Search

Capreomycin and the structurally similar compound viomycin are cyclic peptide antibiotics which are particularly active against Mycobacterium tuberculosis, including multidrug resistant strains. Both antibiotics bind across the ribosomal interface involving 23S rRNA helix 69 (H69) and 16S rRNA helix 44 (h44). The binding site of tuberactinomycins in h44 partially overlaps with that of aminoglycosides, and they share with these drugs the side effect of irreversible hearing loss. Here we studied the drug target interaction on ribosomes modified by site-directed mutagenesis. We identified rRNA residues in h44 as the main determinants of phylogenetic selectivity, predict compensatory evolution to impact future resistance development, and propose mechanisms involved in tuberactinomycin ototoxicity, which may enable the development of improved, less-toxic derivatives.

The ability to rapidly identify small molecules that interact with RNA would have significant clinical and research applications. Low-molecular-weight molecules that bind to RNA have the potential to be used as drugs. Therefore, technologies facilitating the rapid and reliable identification of such activities become increasingly important. We have applied a fluorescence-based assay to screen for modulators of hammerhead ribozyme (HHR) catalysis from a small library of antibiotic compounds. Several unknown potent inhibitors of the hammerhead cleavage reaction were identified and further characterized. Tuberactinomycin A, for which positive cooperativity of inhibition in vitro was found, also reduced ribozyme cleavage in vivo. The assay is applicable to the screening of mixtures of compounds, as inhibitory activities were detected within a collection of 2,000 extracts from different actinomycete strains. This approach allows the rapid, reliable, and convenient identification and characterization of ribozyme modulators leading to insights difficult to obtain by classical methodology.

Antibiotics act as inhibitors of various biological processes. Here we demonstrate that some tuberactinomycins, hitherto known as inhibitors of prokaryotic protein synthesis and of group I intron self-splicing, have a modulatory effect on group I intron RNAs. The linear intron, which is excised during the self-splicing process, is still an active molecular capable of performing an intramolecular transesterification resulting in a circular molecule. However, in the presence of sub-inhibitory concentrations of tuberactinomycins, the intron reacts intermolecularly leading to the formation of linear head-to-tail intron-oligomers. The antibiotic stimulates the intron to react in trans instead of in cis. The phage T4-derived td intron uses the same sites for oligomerisation as for circularisation. Gel- retardation experiments demonstrate that the intron RNA forms non-covalent complexes in the presence of the antibiotic. It might be envisaged that the role of these peptide antibiotics is to bridge RNA molecules mediating RNA-RNA interactions and thus enabling their reaction. The tuberactinomycins are further able to induce the interaction of heterologous introns. The ligation of the T4 phage-derived td intron with the Tetrahymena rRNA intron is very efficient, resulting in molecules composed of two introns derived from different species. The td intron attacks the Tetrahymena intron at various sites, which are located within double-stranded regions. These observations suggest that small molecules like these basic peptide antibiotics could have mediated RNA-RNA interactions in a pre-protein era.

Several examples of inhibition of the function of a ribozyme or RNA-protein complex have shown that certain antibiotics can interact specifically with RNA. There are, however, few examples of antibiotics that have a positive, rather than a negative, effect on the function of an RNA. We have found that micromolar concentrations of viomycin, a basic, cyclic peptide antibiotic of the tuberactinomycin group, enhance the cleavage of a ribozyme derived from Neurospora VS RNA. Viomycin decreases by an order of magnitude the concentration of magnesium required for cleavage. It also stimulates an otherwise insignificant transcleavage reaction by enhancing interactions between RNA molecules. The ability of viomycin to enhance some RNA-mediated reactions but inhibit others, including translation and Group I intron splicing, demonstrates the potential for natural selection by small molecules during evolution in the 'RNA world' and may have broader implications with respect to ribozyme expression and activity in contemporary cells.

The tuberactinomycins are a group of cyclic peptide antibiotics, which are potent inhibitors of prokaryotic protein synthesis. We report the inhibitory effect of viomycin, di-beta-lysyl-capreomycin IIA and tuberactinomycin A on group I intron self-splicing. They compete with the guanosine cofactor for the G-binding site located in the conserved core of the intron. They are 100-fold more active than all other competitive inhibitors described so far (dGTP, arginine or streptomycin), inhibiting splicing at concentrations between 10 and 50 microM. Mutation of the G-binding site leads to partial resistance, and the inhibitory effect of these drugs is dependent on Mg2+ concentration. This suggests that the tuberactinomycins have more than one contact site with the intron RNA: via the G-binding site and via additional contacts with the RNA backbone. Positioning the tuberactinomycins in the three-dimensional model of the td intron core suggests that the charged lysyl side-chain (R1) is in contact with the backbone of the P1 helix. Structure/function analyses with various tuberactinomycin analogues with different activities confirm the involvement of this sidechain in inhibition of group I self-splicing. The demonstration of a new class of splicing inhibitors, the peptide antibiotics, illustrates how antibiotics may interact with catalytic RNA.

A total of 114 strains of Mycobacterium tuberculosis isolated from sputum samples of 114 patients of pulmonary tuberculosis in south India, were coded and tested for their in vitro susceptibility to tuberactinomycin (Tum) incorporated in Lowenstein-Jensen (LJ) medium. Of these strains, 95 (83.3%) and 15 (13.2%) were susceptible to Tum at 25 and 50 mg/l respectively. Only 4 (3.5%) strains were inhibited at 100 mg/l or more. Of the 37 drug sensitive strains, 2 (5.4%) were not susceptible to Tum at 25 mg/l compared to 17 (22.1%) of 77 strains-resistant to one or more of antituberculosis drugs (P less than 0.02). The drug susceptibility pattern of the strains revealed that there was no significant association of resistance between Tum and streptomycin or rifampicin or ethambutol or ethionamide or isoniazid. However, 15 (53.6%) of 28 kanamycin (K) resistant strains were not susceptible to Tum at 25 mg/l. This cross resistance between Tum and K was further studied in 24 and 15 K sensitive and resistant strains respectively, by correlating their proportion resistance at 16 mg/l and it was found to have a significant positive correlation (r = 0.55; P less than 0.01).

Bacteriostatic and bactericidal activities of rifampicin, isoniazid, streptomycin, enviomycin and ethambutol against Mycobacterium tuberculosis, Mycobacterium avium--M. intracellulare complex and Mycobacterium kansasii were studied in different growth phases. Bacteriostatic activities of the drugs were similar in different growth phases, except isoniazid. M. tuberculosis was much less susceptible to isoniazid in the lag phase than in the log and the stationary phases. In contrast, bactericidal activity was influenced by the growth phase. M. tuberculosis was killed by isoniazid, streptomycin and rifampicin. The bactericidal activity of isoniazid was strongest. The bactericidal activity of isoniazid and streptomycin was most marked in the log phase. M. avium complex and M. kansasii resisted the bactericidal activity, but some strains of M. avium complex were killed by streptomycin and enviomycin, and the activities of these two drugs were most marked in the lag phase.

Bacteriostatic and bactericidal activities of rifampicin, ethambutol, enviomycin and streptomycin on Mycobacterium avium-Mycobacterium intracellulare complex (MAI complex) strains were determined. The susceptibility testings were made in Ogawa egg medium, and minimal inhibitory concentrations (MICs) were determined as the lowest concentration of drugs, in which the growth of 20 to 100 colony-forming units was completely inhibited. The MICs correspond to the MICs that can inhibit the growth of 95 to 99% of bacterial population. Bactericidal activity was determined in a modified Dubos liquid medium (1.3 g of Dubos TB Broth Base were dissolved in 180 ml of distilled water and this solution was supplemented by 20 ml of bovine serum). A one ml-sample of bacterial suspensions (10 mg wet weight per ml) was added to 9 ml of the Dubos liquid medium, and the medium was incubated at 37 degrees C for 0, 1, 3 and 7 days under shaking condition (56 strokes per minute; 8 cm amplitude). The bactericidal activity was measured as the number of colony-forming units contained in a 0.02 ml-sample of the medium. The bactericidal activities of rifampicin and ethambutol were weak or absent even in strains 13008 and 13016, which were very susceptible to all four drugs. However, the bactericidal activities of streptomycin and enviomycin could be observed in these strains.(ABSTRACT TRUNCATED AT 250 WORDS)

Treatment of nontuberculous mycobacterial infection should be carried out by the chemotherapeutic regimens most suitable for each species. We performed in-vitro susceptibility tests for various species and determined the probability in which the mycobacteria of each species are inhibited by blood concentrations attainable by the dosages usually used. From such determinations, the following regimens have been recommended: 1) M. avium-M. intracellulare complex, Rifampicin + Enviomycin + Ethambutol; 2) M. kansasii, Ofloxacin + Enviomycin + Rifampicin; 3) M. szulgai, Enviomycin - Ethambutol + Isoniazid; 4) M. fortuitum, Ofloxacin. The effectiveness of the treatment of the infection caused by M. avium complex is less than 15% even using the above regimen, while it is high in the treatment of infections caused by M. kansasii and M. szulgai. There are no effective regimens for the treatment of infection caused by M. chelonae or M. simiae. The above recommendations are useful in practice, because, at present, reliable susceptibility testing of nontuberculous mycobacteria is not yet done commonly.

Tuberactinamine N, the cyclic peptide moiety of tuberactinomycin N, was obtained in a crystalline state through liberation of gamma-hydroxy-beta-lysine from tuberactinomycin N by acid treatment. Tuberactinamine N possesses an intramolecular hydrogen bond in its molecule and showed antibacterial activities comparable to those of the original antibiotics. Conversion of tuberactinomycin N to O was achieved through coupling of diacyl-beta-lysine with tuberactinamine N followed by removal of the protecting groups.

The effects of 61 synthetic tuberactinomycin derivatives on polypeptide synthesis were tested in bacterial cell-free systems. Di-beta-lys-capreomycin IIA was more effective than the natural product. Palmitoyl tuberactinamine N inhibited the growth of tuberactinomycin-resistant mutants and was not a ribosome inhibitor.

Binding studies were performed with a [14C]-labelled derivative of viomycin, tuberactinomycin 0 (TUM O). TUM O bound to 30S and 50S subunits. The binding component was the RNA, since ribosomal proteins did not bind the drug. Other RNAs such as tRNA, phage RNA (MS2), and homopolynucleotides also bound the drug. Striking differences in the binding capacity of the various homopolynucleotides were found. Poly(U) bound strongly, poly(G) and poly(C) bound intermediately, whereas poly(A) showed a very low binding. DNA also bound TUM O, although with native DNA the binding was only weak. Finally the effects of viomycin on the assembly in vitro of the 50S subunit from E. coli were tested. A very strong inhibition was found: when the reconstitution was performed at 0.5 x 10(-6) M viomycin the particles formed sedimented at about 50S, but showed a residual activity of less than 10%. The inhibitory power of viomycin with respect to the in vitro assembly is more pronounced than that found in in vitro systems for protein synthesis.

Phenotypes of isolates of Mycobacterium tuberculosis H37RV showing resistance to the aminoglucoside antibiotics streptomycin, viomycin, kanamycin, capreomycin, tuberactinomycin N, lividomycin and paromomycin could be grouped into the following types: (I) resistant only to different levels of streptomycins; (2) resistant only to a low level of kanamycin; (3) triply resistant, to low levels of viomycin, tuberactinomycin N and capreomycin; (4) triply resistant, to a low level of kanamycin and high levels of lividomycin and paromomycin; (5) quadruply resistant, to a low level of capreomycin and high levels of kanamycin, lividomycin and paromomycin; (6) hextuply resistant, to high levels of viomycin, tuberactinomycin N, capreomycin, kanamycin, lividomycin, and paromomycin. Three modificatied types of the latter were also observed. Appearance rates of the six types were estimated as 10(-6) to 10(-9), 10(-6), 10(-6) to 10(-7), 10(-8), 10(-8), and 10(-8) to 10(-9), respectively, in a total viable population of the parent strain. Mutations to all phenotypes were considered to be produced by single mutations. According to cross-resistance relationships, aminoglucoside antibiotics were classified into three groups: (I) streptomycin; (II) viomycin, tuberactinomycin N and capreomycin; (III) kanamycin, lividomycin and paromomycin. No cross-resistance relationship between streptomycin and other antibiotics was observed. Resistances to viomycin, tuberactinomycin N and capreomycin occurred by single mutation to type 3. Resistances to kanamycin, lividomycin and paromomycin occurred by single mutations to types 4 and 5. Low resistance to capreomycin was produced by mutation to type 5. Therefore capreomycin was considered to be an intermediate between the second and third groups. These two groups had a close relationship, as resistance to all six agents in these groups could be produced by a single mutation to type 6 (and its modified types).

In vitro susceptibility testing to streptomycin, kanamycin and enviomycin was carried out in nontuberculous mycobacteria and the results were compared with those of M. tuberculosis strains. Among nontuberculous mycobacteria tested, only M. xenopi strains exhibited similar minimal inhibitory concentration (MIC) values to streptomycin with M. tuberculosis strains. On the other hand, M. xenopi, M. malmoense and M. marinum strains showed the same MIC values to kanamycin with the MIC values for M. tuberculosis strains. The MICs of enviomycin for M. kansasii, M. malmoense, M. szulgai, M. xenopi and M. marinum strains were similar to those for M. tuberculosis strains. Out of 64 strains of M. avium complex, the MIC value similar to that for M. tuberculosis strains was shown in 16% of strains to streptomycin, in 42% of strains to kanamycin, and in 50% of strains to enviomycin.