Logo of iaiPermissionsJournals.ASM.orgJournalIAI ArticleJournal InfoAuthorsReviewers
Infect Immun. Nov 2006; 74(11): 6037–6045.
Published online Aug 21, 2006. doi:  10.1128/IAI.00970-06
PMCID: PMC1695495

Globally Distributed Mycobacterial Fish Pathogens Produce a Novel Plasmid-Encoded Toxic Macrolide, Mycolactone F[down-pointing small open triangle]


Mycobacterium ulcerans and Mycobacterium marinum are closely related pathogens which share an aquatic environment. The pathogenesis of these organisms in humans is limited by their inability to grow above 35°C. M. marinum causes systemic disease in fish but produces localized skin infections in humans. M. ulcerans causes Buruli ulcer, a severe human skin lesion. At the molecular level, M. ulcerans is distinguished from M. marinum by the presence of a virulence plasmid which encodes a macrolide toxin, mycolactone, as well as by hundreds of insertion sequences, particularly IS2404. There has been a global increase in reports of fish mycobacteriosis. An unusual clade of M. marinum has been reported from fish in the Red and Mediterranean Seas and a new mycobacterial species, Mycobacterium pseudoshottsii, has been cultured from fish in the Chesapeake Bay, United States. We have discovered that both groups of fish pathogens produce a unique mycolactone toxin, mycolactone F. Mycolactone F is the smallest mycolactone (molecular weight, 700) yet identified. The core lactone structure of mycolactone F is identical to that of M. ulcerans mycolactones, but a unique side chain structure is present. Mycolactone F produces apoptosis and necrosis on cultured cells but is less potent than M. ulcerans mycolactones. Both groups of fish pathogens contain IS2404. In contrast to M. ulcerans and conventional M. marinum, mycolactone F-producing mycobacteria are incapable of growth at above 30°C. This fact is likely to limit their virulence for humans. However, such isolates may provide a reservoir for horizontal transfer of the mycolactone plasmid in aquatic environments.

Mycobacterium marinum is a globally distributed pathogen of marine and freshwater fish which also causes skin infections in humans (7, 9). M. marinum is phenotypically distinguished from other mycobacteria by its low optimal growth temperature, light-induced carotenoid production, and relatively rapid growth rate compared to other slow-growing Mycobacterium species. There is considerable heterogeneity among M. marinum isolates, and several subgroups have been described (28, 33-35).

Mycobacteriosis was first diagnosed in fish from the Red Sea in 1990 (5). The infection was initially found in cultured sea bass (Dicentrarchus labrax) in Eilat and has since been found in over 20 different fish species and a hawksbill sea turtle. The Red Sea isolates differed phenotypically from other M. marinum strains by being scotochromogenic (having constitutive pigment production). Whereas most M. marinum strains form colonies on mycobacterial media within 8 days, initial growth was not obtained from these isolates for at least 2 weeks. Similar isolates have also been found in the Mediterranean Sea in Greece and Italy. Molecular characterization of the Israeli isolates from fish confirmed their identity as M. marinum, but analysis of the 16S rRNA gene showed that the isolates formed clades within the species (33, 34). Molecular comparison of the fish isolates with human isolates of M. marinum from Israel showed that human and fish isolates fell into discrete groups based on 16S rRNA, hsp65, and AFLP patterns (34). Human isolates were designated A strains, whereas fish isolates were placed in the B group. Within group B, marine isolates (B1) could be distinguished from freshwater isolates (B2). A growing number of similar M. marinum strains have been isolated from marine fish in various locations along the Mediterranean coast (Italy, Israel, and Greece) as well as from fish from other aquatic environments in Israel.

The Red Sea M. marinum strain is the only mycobacterial species detected in the marine environment in Israel and is associated with unusually severe pathology. Though cutaneous lesions are not always present, the spleens and kidneys of diseased fish are riddled with granulomas containing massive numbers of acid-fast bacilli (Fig. (Fig.1).1). The Israeli marine strain was first isolated from cultured sea bass (Dicentrarchus labrax), but many Red Sea species appear to be susceptible to infection (5, 8, 33).

FIG. 1.
Pathology showing infection with M. marinum DL240490 in sea bass (Dicentrarchus labrax) from an outbreak of mycobacterial disease in the Red Sea. (A) ...

Newly emerging mycobacterial fish pathogens from the Chesapeake Bay in the United States have also been reported (15, 25). In 2005 a new mycobacterial species, Mycobacterium pseudoshottsii, was described in association with epizootic fish disease among striped bass (Morone saxatilis) in the Chesapeake Bay, Maryland. Molecular analysis showed that the 16S rRNA gene from M. pseudoshottsii was greater than 99% similar to that of an Israeli marine M. marinum isolate, DL240490 (25). M. pseudoshottsii was reported to be closely related to M. marinum, to the human pathogen Mycobacterium ulcerans, and to a newly described frog pathogen, Mycobacterium liflandii (24). It was of particular interest that M. pseudoshottsii, like M. liflandii, contained an insertion sequence, IS2404, previously found only in M. ulcerans and in an atypical M. marinum strain isolated from a human patient (25, 30).

DNA-DNA hybridization and multilocus sequence analysis show that M. marinum is extremely closely related to M. ulcerans (28, 31). The two species share 99.8% identity within the 16S rRNA gene, and a growing body of evidence suggests that M. ulcerans evolved relatively recently from an M. marinum ancestor (29). Two major events have been associated with the evolution of M. ulcerans. One is acquisition of a plasmid-encoded macrolide toxin, mycolactone, and the second is major genome remodeling associated with insertion of over 300 copies of two insertion sequences, IS2404 and IS2606. Acquisition of these IS elements has created a large number of pseudogenes in M. ulcerans, leading to the hypothesis that M. ulcerans is adapting to a specialized niche (http://genopole.pasteur.fr/Mulc/BuruList.html).

Despite the close taxonomic relationship, M. marinum and M. ulcerans produce distinctly different diseases (7, 9). Whereas M. marinum causes a granulomatous intracellular infection in fish and humans, M. ulcerans produces an extracellular infection, Buruli ulcer, characterized by severe tissue destruction and the presence of few inflammatory cells at the site of infection (2, 6, 20, 36). Genetic evidence with mycolactone-negative mutants suggests that mycolactone is responsible for the unique characteristics of Buruli ulcer (1, 12). Recently a second mycolactone-producing mycobacterial pathogen, M. liflandii, was isolated from Xenopus tropicalis and Xenopus laevis frogs in U.S. laboratories (24, 32). This organism is also closely related to M. marinum and M. ulcerans.

Mycolactone is a hybrid polyketide composed of a lactone core and a fatty acid side chain (14). All strains of M. ulcerans produce a major mycolactone species along with several minor congeners (4, 16, 23). The first mycolactones were isolated from a Malaysian isolate of M. ulcerans as a 3:2 mixture of two stereoismers designated mycolactone A/B (14). Since then it has been shown that all African and Malaysian isolates of M. ulcerans produce identical molecules. However, there is heterogeneity in the mycolactones produced by M. ulcerans from different geographic areas. M. ulcerans strains isolated in Australia produce a different mycolactone, mycolactone C, whereas Asian isolates of M. ulcerans produce primarily mycolactone D (23). A novel plasmid-encoded mycolactone, mycolactone E, has been recently identified in the frog pathogen M. liflandii (23). The large polyketide synthase genes for mycolactone biosynthesis comprise a 110-kb gene cluster on a large (154- to 180-kb) plasmid (29). The lactone core structure, encoded by mlsA, is conserved in all mycolactones. Molecular diversity has arisen through modifications of the fatty acid side chain structure encoded by mlsB.

Despite the structural heterogeneity, all mycolactone species produce apoptosis and necrosis in cultured cells (1, 13, 26). Mycolactone A/B is the only mycolactone evaluated in vivo. In guinea pig and mouse infection models as well as in human infection, mycolactone A/B causes apoptotic cell death as well as death via necrosis (1). In addition, in humans and guinea pigs, mycolactone is associated with immunosuppression (36, 37).

The close taxonomic relationship of an Israeli M. marinum isolate (DL240490) to M. pseudoshottsii, the relationship of these species to M. ulcerans, and the shared association with aquatic environments led us to investigate further the relationship of M. pseudoshottsii and Israeli fish isolates to M. ulcerans. Results from these studies show that M. marinum strains isolated from fish in several locations in the Mediterranean and Red Seas as well as M. pseudoshottsii strains isolated from fish in the Chesapeake Bay produce a plasmid-encoded mycolactone. Despite the wide geographic separation of mycolactone-producing M. marinum and M. pseudoshottsii, both groups of organisms produce an identical mycolactone, designated mycolactone F.



Fish were trapped, euthanatized, and dissected within 48 h. Dissected tissue was fixed with buffered neutral formalin, dehydrated in ethanol, and embedded in paraffin. Paraffin blocks were sectioned and stained with Ziehl-Neelsen stain as described previously (8).

Strains and growth conditions.

Strains and their sources are shown in Table Table1.1. All isolates were grown on Middlebrook 7H10 agar medium supplemented with 10% oleic acid, dextrose, and albumin supplement or on Bordet-Gengou medium. M. ulcerans strains were incubated at 32°C. All other isolates were grown at room temperature (RT). For determination of optimal growth temperature, mycobacterial isolates were grown at 32°C and RT (23 to 25°C). For determination of pigment production, isolates were grown initially in the dark and exposed to light for 2 h. Plates were subsequently returned to the dark to prevent degradation of mycolactones.

Mycobacterial strains used for this study, grouped by species

Mycolactone isolation.

Lipids were extracted from M. ulcerans with chloroform-methanol (2:1, vol/vol), and phospholipids were removed by precipitation with ice-cold acetone to obtain acetone-soluble lipids (ASLs). Purified mycolactone was obtained using centripetal silica chromatography with a chromatotron as previously described (11, 24). ASLs were analyzed by thin-layer silica chromatography using the solvent system chloroform-methanol-water (90:10:1, vol/vol/vol) and visualized by charring with ceric sulfate-ammonium molybdate in 2 M sulfuric acid (24).

Cell culture and cytopathicity assays.

L929 mouse fibroblasts (ATCC CCL1) were purchased from the American Type Culture Collection and maintained in the lab in Dulbecco's minimal Eagle medium with 5% fetal calf serum and 1× l-glutamine in tissue culture flasks incubated in 5% CO2 at 37°C as previously described (1). ASL samples or individual lipid species were diluted in tissue culture medium and added to cells in a 96-well tissue culture plate to determine cytopathicity. In addition, intact bacteria and culture filtrate were directly added to cells in serial dilutions to evaluate the cytopathic capacity of surface-exposed and secreted mycolactone. Cytopathicity was defined as the minimal concentration of ASL or mycolactone necessary to produce cell rounding in 24 h and loss of the monolayer by 48 h (1, 11).


Fifty-microliter reaction mixtures containing 15.5 μl double-distilled water, 25 μl Failsafe PCR 2× PreMix G (Epicenter) (50 mM Tris-HCl [pH 8.3, 22°C], 50 mM KCl, 200 μM of each deoxynucleoside triphosphate, and proprietary amounts of FailSafe PCR enhancer and MgCl2 per reaction), 1 μM of each primer, 1.25 U Failsafe PCR enzyme mix (Epicenter), and 7 μl template derived from whole-cell boiled preparations were prepared. Cycling was performed in a Mastercycler gradient thermal cycler (Eppendorf) as follows: 98°C for 5 min; 35 cycles of 98°C for 1 min, 55°C for 1 min, and 72°C for 1 min; and 72°C for 10 min. Nine microliters of each reaction mixture was analyzed on 1 to 1.5% agarose gels in 1× Tris-acetate-EDTA stained with 1 μg/ml ethidium bromide.

Gene probing included the following: repA, 5′-TGGACCGGTCCTCAGTAACC-3′ and 5′-ATCGACGCTCGCTACTTCTG-3′; mlsA, as described previously (23); mlsB, 5′-CAGCCAACTGCGCTACTACA-3′ and 5′-AGGAGACACGGTTGGCTATG-3′; MUP045, as described previously (23); MUP053, 5′-ACCCACCTCGTCGTTAGTCATG-3′ and 5′-CGCAGAGTTCGAGTATCACGTCTAT-3′; IS2404 and IS2606, as described previously (26); uhp-1, 5′-GCACCGACGACGAGCTTCTT-3′ and 5′-GGCCGAGCATCTCAATCTCC-3′; uhp-2, 5′-GCGGAGTTCCACATCACGAC-3′ and 5′-ACCGGCTGAGGATCGAATTG-3′; chp, 5′-CGATCCACCACGACTTCACG-3′ and 5′-CAATTCCATGTGCGCCTCTG-3′; zinc, 5′-GAGCGGGAAATGGTCACTGG-3′ and 5′-GCGCTGTTCCCAATCCTCCT-3′; uhp-mem, 5′-ACCCAGGATGAGCGAACACC-3′ and 5′-TGGCTGAGTTTCGGGTCCAT-3′; and esxA and esxB, as described previously (24).

DNA sequencing.

PCR products were cloned into the pCR2.1-Topo vector (Invitrogen) and sequenced using an ABI 3100 automated genetic analyzer (Applied Biosystems, Inc.) and the ABI Big Dye Terminator 3.1 cycle sequencing kit (Applied Biosystems, Inc.).

LDH release and apoptosis.

L929 fibroblasts and peripheral blood neutrophils were assayed for cell death via apoptosis and necrosis. Cytotoxicity/necrosis was measured using a colorimetric kit from Promega as described previously (1). Briefly, cells suspended in Dulbecco's minimal Eagle medium supplemented with 5% (fibroblasts) or 10% (neutrophils) fetal calf serum and 1× l-glutamine were seeded in triplicate at a concentration of 104 cells/well in a 96-well plate. The release of cytoplasmic lactate dehydrogenase (LDH) from mycolactone-treated (15 ng, 150 ng, 1.5 μg, and 15 μg), permeabilized cells was measured at 24 h postinfection by using the colorimetric kit from Promega according to the manufacturer's instructions. Background release of LDH was obtained by measurement of ethanol-treated cells, and maximum release of LDH was obtained by lysis of untreated cells according to the manufacturer's protocol. The percent LDH release was then determined using the following calculation: [(release of LDH from mycolactone-treated cells − background release from ethanol-treated cells)/(maximum release of LDH by cell lysis − background release)] × 100.

Apoptosis of mycolactone-treated neutrophils and fibroblasts was measured at 24 h posttreatment by using the Cell Death Detection Plus enzyme-linked immunosorbent assay (Roche, Indianapolis, IN) as described previously (1). Apoptosis was then determined as fold enrichment of nucleosomes [(measurement of DNA-histone complex from treated cells/background measurement of untreated cells)].

HPLC and MS.

Methanolic extracts of ASLs were analyzed using Shimadzu LC-20AD analytical high-pressure liquid chromatography (HPLC) with UV detection at 360 nm. Separation of various mycolactones was obtained on a Phenomenex-Luna C18 250- by 4.6-mm column at a flow rate of 2 ml/min with a water-acetonitrile gradient (55% acetonitrile for 3 min [isocratic] followed by a 37-min linear gradient to100% acetonitrile). UV-active fractions were analyzed offline on an ion trap ESI Bruker-Esquire mass spectrometer (MS). MS analysis conditions were as follows: dry temperature of 300°C, gas flow of 5 liters/min, and nebulizer pressure of 15 lb/in2. For preliminary screening, the methanolic extracts of ASLs were directly analyzed without chromatographic separation.


Two milligrams of previously purified mycolactone F was dissolved in CD3COCD3 (500 μl). 1H nuclear magnetic resonance (NMR) and two-dimensional (2D) spectroscopy (COSY) spectra were recorded on a Varian INOVA spectrometer at 500 MHz. Chemical shifts were reported in ppm relative to the residual acetone peak at 2.09 ppm.


Israeli isolates of M. marinum from fish exhibit temperature-restricted growth.

The fact that M. marinum and M. ulcerans grow poorly, if at all, above 35°C is thought to limit infection in humans primarily to cutaneous lesions. In hospital laboratories M. ulcerans and M. marinum are grown at 30 to 32°C. Furthermore, M. marinum strains isolated from fish as well as from humans have been shown to grow readily at 30°C (9, 35). In marine veterinary laboratories, however, including that at Eilat, mycobacteria from fish are routinely grown at 23 to 25°C. Also, the fish pathogen M. pseudoshottsii has been reported to be unable to grow at 30°C (25). Since the temperature sensitivity of M. marinum isolated from fish in Israel had not been reported, we investigated the ability of these strains to grow above 25°C. M. ulcerans, M. marinum 1218, M. marinum DL/DK1, and M. marinum M grew readily at 32°C, a temperature consistent with the growth temperature used in clinical laboratories. The frog pathogen M. liflandii also grew at 32°C on M7H10 with oleic acid-albumin-dextrose complex. However, M. marinum isolated from fish from the Red and Mediterranean Seas showed more severe temperature-restricted growth. For these isolates, no growth was observed after 3 months of incubation at 32°C. The two freshwater isolates from Israel, CC240299 and BB170200, produced very scant growth at 32°C, suggesting that the growth temperature ranges of these two isolates may be broader than those of the Red Sea isolates. None of the five M. pseudoshottsii isolates tested produced growth at 32°C.

The generation time of M. marinum is reported to be 4.5 h, and colonies can be detected on solid medium within 7 to 10 days. M. marinum 1218 (fish, United States), M. marinum DL/DK1 (fish, Denmark), and M. marinum M (human, United States) formed colonies by 8 days at 32°C. However, M. marinum isolates from the Red and Mediterranean Seas did not produce colonies for at least 14 days at RT. M. ulcerans, M. liflandii, and M. pseudoshottsii required 4 weeks of incubation at their optimal temperatures (32°C, 32°C, and 23 to 25°C, respectively).

Pigment production in Israeli isolates is variable among strains and is constitutive.

Most M. marinum strains produce large quantities of carotenoid pigments upon exposure to light. Pigment production was variable among Israeli and other Mediterranean strains tested. All of the 11 saltwater isolates tested were deep yellow, consistent with carotenoid production. However, pigment production occurred constitutively without light induction (i.e., these strains were scotochromogenic). Colonies from freshwater isolates CC240299 and BB170200 and two isolates from Italy, DL300/04 and DL272/05, were very pale yellow. Analysis of chloroform-methanol extracts by thin-layer chromatography (TLC) revealed the presence of a major lipid species with a retention factor (Rf) of 0.71 in all strains which produced a deep-yellow pigment consistent with carotenoid production (data not shown).

PCR analysis of M. marinum and M. pseudoshottsii for the presence of mycolactone- and M. ulcerans-specific genes.

All Israeli and Mediterranean Sea isolates of M. marinum as well as M. pseudoshottsii were positive for all mycolactone-associated genes except MUP053, suggesting that mycolactones from these strains would not have a hydroxyl at C-12 on the fatty acid side chain. DNA sequence comparisons of mlsA, mlsB, and MUP045 from mycolactone-producing M. marinum (MPMM) and M. pseudoshottsii with those from M. ulcerans revealed greater than 99% homology for all three genes. In contrast, these sequences could not be amplified from two M. marinum fish isolates, DL/DK1 (Europe) and 1218 (United States) or from the genome sequence strain for M. marinum (strain M). M. marinum isolates obtained from human patients in Israel (34) were PCR negative for all mycolactone-related genes.

The availability of the M. ulcerans and M. marinum genome sequence made it possible to further characterize MPMM and M. pseudoshottsii isolates with respect to M. ulcerans- and M. marinum-specific genes (T. P. Stinear et al., unpublished data). A number of M. ulcerans genes, including the insertion sequences IS2404 and IS2606 as well as a few chromosomally encoded loci, are present in the M. ulcerans genome strain, Agy99, but absent in the M. marinum genome strain, strain M. A further distinction between M. ulcerans and M. marinum is that most M. ulcerans strains have undergone a 2.8-kb deletion in RD1 (24), a region first defined in Mycobacterium tuberculosis (17, 21) which encodes a putative secretion apparatus required for secretion of the small antigenic proteins Esat6 (esxA) and Cfp10 (esxB). Esat6 and Cfp10 play an important role in the granulomatous response to M. marinum and M. tuberculosis, and the absence of these genes in most M. ulcerans strains may contribute to the diminished granulomatous response to M. ulcerans infections.

IS2404 was detected in all MPMM and M. pseudoshottsii isolates tested and was absent from the M. marinum 1218 and M. marinum DL/DK1 isolates as well as the M strain. In contrast, IS2606 was not detected in the two MPMM isolates, CC240299 and BB170200, obtained from freshwater fish but was detected in all marine MPMM isolates. The majority of “M. ulcerans-specific” genes were detected in MPMM and M. pseudoshottsii isolates as well as in M. liflandii (Table (Table2).2). However, an M. ulcerans-specific hypothetical membrane protein gene, uhp-mem, was lacking in all isolates of MPMM and M. pseudoshottsii, and uhp-2, an M. ulcerans-specific phage sequence, was absent from M. liflandii and MPMM CC240299 and BB170200. Finally, esxA and esxB were present in MPMM and M. pseudoshottsii. The presence of esxA and esxB in MPMM and M. pseudoshottsii might provide insight into the evolutionary relationship between these isolates and M. ulcerans.

PCR results for other genes common to M. ulcerans and/or M. marinum

MPMM and M. pseudoshottsii produce a unique mycolactone.

ASLs from 2:1 chloroform-methanol extracts of bacterial pellets were analyzed by silica thin-layer chromatography for the presence of polar lipid species typical of mycolactones. In a chloroform-methanol-water (90:10:1, vol/vol/vol) solvent system, M. ulcerans mycolactone A/B has an Rf of 0.29, whereas M. liflandii mycolactone E has an Rf of 0.53. Silica TLC analysis of lipids from 6 M. pseudoshottsii isolates and 15 MPMM isolates showed that ASLs from all strains contained a putative mycolactone species with an Rf of 0.43 (data not shown). Cytopathicity assays using L929 cells showed that this lipid species caused cell rounding after 24 h and detachment at 48 h at a concentration of less than 100 ng/ml, a phenotype consistent with mycolactone activity. Cytopathicity was associated with this lipid species in each strain tested. Addition of approximately 104 intact bacteria also produced this cytopathic effect, suggesting that this molecule, like other mycolactones, was present on the cell surface. Finally, addition of sterile culture filtrate elicited cytopathicity, suggesting that this molecule was secreted as well as cell associated.

Identification of mycolactone F.

Since all M. pseudoshottsii isolates were phenotypically very similar, showed identical lipid profiles by TLC, and were identical by PCR profiling, only L15, the type strain for M. pseudoshottsii, was grown in quantity to obtain sufficient mycolactone for detailed chemical analysis. However, both phenotypic heterogeneity and molecular heterogeneity were detected within MPMM isolates. For this reason mycolactones were purified from nine of them, representing each of the major groups of isolates characterized by Ucko and Colorni (34).

Previously identified mycolactones and cometabolites share a conserved core structure, with variation only in the substitution of the side chain. Therefore, mass spectroscopy analysis of ASLs has been a very useful tool in screening for new mycolactones. Mass spectroscopic analysis of M. pseudoshottsii L15 and the nine isolates of MPMM revealed a major ion at m/z 723 (M + Na), suggesting the presence of a novel mycolactone, which was designated mycolactone F (Table (Table3).3). The cytopathic activity of this species for L929 cells was 10 ng/ml. This ion was absent from M. marinum DL/DK1. Using ion trap MS/MS analysis, the conserved lactone core can be detected as an M + Na ion at m/z 429 that is produced upon fragmentation of the parent mycolactone. Using this technique, we found that ASLs of all of the MPMM and M. pseudoshottsii strains tested showed a major positive ion at m/z 723 (M + Na) which upon fragmentation produced the conserved core fragment ion at m/z 429. Moreover, the fragmentation patterns in all samples were identical, suggesting that the mycolactones produced by these strains have the same chemical structure (Fig. (Fig.2).2). Deuterium exchange experiments concluded that mycolactone F has only four hydroxyl groups, two of which are located in the side chain. This is consistent with PCR data suggesting the absence of MUP053, which is thought to produce a third hydroxyl group on carbon 12 of the side chain of mycolactone A/B.

FIG. 2.
Mass spectroscopy of ASLs from mycolactone F-producing M. marinum strains. (A) MS of ASLs from M. marinum DL240490, showing the mycolactone F sodium adduct ...
Mycolactones identified in ASLs from 11 mycobacterial strains isolated from diseased fish

Most mycolactones exist as a mixture of double-bond isomers and are isolated along with minor cometabolites. To investigate this issue, reversed-phase HPLC analysis was performed on the ASL extracts. Under the condition described, the major isomer of mycolactone F was easily identifiable at a retention time of 20.5 ± 0.05 min (Fig. (Fig.3).3). Other geometric isomers and minor cometabolites were also identified. HPLC and MS/MS analysis confirmed that all strains produce the same molecule. However, differences in stereochemistry, though unlikely, cannot be ruled out by these methods.

FIG. 3.
Reverse-phase HPLC of ASLs derived from mycolactone F-producing strains. (A) M. marinum DL240490; (B) M. pseudoshottsii L15; (C) M. marinum SA200695; ...

Structural characterization of mycolactone F.

Although mass spectroscopy is an increasingly valuable tool for molecular analysis, data from NMR analysis provides a more rigorous method for structural assignments. The limited amount of mycolactone produced by mycobacterial species as well as the very low growth rate of mycolactone-producing bacteria has made it extremely difficult to obtain sufficient mycolactone for optimal structural analysis. However, the relatively robust growth yield of lipids from some MPMM isolates made it possible to undertake a more thorough analysis of the mycolactone F structure. For these studies, 1D and 2D NMR experiments were performed on purified mycolactone from M. marinum DL240490. 1H NMR revealed a structure very similar to that reported for mycolactone A/B, with the characteristic core signals of protons at C-9, C-15, C-11, and C-5 easily identified at 5.16, 5.03, 4.93, and 4.75 ppm, respectively (Fig. (Fig.4;4; Table Table4).4). In the region corresponding to the side chain, two geometric isomers were distinguished, with an integration ratio of 10 to 1. Both isomers showed five olefinic protons. For the major isomer, the most downfield proton was that of C-3′ which appeared as a doublet at 7.39 ppm and was coupled to the C-2′ proton (doublet at 5.91 ppm). The presence of two singlet peaks at 6.4 and 6.1 ppm suggested methyl substitutions at C-4′, C-6′, and C-8′. This was confirmed by 2D COSY analysis, which showed the long-range coupling of these olefinic protons to methyl groups at 2.06 and 2.03 ppm. The remaining side chain signals from C-10′ to C-15′ overlapped signals from the mycolactone core in the 1D 1H NMR. Thus, 2D COSY was used to assign these protons (C-9′ to C-14′), using the off-diagonal 1H-1H correlation peaks. Mycolactone F (Fig. (Fig.5)5) is the smallest mycolactone so far identified. As with other mycolactone variants, the core lactone structure is identical to that of other mycolactones, whereas the side chain is unique.

FIG. 4.
1D H1 NMR spectrum of mycolactone F.
FIG. 5.
Structural comparison of mycolactone A/B and mycolactone F.
Proton shifts relative to residual acetone for the two major isomers of mycolactone F

Biological activities of mycolactone F.

Mycolactones produce a distinct phenotype on cultured cells, characterized by necrosis at concentrations of above 1 μg/ml within 4 h and delayed apoptosis at concentrations as low as 1 ng/ml within 24 h (1). In order to determine the activity of mycolactone F on cultured cells, purified mycolactone was added to a semiconfluent layer of L929 fibroblasts and tested for the ability to produce necrosis or apoptosis. Although mycolactone F showed typical mycolactone phenotypes, the potency of mycolactone F with regard to apoptosis was significantly less than that of mycolactone A/B (Fig. (Fig.66).

FIG. 6.
Analysis of mycolactone-mediated cytopathicity. (A) Cytopathicity on L929 murine fibroblasts, showing untreated cells (left; total magnification, ×200) versus cells treated with 100 ng mycolactone A/B (middle; total magnification, ×200) ...


For 25 years, M. ulcerans had the distinction of being the only mycobacterial pathogen to produce a secreted toxin. The finding that mycolactone is plasmid encoded led to the hypothesis that other mycolactone-producing mycobacteria might exist, particularly in aquatic environments. Results from this investigation show that mycolactone-producing Mycobacterium species are widely distributed within aquatic environments. It is significant that all mycolactone-producing mycobacteria have been identified as pathogens.

Although the ecology of M. ulcerans is poorly understood, the only well-defined risk factor for infection is association with slow-moving water (2, 3, 6, 35). Of particular interest is the fact that human disease caused by M. ulcerans, frog disease caused by M. liflandii, and fish disease caused by MPMM and M. pseudoshottsii have been associated with deteriorating water quality and/or eutrophic enrichment (2, 8, 24, 32). This suggests that similar water quality parameters in different parts of the world may play a role in the ecology of mycolactone-producing bacteria. The geographic heterogeneity of mycolactone structure, along with highly conserved structure within a specific region, has led to the hypothesis that there is strong selection within a specific environment for a particular mycolactone. The discovery of mycolactone F in two mycobacterial species isolated from fish in widely separated geographic areas suggests that mycolactone F plays a particular role in fish disease. Fish may in fact act as “indicator species” in these environments.

Mycolactones are synthesized by large type 1 polyketide synthases. There is a remarkable amount of repeated DNA sequence within the mycolactone pks genes, which provides the opportunity for formation of new mycolactones through recombination (29). However, the core lactone structure in mycolactone F is identical to that in mycolactones from M. ulcerans and M. liflandii. Mycolactone F is the smallest mycolactone identified so far, and it varies from other mycolactones in side chain structure. Our results suggest that side chain structure plays an important role in potency, since the ability of mycolactone F to cause apoptosis is significantly less than that of other mycolactones. In the past five years there has been an intense search for M. ulcerans in the environment. Those studies have relied solely on detection of IS2404 as evidence for M. ulcerans, based on a report showing that IS2404 was absent from 48 species of mycobacteria obtained from a laboratory collection (27). IS2404 PCR has been used to demonstrate the presence of M. ulcerans DNA in fish (10), frogs, many species of insects, snails, and a large range of aquatic samples from Africa (19, 22). The presence of IS2404 in MPMM, along with its presence in M. liflandii and M. pseudoshottsii, provides a clear signal that IS2404 PCR cannot be used as the sole evidence for M. ulcerans in environmental samples. Nonetheless, IS2404 may be a useful marker for the mycolactone plasmid.

There is considerable heterogeneity within M. marinum species (31, 35). In the past few years investigators have suggested that M. marinum isolates from humans are genetically different than isolates from poikilothermic species. Stinear et al. (28) examined 22 M. marinum isolates (19 from humans and 3 from fish) and used multilocus sequence analysis to separate them into five categories. Based on an examination of 11 human isolates and 6 fish isolates, Van der Sar et al. (35) suggested that the human isolates were more pathogenic in zebra fish and better able to grow in the human macrophage THP-1 cell line than isolates from poikilothermic species such as fish. The majority of these strains were from collections, and the country of origin was not provided, making it impossible to compare isolates within a geographic region. The best data for the association of specific M. marinum genotypes with human or nonhuman aquatic vertebrate strains comes from a report by Ucko and Colorni (34) comparing 20 human isolates of M. marinum with 12 fish isolates from the Red Sea. In that study, human isolates were distinguished phenotypically and genotypically from fish isolates from the Red Sea in Israel. In our study we further distinguish the Red and Mediterranean Sea fish isolates from other groups of M. marinum strains by showing that these isolates produce a mycolactone toxin. This work also supports earlier studies describing discrete populations of M. marinum strains in Israel (34).

Why is it that the vast majority of M. marinum strains do not produce mycolactone? One possibility is that not all strains are capable of acting as recipients for the mycolactone plasmid. This is supported by the fact that MPMM strains form a group distinct from other M. marinum isolates in Israel. An interesting parallel may exist between Shigella species and mycolactone-producing mycobacteria. Although it is convenient to look at Shigella species as Escherichia coli with an invasion plasmid, not all E. coli strains can act as recipients for the plasmid. In particular, the loss of outer membrane proteins is a prerequisite for conjugal competency. What are the host requirements for horizontal transfer of the mycolactone plasmid? One possibility is that insertion of IS2404 in M. ulcerans, MPMM, M. liflandii, and M. pseudoshottsii has resulted in loss of membrane-associated genes, thus rendering them competent for horizontal transfer. Alternatively, these strains may have acquired genes enabling horizontal transfer, although this explanation would be inconsistent with molecular evidence suggesting that M. ulcerans evolved from an M. marinum ancestor. Most of the studies on M. marinum and most of the strains in collections are from infected humans. These strains may not be representative of M. marinum in its natural aquatic environment. We cannot rule out the possibility that a more thorough analysis of nonhuman M. marinum isolates might show that MPMM are “typical” strains.

Given the importance of the mycolactone plasmid for virulence of M. ulcerans, what is the likely virulence of MPMM and M. pseudoshottsii for humans? Although mycolactone F is toxic for human cells, this is relevant to human disease only if MPMM or M. pseudoshottsii strains are capable of colonizing mammalian species. The temperature-restricted growth of MPMM and M. pseudoshottsii may well be a barrier to infection of mammalian species. However, mycolactone-producing aquatic mycobacteria may provide a reservoir for genes important for the evolution of new human pathogens.


We thank Tim Stinear, Stewart Cole, and Julian Parkhill for access to unpublished data from the M. ulcerans and M. marinum genomes.

This work was supported by a grant from NIAID, NIH (R01-1015-084).


Editor: J. T. Barbieri


[down-pointing small open triangle]Published ahead of print on 21 August 2006.


1. Adusumilli, S., A. Mve-Obiang, T. Sparer, W. Meyers, J. Hayman, and P. L. C. Small. 2005. Mycobacterium ulcerans toxic macrolide, mycolactone modulates the host immune response and cellular location of M. ulcerans in vitro and in vivo. Cell. Microbiol. 7:1295-1304. [PubMed]
2. Asiedu, K., R. Scherpbier, and M. Raviglione (ed.). 2000. Buruli ulcer: Mycobacterium ulcerans infection. World Health Organization, Geneva, Switzerland.
3. Barker, D. J. 1973. Epidemiology of Mycobacterium ulcerans infection. Trans. R. Soc. Trop. Med. Hyg. 67:43-50. [PubMed]
4. Cadapan, L. D., R. L. Arslanian, J. R. Carney, S. M. Zavala, P. L. Small, and P. Licari. 2001. Suspension cultivation of Mycobacterium ulcerans for the production of mycolactones. FEMS Microbiol. Lett. 10246:1-5. [PubMed]
5. Colorni, A. 1992. A systemic mycobacteriosis in the European sea bass Dicentrarchus labrax cultured in Eilat (Red Sea). Isr. J. Aquacult. Bamidgeh. 44:75-81.
6. Connor, D. H., and H. F. Lunn. 1965. Mycobacterium ulcerans infection. Int. J. Lepr. 33(Suppl.):698-709. [PubMed]
7. Decostere, A., K. Hermans, and F. Haesebrouck. 2004. Piscine mycobacteriosis: a literature review covering the agent and the disease it causes in fish and humans. Vet. Microbiol. 99:159-166. [PubMed]
8. Diamant, A., A. Banet, M. Ucko, A. Colorni, W. Knibb, and H. Kvitt. 2000. Mycobacteriosis in wild rabbitfish Siganus rivulatus associated with cage farming in the Gulf of Eilat, Red Sea. Dis. Aquat. Org. 39:211-219. [PubMed]
9. Dobos, K. M., F. D. Quinn, D. A. Ashford, C. R. Horsburgh, and C. H. King. 1999. Emergence of a unique group of necrotizing mycobacterial diseases. Emerg. Infect. Dis. 5:367-370. [PMC free article] [PubMed]
10. Eddyani, M., D. Ofori-Adjei, G. Teugels, D. De Weirdt, D. Boakye, W. M. Meyers, and F. Portaels. 2004. Potential role for fish in transmission of Mycobacterium ulcerans disease (Buruli ulcer): an environmental study. Appl. Environ. Microbiol. 70:5679-5681. [PMC free article] [PubMed]
11. l. George, K. M., L. Barker, D. Welty, and P. L. C. Small. 1998. Partial purification and characterization of biological effects of a lipid toxin produced by Mycobacterium ulcerans. Infect. Immun. 66:587-593. [PMC free article] [PubMed]
12. George, K. M., D. Chatterjee, G. Gunawardana, D. Welty, R. E. Lee, and P. L. C. Small. 1999. Mycolactone: a polyketide toxin from Mycobacterium ulcerans required for virulence. Science 283:854-857. [PubMed]
13. George, K. M., D. M. Welty, L. Pascopella, and P. L. C. Small. 2000. A Mycobacterium ulcerans toxin, mycolactone, causes apoptosis in guinea pig ulcers and tissue culture cells. Infect. Immun. 68:877-883. [PMC free article] [PubMed]
14. Gunawardana, G., D. Chatterjee, K. M. George, P. Brennan, D. Whittern, and P. L. C. Small. 1999. Characterization of novel macrolide toxins, mycolactone A and B, from a human pathogen. J. Am. Chem. Soc. 121:6092-6093.
15. Heckert, R. A., S. Elankumaran, A. Milani, and A. Maya. 2001. Detection of a new Mycobacterium species in wild striped bass in the Chesapeake Bay. J. Clin. Microbiol. 39:710-715. [PMC free article] [PubMed]
16. Hong, H., P. J. Gates, J. Staunton, T. Stinear, S. T. Cole, P. F. Leadlay, and J. B. Spencer. 2003. Identification using LC-MSn of co-metabolites in the biosynthesis of the polyketide toxin mycolactone by a clinical isolate of Mycobacterium ulcerans. Chem. Commun. (J. Chem. Soc. Sect. D) 22:2822-28233. [PubMed]
17. Hsu, T., S. M. Hingley-Wilson, B. Chen, M. Chen, A. Z. Dai, P. M. Morin, C. B. Marks, J. Padiyar, C. Goulding, M. Gingery, D. Eisenberg, R. G. Russell, S. C. Derrick, F. M. Collins, S. L. Morris, C. H. King, and W. R. Jacobs Jr. 2003. The primary mechanism of attenuation of bacillus Calmette-Guérin is a loss of secreted lytic function required for invasion of lung interstitial tissue. Proc. Natl. Acad. Sci. USA 100:12420-12425. [PMC free article] [PubMed]
18. Kazumi, Y., K. Ohtomo, M. Takahashi, S. Mitarai, I. Sugawara, J. Izumi, A. Andoh, and H. Hasegawa. 2004. Mycobacterium shinshuense isolated from cutaneous ulcer lesion of right lower extremity in a 37-year-old woman. Kekkaku 79:437-441. [PubMed]
19. Kotlowski, R., A. Martin, A. Ablordey, K. Chemlal, P. A. Fonteyne, and F. Portaels. 2004. One-tube cell lysis and DNA extraction procedure for PCR-based detection of Mycobacterium ulcerans in aquatic insects, molluscs and fish. J. Med. Microbiol. 53:927-933. [PubMed]
20. MacCallum, P., J. C. Tolhurst, G. Buckle, and H. A. Sissons. 1948. A new mycobacterial infection in man. J. Pathol. Bacteriol. 60:93-122. [PubMed]
21. Marmiesse, M., P. Brodin, C. Buchrieser, C. Gutierrez, N. Simoes, V. Vincent, P. Glaser, S. T. Cole, and R. Brosch. 2004. Macro-array and bioinformatic analyses reveal mycobacterial ‘core’ genes, variation in the ESAT-6 gene family and new phylogenetic markers for the Mycobacterium tuberculosis complex. Microbiology 150:483-486. [PubMed]
22. Marsollier, L., R. Robert, J. Aubry, J. P. Saint Andre, H. Kouakou, P. Legras, A. L. Manceau, C. Mahaza, and B. Carbonnelle. 2002. Aquatic insects as a vector for Mycobacterium ulcerans. Appl. Environ. Microbiol. 68:4623-4628. [PMC free article] [PubMed]
23. Mve-Obiang, A., R. E. Lee, F. Portaels, and P. L. C. Small. 2003. Heterogeneity of mycolactones produced by clinical isolates of Mycobacterium ulcerans: implications for virulence. Infect. Immun. 71:774-783. [PMC free article] [PubMed]
24. Mve-Obiang, A., R. E. Lee, E. S. Umstot, K. A. Trott, T. C. Grammer, J. M. Parker, B. S. Ranger, R. Grainger, E. A. Mahrous, and P. L. C. Small. 2005. A newly discovered mycobacterial pathogen isolated from laboratory colonies of Xenopus species with lethal infections produces a novel form of mycolactone, the Mycobacterium ulcerans macrolide toxin. Infect. Immun. 73:3307-3312. [PMC free article] [PubMed]
25. Rhodes, M. W., H. Kator, A. McNabb, C. Deshayes, J. M. Reyrat, B. A. Brown-Elliott, R. Wallace, Jr., K. A. Trott, J. M. Parker, B. Lifland, G. Osterhout, I. Kaattari, K. Reece, W. Vogelbein, and C. A. Ottinger. 2005. Mycobacterium pseudoshottsii sp. nov., a slowly growing chromogenic species isolated from Chesapeake Bay striped bass (Morone saxatilis). Int. J. Syst. Evol. Microbiol. 55:1139-1147. [PubMed]
26. Snyder, D. S., and P. L. C. Small. 2003. Uptake and cellular actions of mycolactone, a virulence determinant for Mycobacterium ulcerans. Microb. Pathog. 34:91-101. [PubMed]
27. Stinear, T. P., B. C. Ross, J. K. Davies, L. Marino, R. M. Robins-Browne, F. Oppedisano, A. Sievers, and P. D. R. Johnson. 1999. Identification and characterization of IS2404 and IS2606: two distinct repeated sequences for detection of Mycobacterium ulcerans by PCR. J. Clin. Microbiol. 37:1018-1023. [PMC free article] [PubMed]
28. Stinear, T. P., G. A. Jenkin, P. D. R. Johnson, and J. K. Davies. 2000. Comparative genetic analysis of Mycobacterium ulcerans and Mycobacterium marinum reveals evidence of recent divergence. J. Bacteriol. 182:6322-6330. [PMC free article] [PubMed]
29. Stinear, T. P., A. Mve-Obiang, P. L. Small, W. Frigui, M. J. Pryor, R. Brosch, G. A. Jenkin, P. D. Johnson, J. K. Davies, R. E. Lee, S. Adusumilli, T. Garnier, S. F. Haydock, P. E. Leadlay, and S. T. Cole. 2004. Giant plasmid-encoded polyketide synthases produce the macrolide toxin of Mycobacterium ulcerans. Proc. Natl. Acad. Sci. USA 101:1345-1349. [PMC free article] [PubMed]
30. Stragier, P., A. Ablordey, W. M. Meyers, and F. Portaels. 2005. Genotyping Mycobacterium ulcerans and Mycobacterium marinum by using mycobacterial interspersed repetitive units. J. Bacteriol. 187:1639-1647. [PMC free article] [PubMed]
31. Tonjum, T., D. B. Welty, E. Jantzan, and P. L. C. Small. 1998. Differentiation of Mycobacterium ulcerans, M. marinum, and M. haemophilum and mapping of their relationships to M. tuberculosis by gas chromatography, DNA-DNA hybridization and 16S rDNA sequence analysis. J. Clin. Microbiol. 61:917-925. [PMC free article] [PubMed]
32. Trott, K. A., B. A. Stacy, B. D. Lifland, H. E. Diggs, R. M. Harland, M. K. Khokha, T. C. Grammer, and J. M. Parker. 2004. Characterization of a Mycobacterium ulcerans-like infection in a colony of African tropical clawed frogs (Xenopus tropicalis). Comp. Med. 54:309-317. [PubMed]
33. Ucko, M., A. Colorni, H. Kvitt, A. Diamant, A. Zlotkin, and W. R. Knibb. 2002. Strain variation in Mycobacterium marinum fish isolates. Appl. Environ. Microbiol. 68:5281-5287. [PMC free article] [PubMed]
34. Ucko, M., and A. Colorni. 2005. Mycobacterium marinum infections in fish and humans in Israel. J. Clin. Microbiol. 43:892-895. [PMC free article] [PubMed]
35. Van der Sar, A. M., A. N. Abdallah, M. Sparrius, E. Reinders, C. M. J. E. Vanderbrouchke-Grauls, and W. Bittern. 2004. Mycobacterium marinum strains can be divided into two distinct types based on genetic diversity and virulence. Infect. Immun. 72:6306-6312. [PMC free article] [PubMed]
36. van der Werf, T. S., W. T. van der Graaf, J. W. Tappero, and K. Asiedu. 1999. Mycobacterium ulcerans infection. Lancet 354:1013-1018. [PubMed]
37. van der Werf, T. S., T. P. Stinear, W. T. van der Graaf, and P. L. C. Small. 2003. Mycolactones and Mycobacterium ulcerans disease. Lancet 362:1062-1064. [PubMed]

Articles from Infection and Immunity are provided here courtesy of American Society for Microbiology (ASM)
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


  • EST
    Published EST sequences
  • GSS
    Published GSS sequences
  • MedGen
    Related information in MedGen
  • Nucleotide
    Published Nucleotide sequences
  • Protein
    Published protein sequences
  • PubMed
    PubMed citations for these articles
  • Substance
    PubChem Substance links
  • Taxonomy
    Related taxonomy entry
  • Taxonomy Tree
    Taxonomy Tree

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...