Neutral-red reaction is related to virulence and cell wall methyl-branched lipids in Mycobacterium tuberculosis

Searching for virulence marking tests for Mycobacteritan tuberculosis, Dubos and Middlebrook reported in 1948 that in an alkaline aqueous solution of neutral-red, the cells of the virulent H37Rv M. tuberculosis strain fixed the dye and became red in color, whereas the cells of the avirulent H37Ra M. tuberculosis strain remained unstained. In the 1950 and 1960s, fresh isolates of M. tuberculosis were tested for this neutral-red cytochemical reaction and it was reported that they were neutral-red positive, whereas other mycobacteria of diverse environmental origins that were non-pathogenic for guinea pigs were neutral-red negative. However, neutral-red has not really been proven to be a virulence marker. To test if virulence is in fact correlated to neutral-red, we studied a clinical isolate of M. tuberculosis that was originally neutral-red positive but, after more than I year passing through culture mediums, turned neutral-red negative. We found that, in comparison to the original neutral-red positive strain, this neutral-red negative variant was attenuated in two murine models of experimental tuberculosis. Lipid analysis showed that this neutral-red negative natural mutant lost the capacity to synthesize pthiocerol dimycocerosates, a cell wall methyl-branched lipid that has been related to virulence in M. tuberculosis. We also studied the neutral-red of different gene-targeted M. tuberculosis mutants unable to produce pthiocerol dimycocerosates or other cell wall methyl-branched lipids such as sulfolipids, and polyacyltrehaloses. We found a negative neutral-red reaction in mutants that were deficient in more than one type of methyl-branched lipids. We conclude that neutral-red is indeed a marker of virulence and it indicates important perturbations in the external surface of M. tuberculosis cells. (c) 2005 Elsevier SAS. All rights reserved.