Open Access

Research Article

• Download Article XML
• Download Article PDF
• Download Citation
• E-mail this Article
• Order Reprints
• Print this Article

Introduction

Buruli ulcer (BU), the third most common mycobacteriosis in humans after tuberculosis and leprosy is an indolent, necrotizing disease of skin, subcutaneous tissue and, occasionally, bones [1]. BU has emerged in recent times as an increasingly important cause of morbidity around the world, and has been reported in 30 countries, mostly in tropical areas [2]. This disease is caused by Mycobacterium ulcerans which is peculiar among pathogenic mycobacteria because it produces a potent necrotizing exotoxin, mycolactone, which is a major virulence factor [3].

Although incompletely understood, the epidemiology of BU strongly associates the disease with wetlands and especially slow-flowing or stagnant water [4]–[6]. Indeed, there is indirect evidence that M. ulcerans is an environmental pathogen transmitted to humans from its aquatic niches; however, modes of transmission are unclear [7]. The initial hypothesis that predatory aquatic insects, including Naucoridae and Belostomatidae, were involved in transmission [8] was later reinforced by reports that the salivary glands of Naucoris were colonized by M. ulcerans when fed on infected grubs, and that bites of infected Naucoris transmitted the pathogen to mice [9]. The observation that non-infected humans exposed to aquatic environments in BU endemic areas have higher titers of antibodies to salivary proteins of Naucoridae and Belostomatidae than BU patients in the same areas [10] shows that these water bugs bite humans in natural settings. However, Naucoridae and Belostomatidae are carnivorous insects that normally feed on other aquatic insects, snails, and small fish and only bite humans incidentally [11]. Thus, the significance of biting by M. ulcerans-colonized aquatic insects in the transmission of BU to humans is unknown, and other forms of transmission, including skin trauma, have been considered [12].

Since the discovery of IS2404 [13], an insertion sequence with more than 200 copies in M. ulcerans [14], multiple studies have detected IS2404 in environmental aquatic samples, indicating that M. ulcerans is probably present in such samples, and supporting the concept that the etiologic agent of BU is an environmental pathogen. IS2404 was found in samples of water and detritus from swamps in Australia [15],[16],[17], and in aquatic plants [18], insects (Belostomatidae, Naucoridae, Hydrophilidae), crustaceans and mollusks (Bulinus sp. and Planorbis sp.), and fish (including Tilapia sp.) in western tropical Africa [8],[9],[18],[19],[20]. More recently IS2404 was detected in mosquitos in Australia [21]. However, the recent discovery of IS2404 in aquatic mycobacteria other than M. ulcerans requires re-evaluation of the use of IS2404 PCR for the detection of M. ulcerans DNA in the environment [22],[23] and emphasizes the importance of the isolation of M. ulcerans from environmental sources.

Numerous extensive studies have failed to isolate M. ulcerans in pure culture from the environment, even in highly endemic areas of BU, e.g. in Uganda [24], the Democratic Republic of Congo [4],[5],[25] and West Africa [19].

Two cultures from salivary glands of wild aquatic insects (Naucoridae) collected in BU endemic areas of Côte d'Ivoire were positive for IS2404 and were considered to be related to M. ulcerans; however, no phenotypic characteristics of these isolates were reported other than their virulence for mice [9]. In 2004, Marsollier et al. obtained IS2404 PCR positive cultures from two samples of aquatic plants collected in a BU endemic area of Côte d'Ivoire [18]. One IS2404 positive BACTEC culture inoculated into mice revealed infection compatible with M. ulcerans. The culture was, however, contaminated by Mycobacterium szulgai and M. ulcerans could not be obtained in pure culture even after passages through mice.

As briefly reported previously [26], a pure culture of M. ulcerans (isolate 00-1441) was obtained from an aquatic insect from Benin. In that report no description was given of the methods employed for the isolation of M. ulcerans 00-1441 and of the phenotypic and genetic characteristics of the isolate. Here we present the detailed results of the isolation and characterization of strain 00-1441, establishing that this mycobacterium is an African type of M. ulcerans with high virulence for mice. Strain 00-1441 represents the first well characterized M. ulcerans strain isolated in pure culture from an environmental source.

Materials and Methods

Environmental specimens

Collection and in vitro culture.

Four aquatic specimens collected in a BU endemic area of Benin and one in Togo that were part of a larger study on the frequency of mycobacteria in the environment (Portaels et al., manuscript in preparation) were cultured in vitro and inoculated into mouse footpads. The specimens were transported from the field to the laboratory in sterile tubes at 4°C and kept frozen until they were identified by an entomologist.

Types of specimens collected and their origins are indicated in Table 1.

Table 1. Ziehl-Neelsen staining, culture and PCR results for 5 aquatic specimens collected in Buruli ulcer endemic regions of Benin and Togo

doi:10.1371/journal.pntd.0000178.t001

The specimens were thoroughly diced with sterile disposable scalpels in a sterile mortar. They were further homogenized with a sterile pestle and suspended in 2 ml phosphate buffered saline (PBS) (Oxoid, Hampshire, England; pH = 7.3±0.2). The mortars and pestles were used only for environmental specimens. Decontamination was performed by treatment of the suspensions for 15 minutes with an equal volume of aqueous 2.0% sodium hydroxide (NaOH) containing sodium citrate (1.45% final concentration) and N-acetyl-l-cysteine (0.5% final concentration). The suspensions were centrifuged for 20 minutes at 3,800 g and fractions of the resulting sediments were used for Ziehl-Neelsen staining (ZN), culture and PCR.

Primary cultures were performed at 32°C by inoculating the sediment on Löwenstein-Jensen (LJ) solid medium and in BACTEC 12B broth (Becton Dickinson Microbiology Systems, Sparks, MD, USA) supplemented with PANTA and 1.25% egg yolk [27]. Growth index (GI) was measured weekly for 3 months with a BACTEC 460 TB instrument (Becton Dickinson, Sparks, MD, USA). When the GI reached 999, ZN was performed, and BACTEC cultures were tested by IS2404 PCR [28]. Inoculated LJ tubes were incubated for 12 months and observed every 2 weeks.

PCR on decontaminated specimens and BACTEC cultures.

DNA was extracted from decontaminated specimens and BACTEC positive cultures as previously described with minor modifications [29]. Briefly, 250 µl suspensions were treated with equal volumes of lysis buffer L6 (5 M GuSCN, 5 0 mM Tris, pH 6.4, 22 mM EDTA, 2% Triton X-100) followed by 50 µl of proteinase K (20 mg/ml). This mixture was incubated overnight at 60°C with gentle shaking. To capture DNA, 40 µl of diatomaceous earth stock solution (10 g diatomaceous earth obtained from Sigma Aldrich Chemie GmbH, Steinheim, Germany) in 50 ml of H₂O containing 500 µl of 37% (wt/vol) HCl was added to the suspension and placed in a shaker incubator at 37°C for 2 hours to avoid sedimentation of the diatomaceous earth.

The pellets were washed with 900 µl of L2 buffer (5 M GuSCN, 50 mM Tris, pH 6.4) [29] followed by 900 µl of 70% ethanol and 900 µl of acetone. The pellet was then dried at 55°C and resuspended in 90 µl TE (10 mM Tris, 1 mM EDTA, pH 8). Tubes were centrifuged and 50 µl of the DNA extract transferred to a new tube.

PCR reaction: A nested PCR was performed targeting the insertion sequence element IS2404 with four primers: PGP1 (5′-AGGGCAGCGCGGTGATACGG-3′), PGP2 (5′-CAGTGGATTGGTGCCGATCGT-3′), PGP3 (5′-GGCGCAGATCAACTTCGCGGT-3′) and PGP4 (5′-CTGCGTGGTGCTTTACGCGC-3′) (GenBank accession n° U38540), [13].

The following procedures were employed:

For the first PCR run, 5 µl of the DNA extract was added to 45 µl PCR reaction mixture containing 20 pmol of each primer (PgP1 and PgP2), 1U of Taq DNA polymerase (Roche Molecular Systems, Brussels, Belgium), 200 µM concentrations of each deoxyribonucleotide triphosphate, 1.5 M MgCl₂, 0.1% X-100, and 10 mM Tris–HCl (pH 8.4) and overlaid with mineral oil. Cycling was as follows: denaturation at 94°C for 5 min; amplification for 40 cycles at 94°C for 45 sec, 64°C for 45 sec and 72°C for 45 sec and a final extension at 72°C for 10 min.

For the second PCR run, 0.25 µl of the first-run product was amplified in a 25 µl reaction mixture with primers PgP3 and PgP4. Cycling conditions were similar to the first run except that amplification was reduced to 25 cycles. Amplified DNA (7 µl) was then submitted to electrophoresis in 2% agarose gel and detected by ethidium bromide staining and UV transillumination.

Mouse footpad inoculation.

The three IS2404 PCR positive BACTEC cultures (Table 1) were inoculated (0.03 ml) into the left hind footpad of three 8 week old female mice (strain NMRI).

Mice were sacrificed after 6, 9 or 12 months (Table 2). Entire feet were placed into 10% formalin for histopathological analysis, or were used for microbiological analysis as described previously [30]. For specimens 97–1455 and 98–443, mouse footpad suspensions (harvested after 9 months) were also passaged (P1) into 3 other mice. A second passage (P2) was done from P1 after 12 months. No passage was performed for specimen 98–447 because only one mouse survived for 9 months and the entire foot of this mouse was used for histopathologic analysis.

Table 2. Results of inoculation of mouse footpads with BACTEC cultures of 2 environmental specimens

doi:10.1371/journal.pntd.0000178.t002

Histopathologic analyses.

Histopathologic analyses of mice feet were made at the indicated times after inoculation of positive BACTEC vials and, in some cases, after passage of inoculated footpads to other mice. The entire feet were intact at this stage. Specimens were decalcified with a solution containing 4% concentrated hydrochloric acid and 4% concentrated formic acid in distilled water or with a commercial solution (Thermo, Shandon TBD-1 Rapid Decalcifier, Chesire, UK) for a period of 45 min. Multiple longitudinal sections 1.5–2 mm thick were cut, processed routinely and sectioned at 4 µm. Sections were stained by hematoxylin-eosin, ZN, Grocott's methenamine-silver and Brown-Hopps Gram methods [31].

Identification of the mycobacteria.

Mycobacteria cultivated directly from the aquatic specimens and the M. ulcerans isolate 00-1441 cultivated from a mouse footpad (see Table 2) were identified as described previously [32].

Analysis of isolate 00-1441

16S rRNA gene sequencing and phylogenetic analysis.

The isolate 00-1441 was also identified by partial analysis of the 3′ end region as described previously [33] as well as the 5′ end region of the 16S rRNA gene by Eurogentec (Liège, Belgium) with an automated nucleic acid sequencer (Applied Biosystems, Foster City, CA, USA).

Mycolactone extraction and analysis.

Acid soluble lipid containing mycolactones were extracted from M. ulcerans strains 1615 and 00-1441 with chloroform-methanol 2:1 followed by back extraction with ice-cold acetone as previously described [34]. Lipids were resolved by silica thin-layer chromatography using a chloroform-methanol-water (90:10:1) solvent system and visualized by charring with anisaldehyde in 10% sulfuric acid.

Mass spectrometric (MS) analysis of the mycolactone extracts was performed as previously described [34]. Ten milligrams of the dried acetone soluble lipid (ASL) extract were re-suspended in 1 milliliter of methanol and subsequently filtered prior to MS analysis. Using a Cole Palmer 74900 series syringe pump, the methanolic extract was perfused into an ion trap ESI Bruker-Esquire mass spectrometer at dry temperature of 300°C, gas flow of 5 liters/min and nebulizer pressure of 15 lb/in².

Mouse footpad inoculation.

The present study was conducted under the guidelines and approval of the Research Ethics Committee of the Life and Health Sciences Research Institute (Braga, Portugal).

Isolate 00-1441 was grown on LJ medium at 32°C for approximately 2 months, recovered from LJ slants, diluted in PBS to a final mass concentration of 1 mg/ml, and vortexed vigorously using 2-mm glass beads. In all the experiments, the number of acid-fast bacilli (AFB) in each inoculum was determined by the method of Shepard and McRae [35], using ZN staining (Merck, Darmstadt, Germany). The suspensions revealed more than 90% viable bacilli as assessed with the LIVE/DEAD BacLight Kit (Molecular Probes, Leiden, The Netherlands).

Eight-week-old female BALB/c or NMRI mice were obtained from Charles River Laboratories (Barcelona, Spain) and were housed in specific pathogen-free conditions with food and water ad libitum. Both strains of mice were infected in the left hind footpad with 0.03 ml of a suspension containing 5.4 log₁₀ AFB of M. ulcerans 00-1441.

As an index of lesion development, footpad swelling was measured over time with a caliper. Bacterial proliferation was evaluated in footpad homogenates of infected mice at selected time points, as previously described [30],[36]. Histological analysis of the feet was carried out as described above.

Infection of murine bone marrow-derived macrophages.

Bone marrow-derived macrophages (BMDM) from BALB/c mice were used as mouse primary macrophages and were prepared as previously described [36]. Briefly, both femurs were removed under aseptic conditions. Bones were flushed with cold Hank's Balanced Salt Solution (HBSS, Gibco, Paisley, UK). The resulting cell suspension was centrifuged and resuspended in Dulbecco's Modified Eagle's Medium (DMEM, Gibco) supplemented with 10 mM HEPES (Sigma, St. Louis, MO), 1 mM sodium pyruvate (Gibco), 10 mM glutamine (Gibco), 10% of heat-inactivated fetal bovine serum (Sigma) and 10% L929 cell-conditioned medium (Complete DMEM [cDMEM]). To remove fibroblasts or differentiated macrophages, the cells were cultured at 37°C in a 5% CO₂ atmosphere for a period of four hours in cell culture dishes (Nunc, Naperville, IL) with cDMEM. The non-adherent cells were collected with warm HBSS, centrifuged and distributed in 24-well plates at a density of 5×10⁵ cells/well and incubated at 37°C in a 5% CO₂ atmosphere. L929 cell-conditioned medium was added 4 days after seeding and medium was renewed on the seventh day. After 10 days in culture, cells were completely differentiated into macrophages. Twelve hours before infection, macrophages were incubated at 32°C in a 5% CO₂ atmosphere and maintained at 32°C until the end of the experimental infection.

For macrophage infectivity assays bacterial suspensions were prepared as described above and further diluted in cDMEM to obtain the selected multiplicity of infection (MOI) of 1:1 (bacteria/macrophage ratio). Cells were incubated for 4 hours at 32°C in a 5% CO₂ atmosphere and then washed with warm HBSS to remove non-internalized bacteria and re-incubated in cDMEM for eight days.

Genotyping methods.

Genotyping methods developed to analyze the diversity among M. ulcerans and M. marinum strains from different geographical areas were applied to isolate 00-1441. IS2404 restriction fragment length polymorphism (RFLP) and PCR restriction profile analysis (PRPA) were performed as described previously [37],[38].

Mycobacterial Interspersed Repetitive Units (MIRU) Variable Number Tandem Repeat (VNTR) analysis was also used to investigate the MIRU-VNTR profile of 00-1441 as previously described [39],[40].

Results

Environmental specimens

Results of ZN staining, culture and PCR studies for the 5 aquatic specimens are shown in Table 1. Table 2 shows the results of the mouse footpad inoculation with the BACTEC suspensions (98–447, 97–1455 and 98–443) that were positive by IS2404 PCR after inoculation with the aquatic specimens.

Specimen 98–447: Histopathologic analysis of one mouse sacrificed after 9 months revealed a few well formed granulomas with minimal necrosis around blood vessels, nerves and in muscle. There were large numbers of beaded AFB in the granulomas.

Specimen 97–1455: Of the three mice inoculated with this BACTEC culture (P1 in Table 2), two were sacrificed 9 months after inoculation. The histopathologic analysis of the footpad of one mouse showed marked necrosis with a mild granulomatous response, inflammation of periosteum and many large clumps of AFB in necrotic areas. The footpad homogenate of the third mouse was positive for AFB and the culture on LJ was positive for mycobacteria (isolate 99–2832) but was lost due to contamination by nonacid-fast bacteria; however, PCR performed on the contaminated culture was positive for IS2404. This mouse footpad was inoculated in vitro and passaged twice into two groups of three mouse footpads. The second (P2 in Table 2) and third passages (P3 in Table 2) were negative for AFB by ZN staining and by culture.

Specimen 98–443: This homogenate of a Gerris sp. aquatic insect ultimately produced the M. ulcerans isolate 00-1441 after culture in BACTEC (positive for IS2404) inoculated in mouse footpads (P1) and followed by two other mouse footpad passages (P2 and P3).

Following the first mouse inoculation (P1), one animal died after 1 month and the other two were sacrificed 9 months after inoculation. Histopathologic evaluation of the footpad of one of these mice showed granulomatous changes with minimal necrosis around blood vessels and nerves. There were large numbers of scattered, short, beaded AFB in the granuloma. ZN stain and culture were negative for the footpad of the third mouse inoculated in vitro. The suspension obtained from the third mouse was used to reinoculate 3 other mice (P2). One P2 mouse died after 6 months and the other two were sacrificed 12 months later. The histopathologic study of one footpad showed minimal nonspecific inflammation. The other mouse footpad was negative for AFB and by culture and was used for a third passage (P3) into 3 mice. Two of the P3 mice were sacrificed after 6 months. One footpad used for histopathologic study showed nonspecific inflammation. The other footpad was ZN-negative but gave a positive culture on LJ (5 colonies) after 2 months incubation at 32°C. The isolate (00-1441) was further analyzed and identified as M. ulcerans (see below).

The remaining P3 mouse, sacrificed after one year, did not reveal any histopathologic changes.

Characterization of isolate 00-1441

Phenotypic characterization.

The phenotypic characteristics of M. ulcerans isolate 00-1441 and those of some M. ulcerans geographic subgroups are given in Table 3. Isolate 00-1441 has the same phenotypic characteristics as M. ulcerans strains belonging to the African subgroup [33], i.e., it was scotochromogenic, did not grow on LJ containing 250 µg/ml hydroxylamine and was acid phosphatase positive.

Table 3. Phenotypic characteristics of isolate 00-1441 and M. ulcerans geographic subgroups

doi:10.1371/journal.pntd.0000178.t003

16S rRNA gene sequencing and phylogenetic analysis.

The 5′ end of 16S rDNA sequence was 100% identical to that of M. ulcerans [41].

Sequencing results of the 3′ end of 16s rDNA for 00-1441 and the different M. ulcerans types are given in Table 3.

Isolate 00-1441 was characterized by a G at position 1248 shared by M. ulcerans type 1, type 2, type 3, and Mycobacterium shinshuense, a C at position 1289 typical of M. ulcerans type 1 and type 2 and CTTT at positions 1450–1452 unique to M. ulcerans type 1 [33]. However, a point mutation was found at position 1317, with a T instead of a C typical for all M. ulcerans types and Mycobacterium marinum.

Mycolactone analysis.

Thin-layer chromatography of ASLs showed that isolate 00-1441 produces mycolactone A/B as a major lipid species (data not shown). Mass spectroscopy provided definitive evidence for mycolactone production as evidenced by sodium adducts at 765.9, representing the intact mycolactone molecule, and the hydrolysis product showing the core ion at 447.3 (Fig. 1). Enrichment for the intact ion using ion trap for positive ions between m/z 755–775 revealed a major peak consistent with intact mycolactone A/B (Fig. 2). Fragmentation pattern of this species following MS-MS demonstrated the presence of the characteristic mycolactone fragmentation pattern with core mycolactone at 429.4 and fatty acid side-chain at 359.3 (Fig. 3).

Figure 1. Mass spectroscopic analysis of ASL of M. ulcerans 00-1441 showing the hydrolysis product of mycolactone A/B at m/z 447.3 (*).

The intact mycolactone can be identified at lower intensity at m/z 765.9 (**).

doi:10.1371/journal.pntd.0000178.g001

Figure 2. Mass spectroscopic analysis of the ASL of M. ulcerans 00-1441 showing the mycolactone A/B sodium adduct at m/z 765.6.

(In this case an ion trap was set to select for positive ions in the range m/z 755–775.)

doi:10.1371/journal.pntd.0000178.g002

Figure 3. MS/MS fragmentation pattern of the mycolactone A/B ion at m/z765.6.

The two main characteristic fragments are those of the macrolide ring m/z = 429.4 (*) and the side chain at m/z = 359.3 (**).

doi:10.1371/journal.pntd.0000178.g003

Infection of murine bone marrow macrophages

As previously described for virulent M. ulcerans strains [36], isolate 00-1441 showed cytotoxic activity against BMDM infected at an MOI 1:1 as deduced at day 4 post-inoculation from the occurrence of mycolactone-associated cytopathic signs [42] namely, cell rounding, shrinkage and detachment of the macrophages (Fig. 4).

Figure 4. Cytotoxic activity of M. ulcerans 00-1441 to BMDM.

BMDM were infected with M. ulcerans 00-1441 at an MOI of 1:1. Macrophages were photographed by phase-contrast microscopy at 4 h (A), 4 days (B), and 6 days (C) postinfection. Macrophage cell rounding, shrinkage and detachment are present at days 4 and 6 post-infection.

doi:10.1371/journal.pntd.0000178.g004

Mouse footpad inoculation.

Footpads of three NMRI mice showed swelling beginning 7 days after inoculation of 00-1441. Isolate 00-1441 was also virulent for BALB/c mice. The proliferation of bacilli, assessed by AFB counts, and the level of pathologic changes, evaluated by footpad swelling and emergence of ulceration, were monitored throughout the experimental period of infection of BALB/c mice footpads. As shown in Fig. 2, swelling became apparent during the second week of infection; ulceration was observed after the fourth week post-inoculation. AFB counts in footpad homogenates increased significantly from 5.15 log₁₀ to 7.50 log₁₀ between days 0 and 27 post-inoculation (Fig. 5).

Figure 5. Footpad swelling and bacterial proliferation during infection with isolate 00-1441.

BALB/c mice were infected subcutaneously in the footpad with 5.2 log¹⁰ AFB of M. ulcerans 00-1441. The degree of pathologic changes was assessed by measurement of footpad swelling (▪) (n = 8). The proliferation of bacilli in the footpad was determined by counting AFB (open bars) of footpad homogenates (n = 5). Significant differences were performed using Student's t test (**, p≤0.01).

doi:10.1371/journal.pntd.0000178.g005

Observation of serial footpad sections of BALB/c mice showed an acute neutrophilic inflammatory response in the subcutaneous tissue early after infection with 00-1441 (data not shown). In the course of the second week of infection, evidence of dermal edema was found (Fig. 6A) along with a mixed inflammatory infiltrate containing mononuclear cells and neutrophils (Fig. 6A and B), surrounding the necrotic center of the lesion (Fig. 6B). By weeks 3 to 4, the necrotic focus expanded progressively, invading healthy tissue and numerous AFB were observed in areas co-localizing with leukocytes, predominantly of the mononuclear type (Fig. 6C). In more advanced stages of infection, and concurring with ulceration, the subcutaneous tissue of BALB/c and NMRI mice revealed extensive necrotic acellular areas with clumps of free bacilli (Fig. 6D).

Figure 6. Histologic sections of mouse footpads infected with M. ulcerans 00-1441.

BALB/c (A–C) and NMRI (D) mice were infected subcutaneously with 5.4 log¹⁰ M. ulcerans 00-1441. At different time points the footpads were harvested and processed for histologic analysis. Dermal edema could be found by the second week of infection (A), along with a necrotic center surrounded by both chronic and acute inflammatory infiltrates (B) (hematoxylin-eosin stained sections). After 4 weeks of infection, bacilli were observed co-localized with cells (C) and in large clumps in the necrotic center of the lesion (D) (Ziehl-Neelsen stained sections).

doi:10.1371/journal.pntd.0000178.g006

An important histopathological finding regarding footpad infection by strain 00-1441 in NMRI and BALB/c mice was the presence of AFB in the bones (Fig. 7A) and bone marrow (Fig. 7B) of the feet. Extensive destruction of the bone was observed with erosion of the cortex and replacement of marrow by inflammatory cells (Fig. 7C and D).

Figure 7. Ziehl-Neelsen staining of bone tissue of mice infected with M. ulcerans 00-1441 (A, B and inset of D, arrows).

NMRI mice were infected subcutaneously in the footpad with 5.4 log¹⁰ M. ulcerans 00-1441. At 6 weeks post-infection, bacilli were found both in the bone tissue and in the bone marrow (A, B and inset in D, arrows). Hemotoxylin-eosin staining of C and D reveals extensive destruction of bone. Outlines of dead cortical bone spicules are seen in C and D (arrowheads). Marrow is necrotic and replaced by chronic inflammation.

doi:10.1371/journal.pntd.0000178.g007

These results indicate that M. ulcerans strain 00-1441 is highly virulent for mice.

Genotyping results.

The IS2404 RFLP banding pattern of 00-1441 was identical to that of African M. ulcerans isolates [37]. Using PRPA, 00-1441 also yielded a profile similar to that of African M. ulcerans isolates [38].

Comparison of MIRU-VNTR profile based on 6 loci showed that 00-1441 has a typical Atlantic African genotype [40].

Discussion

The prevailing concept that BU is associated with wetlands, especially slow-flowing or stagnant water, implies that M. ulcerans is an environmental pathogen transmitted to humans from particular aquatic niches. Historically, the presence of M. ulcerans in aquatic samples, including water, mud, aquatic plants, aquatic insects, aquatic mollusks, crustacea and small fish, has been inferred from the detection by PCR of the insertion sequence IS2404, highly represented in the genome of M. ulcerans [14]. All previous attempts to isolate fully characterized M. ulcerans from environmental samples, however, have failed, and recent evidence [22] indicates that IS2404 positivity alone is inadequate to establish the presence of M. ulcerans in environmental samples.

M. ulcerans 00-1441, isolated from a Hemiptera (Water Strider, Gerris sp.) collected from a swamp in a BU endemic region (Zagnanado, Benin), represents the first fully characterized culture of the agent of BU from an environmental source. Isolate 00-1441 was identified as M. ulcerans by the following criteria:

1. Phenotypic characteristics are identical to those of M. ulcerans strains from Africa [26];

2. Nucleotide sequence of the 5′ end of 16S rRNA gene has 100% identity with M. ulcerans and M. marinum;

3. Nucleotide sequence of the 3′ end of 16S rRNA gene is identical to that of M. ulcerans (African type) [33] except for a single nucleotide substitution at position 1317;

4. Identical profile with the predominant African type for the genotypic IS2404-RFLP, PRPA assays [37],[38], and MIRU-VNTR [40];

5. Identical mycolactone A/B to that produced by M. ulcerans strains from Africa [34];