Selection of characters for an Adansonian analysis of mycobacterial taxonomy

Phenol-Soluble Antigens from Mycobacterium kansasii, Mycobacterium gastri, and Mycobacterium marinum

Many of the demonstrable antigens derived from mycobacteria are common to members of different species. Agglutination tests have yielded the most specific characterizations. An antigen which may be associated with the specific agglutination is soluble in phenol and can be extracted and separated from other antigens by use of this solvent. Phenol-soluble antigens have been extracted from representative cultures of Mycobacterium kansasii, M. gastri, and M. marinum. Most representatives of each of these species yielded an antigen which was characteristic of the species but was distinct from antigen derived from the other two species.

Agents of newly recognized or infrequently encountered mycobacterial diseases

This paper reviews recent information on the systematics and clinical significance of potentially pathogenic environmental mycobacteria. A short history of these mycobacteria is given. Information on species for which clinical and systematic aspects have already been well documented, i.e., Mycobacterium kansasii, M. marinum, M. scrofulaceum, M. simiae, M. szulgai, M. ulcerans, M. xenopi, and members of the M. fortuitum complex, is updated. Although the M. avium complex was extensively reviewed in earlier literature, major new systematic and clinical information is presented in some detail. Species that have received very limited prior coverage, i.e., M. asiaticum, M. haemophilum, M. malmoense, and M. shimoidei, are the main subjects of this review and are discussed in detail. The rare infections attributed to species that are normally considered nonpathogenic, i.e., M. gastri, M. gordonae, the M. terrae complex, and most of the rapidly growing mycobacteria outside of the M. fortuitum complex, are critically reviewed. Finally, suggestions are offered for practical measures that can minimize the risk of failing to isolate or misidentifying some of the more obscure potentially pathogenic environmental mycobacteria that are only infrequently recognized.

Numerical identification of slowly growing mycobacteria

Numerical identification using the range of a species defined as (M +/- 2s)% was shown to be suitable for the identification of slowly growing mycobacteria. In the formula, M is the mean matching coefficient for individual strains of each species to the hypothetical median organism pattern of their own species and s is the standard deviation.

The "atypical" mycobacteria: recognition and disease association

Although techniques based on immunologic or chromatographic analyses have been described for identifying mycobacteria in clinical laboratories, most microbiologists continue to rely on a series of specialized physiological and biochemical tests for this purpose. The recognition of additional significant species over the past decade has required the addition of more tests to the battery used for mycobacterial identification. This paper will review briefly the taxonomic status of species likely to be encountered in clinical specimens and the most useful tests for characterizing them. Strategies will be presented for using these tests in the most efficient way to provide optimal resolution of taxa without use of an unreasonably large battery of tests. A brief survey of techniques that may become more practical in the future will also be included.

Numerical classification of 280 strains of slowly growing mycobacteria. Proposal of Mycobacterium tuberculosis series, Mycobacterium avium series, and Mycobacterium nonchromogenicum series

Numerical classification of 280 strains of slowly growing mycobacteria was carried out by testing each strain for 76 characters. The following fourteen clusters were observed: 1. M. tuberculosis, M. bovis, M. africanum, and M. microti; 2. M. haemophilum; 3. M. ulcerans; 4. M. xenopi; 5. M. kansasii; 6. M. szulgai; 7. M. gordonae; 8. M gastri; 9. M. avium, M. intracellulare, M. scrofulaceum, and M. asiaticum; 10. M. marinum; 11. M. simiae; 12. M. nonchromogenicum, "M. novum," M. terrae, and M. triviale; 13 M. malmoense; 14. M. shimoidei. The clusters composed of M. tuberculosis, M. bovis, M. africanum, and M. microti, of M. avium, M. intracellulare, M. scrofulaceum, and M. asiaticum, and of M. nonchromogenicum, "M. novum," M. terrae, and M. triviale appeared to be reduced to a single species each. The names having priority for each species should be M. tuberculosis, M. avium, and M nonchromogenicum, respectively. However, the clusters may, in practice, be called the M. tuberculosis series (complex), the M. avium series (complex), and the M. nonchromogenicum series (complex). The type species of these series are M. tuberculosis, M. avium, and M. nonchromogenicum, respectively. These series were characterized in this study.

Deoxyribonucleic acid relatedness among species of slowly-growing mycobacteria

DNA-DNA hybridization is a reliable method for determining the phylogenetic relationship between bacterial strains The hybridization kinetics for DNA from different slowly-growing mycobacteria were measured optically in a spectrophotometer. The results indicate that M. tuberculosis and M. bovis belong to one species. M. avium and M. intracellulare are two species, but some serotypes, now designated M. intracellulare, actually belong to M. avium.

Nucleic acid homology in the genus Mycobacterium

Immobilized deoxyribonucleic acid (DNA) preparations from eight species of mycobacteria were reacted with labeled reference DNA from Mycobacterium tuberculosis and M. kansasii. All mycobacteria showed some degree of homology. Immobilized DNA from Pseudomonas multivorans, which has a guanine plus cytosine content within the range established for the mycobacteria, showed no significant binding with either reference system. Relative per cent binding and thermal stability of bound DNA were used as parameters for determining degrees of relatedness among members of the genus. When these relationships were compared with those established by numberical taxonomic methods, a high correlation was found.

Chemical and biological properties of mycobactins isolated from various mycobacteria

Nine different strains of mycobacteria grown on media deficient in iron all produced mycobactins. Most strains produced one mycobactin in great preponderance. Mycobacteria from clearly distinct taxonomic groups gave mycobactins differing in the structure of their nuclei. One group of taxonomically related mycobacteria produced mycobactins having the same nucleus but with different distributions of side chains within the homologous mixtures. Simple methods are described for identifying mycobactins on a small scale; these may be of value in classifying mycobacteria. Structures are proposed for mycobactin A from Mycobacterium aurum, mycobactin R from M. terrae, mycobactin F, produced together with mycobactin H by M. fortuitum, and mycobactins M and N from M. marinum. The first three of these differ from known mycobactins in details of substitution and configuration of asymmetric centres in the nucleus. Mycobactins M and N are substantially different, having only small acyl groups (acetyl and propionyl respectively) at the hydroxamic acid centre of the mycobactic acid moiety. Both are homologous mixtures having long-chain saturated 3-hydroxy-2-methyl acid fragments in the cobactin moiety. All mycobactins so far isolated promote almost maximal growth of M. johnei at 30ng./ml. in liquid medium. The activity of some mycobactins extends to much lower concentrations, mycobactin S showing significant growth promotion at 0.3ng./ml. Mycobactin M or N in combination with mycobactins having a long side chain in the mycobactic acid moiety exerts a mutually antagonistic effect on the growth of M. johnei, the mixture giving less growth than either mycobactin separately. Mycobactin M also decreases the growth of M. kansasii and M. tuberculosis on liquid media. These antagonistic effects are probably caused by a lengthening of the lag phase.

Base composition of deoxyribonucleic acid isolated from mycobacteria

Guanine plus cytosine values of deoxyribonucleic acid derived from 30 cultures representing 14 mycobacterial species or varieties are presented. These data provide impressive reasons for maintaining the separation between the genera Corynebacterium and Mycobacterium; no conclusions can be arrived at from these data with respect to the Nocardia-Mycobacterium relationship. A bimodal clustering, in terms of guanine plus cytosine composition, is apparent within the genus Mycobacterium. In general, all members of any single phenetic species appear to fit into one or another of these clusters. The phenetic separation of species is, in some cases, confirmed by separation in terms of guanine plus cytosine values. The bimodal separation of guanine plus cytosine values within the genus Mycobacterium does not correspond to a division of the species into slow and rapid growers; it thus provides no justification for splitting Mycobacterium into two genera, composed of slow and rapid growers. This is not to say that such a split would not be useful, only that these data do not contribute to such a decision. Any further attempts to correlate phenetic classification with properties of mycobacterial deoxyribonucleic acid will require more specific techniques, such as molecular hybridization.

Diagnostic key to mycobacteria encountered in clinical laboratories

A diagnostic key has been developed which will permit identification of most mycobacteria encountered in clinical laboratories. The key is based on performance of a few simple tests. The efficiency and accuracy of the key was evaluated in terms of correlation between identifications based on the few tests and those arrived at through application of the techniques of numerical taxonomy, which involves a large battery of tests. Of 679 cultures of mycobacteria other than Mycobacterium tuberculosis, 86.5% were correctly identified by use of the key, and only 1.8% of the cultures were erroneously identified. The remaining cultures required further examination.