Mycobacterium ulcerans, the causative agent of Buruli ulcer, is an emerging environmental bacterium in Australia and West Africa. The primary risk factor associated with Buruli ulcer is proximity to slow moving water. Environmental constraints for disease are shown by the absence of infection in arid regions of infected countries. A particularly mysterious aspect of Buruli ulcer is the fact that endemic and non-endemic villages may be only a few kilometers apart within the same watershed. Recent studies suggest that aquatic invertebrate species may serve as reservoirs for M. ulcerans, although transmission pathways remain unknown. Systematic studies of the distribution of M. ulcerans in the environment using standard ecological methods have not been reported. Here we present results from the first study based on random sampling of endemic and non-endemic sites. In this study PCR-based methods, along with biofilm collections, have been used to map the presence of M. ulcerans within 26 aquatic sites in Ghana. Results suggest that M. ulcerans is present in both endemic and non-endemic sites and that variable number tandem repeat (VNTR) profiling can be used to follow chains of transmission from the environment to humans. Our results suggesting that the distribution of M. ulcerans is far broader than the distribution of human disease is characteristic of environmental pathogens. These findings imply that focal demography, along with patterns of human water contact, may play a major role in transmission of Buruli ulcer.

Although several studies have associated Mycobacterium ulcerans (MU) infection, Buruli ulcer (BU), with slow moving water bodies, there is still no definite mode of transmission. Ecological and transmission studies suggest Variable Number Tandem Repeat (VNTR) typing as a useful tool to differentiate MU strains from other Mycolactone Producing Mycobacteria (MPM). Deciphering the genetic relatedness of clinical and environmental isolates is seminal to determining reservoirs, vectors and transmission routes. In this study, we attempted to source-track MU infections to specific water bodies by matching VNTR profiles of MU in human samples to those in the environment. Environmental samples were collected from 10 water bodies in four BU endemic communities in the Ashanti region, Ghana. Four VNTR loci in MU Agy99 genome, were used to genotype environmental MU ecovars, and those from 14 confirmed BU patients within the same study area. Length polymorphism was confirmed with sequencing. MU was present in the 3 different types of water bodies, but significantly higher in biofilm samples. Four MU genotypes, designated W, X, Y and Z, were typed in both human and environmental samples. Other reported genotypes were only found in water bodies. Animal trapping identified 1 mouse with lesion characteristic of BU, which was confirmed as MU infection. Our findings suggest that patients may have been infected from community associated water bodies. Further, we present evidence that small mammals within endemic communities could be susceptible to MU infections. M. ulcerans transmission could involve several routes where humans have contact with risk environments, which may be further compounded by water bodies acting as vehicles for disseminating strains.

Buruli ulcer is a necrotizing skin disease caused by Mycobacterium ulcerans and associated with exposure to aquatic habitats. To assess possible transmission of M. ulcerans by aquatic biting insects, we conducted a fi eld examination of biting water bugs (Hemiptera: Naucoridae, Belostomatidae, Nepidae) in 15 disease-endemic and 12 non-disease-endemic areas of Ghana, Africa. From collections of 22,832 invertebrates, we compared composition, abundance, and associated M. ulcerans positivity among sites. Biting hemipterans were rare and represented a small percentage (usually <2%) of invertebrate communities. No signifi cant differences were found in hemipteran abundance or pathogen positivity between disease-endemic and nondisease-endemic sites, and between abundance of biting hemipterans and M. ulcerans positivity. Therefore, although infection through insect bites is possible, little fi eld evidence supports the assumption that biting hemipterans are primary vectors of M. ulcerans. M ycobacterium ulcerans infection is an emerging skin disease often called Buruli ulcer (BU). Infection results in illness and lasting negative socioeconomic effects in rural areas of the tropics and subtropics (1). The pathologic changes, clinical signs and symptoms, and treatment have been reviewed elsewhere (2-5). In this article we evaluate fi eld evidence for the potential of aquatic invertebrates to be vectors of M. ulcerans.The exact mode of BU transmission remains unknown; however, past epidemiologic studies have associated BU with human activity near, or within, slow-fl owing or standing water bodies that have been created or disturbed by humans (2-4). Although several water-related risk factors have been recognized, none has been consistently reported, making it diffi cult to identify specifi c water-related risk activities (6-8). Most studies suggest that infection occurs through inoculation of M. ulcerans into skin lesions or insect bites (2,4,9-11). Portaels et al. (11) were the fi rst to propose that aquatic insects might serve as vectors of M. ulcerans. This hypothesis maintains that M. ulcerans is found in biofi lms of aquatic habitats and concentrated by grazing or fi lter-feeding invertebrates that are then consumed by predators known to bite humans (11). Initial evidence for this hypothesis used PCR detection of the insertion sequence IS2404 to document M. ulcerans' association with biting water bugs (Hemiptera), fi ltered concentrates of water, detritus, and aquatic plants (4,(12)(13)(14). These studies were important for understanding the possible environmental reservoirs of M. ulcerans. However, IS2404 is now understood to be not specifi c for M. ulcerans because this insertion sequence has been found in a number of other aquatic mycobacterial species, including M. marinum (15)(16)(17). When more discriminatory methods based on detection of variable number tandem repeats were used, many IS2404-positive environmental samples were reported to lack M. ulcerans (18). In light of these recent fi ndings, the relative fr...