Use of genetic mutations for prediction of resistance to pyrazinamide in M. tuberculosis strains. (March 2015)

• Record Type: Journal Article
• Authors: Cirillo, Daniela M.; Miotto, Paolo; Niemann, Stefan
• Is Part Of: International journal of mycobacteriology. Volume 4 (2015: Mar.)Supplement 1
• Page Start: 42
• Page End: 43
• Publication Date: 2015-03
• Subjects: PZA -- Genetic mutations -- Drug resistance; Mycobacteria -- Periodicals; Mycobacterial diseases -- Periodicals; Mycobacteriaceae; Mycobacteria; Electronic journals; Periodicals; 579.374
• DOI: 10.1016/j.ijmyco.2014.08.011
• Languages: English
• ISSNs: 2212-5531

Abstract

Abstract : Pyrazinamide (PZA) is a prodrug that is converted to the active compound pyrazinoic acid by the pyrazinamidase enzyme (encoded by the pncA gene in Mycobacterium tuberculosis ) at low pH. The phenotypic approach for testing pyrazinamide sensitivity is technically demanding (often associated with poorly reliable results) and expensive. For these reasons, sensitivity of PZA is rarely tested and, if tested, when the sensitivity pattern becomes available to clinicians, the drug regimen of the patients is not modified based on the results. The molecular approach to drug resistance identification has considerably changed the capacity to identify and appropriately treat TB cases resistant to Rifampicin, Isoniazid and to some extent Fluoroquinolones and injectables. The identification of mutations on the pncA gene has the potential for rapid detection of PZA-R. However, the genetic variants are highly variable (not all associated to drug resistance) and scattered over the full length of the pncA entangling the development of a molecular test. The TBPannet consortium performed a large study assessing the pncA sequence variations in 1950 clinical isolates, including 1142 MDR and 483 fully susceptible strains. The sequencing analysis identified 280 different mutants. The presence of different mutations was correlated with phenotype, enzymatic activity, structural data, and phylogenetic data. Using an algorithm taking into account all the parameters evaluated, four classes of genetic variants were identified: (1) very high confidence resistance mutations (85% of the genetic variants examined) – always associated with phenotypic resistance, absence of enzymatic activity and protein structure predictive for a lost function; (2) high confidence resistance mutation – highly associated with phenotypic resistance (but not only, probably due to imperfect phenotypic test), absence of enzymatic activity and protein structure predictive for a lost function; (3) mutations with an unclear role found; and (4) mutations not involved in phenotypic resistance (10%). Any future molecular diagnostic assay should be able to target and identify at least the very high and high confidence genetic variants markers of PZA-R. An assay of this kind will have a diagnostic accuracy in the range of 90–99%. This work was supported by the European Community's Seventh Framework Programme (FP7/2007–2013) under grant agreement FP7-223681 to DMC (www.tbpannet.org ).