Premium
Studies of Polymorphic DNA Fingerprinting and Lipid Pattern of Mycobacterium tuberculosis Patient Isolates in Japan
Author(s) -
Chaicumpar Kunyaluk,
Fujiwara Nagatoshi,
Nishimura Osamu,
Hotta Hisako,
Pan Jiong Wei,
Takahashi Mitsuyoshi,
Abe Chiyoji,
Yano Ikuya
Publication year - 1997
Publication title -
microbiology and immunology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.664
H-Index - 70
eISSN - 1348-0421
pISSN - 0385-5600
DOI - 10.1111/j.1348-0421.1997.tb01176.x
Subject(s) - restriction fragment length polymorphism , biology , mycobacterium tuberculosis , dna profiling , glycolipid , rhamnose , tuberculosis , microbiology and biotechnology , typing , genotype , dna , genetics , gene , biochemistry , polysaccharide , pathology , medicine
Abstract Strain differentiation by DNA restriction fragment length polymorphism (RFLP) has been used mainly for the epidemiological purpose of Mycobacterium tuberculosis infection. In this study, we tried to connect the molecular and phenotypic characteristics of M. tuberculosis patient isolates by comparing the DNA fingerprints obtained by RFLP using IS 6110 and lipid patterns using two‐dimensional thin‐layer chromatography (2‐D TLC) with silica gel, since M. tuberculosis has a lipid‐rich cell envelope which contributes to the virulence and immunomodulatory properties. We found that 66 isolates of M. tuberculosis from tuberculosis patients showed that the occurrence of IS 6110 varied from 1 to 24 copies. The IS 6110 patterns were highly variable among isolates. Fifty different RFLP patterns were observed, and 12 RFLP patterns were shared by two or more strains. By computerized analysis of the RFLP patterns of M. tuberculosis patient isolates, we found that 95% of the isolates fell into seven clusters, from A to G, with at least two isolates in each (> 30% similarity). Among the cellular lipids, the phospholipid composition did not differ by strain, whereas the glycolipid pattern differed markedly. Especially, the relative concentration of cord factor and sulfolipid, both of which were known as virulent factors, varied by strain. The fingerprints of some strains showed an association between the DNA and glycolipid patterns, even though some of the same DNA fingerprint strains showed differences in lipid patterns. Among the patient isolates, M. tuberculosis strain 249 possessed a specific glycolipid with 2‐ O ‐methyl‐ L ‐rhamnose and L‐rhamnose, which is rarely found in other strains. This glycolipid showed serological activity against the sera of tuberculosis patients, even if the reactivity was not as strong as trehalose dimycolate. It also showed the inhibition of phagosome‐lysosome fusion in macrophages, suggesting involvement with virulence. These results suggest that RFLP analysis using IS 6110 is useful for clustering the human isolates of M. tuberculosis , however, for further strain differentiation on virulence, a lipid analysis provides more information.