.
Table III. Minimal inhibitory MAIS-complex mycobacteria.
concentrations
M/C Img/l)
Rifam
110 10
5 2.5 1.25
0.6 0.3 0.16 0.08 0.4 0.02
or less
Sensitive (MIC (0.6 No. Percent
of
rifamycins
against
ycin
RMP
RPE
RBU
C29
C70
C27
3 1 3 3 3 4 0 1 2 3 3*
0 1 0 1 4 2 2 6 4 4 2
0 1 1 0 1 5 4 7 5 0 2
0 0 1 1 1 1 8 4 7 1 2
0 1 0 0 1 0 2 4 14 1 3
1 2 1 3 3 3 5 4 1 2 1
mgll) 9
18
18
22
24
13
35
69
69
85
92
50
“Two of the three strains had an MIC of 0.16 or less because the lowest concentration of RMP in the batch of medium used was 0.3 mgll.
26
strains
of
Activity of new rifamycins
181
rifamycins, the MlCs being about four times lower, on average. Using our definition of sensitivity, all 11 (31 %) were sensitive to RBU but only l-4 strains (3-I 1 %I were sensitive to the other rifamycins. However, 9 of the 11 strains had MlCs of 0.6 or 0.3 mg/l RBU, only just within our classification as sensitive. In summary, only RBU had appreciable in vitro activity against RMP-resistant M. tuberculosis, but only against about one-third of the strains and usually with MlCs sufficiently high to raise doubts about their likely response to RBU. MAE-complex The MlCs of the six rifamycins against MAIS-complex strains are set out in Table III. A strain was considered sensitive if the MIC was less than 0.6 mg/l, a lower concentration than in the solid medium tests on M. tuberculosis because the incubation period was shorter. The proportion of sensitive strains was 35 % with RMP, 50-60 % with RPE, RBU and C27, and 85-92 % with C29 and C70. Discussion
The new rifamycins RPE and C27 appeared to have MlCs against drug-sensitive strains of M. tuberculosis about four times lower than those of RMP. Similar results for RPE were obtained by Arioli et al. [3], although Grosset and his colleagues [41 found that the MlCs of RMP and RPE were similar. RBU, C29 and C70 appeared to be eight times more active than RMP. In experiments to be reported elsewhere, C70 was found to be appreciably more stable in vitro than RMP, so that the estimate of activity relative to RMP may have been overestimated for C70 though not for the other four rifamycins. The sixth new rifamycin, FCE 22250, was also included in some of the tests done, but it was found to be much less stable than RMP during incubation. Because its instability was probably responsible for the much higher MlCs obtained, the results are not presented here. Our results suggest that only RBU is likely to have any activity against RMP-resistant strains of M. tuberculosis. Even so, only 31 % of these strains appeared sensitive, and with MlCs slightly higher than the MlCs of RMP against sensitive strains. This proportion is similar to the 36 % estimated by Woodley and Kilburn [I I] and is consistent with the data of Heifets and lseman [141. The 11 strains that we found sensitive to RBU were also the least resistant to the other rifamycins, including RMP, the MlCs for RBU being about four times lower than for the other rifamycins. Considering activity against the MAIS-complex strains, the two rifamycins with the lowest MlCs appeared to be C29 and C70, although the activity of the latter may have been overestimated relative to the other rifamycins because of its exceptional stability. Both of these compounds appeared more active than RPE, RBU and C27, which in their turn were more active than RMP. One cannot assume that a finding of high in vitro activity implies similar activity in human disease, since other factors come into play. The extent of binding to plasma proteins could well influence their activity in animals and man. RPE is bound to a greater extent than RMP ]18,191, RBU is bound to a lesser extent [communication from Messrs Farmitalia Carlo Erba] while the binding of C29, C70 and C27 is similar [communication from Messrs Ciba-Geigy]. The distribution between extracellular and intracellular rifamycin concentrations could also be of great importance. After dosage with RBU, exceptionally high tissue concentrations but very low plasma concentrations are found [2, communication from Farmitalia Carlo Erba]. Although it is six to seven times more active weight for weight than RMP in experimental murine tuberculosis 121, this is a disease where the tubercle bacilli are predominantly
182
Dickinson
and Mitchison
intracellular. In contrast, the location of bacilli in human disease, whether in pulmonary tuberculosis or in the overwhelming infection with MAIS-complex organisms in AIDS, is predominantly extracellular, so that the activity of RBU in man must remain in doubt until it has
been
demonstrated
by
clinical
studies.
Acknowledgement This research was supported by the Medical Research Council. We are grateful to Messrs Merrell Dow Pharmaceuticals for supplying rifapentine, to Messrs Farmitalia Carlo Erba for rifabutin and FCE 22250 and to Messrs Ciba-Geigy for CGP 29861, CGP 7040 and CGP 27557.
References 1 Sanfilippo, A., Delia Bruna, C., Marsili, L., Morvillo, E., Pasqualucci, C. FL, Schioppacassi, G., Ungheri, D. (1980). Biological activity of a new class of rifamycins Spiro-piperidyl-rifamycins. Journal OfAntibiotics, 33, 1193. 2 Della Bruna, C., Schioppacassi, G., Ungheri, D., Jabes, D., Morvillo, E., Sanfilippo, A. (1983). LM 427, a new spiropiperidylrifamycin: in vitro and in vivo studies. Journal of Antibiotics, 36, 1502. 3 Arioli, V., Berti, M., Carniti, G., Randisi, E., Rossi, E., Scotti, R. (1981). Antibacterial activity of DL 473, a new semisynthetic rifamycin derivative. Journal of Antibiotics, 34, 1026. 4 Truffot-Pernot, C., Grosset, J., Bismuth, R., Lecoeur, H. (1983). Activite de la rifampicine administree de maniere intermittente et de la cyclopentyl rifamycine (ou DL 473) sur la tuberculose experimentale de la souris. Revue francais Maladies Respiratoire, 11, 875. 5 Vischer, W. A., Imhof, P., Hauffe, S., Degen, P. (1986). Pharmokinetics of new long-acting rifamycin-derivatives in man. Bulletin of the lntemational Union against Tuberculosis, 61, 8. 6 Vischer, W. A., Gowrishankar, R., Ashtekar, D. R., Costa-Pereira, R., Subrahmanyam, D., Kump, W., Traxler, P. (1986). Antitubercular activity in vitro and in vivo of new long-acting rifamycin derivatives. Bulletin of the lnrernational Union against Tuberculosis, 61. 8. 7 Tosch, W., Ban, E. (1988). Pharmacokinetics of new long-acting riafmycin-derivatives in animals. Bulletin of the lnremational Union against Tuberculosis, 61, 9. 8 Lecoeur, H., Truffot-Pernot, C., Grosset, J. (1986). Activity of CGP 29861. a new long lasting rifamycin-derivative against Mycobacterium tuberculosis in the mouse. Bulletin of the lntemational Union against Tuberculosis, 61, IO. 9 Ungheri, D., Morvillo, E., Sanfilippo, A. (1983). Attivita delle ansamicine LM 427 e FCE 22250 su micobatteri atipici in vitro. Giornale Italian0 Chemioterapia, 30. 97. IO Della Bruna, C., Ungheri, D., Sebben, G., Sanfilippo, G., Franceschi, G., Marsili, L. (1983). FCE 22250: Abstract: A long-acting antibiotic belonging to the novel class of 3-azinomethyl rifamycins. 73th international Congress of Chemotherapy 28th August-2nd September, Eds. K. H. Spitzy, K. Karrer. Part 111, p. 39. Vienna. 11 Woodley, C. L., Kilburn, J. 0. (1982). In vitro susceptibility of Mycobacrerium avium complex and Mycobacrerium ruberculosis strains to a Spiro-piperidyl rifamycin. American Review of Respiratory Disease, 126, 586. 12 Ungheri, D., Della Bruna, C., Sanfilippo, A. (1984). Activity of the spiropiperidyl rifamycin LM 427 on rifampicin resistant Mycobacterium tuberculosis. Giomale kaliano Chemioterapia, 31, 211. 13 Ungheri, D., Della Bruna, C., Sanfilippo, A. (1984). Studies on the mechanism of action of the spiropiperidylrifamycin LM 427 on rifampicin-resistant M. tuberculosis. Drugs Experimental Clinical Research, 10, 681. 14 Heifets, L. B., Iseman, M. D. (1985). Determination of in vitro susceptibility of mycobacteria to ansamycin. American Review of Respiratory Disease, 132, 710. 15 Canetti, G., Fox, W., Khomenko, A., Mahler, H. T., Menon, N. K., Mitchison, D.A., Rist, N., Smelev, N. A. (1969). Advances in techniques of testing mycobacterial drug sensitivity and the use of sensitivity tests in tuberculosis control programmes. Bulletin of the World Health Organization, 41, 21. 16 Mitchison, D. A., Allen, B. W., Carrol, L., Dickinson, J. M., Aber, V. R. (1972). A selective oleic acid albumin agar medium for tubercle bacilli. Journal of Medical Microbiology, 5, 165. 17 Medical Research Council (1973). Co-operative controlled trial of a standard regimen of streptomycin, PAS and isoniazid and three alternative regimens of chemotherapy in Britain. Tubercle, 64, 99. 18 Assandri, A., Perazzi, A., Berti, M. (1977). Studies of binding of C,-substituted rifamycins to human and bovine albumin. Journal of Antibiotics, 30, 409. 19 Assandri. A., Cristina, T., Moro, L (1978). Physiological disposition of a series of rifamycins in the rat: a comparative study. Journal of Antibiotics, 31, 894.