Comment on: characterization of the embB gene in Mycobacterium tuberculosis isolates from Barcelona and rapid detection of main mutations related to ethambutol resistance using a low-density DNA array.

Sir,

We agree with Moure et al.[1] that fast genotypic methods will play an increasingly prominent role in drug susceptibility testing for the Mycobacterium tuberculosis complex (MTBC).[2,3] We would, however, like to point out that the embB (Rv3795) Glu378Ala polymorphism, which is detected by probe 3 of their newly developed low-density DNA array, is not a marker for ethambutol resistance.[4-7] Instead, Ala represents the ancestral amino acid at this codon (Figure 1), whereas Glu is present in all modern MTBC (lineages 2, 3 and 4).[6-9] The MIRU–VNTR data of the 51 ethambutol-resistant isolates from the study by Moure et al.[1] are largely congruent with this finding. All 49 phylogenetically modern MTBC isolates had the embB 378 Glu variant. Isolate 5765 was a representative of Mycobacterium bovis, which is consistent with the fact that it harboured the Ala variant and was pyrazinamide resistant. By contrast, it was unclear why isolate 233R, which appeared to be M. bovis based on its MIRU–VNTR signature, had the Glu variant (experimental error or a homoplastic event might account for this discrepancy).

Figure 1.

Whole-genome phylogeny of 219 isolates representative of all major MTBC lineages.[9] Glu at codon 378 is a marker for modern MTBC, which all share the TbD1 deletion and include the lineage 4 M. tuberculosis H37Rv laboratory strain that is used as the reference/wild-type sequence for sequence analyses.[10]

In light of these data, the results of probe 3 would be predicted to lead to systematic false-positive reports, which calls into question the validity of this probe. This underlines that the entire MTBC diversity has to be considered when designing and validating genotypic drug susceptibility testing assays.[7,10]

Funding

This work was supported by a grant from the Department of Health, Wellcome Trust and the Health Innovation Challenge Fund (HICF-T5-342 and WT098600 to S. J. P.), Public Health England (to S. J. P.), the Medical Research Council (to J. M. B.) and the Wellcome Trust Sanger Institute (WT098051 to J. P. and J. M. B). C. U. K. is a Junior Research Fellow at Wolfson College, Cambridge. I. C. is supported by a Ramón y Cajal fellowship from the Spanish Government (RYC-2012-10627).

Transparency declarations

J. P. has received funding for travel and accommodation from Pacific Biosciences Inc. and Illumina Inc. S. J. P. is a consultant for Pfizer Inc. and has received funding for travel and accommodation from Illumina Inc. All other authors: none to declare.

Disclaimer

This publication presents independent research supported by the Health Innovation Challenge Fund (HICF-T5-342 and WT098600), a parallel funding partnership between the Department of Health and Wellcome Trust. The views expressed in this publication are those of the authors and not necessarily those of the Department of Health or Wellcome Trust.