Evidence of increasing intra and inter-species transmission of Mycobacterium bovis in South Africa: are we losing the battle?

Tuberculosis caused by Mycobacterium bovis is recognized worldwide as a significant health risk in domestic cattle, farmed and wild animal species as well as in humans. We carried out spoligotyping and variable number of tandem repeat (VNTR) typing methods to characterize 490 M. bovis isolates from livestock (cattle, n=230; pig n=1) and wildlife species (n=259) originating from different farms and regions in South Africa, with the aim to further establish the genetic diversity of the isolates, study the population structure of M. bovis and elucidate the extent of interspecies transmission of bovine tuberculosis. A total of ten spoligotype patterns were identified, two of which were novel (SB2199 and SB2200) and reported for the first time in the literature, while VNTR typing revealed a total of 97 VNTR profiles. Our results showed evidence of clonal expansion for some ancestral strains as well as co-infections with two or three M. bovis strains on some of the cattle and game farms, which suggested independent introductions of infected animals from epidemiologically unrelated sources. Five spoligotypes and nine VNTR profiles were shared between cattle and wildlife. Our findings showed that besides cattle, at least 16 different animal species in South Africa are infected with bovine tuberculosis, and highlight a strong evidence of inter and intra-species transmission of M. bovis. Infection of the blue wildebeest (Connochaetes taurinus) with M. bovis is described for the first time in this report. This update in epidemiological information raises concerns that bovine tuberculosis has increased its spatial distribution in South Africa and is also affecting an increasing number of wildlife species compared to ten years ago.

Mycobacterium tuberculosis at the human/wildlife interface in a high TB burden country.

This study reports on an investigation of Mycobacterium tuberculosis cases in mostly captive wild animals using molecular typing tools [Variable Number of Tandem Repeat (VNTR) typing and Restriction Fragment Length Polymorphism typing]. The investigation included cases from (i) the National Zoological Gardens of South Africa (NZG) recorded between 2002 and 2011; (ii) Johannesburg Zoo, where tuberculosis was first diagnosed in 2007 and has since been detected in three antelope species; (iii) a rehabilitation centre for vervet monkeys (Chlorocebus pygerythrus) in which M. tuberculosis was diagnosed in 2008; and (iv) incidental cases in other facilities including a sable antelope (Hippotragus niger), two unrelated cases in chacma baboons (Papio ursinus) (one of which was from a free-ranging troop) and a colony of capuchin monkeys (Cebus capucinus). Identical genetic profiles of the latter three isolates indicate the persistence of a single M. tuberculosis strain in this population since at least 2006. Results of the outbreak investigation in the captive vervet monkey colony indicate that it was caused by two unrelated strains, while all 13 M. tuberculosis isolates from 11 animal species in the NZG showed different VNTR patterns. A substantial increase in tuberculosis cases of 60% was recorded in the NZG, compared with the previous reporting period 1991-2001, and may indicate a countrywide trend of increasing spillover of human tuberculosis to wild animals. South Africa ranks among the countries with the highest-tuberculosis burden worldwide, complicated by an increasing rate of multidrug-resistant strains. Exposure and infection of captive wildlife in this high prevalence setting is therefore a growing concern for wildlife conservation but also for human health through potential spillback.

Evaluation of the discriminatory power of variable number of tandem repeat typing of Mycobacterium bovis isolates from southern Africa.

The usefulness of variable number tandem repeat (VNTR) typing based on limited numbers of loci has previously proven inferior compared to IS6110-RFLP typing when applied to the study of the molecular epidemiology of bovine tuberculosis (BTB) in both livestock and wildlife in southern Africa. In this study, the discriminatory power of 29 published VNTR loci in the characterization of 131 Mycobacterium bovis strains isolated predominantly from wildlife and a smaller number from livestock in southern Africa was assessed. Allelic diversities calculated when loci were evaluated on a selected panel of 23 M. bovis isolates with identified varying degrees of genetic relatedness from different geographic origins as well as M. bovis BCG ranged from 0.00 to 0.63. Of the 29 loci tested, 13 were polymorphic (QUB 11a, QUB 11b, QUB 18, ETR-B and -C, Mtub 21, MIRU 16 and 26, ETR-E, QUB 26, MIRU 23, ETR-A, and Mtub 12). In addition, a comparative evaluation of the 13 loci on a panel of 65 isolates previously characterized by IS6110 restriction fragment length polymorphism (RFLP) typing and further evaluation on 41 isolates with no typing history from Kruger National Park (KNP) highlighted that M. bovis from epidemiologically unrelated cases of BTB in different geographic regions can be adequately distinguished. However, there is a need for improvement of the method to fully discriminate between the parental KNP strain and its clones to allow the detection of evolutionary events causing transmission between and within wildlife species.

High Mycobacterium bovis genetic diversity in a low prevalence setting.

The genetic diversity among South African Mycobacterium bovis isolates from cattle was determined by genetic fingerprinting. The restriction fragment length polymorphism (RFLP) markers IS6110 and polymorphic GC-rich sequence (PGRS) as well as spoligotyping and determination of variable number of tandem repeats (VNTR) were used to characterize sub samples of 91 M. bovis field isolates. PGRS RFLP was the single most discriminatory method and combinations of typing methods, which included IS6110 and/or PGRS had the highest discriminatory power, able to reveal 29 distinct genotypes among 35 farms with no epidemiological link. Three of the farms were co-infected with two genetically unrelated strains. In contrast to reports from European and also other colonised countries on the African continent our findings are suggestive of a high genetic diversity of M. bovis in South Africa's cattle population, implying a variety of unrelated ancestor strains. Despite effective intervention through test-and-slaughter campaigns no indication of a 'founder effect' was apparent in the panel of isolates derived from all infected provinces.