Evaluation of the GenoType MTBC assay for differentiating 120 clinical Mycobacterium tuberculosis complex isolates

Abstract

The majority of mycobacterial infections are caused by strains of the Mycobacterium tuberculosis complex (MTBC). MTBC includes the species M. tuberculosis, M. africanum, M. bovis, M. microti, M. canettii and M. pinnipedii. The M. bovis species can be further divided into M. bovis subsp. bovis, M. bovis subsp. caprae and the M. bovis-derived BCG vaccine strain, whereas M. africanum includes two subtypes, I and II [1–5]. Identification of mycobacteria to the species level on the basis of growth rate, phenotypic characteristics and biochemical tests is laborious and extremely timeconsuming. GenoType MTBC (Hain Lifescience, Nehren, Germany) is a recently developed commercial DNA-strip assay for differentiating MTBC strains isolated from cultured material. The procedure involves isolating DNA from cultured material, multiplex amplification with biotinylated primers and reverse hybridization of the single-stranded, biotin-labeled amplicons to membrane-bound probes. The resulting banding pattern indicates the species of the isolated mycobacterium. The assay permits genetic differentiation of the following species: M. africanum I, M. bovis BCG, M. bovis subsp. bovis, M. bovis subsp. caprae, M. microti and M. tuberculosis/M. africanum II. The aim of the current study was to evaluate the proficiency of this assay for identifying clinical MTBC isolates. The study was performed using 120 consecutive clinical MTBC strains isolated from different patients at our hospital over a 5-year period (2000–2004). Specimens are routinely processed in accordance with international guidelines, and when necessary, an N-acetyl-L-cysteine-NAOH decontamination procedure is used [6, 7]. Specimens are inoculated into BacTAlert 3D tubes (Organon Technika, Durham, NC, USA) and onto Lowenstein–Jensen agar (LJ; bioMerieux, Marcy l’ Etoile, France) and incubated at 37°C for up to 6 and 8 weeks, respectively. All 120 strains included in this study were identified as members of the MTBC using gene probes (AccuProbe; GenProbe, San Diego, CA, USA) and further differentiated to the species level by means of colony morphology and biochemical analysis [nitrate reduction (Nitrati Test; Liofilchem, Roseto, Italy), niacin accumulation (BBL Taxo TB Niacin Test Strips; Becton Dickinson, Sparks, MD, USA), and growth on LJ slants in the presence of 1 μg/ml thiophen2-carboxylic acid hydrazide (TCH; bioMerieux)]. Of the 120 isolates, 119 (99.17%) were identified asM. tuberculosis; the remaining isolate was presumptively identified as M. bovis (negative niacin accumulation and nitrate reductase tests and no growth on TCH). Given this strain distribution, our results can only be applied to M. tuberculosis. Prior to use in this study, the LJ slants of all 120 isolates had been stored in a dark environment at room temperature and a loopful of culture material suspended in 1 ml distilled water had been preserved at −20°C. Four control strains (two M. africanum and two M. bovis) obtained from Institute Pasteur (Paris, France) were used. In parallel, four non-MTBC microorganisms (one M. kansasii, one Corynebacterium jeikeium, one Nocardia abscessus and one Escherichia coli) were tested with GenoType for cross reactions. The GenoType MTBC assay has formerly been evaluated in two German laboratories using control strains along with clinical isolates that were mostly obtained from positive liquid cultures [8, 9]. In the present study, the clinical isolates had been obtained exclusively from solid media. Moreover, these positive LJ slants had been stored for a prolonged period of time (ranging from weeks to years). Eur J Clin Microbiol Infect Dis (2007) 26:151–152 DOI 10.1007/s10096-007-0255-y