Genotyping of Mycobacterium tuberculosis clinical isolates in two cities of Turkey: Description of a new family of genotypes that is phylogeographically specific for Asia Minor

Turkey is a large and densely populated country (area.: 780 576 km², population: around 71 millions). In Turkey, tuberculosis remains an important public health concern with a case notification rate of 26.2/100.000 inhabitants in 2002 [1]. In many of the studies, social aspect of TB is also underestimated. The men/women sex ratio shows a proportion much more important of men (3x) than women which suffers from the disease. Besides, the global drug resistance rates in men are twice higher than that of women. The 17–39 year group accounts for 80% of the studied strains, which shows that a young male population is specifically concerned by tuberculosis [1]. Contrary to the picture of most European countries, where the last years have seen a tremendous amount of studies describing the population-based genetic structure of Mycobacterium tuberculosis, no data of this kind are yet available for Turkey, with the exception of a recent paper describing the diversity of phospholipases genes in 106 clinical isolates of M. tuberculosis [2].

From a geographical, historical and anthropological aspect, Turkey is a link between Europe and Asia, an early region of human settlement, located in the Western part of Eurasia [3]. Various cultural and anthropological influences of early civilizations have left complex scars, making from Turkey an anthropologically rich area. More evidence about the unique nature and complexity of human genetics in Anatolia has been discovered, whether based on mitochondrial or on Y chromosome diversity [4–7]. The existence of a hypothetical proto-Indo-European language, whose link to Anatolia appears likely, is another issue that reinforces the importance of Turkey in human history, especially of the "Indo-European" lineages [8].

Molecular characterization studies on M. tuberculosis in Turkey until recently focused on the use of the "gold-standard" IS6110-RFLP method and on the description and characterization of multi-drug-resistance rates and their mechanisms, two issues of great importance for Public-Health [9–14]. However, to the best of our knowledge, no population-based data were available on the genetic diversity of M. tuberculosis in Turkey. The goal of this study was to get an initial insight into the biodiversity of M. tuberculosis in Turkey by studying a set of 245 DNAs from Anatolia by spoligotyping, a high throughput technique for which large polymorphism databases have been created [15]. These DNAs originated from as many clinical isolates from tuberculosis patients resident in Turkey. We show in this paper that a single genotype or "clonal complex" (ST41) accounts for one fourth of the total TB cases. A second genotype, ST284, may also bear some Eastern Mediterranean or Asian specificity but remains to be more exhaustively investigated. The predominance of a specific clonal complex in Turkey may argue in favour of a long-lasting presence of tuberculosis in this country and open new avenues of investigation to better understand why and how such a clone became predominant.

Tuberculosis remains a great public health concern in Turkey. The resistance to antituberculosis drugs, which represents a specific threat and as such deserves much attention, was recently the focus of many investigations in Turkey [9–14], however, molecular epidemiology is of more recent interest [13, 18]. Indeed, the unraveling of the effect of genetic variability of M. tuberculosis on the presentation of the disease remains a challenging poorly investigated issue, which consists in understanding why a strain may become prevalent in certain communities [19]. Recent results obtained on the polymorphims of genes known to be involved in pathogenicity and virulence (phospholipases) may create a bridge between pathogenicity and population genetics studies [2]. However, the population-based genetic landscape of tuberculosis biodiversity was, to the best of our knowledge, unknown before this study and as such deserved such an attempt to define which genotypes are responsible of the TB cases in Turkey, a subject which, given the highly complex anthropological structure of Turkey, is of great interest for clinical scientists, bacteriologists, and evolutionary biologists.

A total of 245 DNAs extracted from M. tuberculosis clinical isolates from TB Turkish patients were genotyped by spoligotyping. A major genotype, as revealed by spoligotyping ST41, which misses spacer 20–24, 26–27 and 33–36, and had been previously described under the designation of "LAM7", represented up to one fourth of all TB isolates. When these genotypes were further investigated by IS6110-RFLP or by the highly discriminant MIRU-VNTR technique, highly similar profiles were obtained suggesting that these strains define a true genotype family or clonal complex. This genotype is likely to be identical to the one described in Table 2 by IS6110-RFLP and pTBN12 in a recent paper [13]. However, given the highly discriminative power of pTBN12 as a second genotyping method, and the difficulties to compare these patterns, a total of 7 subclusters were described initially (Ia to Ig). How these previous results correlate to MIRU-VNTR-based or spoligotyping-based clustering remains to be further investigated.

The finding of another genotype ST284 that was already detected in SpolDB3 but without origin of potential phylogeographical specificity is intriguing. This genotype is currently under investigation and is also found to be present in Bulgaria (T. Zozio et al., unpublished obervations). Whether this genotype also bears a larger Eastern-Mediterranean or Middle Eastern phylogeographical specificity remains speculative for the time-being.

The incidence of tuberculosis in Turkey was recently estimated around 26.6 new cases per 100.000 inhabitants [1]. For a city such as Malatya (853.658 inhabitants), which has a slightly superior incidence (32/100.000), the total estimated number of new cases per year would be n = 272. Thus, our sampling (n = 145) represents the equivalent of a quarter of a two-year recruitment, which, we assume, is fully representative of the genetic diversity in Malatya. In Ankara, for which the recruitment was less important, similar genotyping results were obtained. In a third city from the border of the Black Sea (Samsun), similar results were also obtained on 100 DNAs by an independent team (A. Sanic and J. Driscoll, personal communication). Thus the genotyping results obtained in Ankara and Malatya seems to be quite representative of Anatolia, suggesting that the ST41/VIT310 and ST41/VIT196 could represent traces of a contemporary and/or historically endemic/epidemic clone in Anatolia. If further investigations on isolates from the Aegean, Mediterranean and eastern sides of Turkey confirm the prevalence of the ST41 genotype, and provided that it is really ancient, one may expect that its distribution will vary depending on the human population structure in Turkey. Thus, the observed geographical variations in the frequency distribution of ST41 may allow to precisely define its presumed origin. However one should be extremely cautious with such historical inferences. Indeed, one should not forget that epidemics by their bursting nature, may rapidly promote the replacement of genotypes by others and that recent human migrations do complexify the issue [20]. Turkey is a country where ancient Central Asian and European civilizations can be seen. Preliminary phylogenetical results (not shown) suggest that the LAM7-TUR genotype family of M. tuberculosis may be related to the large LAM9 superfamily of genotypes; however, another spoligotype, ST353, could also be the ancestor type of ST41 and cannot be excluded at this step as a potential ancestor of ST41. Further studies using combined MIRU-VNTR-spoligotyping will facilitate the finding of the ancestor clone of the LAM7-TUR family.

We described a new phylogeographically-specific clone of M. tuberculosis, designated LAM7-TUR. Further investigations to assess its frequency within all regions of Turkey and its phylogeographical origin and phylogenetic position within the global M. tuberculosis phylogenetic tree will shed new light on its endemicity in Asia Minor.